Unless you plan to chemically weld the mounts to a chassis of the same material I'd very strongly advise you to drill holes up thru the chassis, tap the motor mounts, and secure with suitable screws. You shouldn't take chances with critical items like mounts.

A: Take a look at some antweights. You'll see all kinds of chassis: stick, platform, box, tray, perimeter, you name it. Anything goes with insect class 'bots. Proper design, materials selection and construction technique are more important than style.

A: There have been a great many antweights powered by two Copal gearmotors, so four is more than enough power. Speed and torque will depend on tire diameter and voltage. I suggest examining those possibilities with the Team Tentacle Torque & Amp-Hour Calculator.

Example: an ant with four 30:1 Copals and 2.25" wheels at 11.1 volts has a top speed of 3.8 MPH that is reached in less than two feet in under half a second. Current drain per motor at maximum push is only 0.33 amp. Sounds pretty good to me.

A: I'm gonna say no. The air tank weighs eleven ounces all by itself - too heavy to use in an antweight. Worse, the connector plugs for the solenoid air valves are compatable only with the VEX radio system receiver, which does not have the failsafe capacity required for an active weapon like a flipper.

A: Horses for courses. Different designs have different space requirements and call for very different performance factors from the drivetrain. Also, we make it a point at Ask Aaron to avoid general endorsement of any commercial product. We aren't comfortable labeling any product as our 'favorite', although we will recommend specific products for specific applications.

We haven't used either the BB-3-9 or the tinyESC and cannot directly comment on their performance.

List Of Part's I Already Have
6x BaneBots Wheel, 1-7/8" x 0.4", 1/2" Hex Mount, 30A, Black/Green
4x Hub, Hex, Series 40, Set Screw, 3mm Bore, 1 Wide
2x Hub, Hex, Series 40, Set Screw, 3mm Bore, 2 Wide
4x 16mm Slave Shaft Kit 1.5 inch long 3mm shaft
half inch aluminum angle (to go around the polycarbonate)
sheet of polycarbonate (for the chassis)
big roll of aluminum flashing (for armor, also might add a sheet under the polycarbonate)
Spektrum DX7 remote

A: See #23 for an explanation of the difference between brushed and brushless motors and a short discussion of correct use of each type in combat robots. Many additional posts on this topic are in the archive. We contend that brushless motors are not suitable for combat robot drivetrains. Available hobby brushless motors just aren't designed for the demands placed on them when pushing around a combat robot. Neither are hobby ESCs designed for use in cars or airplanes - for brushed or brushless motors. I strongly recommend that you stick with combat proven components.

You're going to need more gear reduction than a single-stage belt system will provide. The Team Tentacle Torque & Amp-Hour Calculator can model your drivetrain and help you determine the gearing needed for those 1-7/8" wheels. The simple way to handle gear reduction is a strong planetary gearbox for the motor.

You don't need massive power in order to get maximum pushing force from your 'bot. Pushing power is limited by the weight on the driven wheels; in your case, that's the full three pounds of the robot weight. Power in excess of what is needed to spin the tires freely will simply be wasted. As for speed, the realistic maximum speed depends on the size of the arena and the skill/preference of the operator. Anything much over 5 MPH in an insect-sized arena is unuseable.

I'd suggest a pair of BaneBots 28mm RS-385 16:1 gearmotors direct-driving wheels (6mm hub) with timing belts running to the other end of the robot for your 4-wheel drive. Top speed with 1.75" wheels is 5+ MPH and the motors have about 8 times the torque needed to spin the wheels with the robot pushing an imoveable object. More power than that is wild overkill. Maximum amperage draw at full push is less than 3 amps (per the Tentacle calculator), so a single Sabertooth 5 RC Dual Motor Speed Controller should handle the controlling work for both motors nicely.

By the way, the aluminum angle and flashing you buy at the hardware store is a very soft alloy intended to be easily bent and machined. It makes really lousy armor. The polycarbonate by itself is much better.

Q: I'm trying to find a way to use the pulley system instead of the gear motors. I found an RS-380 motor and I could run it with the sabertooth 12 [ESC] and Fingertech pulleys. I'm thinking a 15 tooth on the motor and a 42 tooth on the slave shafts. Do you see any chance of this setup working? Do you think the pushing power be low?

A: Several problems with your proposed setup:

1. The fingertech timing pulleys and belts are too narrow to reliably handle the amount of torque that this drivetrain can produce. They will skip and slip when pushing and you will have real trouble keeping the belts on the pulleys. If you do pop off a main drive belt, you're sunk.

2. A speed reduction of 2.8:1 (32:15) with 1.75" wheels is not nearly enough to provide optimum acceleration, amperage consuption, and speed in a small arena. Optimum gearing would be around 10:1 which, as noted in the answer above, cannot be done with a compact single-stage belt drive.

3. Maximum amperage consumption when pushing with the proposed motor, gearing, and wheels comes in at about 14 amps (per the Tentacle Calculator) which exceeds the published capacity of the Sabertooth 12 ESC. With a 10:1 gear reduction the maximum amperage consumption drops down below 4 amps.

A: The 'best' motors will depend on the arena size, wheel diameter, how exposed the wheels are, and your driving style. The balance between speed, control, and power is largely a personal choice.

The Team Tentacle Torque & Amp-Hour Calculator can provide performance estimates for a wide variety of motors and wheel diameters in arenas of any size, and provides a peak amperage estimate to help in selecting a speed controller. A pair of BaneBots 16:1 24mm gearmotors might be a good place to start your search.

The drive is two banebot 16:1 motors. Armor is 2025 aluminum side plates that are .25'' thick, and top and bottom armor that is .1" thick. Two more banebot 16:1 motors power a hammer. The hammer can put a small punch mark in steel, a larger one in softer metals, and can crack the copal lexan motor mounts.

A: If I could tell which gelding was going to win a race by looking at pictures of horses, I'd be a very rich man. A few comments:

• A hammer that can't hit its target is not of much use, no matter how powerful. Is the 'bot maneuverable enough to hit a fast moving target? Is your driving that good?

• A good spinner is going to have a field day with those exposed hard edges.

• A good wedge will be under you and have you hammering air all the way into the nearest wall.

• You seem to have placed you 'top' armor 'underneath' your weapon motors and drive. That could be a problem.

• How about some protection for the exposed wires and a cover for that switch?

Q: What should I do about the issues you brought forward?

A: It's kinda too late to do much about the major design concept of your 'bot after you built it. You could add a little light protection to cover the exposed weapon bits and at least keep stray junk from jamming the open gears or tearing away your wiring. Even a thin sheet of flexible plastic would be helpful. I'm guessing you don't have much weight allowance left - probably not enough to build shields for your exposed wheels - but maybe enough to add a small wedge to the front of the 'bot?

Q: Well, should I even bother entering it in a competition? All I want is a single win. It doesn't matter if it loses all the other matches. One win is all I need.

A: Absolutely you should enter! Like I said, I can't predict winners from photos, and I haven't seen your competition. Winning is all well and good, but the satisfaction comes from building and competing. Practice your driving skills, run the 'bot into stuff to see what breaks, replace the broken pieces with something stronger, then get out there and fight!

Q: How do you make a wedge?

A: In your case, I'd just extend the bottom armor plate forward a couple inches and hone it to a sharp edge. Your opponent will run up the wedge into your front armor and be in perfect position for the hammer.

Q: Thanks Aaron! After listening to the input of both you and other builders, I think I have a good "shopping list" of modifications I can make. Fortunately, there are actually quite a few areas to cut down weight.

The main reason I built a hammer in the first place was that I decided it was more important to be unique than it was to be destructive. Diversity is a good thing, and I will be a sacrifice in its name.

It also helped when I noticed the thin, plastic top armor many beetles have.

A: Diversity is a good thing.

Precision control is going to be critical if you're going to make best use of the hammer weapon. Tweak your radio settings until the response feels natural to you. Consider reducing the maximum turn rate (via 'ATV' or 'travel adjust') to retain precise control in hectic combat situations. And of course, put in a lot of driving practice.

Remember that scoring is on both damage and aggression: keep pointed at your opponent and keep moving forward for maximum aggression points.

A: Spinning disks are generally made, not purchased. You might find something at the hardware store that could be adapted, but I don't know what type of design you're thinking about.

A beetleweight blade is available to fit the VDD gearbox -- it requires a hub adaptor. The VDD gearbox has a good record in antweight robots, but I would prefer a belt drive for beetleweights and heavier classes.

A: There is ample discussion on insect class lifter servo selection in this archive.

A: Most beetleweights do not have a separate frame structure, and certainly don't require one. With proper material selection and construction technique a sturdy 'stick' chassis can be built using mechanical fasteners -- triangulate everything and/or gusset the corners.

A: For a start, I don't recommend direct drive for a spinner weapon. The design places too much stress on the small motor shaft and gives too high a spin rate for the shell. The design is simple and tempting, but it just has too many drawbacks to be reliable. There have been many direct drive spinners tried, but you will not find them near the top of the rankings at BotRank.com

About the motor: I really can't make a good recommendation for a specific motor without knowing the explicit design details of the weapon shell and drive -- see #29. I'm going to suggest that you put design numbers into the Team Run Amok Spinner Excel spreadsheet and see what type of performance you get from different motors. As a pure guess, the Hacker A20-50S might be a better choice. It's a smaller motor with less torque, but its less aggressive power consumption will make a more reliable weapon.

A: Radio included?? A decent radio for an active weapon beetle is going to eat up 60% of your budget. You really can't go cheap on your R/C gear -- see the previous post on the Exceed radio. A speed controller for your drive and weapon is going to eat up the rest of that budget. Even if you use a hacked toy for the chassis and a steel salad bowl for the dome, you're still in need of a weapon drive. We've said this before: you'll get a much better result if you design a robot you can afford to build from good components rather than a larger or more complex 'bot that will be cobbled from make-do materials. My suggestion is to build something you can afford to build well.

