Twin motor drive

[Peter Withers]

So I just managed a very quick test of the twin motor setup. I have a 2 channel speed controller from Dimension Engineering to drive the motors direct from the RC. It does run very smoothly and quietly at least until I tried steering, then I fell foul of the very basic mixing that goes on inside the speed controller. It very rapidly reversed the inside track and in the brief moments of confusion I managed to shed a track! No damage done but it does point to needing something between the receiver and controller to sort things out.

The plan is to use an Arduino to take in three channels from the RC; throttle, steering and forward/reverse via a switched channel. By getting the arduino to read the RC inputs I can then organised controlled acceleration ramps and basically get it to mimic the steering you get with a braked differential. With forward selected both tracks will only go forward but the inside one will run to a stop at full lock. Same in reverse.

The good news though is that it all drove very smoothly and quietly and was not at all lacking in power. In the few moments of coherent steering that I did manage it turned very nicely. I'm very pleased so far.

The only other issue was that the longer chain has stretched quite a bit which is not at all surprising but the motor mounts have slotted holes for just that eventuality.

Pete

[Peter Withers]

Just tried the twin drives again and managed to shed another track. At least I now know why. It's nothing to do with the drive system other than that I have drive to the inside track while turning rather than coming to a full stop as with the original setup. On short grass the tracks are reluctant to slew. The inside track gets pulled out of engagement from the drive sprocket and the little guide features on the track links run up to the side face of the sprocket, get caught by the teeth and off comes the track. So how come I'm having this problem when Armortek clearly demonstrated the UC behaving OK on grass?

There are two things. First off, if the inside track stops dead as per the original, it doesn't get driven off the drive sprocket. In my case as I said above, the inside sprocket is still driving so can act as the catalyst for simply lifting the track off the sprocket. The second thing is that my UC has home made sprockets that are 8mm pitch because the track with split pins, when under tension, goes to around 8mm pitch. These sprockets are narrow, if the track disengages from the teeth there is plenty of scope for it to slide sideways. The original design is actually quite a bit fatter due in part to the 2-part construction. So, a wider sprocket and stationary track when slewing result in no de-tracking(!).

The track pitch issue is a real issue that affected my UC even with the original brake setup before any mods. Any tension results in the track links moving apart and a very audible clicking as teeth attempt to mesh and fail. That is why I went to my 8mm design (the originals are around 7mm). What I missed is that I need guide rings around either side of the sprocket to hold the sprocket central in the track link. So out with the lathe and turn up a couple of rings for each sprocket that will fill the gap and hold the teeth in mesh.

As an aside it does look like longer grass is potentially kinder to the tracks if it is flattened down. Notice how slippery long wet grass can be compared to a mown lawn...

As ever, none of the above is intended as criticisms of the design. I do however think the track pitch thing is an issue which may emerge with use. It is probably best addressed with some of those lovely pins from Armorpax. My solution was a cheap and quick fix for the problem and I just need to keep developing it
Pete

[Peter Withers]

Further developments. I have now added guide rings on either side of the drive sprockets and now the tracks stay engaged at all times. As a result I now have the kind of performance I was looking for on grass or hard surfaces. The inside track can be slowed down or stopped for gentle or hard steering and because the ramp rates are controlled it all happens very smoothly. The only issue I have left with the tracks is that it looks like the occasional link doesn't want to 'let go' of the sprocket tooth and I get a slight snatching effect. Looks like some links have a bit of moulding flash where the tooth engages and this is causing the track to grip the tooth. I have now filed down all 320 odd links (phew) to remove the flash and on the bench at least it seems to run more smoothly. I'm installing the electronics at the moment so a test drive will have to wait.

I have decided to stay with the chain drive on my UC despite starting off with the idea of using gears. I have upped the chain to an 8mm link which is much sturdier. The chain setup seemed to work OK even with the skinny 6mm chains and my understanding is that roller chains are very efficient in terms of power transfer.

