After running some errands this morning - including picking up some metric hardware - we're ready to get to it, and see if we can't finish up the drumbot build. We need to install the new drive motors/wheels and finish up the wiring. Once this 'bot is done, all we'll have left is the walker. Let's see if we can get ourselves down to just one 'bot left ... Hit Back after viewing an image.

Using the M4 screws we picked up from Home Despot this morning the drive motors mount okay - we were out of alignment in drilling one hole, but we'll blame that on it being metric :-) Now we can assemble and move on to wiring.

The side rails are back on, and so is the curved rear wall. We're finally settled on the placement of the electronics, so we can move on to hooking everything up.

There's just enough room for the power switch, which has been mounted in the rear compartment. We've tied all the ground wires together, and hooked the battery and weapon speed controller to the switch, along with a jumper wire for the two drive ESCs, so things don't get too crowded around the switch.

We've spent so much time in the build space recently that we're now on to the third season of Stargate SG-1! We'll likely finish this boxed set too before we hit the road for Harrisburg.

The first motor has been wired to the speed controller, after spending a little time with jump leads to determine the spin direction, and make sure left is left and right is right. Note that the motor actually isn't spinning in this shot - it's from where the motor rubbed a wire when it was in the featherweight, but it looks like it's constantly in motion now.

And on to the second motor, again with the jump leads to ensure we have the proper spin direction. This doesn't take too long, and we're able to trim the ESC leads to length and make the final connections.

Et voila!
We do need to hook up the weapon ESC to the receiver, and drill a hole in the new rear top plate so we can get at the power switch. While we're at it we may throw a lick of paint on the new panel too.

Here's the 'bot all buttoned up, sans paint and power hole. There's one other thing missing too, at this point ...

A quick weight check shows us at 11.8lb out of the twelve allowed with everything accounted for but one thing: what we're going to stick to the motors for traction. We're debating the use of the grip tape, based on what it did to the Build Space floor with the previous motors. We're wondering whether we should switch to a layer of adhesive-backed neoprene for traction instead, and so we've put some on order from McMaster. Unfortunately it won't arrive until Wednesday, after Toni leaves in the BotMobile, but Lauren's flying in to Motorama, so she'll throw it in her suitcase before heading to the airport. We're not going to be able to call this a "Presenting" post, as the 'bot isn't 100% done, but it's sure close!

Tomorrow the balance of our metal gear servos show up, so we'll be assembling the combat Hexy kit, and that leaves Tuesday to fabricate the weapon for the walker and pack!

After wrapping up our featherweight yesterday we're left with the 12lb drumbot and 3lb walker to work on. As the hobbyweight is closer to completion than the walker we'll start there today. We need to wire up the drive train and take it for a test drive! Hit Back after viewing an image.

We've cranked up the Keurig, and in a caffiene-induced frenzy set about configuring the wiring inside the hobbyweight. Here we're working on finding the right collection of settings so the wheels run in the direction we indicate with the transmitter. After reversing channels, motor connections, and also receiver connections we finally have the right combination - to the test drive!

Hm. That didn't go exactly as planned. You can see the scuff marks on the Build Space floor that the 'bot has left while it was spinning donuts. The original motor is plenty powerful, and is able to move the 'bot. The new motor isn't powerful enough to do the same, and by the end of the test run all we'd managed to make it do is smoke. Note also that the grip tape is not as ideal a tyre coating as we imagined - there are many spots where it simply ground up the concrete floor. It looks like we're going to have to stop and re-think this drive train ...

Before we know what we can do, we need to know where we're starting from. At the moment the 'bot weighs about 10.8 pounds out of a maximum of twelve, give or take a lid and some small screws. That gives us a pound-ish to engineer a solution.

We've disassembled the whole 'bot, near enough, while we scour the Build Space looking for drive train components that might work. Either we need to gear the brushless motors, like we did in the recently completed featherweight, but finding appropriate transmission components is not going well; or we could switch to a brushed setup, but the issue there is space - fitting a brished motor and gearbox is going to mean making a wider chassis, which isn't really an option given the time available.

Option number three: moar power! :-)
This is the 'bot with the two existing drive motors removed and the two G60 weapon motors from our retired featherweight Formidable Fustigator 2.0 switched in. As you can see we're up to 11 pounds, without a revised chassis back end to accomodate the significantly longer motors.

Time for some P.A.D. - paper-aided design. The inside of the 'bot is 9.25" across. The new motors are 2.3125" long, so that leaves 4.625" of space for the chassis components. Subtract two times a quarter inch for the walls, and we have a space 4.125" by 3.5" to work with. We drew a 1:1 scale box, and here we're test fitting the speed controllers and power switch inside that space.

Confident we can make it work, we've cut new top and bottom plates from quarter-inch thick polycarb, and marked, drilled, and countersunk them.

Here's the new rear chassis bolted to the middle wall. The next step is actually fitting the speed controllers, power switch, and all the wire in this small space.

We did sand down the top and bottom plates to match the curve of the rear wall, and it came out nicely. When we're done we may actually be able to drill and tap the aluminium side walls to bolt them to the rear wall for some added stability.

Ugh - already this is looking pretty messy. With the battery and weapon speed controller in the front compartment we need to fit two drive speed controllers and the power switch in the rear component, and then decide the best location for the receiver. Argh - so much wire!

But the good news is that barring a few screws, this configuration will easily make weight, so we're going to press on and see whether we can make it happen!

This blurry shot is where we removed the bronze bushing that was in there for the old motors, as the new ones have larger dimensions so we need to accomodate those.

Here's a test-fit of one of the new motors. One thing we didn't consider is that now the aluminium side walls aren't recessed into the middle wall, so the hole that was there for the old motors is now about a quarter of an inch in the wrong place.

Having marked roughly where there needs to be a clearance hole in the side rails, we handed them over to Milly to cut out those holes.

Much better! The motor fits okay at this point.

Because we've hogged out so much of the inner walls, we need to mount these motors to the outside walls this time around. After using our CAD file to determine the placement of the holes for the 30mm metric bolt pattern we drilled and countersunk a set of mounting holes. They came out pretty accurately as best we can eyeball by holding one of the motors up to them.

Time for assembly! And we came to a screetching halt: the motors use 3.5mm pitch metric screws, and we none. Plus, we're out of ring terminals if you can believe it! Gah ... well, we figured out a shopping list for tomorrow morning, and turned our attention to the last remaining 'bot ...

... The beetleweight walker. All the leg parts are now dry, so we can try assembling one. We have enough parts to make seven legs, plus a spare set of foot parts, as we imagine those will be the ones taking the most dings as our opponents crash into us. Here we've pressed together a thigh, using the arbor press to ensure things are snug. Note to selves: next time we have slot and tab parts cut, underside the tabs a bit. Still, with the arbor press, and given that this stuff is 6061 aluminium, we can make it work.

Here we're in the midst of attaching a foot. The foot servo is connected to the controller board to ensure it stays stationary while the foot is being attached. Our only niggling doubt about this design is that we're using the plastic servo horns that came with the servos for this build - hope that doesn't prove to be a major weak point later on down the road ...

Oops. Further note to selves: pre-drill holes before installing components - do not attempt to drill hole with part in place. We've accidentally drilled through a servo. Well, we have a spare, so we can carry on for tonight, but we need to make a decision: do we pull all the metal gear servos from the plastic kit for this build, or do we wait for our second shipment to show up? USPS shows it cleared customs in Illinois today, so there's a good chance it'll show up Monday, or worst case Tuesday. We have the day to build and pack Tuesday, but will a day be enough? On the other hand, keeping the plastic kit intact will allow us to work on sizing the weapon properly and ensuring our wireless control scheme works. While we try to make an executive decision we continue with the leg assembly.

Here we have the foot installed, and it moves surprisingly cleanly for being heavier than the plastic one, and a much tighter [press-]fit together. Additional note to selves: press all the captive nuts in before continuing with build.

After not too much longer we have a finished leg! All the servos turn freely, so that's a plus, and we haven;t damaged any more during the completion of the leg, which weighs in at 5.86 ounces. Therefore six of them will weigh a grand total of two pounds and 3.16 ounces, give or take. This is one hefty beast - beetleweight-wise anyway!

