Higher pinions = safer MMM's? Or no? Someone's lying....

[BrianG]

Originally Posted by J57ltr

A little over voltage is fine but During Regen if the batteries can't handle it Its going to about 3 times the rated voltage of the FETS in which case the very thin layer of Si is damaged and the FET will short out, usually just after it's called into action again (when you take off), once one goes then the rest are sure to follow. At least that has been my experience with overvoltage and mosfets.

Jeff

Exactly. Some people don't seem to realize failures sometimes don't come from just excessive current or excessive voltage, it's a domino effect. The high voltage "punches" through and damages dialectrics. And once that is done, even if the high voltage condition goes away, high currents can flow damaging it even further, and usually much more noticeable via fire/smoke.

Of course, damage can be done just by high currents without high voltage.

[Pdelcast]

Originally Posted by BrianG

Exactly. Some people don't seem to realize failures sometimes don't come from just excessive current or excessive voltage, it's a domino effect. The high voltage "punches" through and damages dialectrics. And once that is done, even if the high voltage condition goes away, high currents can flow damaging it even further, and usually much more noticeable via fire/smoke.

Of course, damage can be done just by high currents without high voltage.

That's close -- but the actual FET damage during braking is actually caused by current, not by voltage. It's an effect of voltage, but the damage is done by current.

What happens is this -- the voltage rises until the FETs avalanche -- that is, they start acting like a zener diode. On the MMM the FET avalanche voltage is around 44V-48V. When the FETs avalanche, they provide a low resistance path for current, and the instantaneous currents can be thousands of amps. This can (rarely) be high enough to damage the interconnects inside the FET itself.

We RARELY see this type of damage on an MMM though-- because the FETs on the MMM are tough enough to handle repeated avalanche currents without failing.

Much more often we see a power supply issue, where a part in one of the power supplies fails from voltage stress. We have been (and continue to) toughen up the power supplies on the MMM to help prevent these issues.

Patrick

[BrianG]

So, in your opinion, would it be adviseable to run no more than 4s or 5s when braking is enabled in a heavier and/or highly geared vehicle? Since the rise in voltage is at least partially proportional to the supply, this would give a little more leeway before the avalanche occurs. Or how about offering a TVS/capacitor board module so people with more aggressive setups can add something for a little more protection against ripple and braking?

[J57ltr]

Originally Posted by Pdelcast

That's close -- but the actual FET damage during braking is actually caused by current, not by voltage. It's an effect of voltage, but the damage is done by current.

What happens is this -- the voltage rises until the FETs avalanche -- that is, they start acting like a zener diode. On the MMM the FET avalanche voltage is around 44V-48V. When the FETs avalanche, they provide a low resistance path for current, and the instantaneous currents can be thousands of amps. This can (rarely) be high enough to damage the interconnects inside the FET itself.

We RARELY see this type of damage on an MMM though-- because the FETs on the MMM are tough enough to handle repeated avalanche currents without failing.

Much more often we see a power supply issue, where a part in one of the power supplies fails from voltage stress. We have been (and continue to) toughen up the power supplies on the MMM to help prevent these issues.

Patrick

And by power supplies are you talking about the drivers for the fets or the BEC, or something else?

Thanks,

Jeff

[Pdelcast]

Originally Posted by BrianG

So, in your opinion, would it be adviseable to run no more than 4s or 5s when braking is enabled in a heavier and/or highly geared vehicle? Since the rise in voltage is at least partially proportional to the supply, this would give a little more leeway before the avalanche occurs.

No, I'd suggest running BIG, high quality batteries. The best way to prevent voltage rise on the bus is to have low impedance batteries.

Lowering the voltage is an option, but the batteries are more important -- insufficient batteries will cause high voltage rise on the line even with lower voltages.

Thanx!

[Pdelcast]

Originally Posted by J57ltr

And by power supplies are you talking about the drivers for the fets or the BEC, or something else?

Thanks,

Jeff

No, I'm talking about the power supplies themselves. There are three on the MMM -- one for the processor (3.3V), one for the FET drivers (12V), and one for the BEC (6V).

[J57ltr]

Ok kinda what I thought.