I know you didn't ask for an opinion on an insect-class full-body spinner (FBS), but for the benefit of others reading this I have to mention that they don't work well in small insect arenas. An FBS needs spin-up time that you just don't have in a small arena. You've got no place to hide and no way to shield your weapon to get those critical seconds you need. If you do survive long enough to get a hit in, the rebound can send you bouncing off multiple arena walls. Not recommended.

A: Twackbots are not big favorites of judges. It's very difficult to show 'aggression' since you can't move toward your opponent while spinning, and aggression is a big part of the current scoring system. See multiple posts on thwackbots and 'melty brain' in the archive. A thwackbot is not considered to be an active weapon.

Team Velocity did compete with a successful thwackbeetle named 'Trilocity'.

Q: Thanks for the info on thwackbots, but I noticed my bot does tend to move while spinning. Does this mean that if I positioned it right it would move towards the target getting aggression points?

A: I think a little experimentation will show that your spinning 'drift' tends to be downhill along a imperceptable slope. Your motion will vary according to the arena and judges will not be impressed by an uncontrolled drift. If your opponent is dense enough to sit still and wait for your 'bot to drift in the right direction, they deserve to lose.

Q: Would I get aggression points if I put a spike on the end of the weapon and used it as a ram?

A: Yes. You'd also get your head handed to you by a robot with a real weapon.

You don't actually even need the spike. 'Agression' is scored simply by moving toward your opponent. See our You Be the Judge page for some other surprizes about combat robot judging.

A: See #17.

I can tell you that stainless steel alloys are designed for corrosion resistance and high temperature stability -- not for structural strength. There are much better materials for combat robot purposes. I'd go with the 6061 aluminum.

A: As discussed in the archive, the materials used to form composite armor must be carefully selected to compliment each other's properties. Stacking random materials together in hope that they become stronger than either of the individual materials has a very small chance of success.

Lexan makes excellent armor because of its ability to flex and absorb large amounts of energy without permanent deformation. Kevlar fabric has great tensile strength -- a property not greatly valued in robot armor. As a flexible fabric, the material would transfer impact to the framework supporting it rather than absorbing the energy itself. It could also be prone to being 'caught' by a spinning weapon -- not a good thing! I don't see any synergy in combining the two materials, or much benefit in using Kevlar alone for robot armor.

A: T-304 stainless steel is not a top choice for armor. Although it has good 'toughness', it is also quite ductile and will show damage easily. Since any sign of damage scores points for your opponent, I wouldn't recommend it at any thickness.

A: Mark J. here: a classic physics problem! Here's the formula:

Converting that to the type of speed figures given for servos: you'd need a servo that could travel 60 degrees in 0.04 seconds while accelerating a one-pound load vertically at the end of a 3 inch lifter arm. *** There is no such servo! *** Servos simply aren't fast/powerful enough to 'launch' an antweight. They are useable as lifters, not flippers.

Q: Although it may not flip an antweight 6 inches, the Hitec HS-7940TH Servo has good torque and a blazing fast speed. Could you use this servo to "flip" the other antweight robot? Does the length of the arm change the amount of speed a servo could deliver? Couldn't you use a 4 bar linkage to double the speed, while reducing half of the already large torque rating?

A: Power is calculated as the product of speed and torque: you can't gain power by playing with linkages, arm length, or gearing.

That 'blazing fast speed' is at zero load. When you load a servo to half its maximum torque, the transit time doubles -- cutting the speed in half. Using a linkage to increase the zero-load speed will give an equal increase in torque needed under load, resulting in a nullifying speed reduction.

The Hitec HS-7940TH could accelerate a one-pound weight vertically at the end of a 3 inch arm thru 60 degrees of arc in about 0.18 seconds. That's very quick for a servo, but a long way from flipper speed.

A: Read thru the archive for a full explanation of why direct drive bites. Briefly:

• Attaching that heavy bar to the small diameter shaft of the Speed 400 is a real problem;

• The first good hit you get will very likely bend the motor shaft and render the weapon useless;

• A bar spinning at 15000 plus RPM will have VERY little 'bite'.

A: That's something you're likely going to have to make for yourself or have made for you. I'm sure that Team Whyachi's Bot Shop would be happy to make a disk to your specifications. Depending on the thickness you need, you might find something useful by tearing apart an old computer hard disk drive.

Q: Disk guy, how about a disk made of carbon fiber? Otherwise how thick should a antweight spinner disk be( at 3") Just a rough estimate.

A: No! Carbon fiber is light, stiff, and strong -- but it isn't 'tough'. It shatters under heavy impact, exactly NOT what you want your disk weapon to do.

No competent engineer would spec a material thickness knowing so little about the design -- see #17. I don't know how fast you're going to spin it, what additional weight will be added for 'impactors', which alloy you will use, or how strong the hub will be. Therefore, I absolutely cannot tell you to make it roughly 0.125" thick.

Note that there are considerations other than strength to the design of a disk weapon. The purpose of a rotating mass weapon is to store energy for release on impact. The design of such a weapon calls for a balance between motor power and rotational inertia to store the greatest energy possible within a reasonable time period. The Team Run Amok Spinner Excel spreadsheet can assist you in evaluating these design tradeoffs.

A: The first question is why you would want to build an antweight overhead hammer. An overhead hammer is one of those things that does not 'scale' well -- search the [DESIGN] archive for the previous post on 'scale'. Worse, overhead axes aren't all that successful at any size. I'd strongly suggest another weapon design.

Search this archive for 'pneumatic' to find a great many previous posts on antweight pneumatic parts sources. The short answer is 'there aren't any'. You'd pretty much have to build or modify your own actuator and valve system.

Q: Do you know where I can get a pneumatic acutator suitable for antweight use?

A: Search this archive for 'pneumatic' to find a great many previous posts on antweight pneumatic parts sources. The short answer is 'there aren't any'. You'd pretty much have to build or modify your own actuator and valve system.

Q: Couldn't you use Robart valves and pistons for an antweight flipper? Would this give enough airflow to be able to flip well? I have all the parts but am curious to see what you think. I also believe that Go Flip Yourself uses these parts too.

A: I've been asked about Robart pneumatic components previously -- search this archive. The ports on the Robart valves and cylinders are really small. They're built for realistic slow-motion actuation of landing gear and such on scale aircraft, not for speed and power. If Cutting Edge Robotics is using them on 'Go Flip Yourself' I suspect they have been modified extensively for better air flow.

Q: Hi I'm the overhead axe guy. Im switching to a lifter. I'm using the Spektrum A6030 Digital Aircraft Servo. Its an antweight with a 3 inch arm.

A: Very nice servo, but overkill for your application. With a 3" arm on an antweight you won't have any real use for more than 100 oz-in of torque -- the A6030 pumps out 278 oz-in and has a price tag of about $85. The Spektrum A6020 Digital servo is $40 less and still gives 146 oz-in. Yes, the A6030 has titanium gears, but I don't think you'll need those any more than you need the extra torque.

It certainly won't hurt to go for the high-end servo, but in this case I think I'd save a few bucks and opt for 'plenty good enough'.

Q: Servo lifter guy (again) I forgot to tell you that in the near future I will be putting it to use as a cluster bot vs beetle weights. Sorry.

A: In that case, great servo choice!

A: We get a lot of questions asking how thick the material should be for a specific robot part. The required strength for a specific chassis element will depend on the size, shape, mounting method, additional stresses placed on the chassis by your lifter, and exposure to abuse from opponent weaponry. No competent engineer would spec a material thickness knowing so little about the design -- see #17.

With that said, I can tell you that you're in the right ballpark for your CF thickness. I commonly see antweight top/bottom plate thickness in the 0.03 to 0.09 inch thickness range. If it fails, make it thicker.

A: That 'stick' that emerges from the top of Megabyte is an extension of the 'dead shaft' axle that the shell spins on. Such a design in an antweight would require an offset belt, friction, or gear drive to spin the shell -- not a direct-drive from a motor, as is common with antweight shell spinners.

A: Sorry - I don't follow the UK Antweight World Series competitions, don't know either of those robots, and can't find them on the web. I do know that 'walker' rules are very different in the US and UK.

I am leaning toward the T280 motor, but I'm not too sure about the gearbox reduction. What is your opinion?

A: You may have noticed that few spinners use a model aircraft gearbox. There's a good reason for that: the abrupt stress of weapon impact destroys them. I assume that is why 'Ultra Havoc' abandoned the gearbox for their 'latest version' -- A belt drive absorbs shock much better.

Motors without specs are a problem -- everything you do with them is a guess. You'll end up doing a lot of experimenting, which can be both expensive and time consuming. If you want to try one of the motors you mention, the T-280 is the more powerful of the two. Gearing is dependent on your blade diameter and weight, but 5:1 is a reasonable starting point for a 3" blade. I'd much rather start with something that has specs!

A: You must not be very far along in the process if you're just getting to motor selection. Proper motor selection depends on other design and use factors like arena size, wheel diameter, chassis layout, and performance expectations. See #21. The BaneBots 25mm 10:1 FF-180 gearmotors might be a good starting point. A single Sabertooth 5 RC Speed Controller could handle four of these motors driving 2" diameter wheels.

A: Mark J. here: there is more to being a good weapon motor than power. A weapon motor must be reliable and must have enough information available to allow the needed weapon calculations. Without a full set of motor specifications you're just guessing.

As is typical for inexpensive Chinese brushless motors, the motors you reference come without the specifications needed to calculate their output power. If the specifications did exist, there would still be the question of their reliability in combat. The motors are made for model aircaft -- an application which places much different stress on the motor than spinning a robot weapon. My recommendation is to use a motor that has a proven track record in robot combat -- so, of the motors you list, I'd have to go with the Speed 300.