We still haven't reached a decision on whether to disassemble the plastic kit - we're going to sleep on it and see how much time we have tomorrow after we finish the drumbot and fabricate the weapon for the walker. It may be that we wait until Monday anyhow, and if they don't show up Monday, use the day to test the wireless control on the plastic kit, then salvage the metal gear servos from the plastic kit and continue assembly of the combat Hexy Monday night and Tuesday, then reassemble the plastic kit in the evenings while on the road to Harrisburg, PA, later next week. That sounds like the plan - at least for tonight :-)

Almost a third of the intended Motorama 2013 fleet is yet to be completed, and we jump in the BotMobile in five days! Fortunately it's not as grim as it sounds - both the featherweight and hobbyweight aren't actually that far from completion, and the beetle walker is essentially a kit, so fingers crossed that one goes according to plan. Today we'll focus on the featherweight Intrusive Interloper 3.0. Hit Back after viewing an image.

First: to the paint shack!
Having picked up a fresh can of matte black, the leg parts fo the beetleweight walker get a healthy dose of spray paint, and are left to dry while we turn our attention to the other end of the weight class spectrum.

We have one drive pod that refuses to run in reverse, so our first task for today is to replace it with the spare. We cut some axle shaft, asembled the gear, wheel, and core, and were ready to wire the pod to match the one being removed.

Here we're in the midst of replacing the defective pod, and it's not as plug'n'play as we were originally intending. Given that this 'bot is about six pounds underweight, the next iteration will have both weight and space for better modularity. Maybe we can come up with a way to use PowerPoles to click a drive pod in for power, rather than have to mess with these ring terminal stacks.

But success!
Note the motion blur on the wheels and motors - this pic was taken with the transmitter stick at full throttle; which makes quite a racket too! We didn't notice initially, but the back left wheel is actually rubbing on the cardboard box edge that the 'bot is sitting on - oops!

To help clean up, we pulled out the Dyson mini-vac and got all the shavings out. We also had a drive pod that siezed up at this point - the front right one - turned out the set screw in the collar on the brishless motor had undone itself sufficiently to get embedded in the polycarbonate - oops! Another modification we're going to work into the next iteration is a thin polycarbonate wall seperating the wheel from the gears - this will help keep debris and detritus from the floor getting into the gears.

But all in all it was a success - we could actually run straight, and turn on command. Trushfully, we're not sure how quickly we're going to wear through the 45A durometer wheels, but we have plenty of them, and driving with the left/right on the left stick is weird, but that's what driving practice is for.

With some time left yet this evening we turn our attention to the hobbyweight. This one, too, has a brushless drive train, but the idea was that the motors themselves would be the wheels. One of the motors was damaged at Franklin and we never got around to ordering a spare. Instead we mounted one of the slightly smaller ones we ordered erroneously to go in the wheels. Here we're in the midst of laying out the electronics.

The trick is trying to cram this stuff in in such a way that it makes sense and is easy to maintain. With the smaller LiPo battery versus the LiFePO4 ones we used at Franklin we have more options, but the optimal configuration is eluding us.

Finally we have our final configuration and have started wiring the 'bot up, however time has escaped us, and it's time to wrap up for the evening. Tomorrow morning we'll finish wiring the 'bot, and with any luck get it driving!

Tonight's efforts were predominantly directed at the twelve pound drumbot Nihilistic Naysayer 2.5, which sustained damage back at the Franklin Cup at the hands of Fiasco, which damaged a drive wheel. Our goal tonight is to swap the dead wheel and make sure the speed controllers work, now that we've got the hang of programming and calibrating them. Hit Back after viewing an image.

Before we get started on the hobbyweight, we had a quick test-fit of some parts that came in from McMaster today. This is the weapon 'hinge' for the hammer that will go on top of the beetle walker. We still haven't quite figured out where the servo will go that will power this setup, but one thing at a time.

The first thing we did was extract the damaged drive motor. You can see here that the can took a hit, which crunched a magnet. Even after cleaning out pieces of broken magnet and roughly tweaking the can back into shape, the motor was jammed, so it went in the Pile o' Used Parts.

Given that the motor had locked up, we were concerned about the speed controllers. A while back we had tried programming them without the card, and it had not gone well. Now we had the card we decided to try again. The first speed controller wouldn't calibrate though, and despite numerous attempts, we assumed that half of the FETs had blown up, as it would work in one direction, but not the other. Another piece of kit is relegated to the PoUP.

After successfully calibrating and programming the second speed controller we had a brain wave! These speed controllers can be programmed to only run in one direction, although obviously we don't want that, so maybe we had inadvertently set that last time we attempted to program them with the transmitter? We grabbed it out of the PoUP and hooked the card to it. Sure enough, that's what happened. After setting the parameters and re-programming it was working fine - hurrah!

These were the replacement motors we had ordered, but had managed to mess up - the diameter is larger than we had wanted, and they won't fit inside the nifty FingerTech wheels like we wanted. We debated ordering different motors, but the likelihood of finding one that would reach us in time was slim, so eventually we decided to go with what we have.

We used the foam tape trick again to build up a tire around the motor. It didn't hold up to Fiasco last time around, but what are the odds we'll fight that 'bot again? 1 in 16 we guess, but last time the 'bot was essentially immobile - this time it will drive! If we can keep the weapon pointed at the opponent, that means the wheels are out of the way.

A fresh layer of grip-tape was applied to both drive motor/wheel combos, and we turned our attention to the insides. Having yanked the drive speed controllers to test and program them, we decided to stop for a moment and reconsider placement of the internals. For Franklin we had to use Lithium Iron Phosphate batteries, which are considerably bulkier than LiPo ones, but there's no such restriction at Motorama, and we're free to use a smaller, lighter, Lithium Polymer one.

Here's where we wrapped up for the night, with a fresh layout in mind, incorporating the 4S LiPo pack, and a properly mounted power switch. Tomorrow we'll reconnect all the wiring, and see if our freshly programmed speed controllers can make the 'bot move. The brushless setup we (almost) have working in the 30lb'er uses the same speed controllers mated to four G25 motors theoretically rated to 600W, albeit with some gearing. The two motors in this 'bot are rated to 800W, but there's no gearing, as the motor is the wheel - fingers crossed it works tomorrow!

After all the problems we had trying to get the pistol radio working with these speed controllers in Intrusive Interloper 3.0 we're going to take a swing at getting it running with the Spektrum radio instead. If that doesn't work, then we're at a loss as to where to go from there ... Hit Back after viewing an image.

But before we get to troubleshooting, we need to sand off the waterjetting tabs on the parts for the beetleweight walker. Now that we're dealing with them as pieces, not just a sheet of aluminium, the realization kicks in: there are a lot of pieces!

And with the last of the sun, we line all the pieces up for some paint - the blue team on the left will get primer and purple, the rest will get primer and black - the metal hexapod will have the same colour scheme as our acrylic one.

While the primer dries we turn our attention to the featherweight's drive pods. Here we've switched from the pistol radio and receiver to the Spektrum setup, and after messing with end points and throttle curves we have this pod running semi-reliably.

With a close-up you can see why this pod runs so roughly - look at the remainder of the teeth on the pinion gear now that we've had a chance to run this pod in for about five minutes - they're grinding away. But on the plus side, at least it's running now. Guess we were slightly off with our motor mounting holes on this pod - drat.

The next two drive pods have zero issues, however, and with a few minutes of breaking in each, they run smoothly, so we install them into the 'bot. The fourth pod was mostly okay, but again had a couple of moments where it took a bit to get it started, so this pod could be an issue. We're still waiting for UPS to deliver our McMaster order so we can cut an axle for the fifth drive pod and try it out.

With four mostly-functioning drive pods in place we turn our attention to wiring. We're going to cram four 2S 2100mAh LiPo batteries in there, as a 2S2P arrangement for the voltage (14.8V) and longevity.

We've also switched the power switch to the beefier one from the previous version of this 'bot, and have started wiring up the speed controllers. Plenty of ring terminals here! But it makes wiring pretty simple. Hook all the reds together and run a lead to the power switch, then add a couple of PowerPole connectors from the switch for the batteries.

Next hook up all the negative leads from the ESCs to a pair of PowerPoles to make the connection to the batteries - nice and easy, which we like!

