Tracked Cat delta skidsteer

[ddmckee54]

I've been converting one of these Bruder Cat delta skidsteer toys to RC for some time now.  It's been on the back burner for a long time because I've been learning how to use the printer, draw in 3D, and upgrading my Chinese clone Prusa I3 into a usable machine that gives me predictable results - not just pretty looking junk.

I'm working on something similar to this:http://www.scale4x4rc.org/forums/showthread.php?t=48900
I'm using a similar gearbox to what SmallHaul used, but I've changed out the motors from 3 volt motors to 6 volt motors.  I had originally planned on a 6 volt ni-cad pack but now I'm wondering if I shouldn't upgrade them to 9 volt motors so that I could run a 2s LiPo battery pack. The drive motors will be controlled by a T-Rex controller and I am planning on printing the drive sprockets.  I've printed a test section of the hex shaped drive shafts coming out of the transmission and that fits just fine.  I've also printed up a 9 tooth drive sprocket for the track belts but I've still got some clearance issues there.  0.2mm is apparently not enough when 3D printing, at least on my machine it's not.  When I've got a usable prototype part I'll take a picture for Show and Tell.

I don't plan on using servos for the bucket lift/lower or the bucket tilt.  Or is that bucket curl?  I don't know - I'm just a lowly EE, you construction types can correct me.  My plan is to use screw drives for these functions with the motors controlled by a T-Rex Jr. controller.

This little project is one of the reasons that I got my 3D printer, to make parts for this.  I spent most of Labor Day weekend drawing this thing in 3D.  So I guess maybe I better get on the stick and get busy making some usable parts.

Don

[ddmckee54]

Well I printed out a bunch of junk prototype parts over the weekend, even took some pictures.  Unfortunately I didn't remember to download those pictures from my camera, which is still sitting at home on my desk.

I printed the parts for the track drive sprockets.  They look real pretty and they're even a nice snug fit on the 2.6mm hex drive axles from the Tamiya gear box - unfortunately they don't fit the treads.  The treads are a flexible rubber band with hard plastic tread plates bonded to the inside of the band.  I measured the width of the plate at 6mm, I couldn't get an accurate measurement of the gaps between the plates so I measured the length of 5 plates and spaces by measuring from the leading edge of the 1st plate to the leading of the 6th plate, this measured 50mm.  So I've got a 10mm pitch between the links.  When I wrap the tread around the existing Bruder "drive sprocket" it looks like 9 "links" will fit.

So, 9 links at 10mm per link gives me a circumference of 90mm.  Dividing this value by Pi gave me the outside diameter of the "drive sprocket" with 9 teeth.  I used that value when 3D modeling the sprocket.  But when printed, the first version didn't fit.  I didn't allow enough clearance at the bottom of the pocket for the guide tabs and they bottomed out holding the track plate clear of the side plates of the drive sprocket.  I redesigned the pockets and reprinted the sprocket, it fit the track plate just like it was designed that way.  When I tried wrapping the tread around the sprocket to check the spacing it didn't fit.

WHAT THE F----!!!!  I re-measured everything and came up with just EXACTLY the numbers that I was expecting, so why doesn't this stinkin' thing fit.

Late yesterday I had an AHA moment.  I realized that the 4mm gap between the tread plates was not located where I was originally thinking it should be.  I was thinking that it was on the side of the plate that contacts the drive sprocket, it's not.  The contact point between the tread plate and the drive sprocket is at least 1mm closer, the thickness of the plastic plate, to the sprocket axis.  I modeled it again, this time using the brute force method.  This time the diameter to the contact point was about 28.5mm, as opposed to the 30mm-ish that I was using before. 

This time I may just have a winner, I'm going to reprint this tonight and find out.  I'll take pictures of the before and after versions.

Don

[ddmckee54]

Since nobody believes you if you don't gots pictures, I GOTS PICTURES!

The first attachment shows the 1st, 2nd and 3rd failed attempts at making a usable drive sprocket for the tracks.  The group in the lower left was the 1st attempt.  The "sprocket" was SUPPOSED to fit into a pocket.  My printer doesn't make neat and reliable bridges - yet.  The group in the lower right is the 2nd attempt which eliminated the large pocket.  That part was a success but it didn't fit the tracks properly - not enough clearance at the bottom of the openings for the track plates.  In the 3rd attempt, that blue part on the gearbox axle I went from a 2 piece assembly that would be glued together to a 4 piece assembly that is bolted and glued together - still didn't fit as shown in the 2nd attachment.

