Spinning Tortoise Melty Beatle

Al that is left to do is make the S7 knocker and this bot will be ready for testing !!!

It's looking pretty cool......should have a picture or 2 up later today...

Well alrighty then......

Hope to do some testing soon...so far ....it spins...

looks good!

VIDEO. V-I-D-E-O.

Hopefully later... if I get a few minor details ironed out...

The motors are so powerful for this size bot, that I am hoping that it will bash its way out of a corner...if it gets trapped....they hardly draw any amps to spin the little 3 pounder...

Not sure yet....BUT, I spun it up and WOW !!! it gets up to speed like NOW!!!

I'm only running 1 LED, but I need to rasie it up so that it is more visible....unfortunatley, that will also put it a bit more in harms way...

With all the padding, etc...I have weight issues and let me tell you...the G's are thru the roof. After under a minute run time, the battery tried to exit the vehicle between some standoffs...

Will probably go down to an 850ma pack and save a bit of weight ...

The weening period has begun, but it looks VERY promising so far !!!

that was fast

Don D wrote:

that was fast

There's actually not a whole lot to it...lol

Before I started, I counted like 19 pieces that needed to be made or modified...mostly simple and 6 of them are standoffs...

Second attempt....better than the first, but still weening....

I had to modify the code to spin at 20% then even lower at 12% (in config mode), in order to more easily adjust the tracking. Probably not surprising in that top speed could be around 4k...

After 4 or 5 minutes of running at this low speed, the Darlington transistors were very hot.....at the moment they are mounted on delrin standoffs, which began to melt....!!!

So I need something better, perhaps phenolic if it comes in round stock.

Here's a link to the Darlingtons...

http://www.mouser.com/ProductDetail/ON-Semiconductor/MJ11032G/?qs=HVbQlW5zcXVfQ3Hvpr3iTQ%3d%3d

Is it normal for them to get that hot...the motors ween't hot at all and I'm prety sure that amperage draw is not that great???

Any help is appreciated...

Some pictures

With the top off....all the padding at the back is just holding the receiver and a bec....at the front, is an aluminum piece rather than the S7 knocker....just wanted the bot perfectly balanced until I get things working properly. padding and battery sit above the board in the middle...(at the moment)



Here's a picture of the Darlington...it used to be around 1/4" above the Ti plate....now it is almost touching, due to the melted Delrin standoff...

OK, so I'll admit I don't know crap about Meltys right from the start--
but since you ran into heat problems, here is what I've been thinking since you mentioned the darlington-pair controls way back in the other thread.

Why do people use darlingtons in this application? I thought pretty much all motors were controlled with MOSFET H-bridges anymore--clearly they can handle power well, and they switch really fast too. My guess is that the control is not "ON/OFF" (as H-bridges do) but proportional (more typical of transistors, but also more likely to overheat)

Also, those kind of transistor packages are typically mounted in aluminum heat sinks, with heat-conducting paste--did you decide not to do that because of weight?

Just wonderin'

this does only use on or off.

I don't know that much about it either, but I was hoping that heat would not be a big issue, in that amperage is kind of low.

Scorpions use mosfets that are much smaller, lighter, don't get nearly as hot and handle more power than I need.

Weight is an issue..I bought them just to try and get this bot running ...they were on the list of what works with the circuit.

Not much point in adding heat sinks, if I end up going a different route, once I know that the thing acually works...lol

WooHoo.....it's running and translating... altho not all that well...

It runs fine, but the translating is a bit slow for possibly 2 reasons

1) speed is too fast

2) Colsons are too hard

Top speed so far is 1800 rpm according to the LED flashes....

Will try a wheel change or drop to a 2 cel battery to see what happens

Stay tuned.....

i'm just kinda sitting back reading your melty experience... I've never even considered trying it, just because I don't know anything about it... keep updating! Lack of comments doesn't mean lack of readers!

First test....started calibrating the bot....it was mostly only running with 1 motor ...at the end of the video, it launches the receiver.... no damage




time to finish it up, now that I know that the electronics actually works and the bot translates...

wow. that's pretty scary...

Well, the forces are beyond what the motors (and possibly their location) can handle.

Even at calibration speed, I hit an aluminum pot and the armature on 1 motor came right out of the back end of the motor, obviously destroying the rear endbell in the process. (plastic endbell, mind you)

So, I need to do some re-thinking on this wonderful event and how to proceed.

I wisH I knew how to make this run on brushless motors?... anyone know?

Some pictures of the progress....

I solved the motor rotor exiting the vehicle problem, made a decent battery holder and heat sinked the Darlington transistors until I come up with something better.

also I used my 2/3 pound test piece and hit it a few times...althought the camera batteries died before the best stuff. the bot can also beat its way out, if its stalled in a corner.....so far so good!!!