Now on the large pinion smaller pinion issue is it safe to say that having a numerically higher gear ratio will cause more back emf which could exacerbate the problem if the batteries are sub par?

Thanks again,


Jeff

[Pdelcast]

Originally Posted by J57ltr

Ok kinda what I thought.

Now on the large pinion smaller pinion issue is it safe to say that having a numerically higher gear ratio will cause more back emf which could exacerbate the problem if the batteries are sub par?

Thanks again,


Jeff

In 99% of the cases, that is exactly true. Smaller pinion = less ESC stress.

[J57ltr]

Originally Posted by Pdelcast

In 99% of the cases, that is exactly true. Smaller pinion = less ESC stress.

Lol! By numerically higher I meant using the smaller pinion could cause a problem on subpar batteries due to the motor spinning faster from the start of braking.

Jeff

[BrianG]

Originally Posted by Pdelcast

In 99% of the cases, that is exactly true. Smaller pinion = less ESC stress.

Which brings us back to the original question in the thread title. See Freezebyte, we get to the answer eventually.

Patrick, I edited an earlier post apparently after you responded. Any ideas on offering a TVS/capacitor module one could solder to the MMM power wires (not to the PCB)? Definitely not a replacement for good quaility batteries, but would provide some peace of mind.

[Pdelcast]

Originally Posted by BrianG

Which brings us back to the original question in the thread title. See Freezebyte, we get to the answer eventually.

Patrick, I edited an earlier post apparently after you responded. Any ideas on offering a TVS/capacitor module one could solder to the MMM power wires (not to the PCB)? Definitely not a replacement for good quaility batteries, but would provide some peace of mind.

Pretty good idea. I'll look into it.

[RBMike]

Originally Posted by Unsullied_Spy

EDIT: I should mention this was on 20-30C Turnigy 5AH lipos and they were around 100 degrees after the run and didn't seem to cause any issues, not all Chinese packs are bad...

I think your missing the point on the cheap batteries. The point is, they don't screen out the bad cells so you could buy 3 of those packs & have 2 of them work fine but the 3rd one might have a higher resistance cell. It could cause a bunch of ripple to esc & kill it.

So your right, not all Chinese packs are bad but you will never know if your buying the "bad one" or not. The brand name companies (that charge more) are taking the time to test & screen out the bad cells & NOT assembling them into the packs they will sell. That's what your paying for.

[suicideneil]

Ha, so lower gearing is better for the esc- now theres a revelation I didnt believe for a second that using higher gearing was better for the esc on any given voltage, as was suggested earlier.

[ticklechicken]

Originally Posted by Pdelcast

Agreed -- the best way to tell if a battery is good quality (barring using a Phoenix-ICE logger (which logs voltage ripple)) is to measure voltage drop at the battery terminals under a controlled load.

Originally Posted by ticklechicken

Thank you for this information. For the MMM, do you have a minimum voltage drop which is acceptable, like 3.4 volts/cell? Or maybe it's a measurement relative to the packs voltage at that point in the run. For example, require that full throttle only produces a voltage drop of 0.5 volts/cell.

I'm very interested in finding packs that won't harm my MMM, and I want to be able to monitor the batteries through their life to verify they're still meeting the minimum requirements. Even the best cells will eventually start to have more and more voltage drop. How do I know when to throw them away to protect my MMM? I'm sorry to put you on the spot, but I would really like an actual number. I understand it's not a guarantee against anything going wrong, but I would value your opinion on this.

I never got an answer to this. Please?

[J57ltr]

Patricks answer is exactly the opposite of what I asked though.

As far as an external circuit, I am making a clipper circuit that uses a fet that is driven by a Zener in series with a resistor. Once the voltage rises above about 30V it drives the fet on and the load is dumped into a 10 ohm 5 watt resistor (Think I am going to go to a 10 watt though). This is specifically for the motor I use, but I am sure it could be applied in this case as well.

Jeff

Also higher and lower gearing means different things to different people. Higher gearing may mean higher speed (numerically lower) and lower gearing may mean Lower speed (numerically higher).

To me Higher gearing means Numerically low (or shorter gearing) and lower gearing means numerically high (or taller gearing)