Q: Okay, you would suggest the Speed 300 brushed motor. But how come most insect builders use Hobby King brushless motors now? For example, team Twisted Sick Robotics uses Turnigy brushless outrunners with good results. Also, what about the Axi's? Because the Axi's are so expensive compared to the Hobby King motors, would it be a better idea to use the AXI 2208/34 SILVER LINE? What is the difference between the gold line and the silver line, besides having slightly lower specs?

A: The Hobby King motors are false economy, IMHO. Insect builders think they're saving money, but the experimentation and failures are not worth the few dollars saved. Twisted Sick Robotics had poor results from their first experiments with the Turnigy motors. They stuck with them and learned how to best use the Turnigy, but I suspect they would have been better off to have started with a proven motor. Do yourself a favor and save yourself the frustration.

The Axi "silver line" motors are the same design as the 'gold series' but have less expensive magnets which results in somewhat lower performance and efficiency. I would much rather use a silver line Axi than a chinese knockoffs because the Axi comes with accurate specs so you know what you're getting, but if it were me I'd go for the best I could buy. We've said this often: 'Cheap' and 'Combat Robot' don't go together. You'll kick yourself when that cheap motor fails and knocks you out of a tournament.

Q: Hi, one last question. Would it be a better idea to use an Axi 2208/20 gold series motor and gear it down by around 5:1 to spin my sawblade, or use the 2208/34 gold series and gear it down 3:1? I would be running it at 11.1 volts. It seems like the 2208/20 uses more amps and has a higher current capacity than the 2208/34, so which one would have a higher starting torque with the given geardowns?

A: The Axi 2208/20 has both more torque and more speed than the Axi 2208/34 at the same voltage -- even if geared the same the /20 would provide greater starting torque than the /34. I would suggest using the Team Run Amok Spinner Excel spreadsheet to examine weapon parameters and to select optimum gearing and blade dimensions for a given motor and voltage.

A: Mark J. here: you aren't interested in torque or power consumption? The Speed 280 and 300 motors are generally used for model aircraft or combat robot weapons where neutral timing is not required. I don't believe that any of the Speed 300 motors are neutrally timed, and the one Speed 280 motor that I know of that has neutral timing does not meet your speed requirement. If I knew what you plan to use the motor for I could be more helpful.

Background for readers: motor timing relates to the alignment of the commutator and brushes to the magnetic field of the motor. A neutrally timed motor will operate in either direction with equal speed and efficiency. A motor with 'advanced' timing will gain speed and efficiency in one direction and sacrifice both in the other.

Q: I am looking to replace the stock motors on the B-16 gearmotors with something more powerful and since I plan on using ~1.5" wheels, I need a motor that has pretty high rpm in order to have decent speed.

A: For wheels that small I think it would be both easier and better to switch to a different gearmotor. Consider the BaneBots 9:1 24mm RS-370. With 1.5" wheels at 7.4 volts, a pair will give about 8 MPH top speed and break the wheels loose when pushing at just over 3 amps apiece.

A: We've said this before: the weapon is perhaps the least important system on a combat robot. Drivetrain, radio set-up, general construction practice, and balance are all much more important. Too much emphasis on the weapon will result in a poor robot. Antweight 'Metroid' has a weapon well balanced with the rest of the design.

A: Most of the top ranked antweights at BotRank.com are two-wheel drive, but they all have a skid or castor rather than a second set of unpowered wheels. Trying to drag unpowered rubber tires sideways to turn will severely limit the maneuverability of a robot. A two-wheel drive robot may not have the pushing power of a four-wheel drive machine, but if you've got a big weapon you don't need pushing power.

A: I have not used the Pololu TRex Jr ESC but specifications are quite similar to the discontinued Barello Ant 150 -- both have a third channel for weapon control that the Ant 100 lacked. It appears to be suitable for insect class robots.

Somebody out there must be using one -- write in a review.

Which design would you rather have?

A: The servo-powered lifter on 'Warpig' is simple to build, easy to drive, and reasonably effective. I would recommend that design for builders new to building and driving lifters.

Personally, I would rather have a design like 'Wizard Beard'. Their lifter wedge itself could use a makeover into something more like the scoop on Team Nightmare's 'Breaker Box' to avoid being such spinner bait, but the flexibility of lifter action makes for a versatile and tenacious robot.

Q: If you made a much wider tip on the lifter of Warpig, would it increase its effectiveness?

A: A wider tip would change the characteristics of the lifter, but I can't gaurantee that it would be more effective. A narrow tip can follow the contour of the arena floor more closely and may be better at getting under low wedges, while a wide tip might offer more lifting opportunity and less chance of the opponent slipping off. Changing to a 4-bar lifter might be the best option.

A: Mark J. here: the greater the spin speed of a rotary weapon, the faster it must approach the target in order to get a chunk of the opponent deep enough into the weapon for good 'bite'. The only video I've seen of 'Sting' has a single 'weapon-to-weapon' impact against an eggbeater. The relative closing rate between Sting's drum and the opponent's spinning eggbeater would be very fast, which would allow enough bite for a powerful impact. This is a special case.

If you listen carefully to the video you can hear the weapon speed on 'Sting' change throughout the fight. I assume the operator knows that hitting an armor panel at full speed would not be effective and he is moderating weapon speed to match the target opportunity.

A: The experience of the builder is more important than the type of weapon. If you keep jumping to new designs you will constantly be in new territory and will have to make new mistakes to learn from. Stick with what you know. Take a good look at your most recent design, evaluate its strengths and weaknesses, and design an improved version.

A: Doesn't weigh very much? Pretty cheap? Are we talking about the same servo? The only Torxis Ultra Torque servo I know of weighs almost two and a half pounds and costs almost $300. It is horribly slow and will deliver only about 25 pounds of force at the end of a 4" long arm attached to the output wheel. Try pushing a sharp nail thru 1/8" Lexan with 25 pounds of force - not gonna happen.

A: The UK 150-gram antweight 'Anticide' is beatifully designed, superbly made, and tremendously effective. I don't have a way to directly compare it to Chris Sherwood's 150-gram flipper so I can't comment on which is more powerful, but both are powerful enough to be effective.

The 150-gram class in the US is neither as well supported or competitive as it is in the UK. Top 150-gram UK robots would rule at US events.

A: Why not? Nanoweight 'bots are 50 grams apiece -- you could have nine of them cooperate as a one pound (454 gram) antweight. I'm not saying it would be a good idea.

A: I would be a poor designer to recommend a servo knowing so little about your lifter design.

A: I'll assume we're talking antweight here? Sure, why not. Trim away excess material and file the leading edge down flush with the floor. Put in a stop to keep it from folding back under and lifting the front of your 'bot.

Q: about my door hinge question, could that be used on a beetleweight? I'm thinking about getting the tonka flip over racer to turn ito a hackbot.

A: Would that be the Tonka 'Bounce Back' racer you're planning on using? Do you know that it only goes forward and backward -- it does not turn? I'd pick something else to hack.

A door hinge would be pretty small for a beetle wedge, and I'm not sure how you're planning on mounting it securely on an invertible hackbot.

A: Mark J. here: calculation of best gear reduction requires more information than you're giving me. Gearing is dependent on the rotational inertia of the drum design as well as the weight. See the previous post on this topic in the weapons archive for information on how to calculate best gear ratio and spin-up time.

I ran some approximate calculations assuming a 1.5" diameter aluminum drum 3" long weighing 3.5 ounces. A 2:1 drive ratio gives a spin-up time to 30 Joules in 0.5 second. I'd go with that. A significantly larger drum should be geared lower.

The small 2mm shaft diameter of the Speed 300 motor is a problem when selecting pulleys. The smallest available timing belt pulleys and belts I know of are for 3mm diameter shafts. You'll need to shim your motor shafts out to the required size. Brushless outrunner motors have larger shafts, which is one reason for their popularity as weapon motors.

A: High-power servos are a simple and popular choice for insect-class lifter power. See Wallop for an example.

ESC selection is dependent on more factors than just robot weight. Run your drivetrain design thru the Team Tentacle Torque & Amp-Hour Calculator to find out how many maximum amps your motors will consume, then select an ESC that can provide that many amps.

A: The various Deans connectors are highly regarded in all weight classes. Teams usually have an assortment in their pit boxes. The Deans Micro Plug has become the standard in the insect classes because of its very low connection resistance, tight fit, tiny size, and high amperage rating (20 amps). The JST connector (8 amp rated) is OK. I wouldn't replace it if it was working well for me, but don't push it beyond its rating.

A: There are no size limits for any U.S. weightclass. As long as you can fit it thru the arena door, you're good.

A: Mark J. here: if the design for the weapon on 'MowBee' was a good one, you'd see many imitators. A drum with motors wedged inside and direct-driving the assembly supported by their shafts works fairly well with outrunner brushless motors that spin at relatively low speed (~4000 RPM) with a lot of torque, but is really not suitable for high-speed (12,000 RPM) low torque brushed motors like the Speed 300. Your spin-up time would be too great for a small insect arena, and a drum spinning that fast will just skitter off the surface of an opponent rather that 'biteing in'.

Also, the shaft and bushings on the Speed 300 are too small and too weak to adequately support the forces generated by the drum weapon impact. One hit and the shaft and/or motor end plate would likely fail. A better solution would be a belt-driven drum, geared down to spin in the 3000 to 4000 RPM range. This is a much more common and successful design which will isolate the motors from impact damage and provide a much batter spin-up time.

A: While you're backing up in a circle, your opponent is going to be all over you. It's also possible to get stuck in a corner, unable to turn in the direction needed to free yourself. Don't even consider this design.

The insect class robots at HORD are quick and maneuverable. If you're on a really tight budget, you'll be much better off running a hacked R/C toy with tank steering. Take it apart if you like and extract the radio, controller, and drivetrain for your own chassis.

A: There are many inexpensive materials that would be far superior to wood for a wedge -- many of which can be obtained at a dollar store. See previous questions in the archive.