The 4th try was the charm, not the 3rd, as shown in the 3rd attachment.  When I wrapped the track around the drive sprocket and it fit just like it was designed for it, I couldn't resist.  I had to put it on the skidsteer and see what it looked like.  I've got some clearance issues, 4th attachment, but that 's just a matter of getting rid of a little plastic.

Next step will be to get rid of those fake rollers on the track frames and make them working.

Don

[ddmckee54]

Well I've accounted for the clearance problem between the drive sprocket and the existing skid steer frame.  The clearance problem is STILL there but at least I now know why it's these when the model showed plenty of clearance.

It would seem that some dummy drew the track frames 4mm wider than they actually are.  Discovered that one when I printed the frames for the left track rollers and they didn't fit - not even close.  Since I copied, pasted, and modified the left side parts to get the right side parts I compounded my error.  In my defense most of the original 3D modeling for this was done over the Labor day weekend and there were a couple of LATE night sessions in there - maybe I should have quit and gone to bed a little earlier?

Anyway, I'm in the process of fixing my Boo-Boo's and will soon print test parts for the left side track roller frames.

Don

[ddmckee54]

Forgot to mention, both the track roller frames prints failed about 3/4 of the way thru the prints.  I'm working on the reasons why they failed, but with what I have discovered so far I believe that I can safely assign most of the blame for the failures to myself.

Tried some stuff that didn't work, tried other stuff to fix it and just generally mucked things up.  You know how it goes.

Last night I think I figured out the root cause/s and started fixing the blunder/s.

More to follow,
Don

[ddmckee54]

The night of the 12th I also printed half of the left hand side track roller frame.  The extruder gear started "popping" as it was printing.  By this time I was wondering if I was printing too fast, 60mm/sec.; and if I was printing at too low a temperature, 200°C.  So as the part was printing I did two things, I slowed down the feed rate and I cranked up the temperature.  By the end of the print I thought I had found the sweet spot for both the temperature, 220°C, and the speed, about 50mm/sec.

Last night I made the changes to my Slicer setup and I managed to test them on one of my 20mm test cubes, it printed OK.  When I tried to print the other half of the track roller frame the extruder drive gear started "popping" again.  Apparently the 20mm test cube was not large enough to put enough of a load on the extruder for the problem to show up, so I slowed the feed rate down another 5mm/sec.    It printed the rest of the part without any issues.

They were just the wrong size, working too late at night over the Labor Day weekend strikes again.

Don

[ddmckee54]

I spent the weekend TRYING to print a test part of the motor/transmission mount, I SUCCEEDED in printing a bunch of junk.  No matter what I tried I COULD not get that blue PLA filament to reliably print.  Sunday afternoon I threw in the towel, I changed out the blue PLA with white PLA and 4 hours later I had a transmission/motor mount.  It's not perfect but it IS usable.

Don

[ddmckee54]

Maybe the motor mount's not so usable after all.  Last night I started wondering where I was going to put the battery pack and how big was said battery pack going to be?  I might have to lose part of the motor mount to get room for a decent sized battery pack.

I can get a 2S 2400mah hardcase battery pack that will fit in the available volume.  Or I can get a battery pack with upwards of 5000mah that's about 104mm long.  I've got 110mm between the axles but that means I'd have to lose part of the motor mount.

According to Tim Allen more power is ALWAYS better, but I've got no idea how long a 2400mah pack would last in this application.

Don

[ddmckee54]

I printed a bunch of junk over the weekend, and a few good parts, but I didn't take any pictures.  I'll take some tonight when I get home.

I switched my slicing software, I went from Slic3r to Simplify 3D, and I like it a LOT!  With Slic3r I always had trouble printing a thin wall, >1.5mm thick, without leaving a gap in the perimeter.  When Slicer couldn't fit all the required perimeters in the wall it would just leave a gap, I eventually just gave up trying to print a solid wall less than 2mm thick.

When Simplify 3D can't fit in all the perimeters it too will leave a gap, but then it will fill that gap in the perimeter with solid infill - no more hollow walls where they are supposed to be solid.