Lastly we just have to tuck all the radio equipment (receiver, mixer, Y-cables, and PWM leads) into the space that's left between the switch and the PowerPoles. In theory we could ditch the mixer and use the mixing capability with the Spektrum radio, but in order to ensure that when we want to the 'bot to run straight it goes straight, we put the forward/backward on the elevon axis of the right stick, and put the will use the rudder axis on the left stick for left/right. This is the first time we've ever done this, so it will take some getting used to. With the wiring done it's time to take the 'bot for a test run!

Hmm ... not perfect, if we're being truthful. We purposely weren't gunning it, because the Build Space is a bit too messy for a 30lb'er to be bouncing around in it, but it seems that we have 2.5 working drive pods ... we're going to have to put the fifth one together and replace the pod that barely functions, and investigate the dodgy one that works some of the time, but otherwise hey: it moves!

Now that we've actually made a start on the beetleweight walker, we need to seriously consider a weapon for it. The plan all along has been an overhead hammer, so we need to work out how to build one and affix it to the body. We've not made a hammer before, so it's been a challenge trying to figure out the best way to put one together, and here's our initial design for it. We wanted it to be able to rotate, to be able to attack on the sides of the 'bot, but that was a bit too much complexity for the first time out. Hit Back after viewing an image.

To keep things simple, we'll start with the top plate of the body. From there we'll put eight one inch long #4-40 threaded stand-offs in, to give us some space from the mess of cabling that sticks up from the top of the controller board.

On top of that is a simple plate to act as the platform for the hammer. This is 1/8"-thick 6061 aluminium, just because we happen to have some (we think) and we'll cut and drill it ourselves, versus getting it waterjetted.

On to the platform we'll mount a couple of bearings from McMaster, good for a half-inch shaft - we can use a piece of the keyed shaft that's already on order for the fifth featherweight drive axle. We also have a couple of shaft collars in there to lock the weapon shaft into place.

There's the shaft, and there's also a 14-tooth #25 sprocket, which will be hooked to the largest servo we ordered a week or two ago, which will be doing the actuation. We're going to write a simple set of moves in the PoMoCo script for Hexy for up, down-front, and down-back.

Couple of things here - firstly the platform has been trimmed, because we don't need those corners sticking out. Secondly we have a piece of 1" 6061 bar, which has a half-inch hole drilled through it to sit on the weapon shaft, and will have the rear rounded off, and a couple of set screws put in to fix it to the axle. In the front of the block there's a 1/2"-13 threaded hole drilled and tapped, into which is the long dark grey cylinder, representing a length of grade 8 threaded rod.

Here's the business end - another drilled and tapped 6061 block with a length of grade 8 threaded rod in it, only this time the two ends have been ground to points. It's not going to hold up super-well, but at least it'll be cheap and quick to replace if it gets blunted. The actual lengths of the rods will have to be determined once the body has been put together, to ensure sufficient clearance. These weapon components weigh close to two pounds, so with the aluminium body we weighed last night, we've used four and three-quarters of our six pound allotment, with battery and electronics left to go. There may even be weight left over for armour ...

Time is ticking away before Motorama 2013, and we still have some unfinished 'bots to deal with. There's a 12lb drumbot with a dead drive train, a fairyweight that needs some wiring sorted out, a featherweight needing assembly, and a walker that's still very much in kit form. Tonight we're going to start with the 30lb'er and see where we get to. Hit Back after viewing an image.

First job of the evening is to finish wiring the drive pods for the 30lb'er. We've already tested them to ensure proper orientation, so we have a pair of pods that are wired the same for proper forward rotation on one side, and another pair with swapped leads for the other side. We've marked the drive pods to indicate which is which, so we can quickly wire the spare pod to match as needed.

While we had the pods out, we hooked up the programming card and punched in some values we hoped would work. We've upped the reverse power to 100%, and set it to Low timing (for a low rpm/V motor), and also widened the deadband. It doesn't take long at all to set the values we want with this thing on all four pods - well worth the four dollars to avoid having to try it with the radio and beep codes.

Here we have all four drive pods sitting in place. Before we put over a couple of dozen bolts into this to hold down the pods we realized we needed to calibrate the throttle on the speed controllers, and the calibration button would be under the controller on the baseplate for two of the ESCs, so lucky we didn't bolt them all down yet! Here we had issues though. No matter what we tried, we couldn't get the ESCs to recognize the minimum throttle level (i.e. full reverse). We tried with and without the mixer, and also tried using different channels, but we just couldn't get the beeps. This is a problem, because the ESC needs full reverse to get the 'bot to run backwards. After a couple of frustrating hours, we posted online for some assistance, and moved on to something else.

That something else was punching the waterjetted parts out of the 24"x12" piece of quarter-inch thick 6061 aluminium. We have body parts, seven complete legs, and also a spare foot for good measure.

Here's the remnant the parts were punched out from - this may very well have to go up on the Build Space wall - it's not only pretty, but it's shiny too. We could probably have crammed a couple more parts on there as spares, but hopefully we have sufficient.

This is where we are with a full complement of legs and body pieces: 2lb 8.74oz. Factor in 18 x 9 grams = 162 grams or 0.36 pounds or 5.7 ounces worth of servos, and we still have half our weight allowance left for armour, electronics, a battery, and a weapon. Not bad - assuming it moves!

Now we have no doubt this 'bot will do well - the 'Bot Gods have their blood sacrifice! Actually, it looks worse than it really is, but it has to be done at least once per 'bot, it seems, and this one can be crossed off that particular list ...

On an initial check, we're going to have to clean up the parts somewhat with a file - you can see the leg servos don't quite fit in the thigh piece. We took out a file, and worked on squaring the corners inside the piece.

Success! Truthfully it didn't take very long, and wasn't that much effort, so we ran a file over all the other thigh pieces at the same time, along with one of the hip pieces where the third servo on the leg sits. We're also going to have to run all the parts on the belt sander to eliminate the waterjet tabs, but it's too late tonight to do that - we'll save that for another evening.

Moving down to our smallest unfinished 'bot, and here we have Lauren's fairyweight Tenacious Tinkerbell out for a test drive. We need to tidy up that wiring, but the 'bot is essentially done, and drives fairly well.

We crammed most of the wiring into various nooks and crannies in the 'bot, then added sufficient electrical tape to keep it all in place. Came out pretty well in the end. Not bad for a cheap and simple build. If we had to do it over, we'd probably swap the 10:1 motors for our favourite 30:1 motors, but we already had the 10:1s sitting around doing nothing, so there you go.

And a beauty shot. The CCD on our camera loves distorting colours - the 'bot is really more of a hot pink than a Ferrari red, but oh well.

Just for fun (and seeing as how they're all done except for stickers) here's Lauren's portion of the Team Radicus fleet for Motorama 2013. On the bottom, the twelve pound hobbyweight Steel Stiletto, in the middle the one pound antweight Malicious Mule, and on top the new fairyweight. That ought to keep her busy :-)

Tomorrow hopefully we can figure out the darned reverse on the 30lb'er speed controllers - we'll try using a different radio and receiver, to see if that makes a difference. We'll also run the walker parts across the belt sander and file them all to make sure the tabs fit the slots, and then we still have some work to put in on the hobbyweight to get the drive back up and running. Stay tuned - same 'bot time - same 'bot channel!

After running a few errands, including a trip to the hardware store for some more #6-32 taps, it's time to get to work - specifically on the drive train for Intrusive Interloper 3.0, and whatever other tasks we want to cross off the lists. Hit Back after viewing an image.

We picked up yesterday's mail, and in it was a box from Amazon with our monthly Subscribe 'n' Save coffee, plus these two items. A 16GB micro-SD card and a very small camera to put it in. This video camera will be mounted to the Sportsman for some action-cam footage at Motorama.

Skip forward a ways, and we're done tapping the FingerTech wheels without any more broken taps. Aside from the one drive pod we still need an axle for, we've bolted the gears to the wheel hubs, and installed the wheels to the axles with keystock and the nylon cores. Next up are the speed controllers, after a brief interlude for some test driving.

A quick shot of the 'victim' Steel Stiletto was sparring with - a 30lb box of junk. Let it be known to all 12lb builders: cardboard would not be a good armour material :-)

Here we've started to install the speed controllers. They're stuck to the side wall with foam tape, and lead 'C' will be connected to the yellow motor wire in all cases. Red and black will be connected depending on which side of the 'bot the pod is installed in.

We happened to have the table saw out to make a couple of supports for a cat tree (we have hefty cats) so we took the opportunity to cut a second piece of aluminium angle down to size for a second wedge for the featherweight. It still needs to be drilled, but that shouldn't take too long.