A: Mark J. here: the 'CKRC WIGGLE' is a re-branded Cirrus CS704MG Servo. I've used other Cirrus servos and found them to be reliable. I have not had 'hands on' this servo, so cannot compare it directly to the Hitec 645MG. The specifications look very good and the chatter on the web about them is good -- but I don't know any robot builders using them. YPYMYTYC.

You don't neen anything fancy to mount your lifter servo - a couple of blocks of UHMW polyethylene, polycarbonate, even wood will do nicely depending on your design. Spend that money on something more critical.

A: Traction in an antweight arena is an elusive goal. Dust, oil spray, and other contaminants on the arena floor will spoil traction for any set-up, so don't expect miracles. That said, a thin layer of RTV silicone rubber applied to the tread surface can give a significant traction boost under some circumstances.

Silicone is gooey stuff to work with, and you have to work quickly and with a steady hand to get a satifactory job. If you've never worked with silicone before, practice applying a thin layer to some scrap objects before setting to work on your tires. Silicone bonds better to some materials than others, and you may get some peeling in even the best cases.

Clean the tire surfaces very well with alcohol and let dry. Apply a very thin, even layer to the tread of each tire and set aside to cure for at least 24 hours before use.

A: 'Gilbert' is very quick, but I'm not sure how fast. MH Robotics does not share much info on their robots - I don't know what motors 'Gilbert' currently uses.

I was thinking about a lifter, but I don't think it would be very effective either. Most antweights are invertable and I'm not using a 4 bar linkage so it would be hard to flip. Could you use a VDD polycarbonate gearbox kit and put a small S7 bar instead of the discs? Would that work on the inertia labs chassis? Thanks a ton.

A: Don't sell the pushing power of your Inertia labs setup short. Motor power alone does not make a good wedge. Pushing power is limited my the weight pushing down on the driven wheels and the coefficient of friction between the tires and the arena surface. Adding more power beyond the traction limit only succeeds in spinning the tires without generating additional pushing power. An all-wheel drive chassis like yours is a good bet to out-push a 2-wheel drive 'bot of equal weight with any amount of motor power.

Also don't underestimate the effectiveness of a simple lifter. Take a look at the results of our What Weapons Win research. The raw data shows that lifters had the highest win rate of any weapon. Our own beetleweight Zpatula is an electric lifter. Simple lifters are about control and opportunity, not just turning your opponent over.

Most builders want to believe that spinners are the way to win, but the numbers just don't show it. I'm sure you could get a VDD spinner gearbox working with a steel bar weapon, but it might not produce the results you're looking for.

A: Browse around builders websites and look for wedge mounting methods that would suit your design ideas. In general the mounts should extend well back into the chassis and be fastened at multiple points. The answer to 'how thick' is always 'as thick as you can make it and stay under weight'. Making a smooth bend in thick sheet metal is something that requires specialized equipment - make a friend at a local sheet metal fabrication shop.

Q: Way down the archive there is a picture of a hacked Battlebots toy with a scoop. Do you have a side view picture of it or do you know how it was mounted? I'm making basically the same thing with my toy.

A: The robot is Holy-Moly from Robot Dojo. The scoop is chrome-moly steel and I suspect that a mounting bracket is welded to the back out of sight. Such a bracket should reach as far back into the chassis as possible and mount with machine screws and nuts thru the plastic with large washers or a reinforcing backing plate.

A: I like the materials list - good choices all around. I'd fasten the top and botton plates only at the corners to give the Lexan some freedom to flex under impact.

Thickness? A chain is only as strong as it's weakest link, and the same applies to combat robot armor. Don't scrimp anywhere, but the top and bottom plates won't be taking a lot of load in most situations.

A: That depends on the equipment you have available and the type/thickness of metal you want to use. I think most builders would form the scoop from sheet unless they just happened to have some tube scrap.

A: I don't know much about arrow tips, but I do know there are many different types. Pick something designed to punch a hole in something more sturdy than a soft target and you should be OK.

Q: Would Deer County 100 grain broadheads be good for the arrow tips? Also, could I take the blades off for weight purposes? Thanks so much, I love the site!

A: It sounds like you know a lot more about arrow tips than I do. Sorry, can't help.

A: There is considerable discussion on armor materials and thickness in this archive. When in doubt, look at what other sucessful robots of similar design are using.

Q: Also, I'm using 2 Copal 50:1 gearmotors at 7.4v. Is this going to be fast enough?

A: Depends on your wheel diameter and the size of the arena in which you'll be competing. The 6 volt Copal motors are usually run with a 3-cell LiPoly battery at 11.1 volts. With 2" wheels and 7.4 volts you're gonna be slow: less than 2 MPH. With 2.5" wheels and 11.1 volts the speed climbs to a reasonable 3.4 MPH; adequate for a small arena.

A: I think you're in good shape, but it's kinda a funny way to design a 'bot - buying assorted parts and then worrying if you can cram the rest in underweight.

A: There are a number of comments about composite armor in the archive and further down in this archive. In general I don't like composite armor because it is so difficult to match materials that compliment each other's strengths rather than amplify their weaknesses.

In this case, you're combining a material that makes good armor because it is so flexible (UHMW polyethylene) and limiting that flexibility by combining it with a hard and stiff aluminum alloy. The UHMW is too flexible to offer any support to the aluminum -- you'd be better off to go with either of the materials alone.

A: There is considerable discussion on wedge materials in the archive and further down in this archive.

A: So, when you try to drive in a straight line it slowly turns to one side? Several possible solutions:

• If your radio system has 'subtrims', check them to make sure both drive channels are synchronized. See the Run Amok Transmitter Programming Guide.
• If you're using twin-stick tank style control, the 'ATV' function on your transmitter can be used to limit the top speed on the faster spinning motors.
• A peizo gyro can be added to your 'bot to self-correct any directional drift.
• You can move motors from one side to the other to see if some different combination of motors and sides improves the tracking.
• Live with it.

A: If it's a small 'bot, sure. I think the Spinbrush uses the Mabuchi FF180-SH motor -- fine for an ant or beetle. I don't know anything about the gearbox. It could be worth a try.

For a price comparable to the Spinbrush you can buy the BaneBots 25mm FF-180 Gearmotor that has all-metal gears and might be easier to mount.

A: No secrets here - Thomas Kenny packed four 16:1 BaneBots 28mm RS-385 gearmotors into his compact beetle wedge. The motors, gearboxes, and wheels make up half the weight of the robot!

Although not all that fast at about 6 MPH, 'Cloud of Suspicion' is terrifyingly quick in a small arena.

Currently ranked 5th among active beetles at BotRank.com, 'Cloud of Suspicion' has a 8 win, 4 loss record.

A: You want a keyed shaft, not key stock. Key stock is the sliver of metal that slides into the key of the shaft and the hub/gear/sprocket to lock them in place. The smallest keyed shaft available at Robot Marketplace is 1/2" diameter. The lack of suitable small diameter keyed shafts is why I recommend dead shafts for insect class robots.

Q: What's wrong with using the 1/2" keyed shaft????

A: Massive overkill for an insect class 'bot. It's heavy, and the 1/2" bearings to support it would also be heavy and bulky.

A: First, check with the event organizer to see if traction magnets are allowed. Some events allow them, some do not.

You're correct to be worried about overloading the motors. The added traction you get from use of magnets will increase the load on the motors before the wheels start to slip. Everytime you double the traction you also double the amperage draw at maximum push, stressing the motors and the ESC.

Calculating how much extra traction you can get from a specific magnet has too many variables, but those Maxons are currently breaking traction at less than 1/10th their stall torque -- you can add a good amount of traction and still be fine. Experiment!

You might also consider using axial magnetic rings to replace your wheels. This set-up has worked very well for insect robots like 'Wallop'. Note that 'Wallop' was designed for the magnet wheels and runs a high gear reduction to avoid over-stressing the motors.

Q: How would you suggest attaching the ring magnets to the shafts of the 17:1 Maxons? It looks like Wallop used a prop adapter for his magnet wheels, but I only have about 1/2" of space for the width of my wheels.

Also, if I chose to just put a square magnet on the inside of my robot, would there be any chance of the large magnetic force messing with any of my electronics?

A: It sounds like you'd need custom hubs. Magnet wheels are best used on robots designed from scratch around them, but you might have been lucky enough to get away with it.

A stationary magnetic field is unlikely to create electronics trouble. It's more likely to stick you firmly to an arena upright support and leave you spinning your wheels in the air.

A: I'm assuming that you're talking about the 6 volt version of the Copal 50? Four of them with 2.5" wheels would give entirely adequate performance for a beetle. Top speed would be around 4.5 MPH and that speed would come in about 3 feet from a standing start. This would be a fairly average top speed for a beetle, but acceleration would be excellent for a small insect arena. The popular B16 16:1 motors would give a top speed closer to 6 MPH which would be considered fast. Down under the 3 MPH zone is getting pretty slow, but the need for speed depends on your weapon system.

Be aware that the Copal spur gearbox is a little fragile for use in a beetle. I would strongly suggest good wheel protection and I'd carry a couple spare motors in my pit box.

Q: So, the wheels should be fully protected and the lexan mounts in addition to the all metal gears should protect the motors, right? Spares are nice in between fights but I dont want to need one during a fight! Also, how big is an average insect arena and what should I be aiming at for an average 'distance to top speed'? This will be a lifter which will need good speed but also quite maneuverable and quite acceleratory (is that a word)?

A: Yes, protect the wheels. A good hit from a beetle weapon could destroy the Copal gearbox. The gears are all metal but they are small, and a spur gearbox is not as strong as a planetary gearbox with the same size gears.

A typical insect arena is about 6' by 6', but there is considerable variation. Check with the events you plan to enter to determine the size of their arena. Top speed versus acceleration is subject to personal preference, but if you're close to top speed in half the distance across the arena you're in the right ballpark. And yes, acceleratory is a word.