The part that gave me the most trouble was the replacement radiator grill.  On the original Bruder model this is a solid part.  I wanted to open this up and mount a cooling fan under the grill.  Partially just because I could, but also because it seems like I'm trying to fit 10 pounds of parts into a 5 pound sack and some of those parts have a tendency of getting a trifle warm.  Battery packs and drive motors for instance.  A cooling fan won't hurt anything and it just FEELS right.

The first time I printed it the grill ALMOST fit.  The width was a little too tight, so I took off 0.5mm and printed it again.  The curve where the angled grill meets the counterweight was not quite right, but it was acceptable.  However it was TOO long by at least a mm, maybe two.  So I took a mm off the length and printed it again.  IT STILL DIDN'T FIT - too long!  WTF, I cut it off twice and it's still too long?  This started an almost day long adventure into checking/verifying slicer and printer settings and calibration.  The most frustrating part was that everything checked out as being OK.

All right, what is going on here?  This part starts out at a 13.5° angle from the print bed.  I rotate it so that it builds flat and doesn't require support to print.  Is that rotation is causing the issue?  I'm going to take another 2mm out of the length, print it, and fix the fit with a file if I have to.  So I take 2mm out of the length, fix the hole size for the cooling fan mounting bolts and print it - again.  WTF???  It's 4mm too short?  How'd that happen?  I modified the 3D cad file and saved it, check? Check!  I saved the STL file after I modified the 3D file, check?  Ummm... awww-CRAP!!  I wasted most of a day because somebody forgot to save the stinkin' STL file?  It's obviously time to hire some competent help around here.  Crueby gets shop elves that work for cookies, and all I get are stinkin' shop gnomes that sit around claiming they're on their break and threatening to file grievances with the union?  I guess you get what you pay for huh?

Pictures will follow.
Don

[ddmckee54]

Since "It didn't happen without pictures", I brung pictures.

The first attachment is the radiator grill/cooling fan fiasco.  I don't wanna talk about it.  Let's just say that when I screw up, I don't do it in half measures.

The cooling fan that is mounted to the radiator grill will blow air down over the TREX board, onto the drive motors, across the battery pack and out the opening cut in the floor of the cab.

The second attachment is my "helpful" shop gnome.  "On his break"!  Again!! As usual!!!  I've REALLY got to get that union contract out and read the fine print to find out just how many "breaks" he's allowed, and how often.  I think I'm gettin' Flim-Flammed here.

The third attachment is the transmission mount, transmission/motors, TREX driver board mount, and TREX board in what will be their permanent homes.  With the re-designed transmission mount I've got room to slide in just about any capacity 2S Lipo battery pack that I want to use.

The fourth attachment is my "helpful" shop gnome doing what he considers to be helping.  I guess it really is true, if the only tool you have in your toolbox is a hammer then every problem looks like a nail.

Don

[ddmckee54]

I made up a battery adapter cable last night, no pictures, basically it's a 3-way wye.  One power feed to the TREX board for the drive motors.  One power feed to the TREX Jr. board for the bucket lift and tilt motors.  And one power feed for the cooling fan, I'll see if the 12VDC fan that I've got will run off a 2S Lipo pack.  If it won't I've got a 5VDC fan coming that I'll run off the 5VDC battery eliminator circuit for the receiver, it's good for a 100mA load and all that it will be powering will be the receiver - no servos.  I'll just use the battery connection as my power switch for everything.

Next I need to figure out how to set this thing up in either my Unimat or my Sherline to mill away the extra plastic.  The stuff that's no longer needed and will just be in the way.

Don

[ddmckee54]

Good news - Bad news:

The Good news is that I got my battery pack Friday.  I disconnected everything, removing it from the skidsteer, and was able to bind the receiver to the transmitter.

The Bad news is that last night I was in a hurry to see this thing move so I put it back together really fast and plugged in the battery to see if I could actually make the treads move.  Somehow, one of the motor lead plugs and the battery lead plug got interchanged.  I'm pretty sure that it was that lazy good for nothing shop gnome that did the dirty deed.