Here we have all five speed controllers mounted in drive pods, so the next step is to break out the transmitter, receiver, and programming card, and get these things configured.

Here we're in the midst of configuring the speed controllers. We have two per side, but we're missing a V-tail mixer, unfortunately, so we're going to have to hit a local hobby shop tomorrow and pick a couple up before we can really get this 'bot wrapped up. Time flies when you're having fun!

We ran some errands this morning, giving us a later than normal start in the Build Space, but the enthusiasm is high, after finishing the Sportsman yesterday, and we're hoping to keep that going as we finish the featherweight Intrusive Interloper 3.0. We have pretty much everything we need, so let's get to it! Hit Back after viewing an image.

First up, we received a McMaster delivery yesterday, and in it was a bottle of Loctite retaining compound 609. We applied some to the one pinion gear we over-bored slightly, and set the drive pod to one side to let it cure.

Next, we need some axles, so we've broken out the hacksaw and some 1/2" keyed shaft. A bit of elbow grease later and we have four serviceable shafts - we need five, but we're not going to hack up a Team Delta DeWalt shaft just to get a spare - we'll order some more keyed shaft from McMaster for next week. For now, four is enough.

While we have the hacksaw out, we chop up some 1.25" diameter nylon round to act as cores for our fancy FingerTech wheels. Naturally, these are easier to chop up than the steel shaft :-)

Next is a trip to the mini-lathe, as the nylon is oversized, and we need 1.25" proper to fit the wheels, so we throw in a cutting tool, and skim off the excess.

Finally we need to bore the nylon to fit on the axle, but we've hit a size limit on the mini-lathe - we can't fit the 1/2" drill bit in the chuck with these nylon cores, and we don't have a shorter drill bit ...

... But we do have a shorter 1/2" end mill, so we chuck that up and set to it. In minutes we have a set of serviceable wheels cores - hurrah!

They're a bit sloppy, but they work, and we're happy with them. Next we need to look at the other side of the wheel, and figure out what to do with the drive gear. We broke the 1/8" broach a while back, so keying the gears is not currently an option.

So instead we elect to drill and screw the gears to the wheels' aluminium hubs. After drilling the first hole, we put a spare 7/64" drill bit in there to keep the gear in place and drill the second hole. This seems to work just fine.

The wheel hub is tapped for #6-32 screws and the holes in the gear are widened to 9/64" and countersunk. With the addition of a pair of 0.75"-long #6-32 screws we have the gear mated nicely to the wheel.

And here's one installed, with the nylon core inside the wheel, the wheel fixed to the axle with some keystock, and the gear pinned to the wheel and mated to the pinion gear. The whole thing spins nicely by hand, so once we have the rest done, we need to hook up the speed controllers, and apply some juice!

Aaaaand the night comes to a screeching halt as we break a tap in the second wheel - drat! Despite scouring the Build Space thoroughly, it seems we're out of #6-32 taps, so our endevours come to an end for the evening, as all the hardware stores are closed at this point. Grrr ... we were really hoping to get the drive pods installed tonight! Oh well, tomorrow will start with a trip to the hardware store, and then we'll pick up where we're leaving off. Same 'Bot Time. Same 'Bot Channel!

We're reallllly close to having the Sportsman finished, so our goal this evening is to finish the 'bot and cross it off the To Do list. We only have a few things to wrap up, so fortified by a home-made pizza, it's time to hit the Build Space! Hit Back after viewing an image.

Issue number one is that we ended up with two different lengths of chain for the weapon transmission, which means something went awry here. After looking at it, we think we installed one of the weapon motor mounts upside down - they're not symmetrical.

So over to the drill press, and we put a couple of new holes in the mount after carefully taping up the motor so we don't get polycarbonate chips in it.We'll spare you the tapping photo :-)

With the mount reinstalled we took a moment to test the spin direction of the motor, and labelled the extension wires accordingly. We want to make sure the saw blades spin in the right direction after all.

We also took the bolts out of the polycarb mounts for the second and fourth mounting points, allowing them to essentially act as spacers, and that definitely helped eliminate some of the friction in the setup. As you can see in this photo we have grease on the the front panel and DeWalt mount, indicating it spun up, and we have [loud] video to prove it:

Success! There's still a little hesitation in the initial spin-up, but not enough to worry about - one half of the weapon is running finally, so it's on to the other side. We removed some of the bolts there too, and verified that both sides spun up at the flick of a switch on the transmitter. Hurrah!

And there we have it: a little over a year after beginning this 'bot, our first ever Sportsman class 'bot is complete: Palindrome30. This is going to be a really fun 'bot to play with, we can already tell. The weapon arrays may not be particularly effective when all is said and done, but it ought to be pretty durable, and hopefully we can keep it running long enough to win at least one match with it.

As you can see, weight is not at issue, at 26.2lb, but you'll also notice the 'bot sits nicely on it's side. This could be an issue, so with the remaining weight we're thinking we should cut some aluminium angle and bolt it to the sides to prevent it from sitting on the edge.

We also have a mild drive train problem to deal with - the keystock we put in is too small for the driven axle and the sprocket falls off. To fix this we need to disassemble the drive train and add a larger length of keystock. We took the side off, but notice the issue? The Nutstrip stops us from removing the wheels, so that had to come off too.

Here we're in the midst of reinstalling the sprockets, and while we're at it we added a couple of roll pins too, to ensure things stay nice and snug. So there we go, everything's been crossed off the To Do List for this 'bot, and 60% of the fleet is ready for Motorama 2013!

We still have a couple of hours left this evening, so it's time to change tack - from the biggest 'bot to the smallest, and we're working on Lauren's fairyweight Tenacious Tinkerbelle. We started with a weight check, and scared ourselves when the scale read 153 grams - before discovering there was an extra battery pack inside the chassis! With that put aside the 'bot weighs in at 140 grams out of the 150 allowed, so no worries there.

Tonight's efforts are mainly in the soldering arena - never our favourite occupation - so here we have the materials: the drive motors, which need extension leads added, and connectors for the batteries to make them compatible with our chargers.

Soldering the extension leads actually wasn't too painful. We did take the time to 'tin' all the ends, and that definitely helped when it was time to actually stick them together. We even went for some shrink wrap tubing to hide the evidence of the ugly soldering :-)

With all the extention leads done, we drilled the wheel hubs to 7/64" which is a hair under three millimeters, and pressed the wheels onto the motor shafts, then installed the motors back onto the base plate.

Whoo-hoo! UPS just came and dropped off a heavy, flat package. We quickly unwrapped it, and found this: the 1/4" 6061 aluminium plate with all the parts for the combat version of the hexapod walker - in order to save time and be able to ship this to us for the weekend, we asked Jake over at Westar Manufacturing (Team Whyachi) to leave everything unsnipped and unsanded - we can do that ourselves, and the resulting piece of waterjetting looks fantastic!

Anyway - back to the fairyweight. Here the receiver has been plopped into place on a piece of double-sided foam tape. We need to corral all the excess wire, and hook the speed controllers to the wheels and the battery, preferably incorporating the demo FingerTech power switch as we go.

Sorry for the blur - we were moving fast :-) This was supposed to be a shot of the new connector being installed to the battery so we could hook it up with the connectors we had lying around, rather than making a special purchase. We have two batteries to modify, and it doesn't take too long.

Here's a close-up of the FingerTech power switch with leads soldered to the tabs. It's actuated with a 3/32" hex key - clockwise for on and counter-clockwise for off, which is the reverse of the Whyachi power switches we're accustomed to, so we'll need to make a mental note of that.

Mid-way through the soldering, and things are going well, but there is a awful lot of wire to cram under the lid in this 'bot - we may have to add a few micro-sized zip ties to help keep things organized.

We're almost done - we have one side of the drive train soldered up to the ESCs, and the ESCs soldered to the wiring harness to connect to the battery. A few more connections for the other side of the drive train, and we'll be about done.

Ta-daa! Everything's been soldered together. We do need to drill a couple of holes in the back wall and mount the power switch properly, because the foam tape isn't holding up to screwing the switch on and off. We fired up the transmitter and calibrated the speed controllers, so that was a pretty good indication that things are soldered together properly.