A: Have you never been to a beetle fight, or were you just not paying attention? Try stopping in at Builders Database and browse pictures of beetles from recent events. I think you'll see a fairly even mix of exposed and protected wheels that depends on the overall concept of the robot. You'll also see a mix of foam and solid tires.

Q: Hey Aaron, I thought a beetle was 1 pound so I am actually building an ant. Sorry for the mistake -- thanks.

A: So, you don't know how much a beetle weighs, and don't know what type of wheels to use or where to put them, but you think you're ready to build your robot? I'd suggest doing quite a bit more research before you start ordering parts.

A: An effective weapon is integrated into the structure of the robot. The time to think about a weapon is at the start of the design process, not after the 'bot is built and you discover that have a little weight allowance left.

I'd spend the extra weight on armor, but if you're determined to tack on a weapon you'll need to take a look at your design and component layout and think about what type of weapon would make sense. Consider if there is room for a servo-powered lifter and if that would make sense in your design.

Q: Hi again -- I had the 7oz beetle weapon question. Is it possable to build an effective beetle spinner with 7 oz? Thank you!

A: I'm gonna say no. You'd want at least 5 ounces of rotating mass for an effective beetle spinning weapon. That doesn't leave much for motor, drivetrain, and support structure.

You really need to decide what your robot is going to be in the design phase and build the robot to meet that goal. If you just keep adding things on 'til you top out on weight you're going to have a robot that tries to do too many things and ends up being good at nothing. Use that extra seven ounces of weight allowance to make the robot better at what it already does well.

A: Maxon makes a large range of high-precision high-cost motors. Once in a while some of their motors are available on the surplus market at prices affordable by combat robot builders. I don't currently have a source for any surplus Maxon motors suitable for insect class robots.

A: I say this all the time: 'best' depends on your expectations and design needs. There is no single 'best'.

There's nothing wrong with the BaneBots gearmotors, but they aren't indestructable. Many builders stick a pair of wheels on the end of the shaft, bolt the motors down, run with the wheels unprotected and without additional support, and are disappointed when a spinner hits a wheel and destroys the gearbox. Protect the wheels and/or add a support bearing to the axle shaft outside the wheel and the BaneBots motors are great.

A: The Banebots wheels are fine for a beetle. I'd use the 0.8" wide version if the wheel is exposed, and carry a few spares in your supply box. You might want to have a look at the Colson Performa Wheels as well.

I'm not a big fan of the BaneBots hubs. Using set screws to hold a hub to a small shaft on a combat robot is a weak point that will sooner or later give trouble. You'll need to check the tightness before every match.

A: The gear ratio on the B62 is too large to give good speed for a beetle. With 3" wheels and running on 12 volts, the top speed would be under 3 MPH.

Pololu has small right-angle gear motors, but their output is much too small for a beetle unless you're just moving a huge weapon around slowly. I don't know of any suitable high-power right-angle gearmotors of the correct size.

A: 'Best' depends on what you want from the motors and what your design calls for. A good general purpose gearmotor for a beetle is the BaneBots 16:1 28mm RS-385. It weighs 5.1 ounces and has a nominal 7.2 volt rating. With 2.25" wheels the top speed would be near 7 MPH with top speed reached in less than a second -- just about perfect for a fast wedge in a small arena.

A: Mark J. here: Hard disk platters are made from thin aluminum or a glass composite material. They are designed for loads very different from the sudden impact they would receive as a spinning disk weapon and they are unsuitable for that purpose.

As a general rule I recommend that the spinning mass for a primary spinner weapon equal at minimum 10% of the robot weight. That's just about what the small VDD3 blade is when used on an antweight. For a beetle, you're looking at about 5 ounces worth of blade.

Update: sorry, you can scrap that question. I looked back through the archive and after reading some other answers in which you said "Make it as thick as you can and still make weight", I'm just going to do that. Thanks.

A: Mark J. here: thanks for checking the archive. The 'make it as thick as you can' rule is generally a good model to follow. I don't think that .0625 aluminum would be thick enough. Fiasco uses .375" 7075 aluminum. For a beetle you won't need to go that thick. The application of some very sketchy engineering formulas leads me to believe that a high-strength aluminum alloy at least 0.125" would be in the right ballpark. Thicker would be better if you have the weight to spare.

A: The basic components for a combat robot are pretty much the same for all weight classes. Take a look at #19 for a diagram that shows the components and how they connect.

A: I always encourage first-time builders to keep their robot design simple. You'll have enough problems to sort out without adding an active weapon into the mix. A well-built wedge or scoop is always competitive.

I can't tell you if a trip to England for an antweight tournament would be worth your while, but I can tell you that I've travelled greater distances for a good robot fight. You'll either love it or hate it, and the only way to find out is to go.

I have the 75 MHz Pico receiver right now but I am going to upgrade to 2.4 GHz. Do you know of any ways to fit the Spektrum BR6000 receiver into the Inertia Labs chassis and still be able to drive inverted? Can you strip the BR6000 of any excess plastic so that it can fit?

A: There isn't a lot of space inside the Inertia Labs chassis. The guys at Dimension Engineering have some photos of the standard set-up with the GWS receiver. They compare fitting the components in place to a game of Tetris.

You're not going top save much room by modifying the case of the BR6000. The problem isn't the thickness, it's the rather large footprint. Unless you come up with some creative solution to reduce the footprint of the battery (custom NiMHd pack?) I don't see a way to cram everything into the chassis and get the lid on.

Another question: I have a Sabertooth ESC. Is it possible to connect this ESC to a 7.2 volt battery on the brushless system without an enternal ESC battery?

A: Hi, Anthony. I can't recommend that a beginner build a Full-Body Spinner for their first robot. It is possible to make an effective FBS, but there are very few that are successful and a whole lot of them that are not.

The spinning shell must carry as much weight as possible and be very well supported. When the weapon makes contact there will be as much force placed on your robot as is transmitted to your opponent. It's likely that your 'bot will fly off in an unpredictable direction and bounce off several walls in a small arena. The 'bot must be very well constructed to survive this sort of treatment. Just bolting a mixing bowl to the shaft of a motor is not going to be nearly sturdy enough.

If this is an insect-class 'bot you've got another problem. Insect arenas are very small and you will have very little time to spin your weapon up to an effective speed before your opponent is on top of you. You'll need to be very good at dodging their first attack if you're going to survive long enough to stay in the match.

There are several previous Q&A about spinners in the archive that should be helpful to you. I'd also suggest reading the #26 for my answer to the question about what type of weapon is best.

I'm not sure I understand the ESC question. If you're asking if you can hook the Sabertooth up to the same battery that powers the brushless ESC for the weapon -- yes, you can do that. Just make sure the battery has enough capacity to power both systems. There is a wiring diagram in the (#19) that shows how everything goes together.

Q: Dear Aaron, it's Anthony again. I decided to build an antweight horizontal disc spinner instead of an FBS. How do I make it effective? Is the new polycarbonate VDD weapon gear box sturdy anough for competition? If it isn't, will I need a tougher shaft then the kit shaft from the Robot Marketplace?

A: We have LOTS of information on spinner weapons and how to make them effective in the archive. Browse around in there. Horizontal spinners are a bit of a challenge because they 'kick' off to one side on impact just about as far as they 'kick' the opponent the other direction. Not very popular for that reason - sure you want to build one of those?

The VDD weapon parts are all combat tested and will perform well.

Q: How exactly does 'Micro Nightmare's main 14.8 volt battery power both the Hacker brushless system and the SozBots speed controler at the same time? Can the same be done with the Scorpion HX? I'm confused.

A: I pointed you to the wiring diagram in the last time, Anthony. The diagram there shows how to wire a weapon ESC and a drive ESC to a single battery pack. They are wired in parallel: connect both '+' inputs on the ESCs to the '+' battery terminal, and both '-' inputs to the '-' battery terminal. That will work just fine with your Scorpion HX.

A: This question keeps popping up and I keep answering it.

A: Sure -- you can use a nano servo to operate the Lego pneumatics switch and control the system. However, the Lego pneumatics components do not provide enough power to be very useful in even an insect class combat robot. CSSOH'S Lego Pneumatics Page (archived) has many examples of Lego pneumatics construction and a primer on Lego pneumatics. I can't recommend a articularly good place to purchase the sets, other than the usual places like amazon.com.

Q: I recently asked about Lego pneumatics and you said they didn't provide enough power for an antweight fliper. Well, I searched and found that the larger Lego pump can supply enough force to lift about 3 pounds. How much more do I need for an antweight flipper?

A: Mark J. here: Power is the product of two components: force and speed. The Inertia Labs micro pneumatic system delivers 16 pounds of force and delivers it very quickly thanks to a high-flow valve and pressurized storage tank. The Lego system operating off of an electric air pump is designed for slow and safe actuation; it takes way too long to develop force and move it over enough distance to make an effective flipper. The best you'd get is a slow and very gentle lifting action -- absolutely not a flipper.

A: Scroll down a little to the 'Grabby' question for a discussion on antweight crushers. Nobody has built a successful antweight crusher yet, and it may not be possible. It certainly wouldn't be easy!

A: Sorry, no off-the-shelf CO₂ systems that small. Inertia Labs made a small low-pressure air pneumatic system for insect-class robots, but they are no longer produced.

A: Mark J. here: let me take this one, Aaron.

It's really difficult to discuss just one part of a weapon system. A motor that would work well for one clamper design would be completely unsuitable in another. That said, the Tamiya 72001 is not really 'combat quality'. The gearbox housing and gears are plastic and the output shaft is poorly supported and of too small a diameter to properly fasten a clamp arm. You'd be much better off with a more robust gearmotor like the Beetle B231.

A: The photo of Grabby at the Inertia Labs website shows a single standard-size servo connected to twin pushrods that activate the claw mechanism -- pulling back to close the jaw and pushing forward to open it. Two pushrods are needed because the ends of the claw mechanism diverge as they move. There is a second 'thin wing' servo attached to an arm apparently used as a SRiMech.