Anyway, the magic smoke got let out of the TREX board.  Along with smelling REALLY-REALLY bad, once you let the magic smoke out of the electronics they just doesn't seem to function QUITE the same as before.  Once it quit glowing and cooled enough to touch, yes I saw a red glow so something on the board was actually glowing red, I looked it over but couldn't see any obvious damage - even though I knew there had to be a problem.

Since it already smelled like it was fried I decided I couldn't hurt the board much worse and hooked everything back up, this time making sure that motor leads went to motors and the battery lead went where it was supposed to go.  Amazingly enough it mostly ran.  The left track runs forward and reverse, while the right track runs forward - no reverse.

The TREX board was overkill for this application.  It will drive motors of up to 13 amps, my little 130 main drive motors are way under that - I'm guessing somewhere in the 0.5-1 amp range.  The TREX Jr. will drive motors of up to 2.5 amps, more than I need and it's 60% of the cost of a TREX - so I've got another TREX Jr. on its' way to me.

I also found out that I need to print a couple of shims so that I can raise the transmission mount a couple of mm.  The treads are a little on the loose side yet.

Don

[ddmckee54]

I printed out a mount for the TREX Jr. board last night.  While it might be about the same length and width as the TREX board, within a mm, it's got different mounting hole locations.  I really wanted to see the loader move last night, but then I remembered the magic smoke incident and thought "NOPE, not gonna do that again!"  Not sure if I'm going to get a chance to make it move tonight either.  Its' finally quit raining here, at least for today, and I've got to get the lawn mowed and the leaves picked up.

I also printed out and installed 3mm worth of 1mm shims under the motor/transmission mount to tighten the tracks.  They look better, but now it looks like I may need to move the transmission forward a couple of mm as the tracks are AWFULLY close to the upper part of the body.  Close as in touching when the raised part of the tracks go by.

As you can see in the "Couldn't resist" shot in Reply #2 there's not a lot of extra room anywhere on this machine.  That includes on the inside, there's a bulkhead that will probably interfere with the transmission if I try moving it forward.  I could take the shims out and move the track rollers, that I have yet to install, down a few mm.

That would tighten the tracks AND not interfere with any internal structures.  It'll give me a couple more mm of ground clearance too.  Good thing I haven't printed out BOTH sides of the track roller frames yet.  The problem with doing that is that it will effectively reduce the track length by about 1/3, making the machine unstable under load.  It would be more likely to rock forward and backward.  If you've ever watched a video of the "pros" dancing with a skidsteer while the back end of it is in the air you'll know that they are unstable enough already.

Maybe I'll just take some of the shims out and say "YUP, I've got loose tracks."

Don

[ddmckee54]

It's alive, IT'S ALIVE - sort of.  It's making a lot of the Nnnnn noises that the monster in "Young Frankenstein" made.  (That's pronounced Fronk-n-steen!!!)  It's moving about as well as the monster did when first animated too.

It moved forward - slowly, and it didn't want to move backward at all.  I thought that the problem was the tracks rubbing on the body, so I removed the 3mm of shims that I recently put in.  It seemed to help but is still moving slowly, in fits and jerks.  Tonight I'll take the tracks off and see how things work with no load.

The tracks still look awfully loose though, maybe it's a GOOD thing that I haven't removed the fake track rollers yet.  I may have to do something to turn the front roller into a track tensioner, just like on the real deal.  Where the track roller would have been is about where I'd need to put the track tensioner.

I'm wondering if maybe I just have a dead battery.  I don't have much experience with Lipo's. just the little 100mah 1S packs for my micro helicopter.  I'm more used to NiCads and NiMHD batteries.  When I checked the battery pack voltage I measured 7.64 volts across the cells.  I thought a Lipo's nominal voltage was 3.7 volts across the cells, so I thought that I was OK at that voltage?  Maybe I'm just farther down on the discharge curve than I thought.

Tonight I'm going to get the charger out, decode its' Chin-glish instructions, program it for a 2S Lipo and run a discharge-charge cycle on the battery pack.  It was after 11:00 when I quit last night and I did NOT want a repeat of the magic smoke incident while trying to charge the battery pack.  That might have let the magic smoke out of my house too!  Hopefully charging the battery will cure all my problems?

Don

[ddmckee54]

OK, progress update.  Been very busy at work so by the

I got the battery charged and didn't let the magic smoke out of anything.