A final piece of soldering for the night is in making a charging cable for the 'bot's battery. These batteries won't be balance-charged, but then again they're under five dollars apiece, so essentially they're considered disposable - if they can't take being charged regularly, we'll replace them.

Finally it's time to wrap up for the night. As you can see from the To Do Lists we've made quite a lot of progress - Lauren's fairyweight is ready for a test drive, we have a couple of tweaks to make to the other insect-'bots, a drive train to repair on the drumbot, a drive train to build for the featherweight, and a walker to assemble in the beetleweight category. Sounds like a lot, but we still have two weekends to go, so this ought to be eminently doable!

We got a late start tonight, but really wanted to make more progress on the Sportsman Palindrome30 as it's the next 'bot closest to being finished, and one that we have everything we need for. Hit Back after viewing an image.

There are essentially three things to do: (1) Finish wiring the weapon speed controls; (2) Fix the slipped keystock in the drive wheels; and (3) Eliminate the friction from the weapon axles. We started with the wiring, and with a prodigious number of ring terminals, hooked up all the weapon ESC-to-motor connections. They tested fine, despite the weapon axles not actually spinning.

As it's getting late, we don't want to start disassembling the weaponry and drive train right now, but to all intents and purposes this 'bot is about ready! The wiring fits nicely in the chassis, which is something that hardly ever happens first time out.

A parting shot of our To Do Lists. Walking beetleweight aside, there's not a whole lot to do as we come down to the penultimate weekend before Motorama 2013. There's some troubleshooting on the Sportsman, drive train assembly on the featherweight and 12lb drumbot, soldering for Lauren's fairyweight, and a little troubleshooting on a couple of the small 'bots. The plan for this weekend is to have all the 'bots except the walker finished, leaving all next week and weekend to work on that one, as it's the most complicated.

We knew the moment we woke up this morning that we were finally going to receive our fancy new parts from FingerTech Robotics (despite obsessive refreshing of the USPS tracking page not showing any updates) and we wanted to finish another 'bot tonight. Things had stalled a little in our timeline, although things were getting crossed off the To Do Lists. To the Build Space! Hit Back after viewing an image.

Ta-Daa!
Three TinyESCs (two for Lauren's fairyweight and a spare), two prototype power switches (one for the same fairyweight and one for the beetle Belligerent Battler 1.1), and ten - count 'em - ten custom urethane wheels (two for the hobbyweight drumbot, five for the featherweight brick, and three spares). They're Shore A45 hardness, which is actually softer than Colson wheels, and in theory a bit grippier, so we'll see how they perform. Thanks to Kurtis over at FingerTech for getting these to us!

Here's a test-fit of one of the wheels. They're a larger version of sumo wheels, and as such they're open on one end. It took a while but we figured out what we were going to do: turn the wheel around so the closed end meets the hub of the gear; drill both the gear and wheel hub for four #6-32 bolts to couple them together; order some 1.25" OD plastic from McMaster, and when it gets here, cut some 1" lengths, bore them on the lathe, and insert them into the aluminium hubs. We don't see what can possibly go wrong :-)

Despite our enthusiasm, we're going to have to wait on further featherweight progress, it seems, until the McMaster order shows up on Friday. So let's turn to the Sportsman Palindrome30 instead. The last report we put up on this 'bot received nineteen Facebook Likes as of writing this report - we have no idea who all those people are, but thanks! First up today, we drilled and tapped the weapon motor mounts.

And they came out quite well - we did manage to bore one mounting hole while running a drill bit part-way down to recess the bolt heads, as the part 'jumped' up. Still, pretty good show.

We have a piece of 1/2" outer diameter aluminium round pressed on to one motor shaft, but when we attempted to press the sprocket onto the round, all we achieved was to cause the motor to lock up. After unjamming the motor, we decided to turn down the round a little, so we hooked up the weapon power, turned on the radio and receiver, and flicked the gear switch to activate the weapon motor.

After applying a file to the running weapon motor for a little bit, we were able to push the sprocket on successfully. The other motor had already been successfully installed with round and sprocket.

Although it's not recommended, we're going to be using set screws and a little LocTite blue to secure the sprockets on the aluminium rounds. At this point, we're ready to install the weapon motors in the 'bot and think about coupling them to the weapon axles.

In order to power the weapon motors we're going to need a 4S LiPo battery and some speed controllers. We're using cheap HobbyKing 60A ESCs we had laying around, and a Turnigy battery pack. Remember the goal is to not have to buy anything special for this 'bot.

Zac O'Donnell gave us a great idea for getting the weapon blades pinned - genius in its simplicity: Take a piece of 1/4" keystock, drill a short hole in the end big enough for the roll pin to sit in, then hammer the keystock to seat the pin, because we can't get the hammer close enough to the pin itself. Thanks Zac!

With all the weapon blades pinned, we moved on to the transmission, which is #35 chain. Why is it chain always needs an offset link?! It's been a while since we used a chain breaker, and in the end went back to how we cut #25 chain - Dremel off the top plate - because the chain breaker was leaving us with warped top plates on the next link.

Et voila!
The weapon transmissions are in place, so we now need to move to our least favourite part of 'bot building: wiring. We started by attempting to solder bullet connectors onto the two weapon speed controllers. It did not go well. Despite our best efforts: tinning wires, using the helping hands gadget, etc., we ended up frustrated.

So out come the ring terminals, of course. Here we've installed ring terminals on the two speed controllers, and our next move is to do the same with the weapon motors. We'll need some extension wiring between the motors and ESCs to let us put the ESCs where we want them, but that's not a big deal.

The ring terminals are on the weapon motors, and we begin to wire them up to the power switch. We decided at this point it might be worth testing to make sure we're getting this right. The weapon ESCs don't have a battery eliminator circuit (BEC) in them, so in order to test, we'll need to turn on the drive power too, which has a dedicated BEC on it.

Drat. With some temporary wires in there and the power on, there's too much friction for the weapon axle to spin up. We tried both front and back axles, and they both have the same issue - no doubt because we're using aluminium round in bronze bushings, and the round may be slightly oversized. It's getting late, but we're going to have to do two things when next we build: (1) Finsih the weapon ESC wiring; and (2) disassemble the weapon axles so we can sand down the aluminium axles where they sit in the bushings. The other thought that we had, which may eliminate the issue, but adds time to the equation is to replace the aluminium axles with steel, which are more likely to not be oversized. For now, though, we're going to try the sanding route, and see where we end up. Hopefully with one more productive build session like this evening was, this 'bot will be done!

Now that we have a mess of metal geared servos, it's time to perform a servo transplant on the Hexy Kit, and upgrade him. This will, by all accounts, mean replacing servo horns, and then we need to test whether these servos have the torque to move the 'bot if we double its weight. Hit Back after viewing an image.

The starting point - a dejected-looking hexapod with a few bad servos. We need to pretty much totally disassemble the kit for this procedure, so we get cracking with the screwdriver.

The kit has been stripped down to its components and we're ready to de-bag eighteen of the new servos to swap in. We've kept the old servos, and if the new ones work, we'll use them in the combat version and put the old servos back in the plastic kit (with repaired gears as necessary) so both walkers will [hopefully] be at Motorama - the combat one to compete, and the plastic one to fill time if needed.

Part-way through, and we have both thigh servos in each of the six legs, without incident. We're using a pin vise with a 1/16" drill bit to pre-drill screw holes in the servo horns, to help prevent splitting. As each servo went in, we centered it using the Servotor32 board, and we have noticed that these servos and horns seem to want to sit centered better than the previous ones.

A while later, and we also have the hip servos mounted. So far the reassembly process has taken about three hours - and we still have a ways to go - this is going to be a late night!

"The thigh bones connected to the hip bone"
The legs are now assembled, but there was an iatrogenic issue - in the top-left you'll see a leg that's been splinted with a clamp, while the superglue sets - we accidentally snapped a femur, to continue to medical operation analogy. It's the same piece we snapped during the initial assembly so we've already used up the spare.

In order to get the new servo horns to sit properly on the baseplate, we decided to drill out the holes to 19/64". Time for the second oops moment of the night - the bit caught on the acrylic, and well, you can see the broken spar in the bottom-left. We think we can salvage it though ...

Using the pin vice we drilled a new bolt hole, and while attempting to drill a second, we managed to drill straight into a finger - ouch! With a bolt, and some electrical tape, though, the servo horn has been mounted. Phew. Time to sew the patient up.