Given the lever advantage, the claw will actuate quickly, but will not have much grabbing force. Certainly not a crusher.

Q: Have "Grabby', 'Lifty', and 'Cutty' competed or are they just examples on the Inertia Labs website?

A: All three robots are credited as being built by The Machine Lab. I'm not sure that they were built as combat robots, and I can find no record of any robots from 'The Machine Lab' in combat competition.

Q: You said that Grabby's design was meant for speed, not force. What design would work for force? What would it take to build an antweight crusher with the Hitec 645-MG servo implementing the design of Grabby? Would the servo supply enough power of would you have to use something else like small hydraulics?

A: Mark J. here: contrasting Grabby to a crusher was a facetious exaggeration. Direct servo-powered clampers can have a large range of travel and little force like Grabby, or you can change the lever pivot point to reduce the range of travel and increase the clamping force. Every time you double the clamping force, you cut the amount of jaw travel in half. By the time you got enough force to do any damage, the jaw travel would be unuseably small.

You could modify the servo for continuous rotation and design an external gearbox to multiply the available force. The problem then becomes that force and speed are a trade-off: more force gives less speed and vice versa. Jaws with significant force and travel would be agonizingly slow.

There is nothing magical about hydraulics; you still have the problem with the inverse relationship between force and speed. If such tiny hydraulic components existed, their only advantage would be that high force can be generated without the need for heavy gearboxes and linkages.

An additional problem with crushers is the enormous stress placed on the chassis of the robot deploying the weapon. A force equal to the force applied to your opponent must be borne by your own chassis. I've seen several would-be crushers that succeeded in ripping themselves apart.

A: Good servo choice. The 645-MG will deliver 38 ounces of lift at the end of a 3.5" arm. That's plenty to give quick and reliable lifts.

A: The kilobot version of 'Mars Attacks' was a thwackbot spinner made to look like a shell spinner. A thwackbot can't move when attacking (unless you have some fancy melty brain electronics) so all it can do for offense is sit and spin. There have been very few successful thwackbots -- judges hate them -- but they are fun to drive.

A: The HXT 12kg (AKA Towerpro MG995) has Futaba splines and is quite inexpensive, but I strongly recommend that you buy something else.

All the reviews I've found say they're junk. They're heavy, poorly made, unreliable, and don't come close to meeting their advertised performance figures. One review thinks they may be the "world's worst servo." I wouldn't modify one of these to throw itself away. Avoid.

A: If you want to keep it simple, try the Hitec HS-775MG servo. It's reasonably priced, has 200 oz/in torque, full metal gears, and easy mounting tabs.

Q: Which will be better for a beetle clamp: the servo you sugested running at 7.2 volts? or a BaneBots 28mm 25:1 gearmotor at 7.2 volts? If I use the BaneBots should I support the shaft?

A: I considered the BaneBots motors before I made my recommendation. The Hitec servo will hold better at a lower amperage draw, is an easier shaped package to mount, and requires no external speed controller. If you use a clamping arm that has some side flex (Lexan, maybe?) and mount close to the output bearing, either the Hitec servo or the BaneBots gearmotor should be OK with no additional shaft support.

Note that the Hitec servo is rated only up to 6 volts. I've never used this servo, but suspect it will be OK at 7.2 volts.

A: Mark J. here: there are several posts on the design of 4-bar lifters in the archive, including a link to design software.

For an antweight the servo output can replace the entire rear (or front) bar of the mechanism to simplify construction. The diagram at right shows one possible configuration.

A: Composite armor is tricky -- you can end up with the worst properties of the individual materials rather than the best. Even if you got the best properties of those three materials you still wouldn't have much. If you have to stick with stuff you find around the house, polypropylene plastic ('PP' in the recycling label) from snap-on lid kitchen storage containers is quite tough and easy to work with. It would be a better choice.

A: I'm not sure what you mean by 'self contained', but there is a discussion on different servo powered lifters in this archive. Search for "using a servo". The simplest approach is to just attach a lifter arm directly to the servo output wheel.

A: The Hitec 645MG is a great servo for an antweight lifter. The Futaba S3305 is comparable: a little lighter and a little less expensive with all metal gears and dual ball bearings.

A: The wheels usually used on the Inertia Labs chassis are the 24mm diameter Ultra-Light wheels. They are a press fit onto the 3mm diameter Sanyo shaft -- no hub needed.

A: Sounds like a good choice. For use in an 8-foot arena, I'd go with the 20:1 gear ratio: 0 to 5.5 MPH top speed in 3.5 feet, and cross the arena in 1.2 seconds. Maximum power draw would be about 2.2 amps per side, well within the Ant 150 capacity. If your arena is larger, consider the 16:1 gearing for a little more speed.

A: Mark J. here: we've never worked with 15-3 (AKA 'Timetal'), but I've looked up it's properties and it appears to be an excellent armor material. Complex metal forming can locally 'work harden' titanium and lead to cracking under later stress; keep the bends simple and you should have no problems.

In aerospace applications, Timetal is re-annealed after forming, 'age hardened' at high temperature, then air cooled. This is the ideal treatment and it will add about 40% to the strength of the finished piece.

A 21 gauge sounds appropriate, but armor performance depends on factors other than just thickness. The size and shape of the piece, mounting method, and support from other chassis elements all count toward performance, but unless you're doing something odd the 21 gauge should be great.

A: Mark J. here: a few thoughts:

• 30K RPM is WAY too fast to spin a rotating blade -- see previous question on blade speed in this archive;
• the drivetrain sounds fine;
• a solid machined titanium antweight chassis is expensive overkill.

Q: What happens if you spin a blade to fast?

A: Like I said, see previous question on blade speed in this archive. Search the page for 'too fast'.

A: Chassis or wheel magnets are left to the discretion of the event organizer. Some allow them, some do not, and some arenas don't have a magnetic floor so it doesn't matter. At least a few builders provide their 'bots with a set of magnet wheels to use where they are allowed and a set of rubber wheels to use elsewhere.

A: Yes. Inertia Labs had an ant named 'KanunDrum' with a internal combustion powered drum. The 'bot was never mentioned on their website. It fought at the Marin Ant Wars in 2004 and 2005. Botrank.com shows 1 win and 3 losses.

Q: How does KanunDrum's weapon work? Does the ICE power the wheels too?

A: The drum is belt driven by a model aircraft internal combustion engine. It looks like a servo is used to engage the belt with an idler wheel. The engine is manually started via an extension of the shaft that sticks out to the left of the 'bot. The wheels are driven by electric motors.

A: See my warnings about overvolting in #25. The FA-130RA motors that come with the Tamiya gearbox are particularly vulnerable to melt-down if allowed to overheat. I would not run them higher than 6 volts even with a low enough gearing choice to spin the wheels well before stalling.

A: What is it about an FBS that makes it so attractive to builders? If you take the time to research the combat record of FBS robots you'll find that, as a group, they have a very poor win percentage in any weight class. They are often as dangerous to themselves as to their opponent -- ricochetting off at odd angles and into trouble. I can't recommend building one.

Your design questions all depend on specific information about the dimensions of the shell, the material used, the torque of the motor, and the size of the arena. The Team Run Amok Spinner Spreadsheet for Excel can perform the calculations you'll need to determine the power and gearing requirements for a specific shell design. I have seen direct-drive antweight FBS, but they had small diameter shells and very powerful motors.

Q: How come there are not many full body spinner antweights?

A: Read the answer above: there aren't many ant FBS because ant FBS have a poor win percentage. Most builders want to build something that they think can win.

A: Several reasons:

• No high-pressure pneumatic equipment is available in antweight sizes;
• Antweight 'bots can generally survive repeated flipping without damage;
• Damage/aggression scoring systems don't give credit for flips;

A: 'Hummer' is a beautifully designed and constructed antweight. It is also very successful: 26 wins and 6 losses, seventh in the Botrank historic rankings. 'Pele' is more of a show-off 'bot -- fun to look at and an interesting challenge to drive, but not really competitive: 10 wins and 14 losses.

I don't believe that the Inertia Labs pneumatic parts are available anymore. If you could scrounge enough pieces to put a system together I'm sure you could have some fun with it, but don't expect to win tournaments.

A: I don't like a hinge on a front wedge. The hinge would be a weak point in the wedge mounting, and a hinged wedge can fold back under the robot and lift the wheels of the robot off the floor. If you want the wedge to scrape the floor, you can build a subframe for the wedge and pivot it from the center or rear of the robot, like the scoop mount on Breaker Box.

Small hinging mounts for antweight side skirts can be made from heavy package tape or from small mechanical hinges available from your local hobby shop. The hobby shop will also have special 'hinge tape' used for model aircraft control surfaces. You could also use small metal jewelry box hinges available at craft stores.

A: I don't know of any ready-made wheels that small in diameter with that large a bore. It's time for a trip to your local hobby shop to see what they have. Check slot car wheels and R/C airplane landing gear wheels. You're going to have a hard time finding a small wheel with enough material in the hub to allow enlarging the bore to 6mm.

You've probably figured out that you aren't going to get much speed out of 3/4" diameter wheels on standard ant gearmotors, but I thought I should mention it.

A: Oddly, there seems to be some argument about what 'best' means. To me it means the weapon that gives you the greatest chance of winning. Other builders would rather put on a spectacular display of destruction, even if it means they seldom win a tournament.

We did some research to find out what weapon types win the highest percentage of matches. Overall, robots with passive weapons (wedges, bricks, rammers) win more often than robots with active weapons (spinners, flippers, lifters). However, if you look carefully at all the sub-classes of weapons, it's lifters that do the best. They don't win many matches by knock-out, but they can dominate a battle and make an oponent look weak and ineffective.

Antweight lifters are relatively easy to construct because R/C servos have enough power to function as your lift motor.