Here's some pics and a video....

quick change battery setup....1standoff is removed and the polycarb battery holder swings out



Darlington transistors are now on teflon spacers, keeping them isolated , but heat is transfered thru some cpu gook (non conductiive) to the Ti bottom piece. Works well so far, heat has gone down a lot, altough now the Ti is hotter...lol




Here's the overkill tie rod bracket, which houses a screw to control motor shaft endplay...when my new motors arrive, I'll have to start all over again...but it seems to work so far!! Sorry for the blurry picture...



and finally a link to the second test video...best parts happened after the camera died...sorry...will try again soon

can you control where it goes

Don D wrote:

can you control where it goes

hahaha obviously , from your comment...YES, but at this point not very well....lol

I need to do some changes to the code that controls it, and once I decide on the drive motors, then it needs to be balanced as well. After all that, possibly changing wheels could help...or so I'm told.

btw: even when running properly, it won't break any land speed records...it's just supposed to kind of drift off in the direction that you want it to go in....

I have some better 540 motors hopefully arriving tomorrow...If so, then stuff should be a flyin shortly thereafter...

After frying the Darlingtons, I move on to a pair of chips out of Robot Power's Scorpion speed controller. 540 can motors on 3s lipo Here's a video of Melty on ICE !!!

Water is oozing out of the floor seem and everytime I go over it, the bot skates for a bit. I turn it around and try to move forward, but in many cases, it slides or skates around until the wheels dry off a bit,,,lol


https://www.youtube.com/user/ReadytoRumbleCombat#p/u/6/T8zVinq973g

Starting to hit things, but one of the 540's didn't like it

https://www.youtube.com/user/ReadytoRumbleCombat#p/u/6/iKKRM9wHWCk

So, how's the heat now that you ditched the darlingtons?

btw--you have done, in a very short time, way more with this than I think I could in an entire school year! Your knowledge of machining I already knew, but I had no idea you had talents in circuitry and programming as well. Very nice work.

Hey Bob, how's it going?

I won't take any credit for programming or circuitry, in that I just started with Rich Olson's circuit and code....but thanks for the kind thought.

I can however build a simple circuit without blowing things up.....

I used to compete in computer overclocking benckmarks, which required soldering under a microscope....I also have programmed in a few languages (mostly C), and will probably have to do a little with this as well, before the smoke clears. My actual knowledge of electronics is limited, but in starting this, it's getting better...slowly ...lol

The Darlingtons are a cheap , simple way of getting things up and running with very little effort, but their efficiency is not great, especially without good heatsinking, which in this case means adding more weight that I don't have.

I am working with 2 setups at the moment. One was controlling the Darlingtons and the second is using chips from a Robot Power Scorpion esc.

The scorpion chips can also benefit from heatsinking, but so far they are fine for controlling 540 can motors. with better heatsinking, they might be able to control more powerful motors as well.

The difference between the 2 setups is that one turns motors on and off, but the motors freewheel when off.

The Scorpion chips actually brake the motors, but whether that is better or worse is yet to be totally determined. So far it seems to translate reasonably well and at fairly high speeds as well...latest version seems to be spinning at 3400 rpm, but I need to tach it to be sure.

Circuit #1 is now using mosfets rather than the Darlingtons. Heat seems to be a lot less (on the bench), they are small and light weight and can be easily heatsinked to the little bit of aluminum that I have in this bot....

If things go well, I should be able to control my hot 540 motors and hopefully get this thing up into the 4 to 5k range, where it should hit hard, but probably self destruct as well..

This is a great project for students. The physics is off the wall, and the eletronice can be simple and cheap enough that kids could do this very easily.

Also, a test platform can be slapped together in very little time and without a whole lot of effort.

On a side note: The popular thought is that Colsons are too hard and translation suffers.

Well, I just managed to graft some much softer rubber onto the colsons. I think the process is called vulcanizing...

Now you can feel that the contact surface of the wheel is much softer and seems to have much more grip.

This could be really cool, if it works, not only for melty, but for any bot using Colsons and looking for a bit more grip.

Some real world testing coming soon, I hope...just finishing up the latest motor driving circuits using mosfets.

Very cool--thanks for the detailed reply!

Grafting onto colsons seems to me like a lot of work when Banebots wheels are so readily available. I know we all have our gripes with BB, but if you get a chance to look at the wheels and hubs you might like them. The compound is way too soft for anything over 15lbs (in my opinion), and it is possible the melty would 'grind' the soft stuff too, but the hub setup makes switching them out really easy and they are not too expensive. . . .
just a thought