Et Voila! Better than new :-)
We've connected up all the servos, upgraded the Servotor32 firmware to version 2.0, and slapped on a battery pack. Let's see what happens when the patient wakes up:

That's pretty smooth. Note that the kit doesn't move quite as fast as it used to - that's not because of the servos, but because the rate of movement was toned down in the latest version of the software. Still, it does seem more solid than before - after a few minutes of moves not one servo was showing any signs of breakage. Unfortunately it had gotten pretty late, and it's a school night, so our testing of the kit with additional weight on it will have to wait until tomorrow.

Having had fun with our new lathe yesterday, we decided to put some effort into our other big 'bot - the Sportsman Palindrome30. We need to get the drive train wrapped up, which means installed and chained up. Then there's the weapon and wiring. After fixing up some work issues, we finally headed out to the Build Space. Hit Back after viewing an image.

In order to get the drive train into place, we need to mount some home-made bearing blocks. Here we've drilled the mounting holes, and also 'countersunk' them with a larger drill bit, as we don't have a countersink with a 3/8" shaft, and the 'bot wouldn't fit on the drill press as is.

As it happened though, we had to take the chassis apart, so yes, we could have countersunk the side walls, but moving on - here we've verified that the total width of the axles and bearings is the requisite 2.75". Time to mount the bearings!

They may be ugly, but they work! It didn't take too long to mount all four outer bearings and Dremel-ish down the bolts. So then the next step is securing the drive sprockets to the axles.

For this we cut four very small pieces of 1/8" keystock with the Dremel-ish, and finagled the tiny pieces into place on the shafts. Everything is looking pretty good at this point, so we moved on to chaining the sprockets together.

Rummaging around, we found some random length of #25 chain, and strung a piece across the first side of the drive train, marked where the connecting link should go, and then ground off the link with the Dremel-ish. A bit more rummaging around turned up a couple of #25 master links, and one side was ready to roll.

Here we are with the first side installed. Next we need to mount the inner bearing on the inner rail before we can get too much further, so it's over to the drill press, which didn't take very long at all. We're getting excited - we might actually be able to get this 'bot moving today!

Another random length of chain being cut to size. The amount of junk we have lying around is pretty astonishing - aside from the black UHMW, we actually haven't bought anything specifically for this 'bot - it's all stuff that we have laying around the build space. Granted, we only have more more Victor 883 speed controller left, but if push comes to shove, we can swap in a BotBitz ESCheap85 to run the DeWalt on - we're not messing with gyros in this 'bot, so the incompatibility won't be a problem.

Ta-daa! The drive train has been assembled. Next we want to add some wiring, batteries, and a receiver, and see if the wheels will spin. There are two power switches because the drive and weapons run at two different voltages - 6S (22.2V) and 4S (14.8V) respectively. We ought to be able to wire up the drive, and not have to make any wiring changes later when it's time to wire up the weapon.

Before wiring, we reassembled the chassis. The 'bot bears a tiny resemblance to its inspiration, but is distinctive enough to stand apart from it. We're happy with it thus far :-)

Time for a quick weight check - with all the parts and pieces piled on, the 'bot weighs a scan 25 pounds even. No worries about making weight here then - just like the featherweight we don't seem to be able to use up the entire allotment this time around.

Success! We didn't take any video, but after plugging a battery into the drive wiring harness the receiver was able to bind to the radio, and we made the wheels work! It took a little bit of channel reversing to get things turning in the proper direction, but after a few minutes of fiddling the 'bot drove under it's own power. Then we lost two wheels. Apparently we're going to need a bit more keystock in those sprockets to keep everything engaged - no matter, that's a pretty easy fix. All told, we're in a good mood, and we're looking forward to wiring up the weapon tomorrow!

We decided to capitalize on our productive spurt, and cut and crimped connectors on all the drive motors for the featherweight Intrusive Interloper 3.0, using the one drive pod we assembled the other night as a reference point.

Finally for tonight, we also put connectors on the remaining speed controllers, so there's one less thing to do when the wheels show up and we can do some final assembly on the drive pods! All in all, a pretty productive day :-)

There are a number of pieces we need to chuck up on a lathe in order to finish the remainder of the fleet, and we feel bad pestering Chris every time we need to make circular operations, so we checked out our lathe-buying options this week. Technically, Milly can lathe things for us, but to convert her from milling to lathing means a pretty big clean-up job, and we don't have the time for that. So we made a purchase ... Hit Back after viewing an image.

Given $100 off, plus a 20% off coupon, we knew what we were going to order, but we didn't know where we were going to put it, so purchase number one (with another 20% off coupon) was a work table. This photo doesn't show the 75 minutes we spent trying to assemble this darned thing, but once it was done, it was rock solid.

This was actually a decent deal price-wise: normally $599, on sale for $499, plus a coupon for 20% off, and the price before tax was $399. We bought the $80 2-year No- Questions-Asked Warranty, which gave us another $25 off, so for $375 we had a $600 mini lathe with some pretty decent reviews.

After taking the shipping bolts out the bottom of the package, we opened the top of the box, and found a few tools and parts. This seems to be a beefy package - 90lb - so hopefully this is a solid piece of kit.

After pulling out the top packaging we get to the meat of the sandwich: a 7" x 10" mini-lathe. There's still some heft to that box, so we're not risking our backs just yet - we want to make sure everything we payed the [discounted] price for is ready and available.

We've liberated handles, tools, outer chuck jaws, safety guards, rubber feet, and a spare fuse - sounds like we're about ready to roll. We did buy an MT2 drill chuck and a few pieces of miscellaneous tooling when we picked up the lathe, so maybe we can do something productive today.

We originally planned on mounting the lathe to the workbench, but didn't have long enough M6x1 bolts, so we decided to mount the rubber feet to the chip pan, and set that on the new work surface.

Turns out that works fine - the instruction manual suggested we break the lathe in - normally we're sceptical of such moves, but what the hey! We ran the lathe for 60 minutes at varying speeds in High, Low, Forward, Reverse, and every combination thereof. It worked well.

So - to productivity! Here we've just run a #11 drill bit through a pinion gear for the 30lb'er drive train. A #11 drill bit is nominally 0.1910", and our calipers tell us the inner diameter is now 0.1915" so that's some pretty tight performance! Unfortunately still not big enough to press fit on to the 5mm drive motor shafts, so we moved up to a #10 drill bit.

The #10 drill bit took us to 0.1940" inner diameter, and we needed closer to 0.1960" so we then moved up to a #9 drill bit, and that was just a hair too much - this gear could be pushed on to the pinion just by hand, so we're going to have to pick up some Loctite 609 retaining compound for this one, and decided to stick with boring the rest of the gears with the #10 bit, and see how they fair being pressed on to the shaft.

The operation went very well! Using a single washer on top of the gear gave us the right spacing when pressing the gear on to the shaft to align with the wheel gear, and we were extremely happy at this point. So happy, we decided to up the ante ...

... and worked on the wheel gears, enlarging the bore in 1/32" increments from 3/8" to 1/2".

The finaly pass with the 1/2" drill bit was tricky - it was riiiiight at the length we could push the tailstock back, but we persevered, and our patience was rewarded.

Here we have it: five wheel gears with 1/2" bored sitting nicely on the keyed shaft - the next step is to cut the keyed shaft to length, and then wait for the wheels to arrive from our friend at FingerTech Robotics. It's going to be a long wait, according to Canada Post tracking :-(

As you can see here, things line up very nicely. Despite the assault from the arbor press, the motor spins freely, and the teeth are well meshed between the gears. We can hardly wait to see these drive pods in operation! All in all, a good day's work :-)

We knew we weren't going to have a whole lot of time this evening, but we did want to get at least one speed controller installed in a drive pod for the 30lb'er, just to verify they'd fit. Plus we got a package in the mail that needed investigating ... Hit Back after viewing an image.

And here it is: an Arduino starter kit from Pololu, put together by SparkFun. All we had time for today was to verify the contents.

We have a number of resistors, presumably for the LEDs, a stretch guage, a potentiometer, a buzzer, a transistor, another USB cable, a small breadboard, a bunch of wires, and the Aduino Uno R3 board itself. We're looking forward to hooking that up and playing with it in the very near future, but on to drive trains!