I have an Arduino Nano between the receiver and motor speed controller doing the ramp rates, steering decoding etc which seems to work very well. I need to add reversing and I want to make the stop points adjustable (one motor still has about 0.5V across it when it should be stopped). If anyone is interested I'd be happy to share the code.

Photos will follow once I have some time.

Pete W

[Sam Hough]

Hi Pete,

My Dad is going down this route (I wanted to play with the original and shave the lawn) so it might be useful for him.

Does your code provide a nice output for the sound unit? I've not used one but presumably it wants a single "servo" link to indicate engine revs?

So might take you up on the kind offer to share code. I have some 328s in DIP without anything to do with them. I'd offer to help but my C is very rusty and I'd probably just slow you down.

Cheers

Sam

[Peter Withers]

Hi Sam, the Arduino (its actually a clone from Pololu) takes in throttle and steering channels from the receiver so the throttle can still go via the sound unit. What the code does is set both motors to the current throttle setting and then slow down the inside motor according to steering input. I need to tidy up the code and add reversing, you are welcome to a copy once I've done that. If you PM me with an email address I'll send it to you.

Thanks,
Pete

[Sam Hough]

Hi Pete,

Ah. Not sure what sticks my dad wants to use. My guess was two throttles and one for track bending. So mixing in the tx for throttle sound.

I should probably try not to get too sucked in as my tiger 1 madness is out of hand enough as it is

Cheers

Sam

[Graham P Hough]

This is my set up with timing belts and sprockets. Not tested with tracks yet. I am busy changing the split pins for proper pins supplied by Dave Dibbs, which are reported to suit the 'standard' sprocket better. I will report as I progress.

[Jerry Carducci]

......
Jerry - the supplied motor is a 550 size, rated at 12V but run at 7.4V (with a 2S battery pack). Looks like a generic Chinese 550 motor with flux boosting sleeve. The motors from my Wickes drills are much smoother and quieter running. There is no doubt that it has the grunt to move the UC around quite effectively as shown in Armortek's video. However, my own personal view is that it is working quite hard to do so and the somewhat on-off nature of the OEM steering brakes makes for dramatic direction changes. My original brake modifications made the steering more progressive but I still felt that power was being thrown away by braking the differential. At part throttle it really bogged down, hence the idea for twin motors. My gearbox is draggy due I suspect to manufacturing tolerances. Planetary gearboxes can be a bit variable in this respect. We use planetary gearmotors in equipment that goes down oilwells and believe me, even motors from a good supplier (aerospace, military etc) can suffer from surprising variations.

The bottom line for me is that I enjoy the effort . I feel that the UC could be a better performer than out of the box but as I have made clear before this is not a criticism of the product, simply an attempt to make my example behave as I want it and have some workshop fun in the process..

Pete

This certainly is an interesting thread! I'd agree that the braking effort does bleed off power from the power source as heat. Prior to starting on these
larger models a great deal of my experience was with the glow (gas) powered 1/10 models that are in the 25-45kg weight range. The one
transmission that I've done a lot of work with is the 3 speed (+reverse) braked differential originally sold by a German maker many years ago.
The unit is a very heavy, robust unit for the scale and within that context nearly indestructible. It runs in a thin oil bath and uses drum brakes
to effect steering. Power is by a Webra 6,5ccm engine that at peak rpm produces approximately .7bhp. For cooling sake I never run them lean enough
to attain peak power. For all the model's weight I never had any problem steering as long as the brakes were kept clean and oil free.

I really need to look at the UC's drive train as it is out of the box. I'm certainly not opposed to following in your and the footsteps of others in
the dual motor conversion if that proves to be the better approach. Also interested in your arduino solution for control, have you already written the code to drive the processor? Are you using a flavor of Eclipse as your IDE?

I am concerned about one other thing, perhaps I already mentioned it. With the increase in available power do you feel there's any chance
that there will be increased risk of breaking a track or some other drive line component? I know some will say that that's always a possibility
but with changing a variable in the equation something else is invariably affected. Again as you stated this isn't in any way a criticism
of the product. What is the maxim? ? "With increased power comes increased responsibility"?

