MMM V3 problem with heatsink
 

Today I received my MMM V3 :

No comment :
http://www.honscha.de/bilder/mmm/mmm1.jpg
http://www.honscha.de/bilder/mmm/mmm2.jpg
http://www.honscha.de/bilder/mmm/mmm3.jpg
http://www.honscha.de/bilder/mmm/mmm4.jpg

Comment :

Two connectors touch the heatsink at the same time ... I think you can imagine what will happen if I connect the LIPO to the ESC...

As I am from germany it is very expensive to send the ESC back to CC-service ...
I thought the MMM is ready to run. What should I do ?

Again, Patrick (from CC) said they were unable to simulate problems with the HS touching the bullets, even when they removed the HS anodizing.

If you still feel nervous about it (as would I), see if you can slip a piece of lexan between the bullet and heatsink to act as a barrier.

or shrink wrap the bullets! i dont know why castle does not solder wires to the pcb. then put there oversized bullets on the wires. its a proven method! cough (mamba max) cough.

More solder joints? More radio interferience? Less flexibility in the wires going to the motor? Less spectacular looking? Maybe that's why.

It's not rocket science to prevent a short and the anodizing of the heatsink is not even electrically conductive. Patrick @ Castle creations insists that the bullets touching the heatsink is not a problem and never was.

He usually admits to issues they have had, but a novice at this hobby understands the short problem. If they touch at rest, they will rub dynamically and eventually short after the coating is worn off.

I have two friends that just sent theirs back because of the short. One even hurt the battery. Some of these don't have any room to put lexan like mine, so I just removed the bullets all together.

No offence, but after all the issues the MMM has had and all the work Patrick/CC have done to get them up to scratch, I *seriously* doubt that Patrick's about to let thousands of MMM V3s leave the factory if they apparently have such an obvious flaw. If there actually was a problem, there's no way he'd be daft enough to say there wasn't after they've tested the issue to death.

Besides, electricity will always take the path of least resistance. The MMM has 0.0003 ohms per phase resistance, which is probably considerably lower than that of the aluminium heatsink (even with the anodizing worn away), so won't be an issue. What happened to your friend's battery?

Sorry, I'm not trying to get on your case! It's just that Patrick has explained and stated that this isn't a problem and has never denied any of the issues that presented themselves in the past...so I just think we should perhaps (IMO) take his word on this one. He's the electronic engineer.:wink:

For the second time in this thread.
From the President of Castle creations, Even if the coating is worn off, It will not short

Why ? In my case it will short ! There is not enough room to put a piece of paper between the bullet and the heatsink....

Never mind - last night I tried the MMM instead of my MGM 22418 in an Losi 8ight-E. I experienced a little more cogging in comparison to the MGM - but it is not bad at all.

Just because there seems to be metal contact between the bullets, it doesn't mean it's going to short.:wink:

Do you have the latest software? With 1.20, the MMM should be smooth as melted butter.

just bend heatsink, so that it doesnt touch battery leads, as that could cause your monster and lipo to burn, as they shortcircuit, but leave that motor side as it is.

My opinion on this..... castle whether or not it will short needs to work on its quality control. there is no reason some one should have to bend a heatsink or stick something in between it. how hard is it to take 30 seconds to look at each unit before the case gets put on. i give them credit for fixing alot of the issues but there are still issues that are present.

Patrick said that they could not duplicate the short in this case, and hadn't witnessed it. That doesn't mean it can't happen (and IMO eventually it will happen if both bullets are touching the aluminum conductor - it is only a matter of time). Personally, I would not even consider running a controller that has this potential issue beyond initial testing - I would address it before any extensive driving. I don't care who says its OK - it looks like a short waiting to happen. I suppose if(when) it does short, it will be covered by warranty, but the batteries won't be covered(or the chassis and melted Lexan body). Just my opinion based on 8 or so years of brushless car/truck use. :)

i know you said it would be a pain to send it back but if you adjusted the software and you still get cogging it is not right. that esc is so smooth you could rub on your babies ass. Figuratively speaking of course.

pedifile...







...jk :lol:

The esc is very smooth for what it is. If you do have cogging, look for your connections etc.

As far as what Patrick said....That's nice but with then MMM on Version 3 since this past May, they still may not have gotten all the issues worked out.

As far as the resistance question with the path of least resistance, if your battery side bullets are short to the heatsink and you plug in your battery with the switch on the controller off ..... the esc circuit is virtually open.

Dear Mike,

thank you very much for this statement. I drive brushless boats, cars and flying objects ;) for years and I always use my own brain.

Look at the images I have taken of my V3 MMM and you will see that the battery side of the bullets hit the heatsink. This is a very serious problem - and that is the reason why I wanted to warn other users.

That's a good point, I hadn't thought of that. Touching bullets shouldn't be a problem on the motor side then, but if both battery bullets are touching when the ESC's off/not moving, then I guess it's a potential hazard.

Yes it is smooth .... but compared to the MGM it could be better ;)

If your MMM is not smoother than your MGM, you are doing something wrong. I have owned both, and the MMM is far smoother period. The MGM could never even run my 6 pole align motor and had issues with the 4 poler even with the latest firmware. The MMM ran the align as smoothly as it runs the neu. Not to mention that the braking on the MGM's leave much to be desired as do many of the other settings.