First question: does the ESC physically fit inside the drive pod? Answer: yes - just. We knew it had capacitors hanging off it, and they fit nicely in the wire entry/exit hole. We will need to trim a bunch of wire off both the motor and the ESC, but at least they do indeed fit, and with just enough clearance too.

First task with these speed controllers is to ditch the on/off switch - these puppies do us no good switched off! We broke out the soldering iron, and put a small blob of solder across the switch pins on all five speed controllers. In hindsight, we should have pulled out the spare for Nihilistic Naysayer 2.5 and done that one too, but oh well.

Second task is securing the PWM cable and the fan power cable. A thin-ish smear of hot glue does the job nicely. Having prepped them, it's time to sort out the wiring.

The basic idea here is repeatability, but with a twist. We can't solder the wires together, because to reverse the motor for left/right drive pod placement of the spare we need to be able to flip two of the three connections. So we decided to go with ring terminals - surprise! The yellow motor cable is the shortest and will always be connected to lead C from the ESC. The red and black motor wires are left longer, and will be the pair swapped if necessary to reverse motor direction. Having established a reference model, when we next come back to work on the featherweight's drive train we can move a little quicker with the remaining four ESCs.

The first of our two Hobby King packages showed up today, featuring some electronics and motors - hurrah! The second package, with metal gear servos for the Hexy kit is looking like it might show up by Friday - fingers crossed! We also put in an order at Pololu for an Arduino Uno Starter Kit, as we will need something to take a PWM signal from the receiver on the walker and turn that into something the servo controller board can work with. We were also notified by our friend up in the frozen wastes of Canada that our custom wheels had shipped with speed controllers and power switches - Wheee! Hit Back after viewing an image.

First, half a dozen speed controllers. Five are destined for our featherweight Intrusive Interloper 3.0 to control the motors we mounted yesterday. The other is for the drumbot Nihilistic Naysayer 2.5, just in case we blew one up at Franklin trying to run a siezed wheel-motor. We still need to check whether it's operational or not.

The rest of the box contents are three replacement motors for the drumbot. We've cunningly decided that we're going to skip the urethane molding, as Kurtis over at FingerTech Robotics has done that for us, creating a 2.5" outer diameter 45A wheel, with an inner diameter of 1.25". Guess what the outer diameter of this motor is? Yup - a touch under 1.25", so we're going to mount this motor inside the wheel, and try running it directly, once we've programmed the speed controllers to run as hot as we can get them to.

We've also been thinking about the walker. The basic plan is going to be to install the metal gear servos into the plastic kit as soon as they show up - multiple people on the Arcbotics Hexy Forum have stated these are drop-in replacements. Then we're going to weigh down the kit with enough stuff to make it weigh six pounds, and then we're going to see if it can move. Truthfully, it may not, because these servos have the same torque as the plastic-geared ones, and we guestimated we would need twice as much, but we're going to try anyway. The reason is that if these servos can move the 'bot, we don't have to redesign any of the parts - we can simply have the existing design cut out of aluminium. If the servos can't handle it, we'll have to scramble and get the original servos we were planning on switching to, and redesign around them.

As far as controlling the walker goes, we've decided against trying proportional speed when making the 'bot move - it's going to be an all-or-nothing approach: depending on where the transmitter stick is located, the 'bot will perform one action continuously until the Arduino board receives PWM pulses indicating the transmitter stick is in a different location. For example:

The right stick will tell the 'bot either: move forward; move backward; turn to the left; or turn to the right - but only one operation at a time. The left stick controls the hammer using the rudder channel (left/right) swinging it either in front of the 'bot or behind the 'bot. The top of the throttle channel (up/down) is a special move, in that if the Arduino detects full throttle, it will ignore the right stick, and put the 'bot into "Defensive Mode" where the two front legs come together, to allow an impact, for example, to be received by the titanium plates on the front.

For kicks we might add the "Dance" move to one transmitter toggle switch, and the "Wave" move to the other toggle switch :-)

After researching on the Interwebz, there are basically two ways the Arduino board can read what a receiver is putting out: by using the PulseIn() method, or by using interrupts. The PulseIn() method is simpler to implement, but apparently wastes a lot of the compute time on the board. The interrupts are more complicated, but allow you to change the commands faster. Now, given that Hexy will only be doing one thing at a time (move forward, turn left, defend, etc.) we're thinking that PulseIn() ought to be fine. The basic command program will be something like:

Initialize the servo controllers
Initialize the servos (stand up)
Begin main routine
Loop {
   Check to see if Throttle channel pulse is > 2000µs {
      Position the servos for defensive posture
      Restart Loop
   }
   Check to see if Elevator channel is > 2000µs {
      Move the servos to move the 'bot forward a step
      Restart Loop
   }
   Check to see if Elevator channel is < 1000µs {
      Move the servos to move the 'bot backward a step
      Restart Loop
   }
   Check to see if Aileron channel is > 2000µs {
      Move the servos to turn the 'bot right a step
      Restart Loop
   }
   Check to see if Aileron channel is < 1000µs {
      Move the servos to turn the 'bot left a step
      Restart Loop
   }
}


Anyhow, that's enough rambling for now - not sure if we're building tonight, but stay tuned for more in the run-up to Motorama 2013!

We have a bonus build day today, as our day job closes to observe Martin Luther King day, so it's out to the Build Space to see if we can cross some more items off the To Do Lists. We're going to focus on the two biggest 'bots today - the featherweight Intrusive Interloper 3.0 and the Sportsman Palindrome30. We're a bit stuck on the drive train for the Sportsman, and pretty much out of parts for the featherweight, but we'll see what we can do! Hit Back after viewing an image.

Ah, the mighty To Do Lists ... they keep us on track, and provide much enjoyment when it comes time to cross something off. If you look closely, you'll see there's actually a couple of items for 'completed' 'bots, but we save small tasks like that for weeknight build sessions so we can maximize the daylight time for marking, drilling, etc. For some reason, it just seems to come out more accurate in sunlight ...

Speaking of accurate drilling - it's time to do some. Using our CAD model of the featherweight, and knowing the mounting holes for the drive motors are 25mm apart, we marked and drilling one of the drive pods and test-mounted the motor. Came out well.

Well enough that we went ahead and drilled and mounted all the drive motors. The all seem to be pretty darn accurate, so that's good. Now we wait for gears, wheels, and speed controllers.

Some more accurate drilling later and we can actually mount the drive motors for the Lauren's fairweight - they're a bit cramped, but they fit, so that's good. We have a battery and a receiver for this 'bot, and we just need a couple of speed controllers.

Turning our attention to the Sportsman, and one area we've been stumped is the drive train - never exactly our speciality by any means, but without a decent lathe (sorry Milly) or even a broach any more, we can't make new ones to fit these 6"x1.5" Colson wheels. Instead we decided to get creative. We started with the pre-made hubs used in the last dismal version of Steel Stiletto. They're not long enough to go through the entire bore, but they're a good start.

Then we broke out our parts bins, and started looking for things that would help fill up the bore in the wheels, yet accomodate a keyed 1/2" shaft. We were willing to take any combination of bearings, bushings, washers, and shaft collars that would work!

While we were fiddling around trying different combinations of the above, UPS dropped off a package - gears for the featherweight! And a bonus LED light - handy. In the foreground are the 12-tooth 32 pitch steel gears; they have a 3/16" (0.1875") bore, and the drive motors have a 5mm (0.1969") shaft, so we're going to need to figure out how to enlarge the bore to fit the motors. In the background we have the 54-tooth gears that drive the wheels, and their 3/8" bore will need to be enlarged to a half an inch at some point. When the wheels arrive, we'll decide whether to go get the gears broached for a keyway, or whether we want to mount the gears to the wheels themselves with bolts.

Back to the Sportsman, and we need to fix the saw blades to the shafts and cut holes in the front and rear walls for the weapon chain. To that end we removed the front wall from the 'bot to make it easier to work with. We never created a CAD model for this 'bot, just started winging it a while back, and made a few poor decisions along the way, which is why the weapon axle mounts overlap the bolts that secure the front wall to the inner rails - oops.

We broke out the Dremel-ish and ground flats into the aluminium axle to make drilling the pin holes easier. It was a bit cramped in the outer spaces, but still relatively trivial to do.

In a moment of inspiration, we figured out the mounting solution for the Colson wheels! They have a 1-3/16" bore diameter, and in addition to the bits of axle core we retrieved from the former hobbyweight, these 1/2" shaft collars have an outer diameter of 1-1/8" so with the addition of a circumference-worth of double-sided sticky foam tape, they pressed in nicely! We're going to pin them through the tire too, but we have a solution!

