View Single Post
  #1  
Old 04-16-2021, 11:20 PM
dremu dremu is offline
Newbie
 
Join Date: Aug 2019
Posts: 97
dremu is on a distinguished road
Default 1:10 M1 Abrams scratch build, not great, but fun

[Side note: if images don't load, see https://rctruckandconstruction.com/s...7&postcount=16 ]

Bit of background: I've been building things of many sizes for years, ranging from scale models to RC vehicles to a 1.5ton ride-on backhoe with working hydraulics. I hate hydraulic fluid.

Always had a thing for Cold War-era stuff more than, say, WWII, and got an idea in my head to build an Abrams. I also hate tracks. Had to weld soooo many grousers for that backhoe, and somehow I forgot how tedious it was. And how complicated tracked vehicles are. Yet I thought,



As it turns out, harder than I thought. This one didn't turn out as well as some of my others. The biggest problem is that I over-engineered it more than my usual over-engineering (over-over-engineered?) and it's absolutely bloody unwieldy. Stupid thing weighs over fifty pounds, so it's dang near impossible for me to move. There is a M1070/M1000 HETS for it, with its own set of issues, but we'll get there in due course.

As with all my projects, I started drawing the thing up first. In this case I had to take a LOT of liberties with the shape to get the suspension and motors to fit.



The hull is plywood, with 1/2" square dowel as reinforcement inside the corners.



The tank treads are ANSI roller chain, basically really, REALLY big bicycle chain. This is actually double-width chain, so that the motor sprocket can go on one set of rollers, and then the 3D printed track pads snap into the other set of rollers. This is #60-2 size, which in retrospect is too big. Each side of the chain is something like ten pounds, so the tracks alone are almost half the weight of this thing.



The motors are right-angled gear motors as used in garage door openers. I opted for 24V versions to use a 6S LiPO, as I knew this thing was gonna be heavy. The ones I got are spec'ed at 250rpm unloaded; in retrospect I should have gotten like 150 or even 100rpm units. At full bore, as well see later on, this thing is scary.

The suspension took a lot of doing, and I'm actually quite proud of this part of the thing. The spring rods are music wire from the LHS, which are in fact spring steel. These are welded into an M8 bolt which has been center-drilled on the lathe:



and then a similarly center-drilled chunk of M8 threaded rod goes on the far end. This allows the threaded end to poke through a lever for the road wheels, and the bolt's hex slots into that lever to hold it in place.

On the far end, the threaded rod has two nuts right up against each, loc-tited into place, which hold it in place. These spring assemblies ride on skateboard bearings, which allows them twist ever so slightly.



This is the right side suspension. As with the 1:1, there are two sets of interleaved springs. The ones to the left in this picture, forward on the vehicle, go to the left side road wheels. The ones to the right in the pic / rear of the vehicle go to the right side road wheels. To the upper right of the pic is the rear right idler wheel, which is another M8 bolt, all riding on the same bearings. There's several dozen of those bearings in this thing; good thing they're cheap.

All of those have one of a few types of pillow block to hold the bearing; there's such a block on each side of the wood, sandwiching it for strength. Similarly, the motors mount to the larger round blocks, one inside the hull and one out.

The frontmost set of road wheels act as tensioners. Their springs are not interleaved; tension is set by moving the road wheel arm into place and then tightening the nut at the end in the center (under the barrel in the pic above.)

Note that, as in the 1:1, the spacing between the road wheels is not the same from one row to the next (which in turn changes the angle of the arm they're attached to.) This complexity is compounded because the road wheel spacing is not symmetrical from side-to-side -- the springs being interleaved means they're offset.

In short, headaches abound.

However, after much tearing out of my very limited supply of hair, it starts to look like a tank:



The roadwheels are printed on the two-headed printer, so the black parts (lugs and outer rim) are "molded" as one in with the khaki. Same for the outer faux sprocket in the back. (The drive sprockets are very real steel, but very boring, and nestled inside.) For this pic only about half the the track pads were printed, thus the every-other-one look.



More track pads, and the electronics. As with my other builds, I use the FlySky FSi6 TX. It has a mixing option for the right stick which makes it perfect for tracked vehicles. The RX goes straight to an Arduino, which has a dual-channel high-current motor controller. The lipo's up front, only place it would fit, with an emergency off switch ("on off") and then wired underneath the turret to the back.

The turret rides on that black turntable in the center, and is driven by a smaller gearmotor with the blue gear. You can just see a small red circuit board with a heatsink at the top of the pic, by the right side rear idler wheel. That's a smaller two-channel motor controller, which controls turret rotation and barrel elevation. The TX has been modified with an auto-centering stick on the left side, so the left stick controls rotation (left-right) and elevation (up-down.)



The turret is my usual mixture of print and fab. Barrel is epoxy tubing from the local plastic place (light, stiff), with printed fume extractor and tip. The black and blue panels are printed as they are STUPID shapes. The Abrams turret is asymmetrical, which was a challenge to engineer.



However, Rustoleum (or in this case, Krylon) hides many sins. It's almost a shame to put the side skirts on, but they do help keep gravel from getting into the tracks. Not that the chain cares, that stuff is indestructible, but especially at speed the tank does like to shed track pads.



Detail is about half fabbed, styrene sheet and bar, and half printed.

Here we have the turret rotation and barrel elevation

https://youtu.be/-2_Az4Y9DxY

The turret will actually rotate over 360*; the wiring inside gets wound up at about two turns. Don't ask me how I know. Even with a very slow gear motor, the barrel kept wanting to fall down as all the weight is on the one side. Ended up counter-weighting it with tungsten as used for pinewood derby cars and got it passably balanced.

Full blast down the driveway

https://youtu.be/Of-jbicl8Bk

And veeeery carefully rolling over some 2x4's. Pay no attention to the gun shield popping off the 50cal as it lands. CA fixes that.

https://youtu.be/SN2jWiE_Wck

So, yeah, not perfect, not cosmetically accurate, and absolutely bloody unwieldy to manage. But, I can say yes, I built that, and it mostly works, sort of. If doing it again, besides making it lighter, I'd actually use slower motors. These are too fast at full throttle and make it hard to control even at slower speed.

Last edited by dremu; 04-22-2021 at 03:57 PM.
Reply With Quote