Jerry

[Peter Withers]

Hi Jerry, the code is partially written, I don't have reverse yet. Also, I'd like to get the zero and full throttle points adjustable so differences in the motors and controllers can be dealt with. I'm using the Arduino IDE with an Arduino Nano clone (the Pololu A star 328PB in fact). The speed controller is a Sabertooth 25V2 from Dimension Engineering.

I am concerned at the extra stresses imposed on the rest of the running gear but so far I'm not seeing anything too worrying. What was interesting was that when I stripped out the rear axles for modification, one of the bearings in the dome shaped housings was actually running rough. I'm not sure why, maybe it was unable to take the side load imposed by the brake setup. Anyway, they have been replaced and the drive train now runs very smoothly.

My own view on the power vs reliability thing is that with the dual motor setup as it is I have more overall power but am able to control it more gently so hopefully things will be more reliable, provided of course that I don't abuse that extra power!

Pete

[Peter Withers]

More photos of the twin motor ambulance UC. The things of note are the rings added to the drive sprockets to keep the tracks centred and engaged at all times. These work well and are definitely a key part of the evolution so far. I have also added braces between the bearing housings and the rear plate of the chassis, otherwise there is way too much flex when the tracks are under load.

I have modified the engine cover to fit over the motors, still a work in progress but it should look OK when painted. Hopefully, one picture shows the meshing of the track with the 8mm pitch sprockets. These sprockets work well with the OEM split pin tracks, there is no sign of slipping or skipping now.

The view of the electronics shows the two channel speed controller and the receiver mounted up near the engine cover grille (for better reception?). The gap between them is for the Arduino that does the decoding for steering and so on. The motors are secured via screws in slotted holes in the base of the chassis to allow easy tension adjustment.

Hopefully the 8mm chain will be a bit less stretchy than the 6mm chain I originally used. The small threaded pillars you can see in the engine cover view are for the stretcher rails which are part of the ambulance conversion.

I still have to work out a mounting for the battery (a 3Ah 18V drill battery). The plan is to have a charge connector under the removable engine cover to make life easier.

Pete

[Dennis Jones]

Hi Peter,

That is looking good. I will put charging lead in the box at the back.

Dennis.

[Peter Withers]

Just finished tweaking Arduino code and rewiring the UC ambulance and it is now driving very well indeed. I now have an 18V 3Ah Makita drill battery providing the power. I had to make a connector to fit but given that they use straight blades it was simple enough. I have added suppression to both motors which should help avoid interference issues and give the speed controller an easier time.

I moved the main switch closer to the battery to shorten the connections which helps the speed controller too (lower inductance means the storage capacitors get an easier time). The switch is operated via a bicycle spoke and is push for off (I'm considering writing Push Off next to the switch...). Hopefully the setup is evident in the photos.

The engine cover has been modified to cover the motors and still allow the original removable covers to work (one of them being slightly shorter now of course). I have made the engine cover itself removable, the radiator shroud stays with the chassis. It locates fairly well anyway but I will add some magnets or possibly some other quick release fixing to allow easy removal. The only glitch is that the stretcher rails (it's an ambulance..) will need removing first.

Is there any way I can post a video on here? I'd really like to show how it drives now on all surfaces. It really is very smooth on grass or hard surfaces and can be made to slew as with the original brakes or gradually turn as with the track warping (which is completely absent on mine).

The paint has suffered and I'm really not happy with the red crosses so some more paintwork is required before it is finally finished.

Pete

[Graham P Hough]

Lovely job. Yes it will be good to see a video. I am paused on my twin motor version it is a winter job really, summer is for motorcycling, gardening and more motorcycling!

[John Clarke]

Great job Peter Looking forward to seeing a video. Interested to know what the endurance is.
Real life.

Still think the Big Red Cross will come in handy when someone works out a fully gun stabilisation system

[Dennis Jones]

Hi Pete,

Looking good, I think you just upload the video or if not then load it on to Youtube and post the the address.

Dennis.