I've taken the tracks off and put them back on a number of times, fixing a number of little issues along the way.  The only problem is it didn't help the speed much.  I realize this thing isn't a race car, but even at scale speed it should move a LITTLE faster.  I re-motored the transmission/gearbox quite a while back, went from the original 3VDC motors to 6VDC motors.  Right now, I'm reconsidering the wisdom of that move.

I've still got the original motors stashed someplace.  I might try putting them back in, overdriving them will help with the speed.  I might also do a little poking around on the Inter-web and see if I can find some higher RPM, 130 size brushed DC motors.

I also received the TREX Jr. board that I ordered.  Now I can see if I can fit everything into the available space in the cab without having to amputate anything from the designated driver.  So far he's sitting about a scale 4-6 inches higher than he should be, but he fits.  But that's without any cables plugged into the receiver or battery, waiting on a low profile battery harness.  He just might wind up minus part of his butt and getting cut off at the ankles.

If my work day doesn't run too long I'll try getting some up to date pictures.
Don

[ddmckee54]

In my stash of "Really Important Stuff", read that as junk that I haven't thrown away because I MIGHT find a use for it in the future, I found the original 3 volt motors that Tamiya supplied with the gearbox.  Now I just have to fabri-cobble together something that will allow me to solder the motor leads onto the motors.

I have started to fit the cab and driver.  The cab I got to fit around everything without doing too much of a hack job on it.  I only had to remove most of the seat to fit around the 2nd TREX Jr. board and the receiver.  All of the cab floor had to go, plus a couple of mm in order for the cab to fit over the battery.

The driver's not gonna be so lucky.  While it looks like he's not going to have to lose any of his butt to fit over the TREX board and receiver, he's probably going to have to lose both his feet at the ankles to clear the battery.  Not to mention that sitting on those servo connectors is going to be damned uncomfortable.  Probably seem like some Medieval torture device.  He's sitting higher in the cab and farther to the front that he was originally, but it doesn't look out of place - he just looks very tall.

I haven't given him the good-news/bad-news yet, I dunno - waiting for divine inspiration to offer up another solution?  (So far Nada, Nuttin, Zipski.)

Don

[ddmckee54]

SHEEESHHHH....

It's been almost a year since I worked on this thing last, obviously no pictures since it hasn't changed since last time.  I wasn't happy with the way things were going on this project, I just kept running out of room in the little beasty - there's not much in there to start with.

It's been sitting on the shelf of my computer desk and I've been glaring at it, because I just know it's laughing at me - or maybe that's just the driver laughing?  I'd been trying  to sorta copy the layout that Smallhaul used in his build over at RC Trucks and Construction, but it's just not working out for me, I'm losing too much of the original Bruder for me to be happy.  I think it's time to go back to the 3D model and start rearranging things.

I've also seen the work that RC Modder has done on YouTube and got a quite a few ideas from him.  I especially like his homebrew linear actuators mostly built from 3D printer parts.  I ordered some 5mm and 8mm leadscrews and lead-nuts, so we'll see what I can come up with.  They are literally on a slow boat from China so it'll be a couple of weeks before I see them.  I also like RC Modder's idea of using small individual ESC's and doing the mixing in the  transmitter instead of the specialized TREX mixer/ESC that I had been using.(Be a HELLUVA lot cheaper when I fry one too - don't wanna talk about it.)  I'll have to get some of the small 10A brushed ESC's to see how big they are and if they'll save me any space.  They might work out better than the TREX just because they'll be smaller and easier to tuck away in the nooks and crannies that are available.

I'll let you know how this works out, I promise it WON'T be another year before that happens.

Don

[ddmckee54]

I started looking at rearranging stuff in the skidsteer last night, I'll probably end up moving everything,.  I started with the gearbox.  I knew that the skidsteer was moving too slow and I WAS considering replacing the original 3V motors.  Until I starting looking at the current draws of the 130 series motors, the 6V motor pulls 800mA stalled, while the 3V can pull up to 4A.  I'll keep the 6V motors thank you very much.  I thought that I had the Tamiya 4 speed gearbox in this thing so I figured that I'd just change the gear ratio in the gearbox.  Turns out that I've got the Tamiya Twin gearbox.  That's got two ratio options too slow, 202-ish:1, and WAY too fast, 58-ish:1.  But I DO have the option of two different output shaft locations in the high speed range.