Ta-daa! With some nylon washers, we're able to run chain pretty smoothly on the sprocket, and the whole solution fits the two inches across that we have between the inner and outer rails. There's still pinning to be done, and we need to put a tiny piece of keystock in each of the sprockets.

But hey - we have a workable solution, and quickly have all four wheels mounted on makeshift hubs - they're also much less wobbly than the ones we previously had in Steel Stiletto, so that's a bonus! Mounting the tires had been a major sticking point, so we're really happy to have this problem licked :-)

Back to the weapon assembly, and here we've drilled the pin holes for the 1/8" roll pins, but then realized that the one inch long pins we have we going to be far too awkward to hammer in, as there's only an eighth-inch of clearance beside the saw blades. We're going to have to order some longer pins and then Dremel-ish them off after inserting them.

While we had the walls off, it was an opportune time to cut holes for the weapon chain. We started by drilling some 3/8" holes in the corners, then broke out the jigsaw to finish them off. Fairly quick and painless to do.

So here's where we're wrapping up for tonight - we need longer pins for the weapon array, then those will be done. We need to drill the side walls for the drive bearing blocks, and install the drive train. We need to drill and tap the weapon motor mounts, and install those. Then it's a matter of wiring the 'bot up, and taking it for a test drive. With any luck we'll finally have this 'bot done by next weekend!

There's not much of a plan for today - just a desire to check things off our To Do Lists. We're waiting for parts at this point, but we ought to be able to make some progress anyhow. Hit Back after viewing an image.

We're going to pick up from where we left off yesterday, working on the chassis for Lauren's new fairyweight. We've screwed down the straight edges of the top and bottom plates, and they fit okay, so that's 50% of them done.

And following a trip around the belt sander, all the edges are flush with the chassis rails. Without the benefit of a waterjet, this was the easiest way we would think of to get these panels right. At 1/16", they're too thin to cut accurately on the table saw.

Next we scored a number of lines into the top and bottom, including additional bolt holes, motor mount locations, and wheel hole dimensions. Then we went on to secure the additional bolt holes, drill the motor mount locations, and begin drilling out the wheel holes.

It dawned on us we could do the wheel holes on Milly - we decided to go manual, because it would be faster than writing a script, lining up the 'bot, etc.

We drilled and tapped the motor mounts, and next looked at scoring them for the motor bolts to fix the motors to the mounts.

Meanwhile, we bolted together the featherweight chassis that we painted yesterday, and as you can see, we're only out of alignment on two bolt holes, so that's pretty good.

We also removed all the drive pods, as we need to mark and drill them for mounting the drive motors. Hopefully our gears will show up tomorrow, and we can assemble the mechanical portion of the drive train. The speed controllers should show up later in the week.

The drive motors for the fairyweight have been mounted, but we're beginning to see a problem ...

Oh, those darn wires - there's not sufficient clearance between the motors for the tabs and wire, so we decide to rotate the motors on their mounts 90 degrees on one side, and cram them in.

A final shot for tonight of the featherweight drive motors with mounting bolts. It doesn't seem like we did much today, but we definitely made progress. Tomorrow we're planning on using our 'bonus' build day to mount the drive motors in the featherweight, finish the drive train on the fairyweight, and see if we can make some Sportsman progress too!

This time in four weeks we'll be in Harrisburg, Pennsylvania, fighting it up in the arena, but to get there from here we need to finish putting together the rest of the fleet. Granted we're fifty percent of the way the there, but we really do want to get the remainder of the 'bots ready. We have three weekends, a spare day (Martin Luther King Day) and plenty of evenings between then and now, and our To Do Lists are rapidly being crossed off. This weekend we want to get as much done on the featherweight as possible, so we're left with the drumbot, Sportsman, and beetle walker for the reaminder of the build time. Hit Back after viewing an image.

We start out today with a can of grey primer and some of the featherweight's walls. The back and sides get a coat, meanwhile we're working on the front wall, to be able to mount the wedge.

Unlike last time, we're not going to both making the wedge adjustable; instead we're going to make two wedges: one with virtually zero ground clearance, for fighting other wedges, and one with an eighth of an inch of clearance, for general purpose. Here we've drilled pilot holes in the front wall, that will be enlarged then tapped for 1/4"-20 bolts.

We'll spare you the drilling and tapping photos and take you to the [almost] finished product. We're pretty pleased with ourselves as far as accuracy goes, with all ten bolts going in without having to enlarge any holes in the wedge. We're going to countersink the wedge and use flathead bolts - this shot was just testing to make sure everything lined up.

On the other end of the size spectrum now, and we've cut rails for a 150 gram fairyweight. Lauren decided she wanted another 'bot to run on the Friday competition at Motorama, so we designed a scaled-down version of Malicious Mule, using the Pololu HP 10:1 motors we used at Franklin. The 'bot will be 3.5" square and five-eighths of an inch tall when completed, running on 1" foam wheels.

A subtle change in this photo - we've milled slots in the 1/8" 6061 aluminium front and rear walls. Next comes drilling and tapping for 0-80 screws.

Drat it! With two holes to go, out of 24 total we broke a tap - grrr! Fortunately we have one more, and carefully we finished up the tapping. There's enough of the broken tap sticking out that it can act as a 'peg' in the hole of the rear rail, and there are plenty of other screws helping to hold the 'bot together, so we're not too terribly worried about this one.

Flipping the scale switch again, and the paint has dried on the front wall for the featherweight, so we set about reassembling the chassis rails.

While the reassembly was going on, we checked weight on the fairyweight - we're suspecting we're going to be quite a bit underweight! Missing from the picture is the battery pack, which should be about 12 grams, power switch (a couple of grams) and motor mounts (a gram or two each), but there's also way too much polycarbonate there, so we'll see how it goes.

Here are some parts that showed up - the 180mAh 2S LiPo batteries for the new fairyweight, and a Car ESC programming card, for the ESCs we'll be using in both the drumbot and the featherweight. The instructions are amazingly amusing to read, and are in classic 'Engrish':
"The two of all parameters can be set quantificationally". What?!

This is the remainder of the package's contents - five G25 710kV motors for Intrusive Interloper 3.0. Originally we were planning on using 610kV motors, but for some weird reason, Hobby King only allows you one per order, and we didn't want to have to pay five shipping charges, plus, as you'll see in a moment, these motors have slightly different dimensions than the 610kV ones.

Just for fun, the assembled chassis is sitting on top of a 3.5" floppy disk we scrounged up. For our younger viewers, a floppy disk was what we used before Thumb drives and cloud storage when we wanted to save documents or install programs. We have fond memories of installing Windows NT 3.51 from 33 sequential floppies ... those were the days! hehe

Here's the shiny new wedge installed on the featherweight. Although we have the motors, we're waiting on three deliveries before we can really get this 'bot up and running: ESCs from Hong Kong, gears from SDP/SI, and wheels from Canada. But hey - at least it was assembled in the USA :-)

Speaking of motors, we're flipping the orientation of the output shaft, which means removing the can, grinding new flats for set screws, and pressing the shaft down to the can with the arbor press. It's a fairly straight-forward operation, although we did find one of the set screws was stripped, so we're just going to have to brute force that shaft.

Here's one of the modified motors sitting in a drive pod. Notice that there's clearance between the end of the motor and the rear wall, so no need to mill pockets or drill holes for clearance. We do, however, need to drill mounting holes for the motors in all the drive pods, so we'd better bring our A[ccuracy]-Game to the marking and drilling tomorrow.

We don't have the benefit of a waterjet to cut the top and bottom plates for the new fairyweight, so instead we snipped out rough shapes from 1/16" polycarbonate with some scissors and then hit them with the belt sander to get them almost to size. The intention is to mount them to the chassis, and then sand off the excess in place, to ensure a perfect fit. We also need to drill out some wheel holes, of course.

While the belt sander was out, we took advantage of it to sand down some spare polycarbonate motor mounts to a half-inch tall, so we have them ready to drill and tap for the fairyweight.

Finally, we've finished modding all the drive motors for the 30lb'er, and we think we'll call it a night. All in all it's been a pretty productive day, and we still have two more days to build! Let's see where we're at come Monday evening ...