Q: I was wondering if I should skip the antweight an go straight to a larger bot? I know it will be more expensive but maybe I can save up. Will I find more or fewer competitions? [pyrobug]

A: Generally, the heavier the 'bot, the harder it is to find a match. In 2007 the number of RFL sanctioned matches by weight class looked like this:

A: Mark J. here: the large scale R/C airplane guys use pneumatic systems to raise and lower landing gear in a realistic, slow manner. Their pneumatic cylinders can produce good force, but the critical ports and valves flow air too slowly for a flipper.

A: The amount of force at the end of a simple arm attached to a servo is:

Example: a servo with 60 oz-in torque with an arm extending four inches from the central axis of the servo will have:

Note that the servo would be able to 'hold' 15 ounces at the end of the arm. It would be able to slowly lift 14 ounces, speed increasing with lessened weight.

A: Mark J. here: as noted in the all the tools you need to evaluate spinner motor performance with specific weapons are in the archives. In particular, the Team Run Amok Spinning Weapon Excel Spreadsheet will detail and graph the performance of a specific motor spinning a specific weapon at a specific gear reduction.

There's an infinite number of combinations of motors and spinner weapon designs. I've provided the tools, you run the numbers. Fair?

P.S. -- You'll need to know the width and thickness of the bar in addition to the length and material.

P.P.S. -- Oh all right: assuming that the bar is 1" wide and 1/4" thick (about 6 ounces), the GWBLM001 2205/15T geared 4:1 will spin it up to about 2500 RPM and 45 joules of energy in 2 seconds. Killer. But seriously, that's the last spinner analysis I'm gonna do!

Buy a couple of those motors -- BaneBots is closing them out at a very good price.

A: I've seen many competitive antweights built around the Tamiya Twin Motor Gearbox. It is inexpensive, sturdy, and provides adequate power for an ant. The main drawback is that it's a little bulky.

A: Mark J. here: It's not even close to being a 'complete kit'. You get a chassis, motors, wire, and wheels. You need to come up with radio gear, an ESC, a battery pack, charger, and some sort of weapon. The machined aluminum chassis is way cool, but not easy to repair if it gets pranged. You might want to order a spare for your pit box. I don't really consider this to be a 'first time' kit.

Q: Were can I get the 'A:1 Complete Antweight Kit' now that Composite Labs shut down?

A: Weak! I hadn't heard that Composite Labs was gone. I had been recommending their kits. They were the manufacturer and only source, so they're gone for good.

It's difficult for me to recommend any of the currently produced antweight kits for a beginner. You're going to spend $300 or $400 and end up with a marginal radio, a low-end battery charger, and a pile of parts with no specific instructions. I'm gonna go back to suggesting that a beginner find an old BattleBots R/C toy on EBay, hack the radio and drive motors, add a wedge, and go beat up some $400 'bots for $50.

A: Way too flimsy. The first spinner you came up against would shred it. Try titanium, and three times that thick.

A: Sure! There are many good antweight lifters powered by servos. See previous posts on selecting an antweight lifter servo and designing 4-bar lifters.

A: Match the cooling fan to your voltage and it'll work, but it's overkill. A pair of them would weigh half a pound -- a big chunk in a three pound robot. Why would you want heavy 60 amp ESCs in an beetle?

A: I won't post information here that could lead an inexperienced builder to injure themselves. See the previous post on flamethrowers for reasons to build something else.

A: Magnetic wheels are extremely effective -- if you happen to be fighting in an arena with a steel floor. I've only seen them used on insect-class bots. See Team Nightmare's beetle page for some good photos of the 2004 beetleweight national champion 'Wallop' with its magnet wheels.

The wheels are simply round rare-earth neodymium magnets with a hole in the center. These can be very expensive, unless you happen to find a suitable set at a surplus outlet.

Don't leave magnet wheels exposed. Put them under your armor or use side guards to keep them from clamping onto steel rails or other vertical structures. It would be embarrassing to end up stuck to a railing. Also, be sure to check with your local event organizer to see if magnet wheels are even allowed!

A: First, find another hobby shop - one that knows something about 'bots. Second, servos make poor 'bot drive trains. Save up your money and buy a real speed controller and drive motors. See the servo discussion for more info.

A: I like to fight big 'bots! The bigger the 'bot the better -- but even tiny 'bots rock.

A: The side-wheel spinner design on Run Away is unique and showy, but a pivot steer 'bot like Run Away would be very difficult to maneuver in a small arena.

A: No. Internal Combustion Engines (ICE) are tempting, but getting them to work well and reliably in combat can be a huge problem. Also, not many insect class competitions allow ICE. Check the rules for your competition before you commit.

A: You could do a lot worse than the components used for the Robot Marketplace Basic Antweight Starter Package, Nick. The Tamiya dual-motor gearbox is quick and tough, and the 56 mm 'sports tires' slide right onto the shafts of the gearbox. The Ant 100 ESC is reliable, but I'd upgrade to the Ant 150 ESC for a weapon control channel that could come in handy on a later 'bot.

Radio and battery/charger are up to you. I use inexpensive NiMHd cells for ants because chargers for them are available and cheap -- but LiPoly cells are lighter. Bolt everything down to a simple flat chassis, add your armor and a sturdy wedge. Keep your first 'bot simple and have fun!

A: Servos are slow! The speed of a typical 'fast' servo modified for continuous rotation is about 60 RPM. With a 2" diameter tire that gives you about 32 feet per minute, or less than 0.4 miles per hour. That's not even a crawl!

It is possible to 'speed mod' a servo by locking the first and third reduction gear clusters together and eliminating the second reduction stage altogether. That will improve the speed by a factor of fifteen or more, but will reduce the torque by the same factor. It will make the motor prone to stall and an early death. It isn't worth the trouble -- don't use servos for drive motors.

The Futaba S3110 is a really tiny servo. The gear train is way too fragile to even consider using it as a drive motor in an antweight. It would be slow, weak, and vulnerable.

A: Mark J. here: I can't figure out why a servo would be your only option, but if you wanna do the servo thing right my choice would be the Hitec HS-5996TG.

• The titanium gears are next to indestructible,
• A beefy 111 oz/in stall torque is un-stallable,
• 60 degrees rotation in 0.10 seconds is nasty quick (for a servo),
• Hardened gear pins and intermediate oilite bushings with ball bearings on the output shaft will take serious punishment, and
• The coreless motor shaves the weight down to 63 grams.

Q: Is there a good servo to power a 'bot for less than $40?

A: It's funny how fast a price limit gets set when somebody finds out what 'absolute best' actually costs.

I really don't want to encourage you to build a servo-powered 'bot - they're really slow! For less than the $80 you're willing to pay for a pair of servos you can buy a Tamaya dual-motor gearbox and a good speed controller that will run rings around any servo-bot. Reconsider?

A: 'Cheap' and 'robot combat' really don't go well together. You're gonna spend your time and effort to build something that's going to let you down in battle. Why bother?

If you insist: buy a Futaba '2DR' 2-channel AM R/C setup. It comes with a transmitter, a receiver, two servos, crystals, mounting hardware, and a battery holder for about $40. Hack the two servos for continuous rotation. Find the centers on two plastic mayonnaise jar lids and mount them to the servo horns with self-tapping screws that come with the radio -- instant wheels. Stretch a wide rubber band around the outside of each jar lid for tires. Foam tape the receiver, battery holder, and servos to one of those free AOL CDs that come in the mail. Insert 4 AA cells in the battery holder and 8 more in the transmitter.

Congratulations, you've just built the world's cheapest scratch antweight. Now, throw it away before you embarrass yourself.

Q: In a earlier post you mentioned the complete Futaba 2DR AM R/C set for around $40. Where can I buy one?

A: Futaba 2DR at Tower Hobbies. This is a two-channel AM system. Check with your local event rules to make sure they allow AM radios for the weapon type and weight class you plan to build. FM radios are more expensive for a reason: they work better.

A: Rat trap - $1.79.

Q: How can I make the rat trap into a weapon?

A: Don't be too anxious; the rat trap is cheap but I'm not saying it's good. Set the trap and drive toward your opponent -- you get one shot. Check legality with your local event organizer and watch your fingers!

Q: I'm having trouble with my rat trap weapon. Every time I bump something the trap snaps shut. Is there a way to make it less sensitive?

A: Examine the way the trap latches to the trip plate. The bar is held by a very narrow dimple of bent metal. Bend the dimple to make it deeper, and reshape it to hold the bar more firmly.

My dad's antweight has a tiny servo that latches the trap shut 'til he releases it -- but that adds too much expense for it to qualify as the 'cheapest active weapon'.

Q: How did you turn Rat Amok into a flipper 'bot?

A: You must have seen the 'Rat Amok' video. Rat Amok isn't designed to be a flipper, but if an opponent climbs far enough on top they can be thrown forcibly upward then the trap is released. Other times, the trap ejects the opponent out the front of the trap at high speed and Rat Amok may flip over backward in reaction (she drives fine upside-down). Only occasionally does the trap actually 'catch' a part of the other 'bot. Teeth or a rubber 'gripper' added to the trap bar would modify the effect of the impact.

Q: What was the total cost of Rat Amok? What servos did it have?

A: Mark J. here: Let's see --

• Rat trap - $1.79
• Futaba T6XAPS R/C system - $229.95
• 2 Futaba S3004 ball bearing servos - $0.00 (came with R/C system)
• Cirrus 'nano' servo trap release - $29.95
• 5 700 mAH NiMHd cells - $9.95
• 2 front wheels from Tamiya R/C dune buggy - $7.95
• Dollar Store knife for front blade - $1.00
• Polypropylene cheese box from thrift store - $0.50
• Assorted hardware - $2.58

A: Servos are way too slow to directly power a flipper. I suppose you could use a servo converted to continuous rotation as a winch to wind a spring for a flipper, but that's too complicated to write about here.