I started looking at the gearbox last night, and I started thinking that I DO have a 3D printer...  I could reuse the gears, print my own gearbox and get my own ratio that's somewhere between 58:1 and 202:1 with the output shaft where I want it.  I'm not sure what the actual gear ratio will be.  I'll probably wind up counting teeth to figure that out as I can't find any info in the instructions regarding gear tooth counts.  I can probably even lose several mm on each side of the gearbox, there's a LOT of air in this one.  I'll have to stretch out the gearbox, but using one of the existing gearbox configurations I found I've got 40+mm of clearance behind the motors.

I also ordered several 10A brushed car/tank ESC's with no brake.  The size those things are I could put 4 of them in the space occupied by 1 TREX board.  I'll clip the hot lead from 3 of the 4 ESC's so that only 1 ESC will act as the BEC for the receiver.  I can stretch out the gearbox and STILL have room for 2-4 ESC's behind/beside the motors.  I'll let the angry pixies in the transmitter do all the required channel mixing, no need for the TREX.

I also ordered a smaller 2S LIPO battery, 2400mAH.  I've currently got a 3500mAH pack that takes up most of the bottom of the skidsteer.  I ordered it because I wasn't sure how long the battery would last, never really thought about to tell the truth.  I've got a Heng Long Walker Bulldog tank.  Cut me some slack, I wanted an RC tank and didn't want to spend a lot of time building it.  I also didn't want to have to sell an arm, leg, kidney, or other appendage to buy one.  It came with a 2200mAH NiCad pack in it.  Couple of weeks ago I took it along with me to one of our group functions, to keep the kids of various ages busy.  That thing was running pretty much constantly for a couple of hours before it got too dark outside and we had to call it quits.  I didn't see any signs that the tank was slowing down, so I think a smaller battery in the skidsteer will be OK.

Don

[ddmckee54]

Pololu lists the gear ratios of my gearbox at 204:1 and 58:1, my numbers don't jive with that.  Counting the teeth on my gearbox, I get ratios of 209:1 and 63:1.  A 6V series 130 motor has an 11,000rpm no load speed.  Using my ratios I get output speeds of either 53rpm or 175rpm.

As a sanity check I decided to work it the other way around and see what the output speed should be for a realistic scale speed.  The speeds I found suggest that a tracked skidsteer will have a top speed of 10-12mph.  [Insert RANT here.]  No I'm not going to convert it to metric for you.  You guys can convert to metric measurements just as easily as I can convert to Imperial measurements. [End RANT.]  This scale 12mph converts to a linear speed of 66 feet/minute. The tracks on my Bruder skidsteer have a 10mm pitch and my drive sprocket has 10 teeth, so this means the tracks will move 100mm per revolution of the sprocket - close enough to 4 inches for me.  Crunching the numbers... drum roll please, we get a total of 198rpm of the drive axle to move at 66 feet/minute.  That's close to my 175 calculated rpm for the 63:1 reduction ratio.  The 175rpm works out to a scale 10.6mph, right about where it should be.  I guess the 63:1 ratio isn't way too fast after all.

Looks like I may have to buy another copy of this gearbox, I've already got 2 of them, one in the model and another in a parts box.  Unfortunately I haven't been able to locate the 12 tooth pinion gears or the additional 3mm shafts that I'll need for the 63:1 ratio output.  The 3mm shaft I'm not worried about, but that 12 tooth pinion needs to be a press fit on the hex output shaft.  I've still got a couple of parts boxes that I need to look in yet, if I can't find them I'll need to get ANOTHER gearbox.  The 2400mAH Lipo pack arrived Monday, the smaller physical size will make finding a new home for it much easier.

Don

[ddmckee54]

I'm going to put this project in indefinite hold, not due to a lack of interest in it on my part, but rather a desire to do the best I can.  I've stockpiled a number of other Bruder models that are a little larger and they should easier to convert.  The stockpile includes a Manitou MRT 2150 telehandler that I picked up used on Ebay for about $10 because it had some broken and/or missing parts.  I'll practice on them and come back to the Cat later, work my way down to it as it were.  I think that I'll start with the Manitou.  That way if it gets irreparably trashed during the conversion I won't be out that much.   I'll start a new thread when I've actually got something to show.

Don