My opinion is if you don't like the MMM design still, sell it and run the MGM. All of the hand wringing is just a waste of time. Meanwhile I really enjoy running my MMM V2 with the latest firmware immensely.

Either way, it's not like the MGM's never fail either.........

That statement is not totally true. Assuming the battery is not completely shorted (causing all the battery voltage to drop across the battery's internal resistance), a partial short will cause some current to go through the heatsink and some to go through the load (ESC/motor). True, more current will go through the path of least resistance, but not all.

Look at it like this: If the load presented a 0.25 ohm load, and the heatsink partial short is 0.01ohms, 96.15% of the current will go through the 0.01ohm heatsink "load", while the other 3.85% will go to the real load. Of course, in reality, the battery voltage will drop substantially even with a partial short...

I still don't think the HS/bullet proximity is "safe" IMO. But, even if the HS did partially short out the bullets, the contact area would not be ideal and the resulting arcing will eat away at the offending contact.

Ah, ok. I stand corrected. :smile: I must admit, I still prefer the V1 setup of wires with bullets; it's less of a problem if you have a motor with short leads.

Just bend a couple of the heatsink fins. Put some shrink tube on the bullits, you'll be fine & not have to think about it all the time.

just fix the problem and remove the bullets

That will probably void the warranty...

Nothing against CC, but they have still not cured the fan wiring issue. Simple fix to do when you get the esc, but something simple like that should be addressed before the case design was finalized. The MMM design and case are quite a bit different to CC other products, save for the sidewinder. No enough clearance, or a specific area for the fan wire.

Not trying to say that you are wrong Dagger, but the MMM is not perfected yet... I still see flaws that can cause the unit to stop working, or at least not survive for a sufficent service life without failure.

Well put Mike, all the testing in the world will only show up some problems. The proximity of the bullets to the heatsink is an issue, the heatsink needs to be smaller, or the board needs to have a bit more room in those areas... I like the bullets on the pcb, at least on the motor side.

"1. pedifile
Michael Jackson, R. Kelly
Dont go to Neverland Ranch "

First entry in the urban dictionary, made me laugh a bit...

as castle said, it doesnt matter if those plugs on motor side touch heatsink, so lets leave that side as it is.
but certainly do something about those battery plugs, as they will shortcircuit and will damage your batterys.

why cant they just replace those damn plugs with wires...

Huh? If the HS does happen to short against the HS, it would definitely matter. You'd be shorting the offending phases together and probably damage the FET(s). I don't think Castle is saying it's ok for the motor connections to touch, they are just saying they could not get the HS to actually short the phases.

But yes, I agree, get rid of those silly plugs and use wires instead. I think they want to use bullets because their CC-Nue motor will have wires with bullets on the end and will offer more of a plug-n-play setup as opposed to soldering anything.

Or maybe use solder-posts (like Mtronics) or thick tabs (like Novak) on the motor outputs instead since they will be smaller in diameter and have much less chance of touching the HS. That way, people wouldn't be soldering directly on the PCB...

why cant they just put a SLIGHTY shorter heatsink on it? that basically sets it 3mm away?

Probably because they already have a bazillion of them made...

Hey Brian,

Well, when you consider that our motors have lower resistance than the heat sink (by about 1/2) -- you can see why the heat sink shorting a phase really isn't an issue. The heat sink actually ablates (burns) away if the bullet manages to get pushed into the heat sink, without damaging the FETs.

One of the tests we do when we stress test our ESCs is to short the output phases during operation -- at low, mid and full throttle. They are designed to survive. :smile:

In another post somewhere, I do mention that the HS would most likely ablate (although I used a less fancy word :smile:) if it does happen to touch. I'm glad that these things are tested so vigorously, but shorting phases on purpose can't be good!

And I understand your point about the HS having more resistance than the coils, but the HS does not have any back-EMF that coils do. So, does that rule still apply?

Also, if a motor has X ohms of resistance, and the HS has X*2 ohms of resistance, then the two (effectively in parallel during a potential short) has X*0.66 ohms of resistance (until the Al ablates). To me, that's just pushing it for no reason. Even if the FETs don't mind this, this can't be good for the batteries.

As far as people modding things based on advice from these forums, I can take partial blame I guess. But, that was for previous versions and I did provide disclaimers (more than once). The only mod I have suggested to a V3 is to place a thin piece of lexan between the bullets and the HS, which of course needs no disassembly or soldering, and provides a little piece of mind (to me anyway). Believe me, I don't claim (not even close) to know as much about this stuff as a true engineer such as yourself, I just get a little apprehensive when I see things I don't feel is right.

Hey Brian,

Well, in the experiments we did here, the heat sink ablates so fast it doesn't even warm the FETs much. (short term surge currents of 1200A don't hurt the FETs or batteries, but it sure vaporizes some aluminum pretty quickly!) :happy:

Well, I wouldn't say that about all batteries. Some are, shall we say, optimistically rated... :whistle:

*cough, cough* maxamps..

Facinating stuff- I still like the wires vs bullets option though- fewer issues, even if its purely all in the mind.

lol....crapsamps...