You can use a servo to power a small lifter for an insect-class 'bot. You can simply connect an arm directly to the servo output wheel, but you'll have more interesting leverage options if you connect the servo to the lifter with either a rod link or a full 4-bar linkage. Search down the page for "4-bar" for tips. Your local hobby shop will have everything you need.

Q: What should I look for in a servo that's gonna power an antweight lifter? How powerful should it be (in/oz)? Are there any successful antweight lifters?

A: Mark J. here: I've looked around at ant lifters and the only things their servos have in common is high torque. The drive gears are usually well protected so metal gears are not needed. Speed isn't really an issue either. You don't need to spend a lot of money here.

Assuming that you're using the servo on one end of a simple lifter arm, you need to have enough power at the end of the arm to lift 16 ounces. Example: if you have a 3 inch long lifting arm, you need 3 in. X 16 oz. = 48 in/oz torque, plus about 25% to keep it from stalling. More torque will lift faster. There are other factors to consider, but that will get you in the ballpark.

My favorite (and very successful) antweight lifter is Adam Conus' Mission Control. Technically it's a clamp-bot since it has a second servo on the lift arm that clamps the other 'bot in place. Adam uses a Hitec high-torque standard servo [HS-545bb] with 76 in/oz of torque for the lifter and a Hitec 'Mighty Mini' servo [HS-225bb] for the clamp.

Q: In an earlier post you mentioned that a servo could be used as a winch to wind a spring for an antweight flipper weapon. How would you put that all together?

A: I also mentioned in that post that the winch was too complicated to write about here. I guess you didn't believe me!

The servo must be converted to continuous rotation and a small bobbin added to wind in string. The string passes around a pulley and attaches to the front edge of a mousetrap bar to pull it down. The complicated part is disengaging the bobbin so that it will spin free when the trap is released and re-engage when you're ready to reset the flipper. That requires a second servo and a better drawing than I can make. A pneumatic flipper would be less trouble.

A: Mark J. here: Chris Sherwood has been building robots in England for a long time. He was able to build his 150-gram pneumatic 'bot out of odd bits and pieces because of his extensive building experience. You can find pictures of 'Flipper': here. I don't believe his design would be legal (or competitive) in the U.S. under current RFL rules.

Very briefly, Chris uses a plastic egg for an air pressure reservoir and syringes for pneumatic cylinders. The system fires the 3-way pneumatic valve when a robot hits the microswitch on the flipper. Take a look at the Inertia Labs Micro Pneumatics FAQ and read the Micro Pneumatics PDF for info on building a similar system that would be safe and legal.

You might also be interested in using Lego Mindstorm Pneumatics parts for a low-budget insect flipper 'bot.

A: If this is your first 'bot, keep it simple. An antweight wedge would be a great choice for a first scratch-built 'bot. Search down the page for an antweight parts list and lots of tips. You can get all the supplies you need at Robot Marketplace. I just counted, and that's the 18th plug for Robot Marketplace. Doesn't anybody read the old questions?

A: That's not much of a description. The motor must fit in with all the design elements of the 'bot and work with them as a whole. What's your weapon? How large is the arena you'll be fighting in? What is the arena layout? Is speed more important to you than pushing power? What's your budget? Two wheel drive? Four wheel drive? Wheel size? Are you willing to make custom hubs, or do you want off-the-shelf parts only? Do you plan to use a lithium battery, or NiMH? How much weight allowance is available for the motors?

You can search down the page for 'fairyweight wedge' for a general recommendation of a motor that would make a good starting point for a 'bot.

A: You really have to make sure your wheels are securely fastened to the gearbox shaft. Having a wheel fall off is pretty much the most embarrassing thing that can happen in combat, so I really can't advise scrimping on hubs. Robot Marketplace has nice 3mm aluminum hubs to fit Lite Flights and other wheels for about $5 each. If you're really cheap, Tamaya wheels come with 3mm press-on plastic hubs for both round and hex shafts -- but if they shatter or just fall off, don't say I didn't warn you.

A: We used car-style steering on our heavyweight 'bot 'Run Amok'. Car steering is better than tank-style 'differential' steering at holding a straight line course and at making smooth turns, but tank steering gives much greater maneuverability in tight spaces. Car steering can work in a wedge or rammer in any weight class, but it certainly isn't popular.

P.S. - Alex Udanis wrote to tell me he thinks car steering bites. He didn't actually say 'bites', but I don't want to use the word he picked.

A: Mark J. here: just like there are different alloys of metals, there are lots of types of fiberglass. The type of resin used is very important to the properties the material as a whole exhibits. Some resins are hard, but brittle. Others are more resilient. See the earlier post on garolite in the archive.

A: No 'product' needed -- just solder three ESC wires to each output pin on the battery connector plug. I use Deans Power Plugs for ant and beetle connectors, but any good R/C hobby power connector is fine. Buy them at your local hobby shop. Don't use automotive connectors (bullet connectors) -- they give trouble in robots.

A: Neither SozBots nor RFL rules specify a maximum size for U.S. antweights or fairyweights. It's the British '150 gram' antweights that must fit in a 4" cube.

A: Tech question, Mark J. here: Outrunner style motors are a problem to mount because the entire outside 'can' rotates. Some outrunners (like the AXI) have a shaft that runs thru the rear mounting plate that makes for a simple mount. Others (like the Park) don't have a shaft in the rear and you have to get creative.

Take a look at Team Basenji's antweight 'Bitsy Blade' at the Robot Riots 5 photos page. It's an ant vertical spinner with an outrunner. They simply mounted the rear motor plate to the chassis and left the rest of the motor and pulley drive dangling in the air! It isn't good engineering practice, but it works for them. A better solution would be to add a support to the motor shaft as close to the weapon pulley as possible with a bearing on a strong support anchored to the chassis -- kinda the same as their weapon support.

Technical question, Mark J. here:

• Robotmarketplace.com claims that the Park 370 Outrunner has "awesome torque" and that "in most circumstances" the motor can be used direct drive for an antweight weapon. I'm not able to find actual torque figures for the motor to back up those claims.
• The power needed to spin a 7 oz. bar depends on the length of the bar. A short bar has less rotational mass than a longer bar and will require less torque to 'spin up'.
• A spinning bar weapon needs to spin slowly enough for the bar end to grab and catch your opponent. The Park 370 Outrunner has a no-load speed close to 10,000 RPM at 7.2 volts! That's much too fast to spin a bar weapon.

A: Unless you're running a very powerful weapon motor or a very small drum, you will need some gear reduction to spin the weapon up properly. The Team Run Amok Spinning Weapon Excel Spreadsheet can give you some help in designing your weapon For a reasonable motor and a mid-sized drum, try a belt drive with about a 3:1 reduction for a starting point.

A: The number of flips you get is dependent on the size of your air storage tank, the volume of your pneumatic actuator, and the pressure of the air in your system. You can design in as many flips as you like.

Antweight flippers are not generally effective at causing damage. They are popular in 'Sozbots' style arenas where a 'bot that leaves the platform looses. Flip 'em over the low arena wall and you win!

A: Lithium polymer battery packs require special chargers that generally run from an external DC power supply. If you want to get off cheap, you can use the battery from your car as that DC power supply. Alternately, you can run NiCad or NiMHd cells that are less picky about charging and buy an inexpensive AC/DC charger from EBay.

If you want cheaper still, run AAA alkaline cells and replace them every few matches.

A: A few years ago it was common to see antweight robots that used 'hacked' R/C servos as drive motors. Inexpensive R/C servos contain a small speed controller, a small DC motor, and a reduction gear train. They can be modified to provide rotating output rather than the stock back/forth motion (see: hacking a servo). Bolt on a wheel and you were in business!

Servo drives are nearly extinct now. They are slow and weak compared to other drivetrain options. You can build a robot with them, but you won't win many matches.

A: Flippers are powered by compressed gas. You'll need a cylinder to store the compressed gas, an actuator to move the flipper arm, an electric control valve to release gas into the actuator (and vent the pressure to lower the flipper), an interface to control the valve with your remote control, and hoses/fittings to connect everything up.

Getting all of this equipment small and light enough for an antweight isn't easy. Inertia Labs makes a full pneumatics kit for antweights, and they have a FAQ page that answers a lot of questions about antweight pneumatic systems. Give it a read!

A: Gee, everybody wants to build an antweight! I really like modifying R/C toys for antweights -- see the Toy Hacks and Kits section of the archive.

If you want to build from scratch you'll need an R/C radio set-up, a two-channel motor controller (three if you want an active weapon), two motors with gearboxes, a small battery pack, a battery charger, and wheels. You'll need to make a chassis to bolt all that stuff onto -- aluminum, Lexan plastic, and carbon fiber panels are popular. Then you'll have to get enough armor together to keep it from being cut in two. This is getting expensive!

There's an antweight discussion forum at forums.delphiforums.com/antweights/start -- dig thru the information there for more tips.

A: The same armor materials used in larger 'bots works great on antweights. Titanium is the top-of-the-line, but is expensive, not easy to find, and can be difficult to work with. Polycarbonate plastic (Lexan) is inexpensive, widely available, and simple to cut and drill. I've used ultra-high molecular weight (UHMW) polyethylene plastic with good results on my beetleweight.

A: Combat robots are usually built to a maximum weight rather than size. Popular weight classes for small 'bots include:

• 1-pound (antweight);
• 3-pound (beetleweight); and
• 12-pound (hobbyweight).

A: Wow -- that's really light! In the U.S. we get a full pound for antweights - 454 grams.

I live near some places that build parts for airplanes and I can get scrap pieces of thin titanium that are great for robots. Titanium is light and very tough, but it's usually expensive and hard to find.

You might want to consider thin carbon composite material instead of metal. Robotcombat.com marketplace sells small sheets of it. It's very stiff and strong.

Some types of plastic make good armor for light robots, too. Polypropylene is used for kitchen storage containers and is very thin, light, and tough.