VESC from Enertion: DRV8302 spectacularly blown, what went wrong?

Report your hardware problems here. Include as many details as possible with pictures to get the best possible help and feedback.
cyril.leroux
Posts: 5
Joined: 17 Aug 2016, 19:05
[phpBB Debug] PHP Warning: in file [ROOT]/vendor/twig/twig/lib/Twig/Extension/Core.php on line 1236: count(): Parameter must be an array or an object that implements Countable

VESC from Enertion: DRV8302 spectacularly blown, what went wrong?

Postby cyril.leroux » 17 Aug 2016, 19:25

Attachments
IMG_4358.jpg
IMG_4358.jpg (723 KiB) Viewed 5236 times
IMG_4352.jpg
IMG_4352.jpg (401.88 KiB) Viewed 5236 times

rew
Posts: 943
Joined: 25 Mar 2016, 12:29
Location: Delft, Netherlands.
[phpBB Debug] PHP Warning: in file [ROOT]/vendor/twig/twig/lib/Twig/Extension/Core.php on line 1236: count(): Parameter must be an array or an object that implements Countable

Re: VESC from Enertion: DRV8302 spectacularly blown, what went wrong?

Postby rew » 18 Aug 2016, 08:30

FYI, the thermal conductance to the top of the case is about 14K/W. (and about 1 to the base of the FET). So if we assume those heatsinks are perfect (0 K/W), you can now dissipate 7W. Not impressive.

The side of the DRV that's blown is the side that does the current amplification. Could it be that one of your current shunts came loose? In that case, you can easily get over 100V on that part of the chip that is built to withstand less than 5. (The DRV contains at least two but maybe three different dies. In any case, the step down converter is separate. If the DRIVER (12V power electronics, 60V linear voltage regulator) is made in the same process as the precision current amplifier.... I don't know. )

Dornacht's VESC looks as if the cap C51 blew. That one is directly on the battery line, but rated for 100V. (expensive M*****F*****...).

cyril.leroux
Posts: 5
Joined: 17 Aug 2016, 19:05
[phpBB Debug] PHP Warning: in file [ROOT]/vendor/twig/twig/lib/Twig/Extension/Core.php on line 1236: count(): Parameter must be an array or an object that implements Countable

Re: VESC from Enertion: DRV8302 spectacularly blown, what went wrong?

Postby cyril.leroux » 19 Aug 2016, 08:11


rew
Posts: 943
Joined: 25 Mar 2016, 12:29
Location: Delft, Netherlands.
[phpBB Debug] PHP Warning: in file [ROOT]/vendor/twig/twig/lib/Twig/Extension/Core.php on line 1236: count(): Parameter must be an array or an object that implements Countable

Re: VESC from Enertion: DRV8302 spectacularly blown, what went wrong?

Postby rew » 20 Aug 2016, 06:56

These are the shunts I'm talking about:
Image
(I had to use an image of someone else's VESC).

Yes those heatsinks are "a waste of time". The heat is generated near the bottom of the FETs, and you need to cool the PCB, possibly from the other side. Unfortunately, there are FETs there as well... Some people have a big trace conduct heat away from the mosfet sideways and solder a heatsink on there. There is no place for that here.

cyril.leroux
Posts: 5
Joined: 17 Aug 2016, 19:05
[phpBB Debug] PHP Warning: in file [ROOT]/vendor/twig/twig/lib/Twig/Extension/Core.php on line 1236: count(): Parameter must be an array or an object that implements Countable

Re: VESC from Enertion: DRV8302 spectacularly blown, what went wrong?

Postby cyril.leroux » 05 Sep 2016, 12:54

I finally came round to taking out the now defunct VESC for some proper post-mortem and there's no sign of the shunts being loose.

As suggested by another user in the electric-skateboard.builders forum, I'm investigating whether the soft switch could have played a role.

What would happen if the soft switch opened the circuit after I started braking, for instance because it's overheating? would the current generated by the braking continue to power up the VESC until it blows up because this current has nowhere to go? Would that be consistent with the failure that I experienced?

Equally, what if the power switch temporarily prevented battery regen for some reason, would the result be consistent with what I got?

Specs of the soft switch here in case it helps. My model is the discontinued SafetyPowerSwitch 60V 60/120A.
http://wiki-en.rc-electronic.com/index. ... owerSwitch

Any further help to understand what's wrong with our design would be greatly appreciated.

regards,
clr

rew
Posts: 943
Joined: 25 Mar 2016, 12:29
Location: Delft, Netherlands.
[phpBB Debug] PHP Warning: in file [ROOT]/vendor/twig/twig/lib/Twig/Extension/Core.php on line 1236: count(): Parameter must be an array or an object that implements Countable

Re: VESC from Enertion: DRV8302 spectacularly blown, what went wrong?

Postby rew » 06 Sep 2016, 07:56

When you "lose" your battery during "regen", the VESC will start charging your capacitors. Once the voltage reaches the configured maximum voltage, the VESC should turn off.

Of course, when you are braking at 2kW, and you have 1000uF of capacitance (2x470), the voltage will rise around 10V in every quarter milisecond. I don't know how fast the "overvoltage on input" reacts to such a situation.

cyril.leroux
Posts: 5
Joined: 17 Aug 2016, 19:05
[phpBB Debug] PHP Warning: in file [ROOT]/vendor/twig/twig/lib/Twig/Extension/Core.php on line 1236: count(): Parameter must be an array or an object that implements Countable

Re: VESC from Enertion: DRV8302 spectacularly blown, what went wrong?

Postby cyril.leroux » 06 Sep 2016, 09:07


pf26
Posts: 310
Joined: 28 Mar 2016, 14:37
Location: FR Valence
[phpBB Debug] PHP Warning: in file [ROOT]/vendor/twig/twig/lib/Twig/Extension/Core.php on line 1236: count(): Parameter must be an array or an object that implements Countable

Re: VESC from Enertion: DRV8302 spectacularly blown, what went wrong?

Postby pf26 » 06 Sep 2016, 10:32

Are you using 12s battery config ?
I cannot find the regen caracteristics of your power switch. I suspect the regen function might come from diodes, and this is likely to cause voltage "drop", in this case, an increase in Vesc voltage, by up to 1 volt - and quite a lot of heat.
Is there an internal temperature protection in this power switch ?
If the power switch opens during regen, I am quite sure this will destroy the VESC (no way it can react as fast as needed to prevent overvoltage, especially if using 12s battery, which is near the voltage limits, especially during regen and with voltage increase along the lines.)

High temperatures reduce junctions and breakdown temperature thresholds in general, and this makes hardware more likely to blow.

rew
Posts: 943
Joined: 25 Mar 2016, 12:29
Location: Delft, Netherlands.
[phpBB Debug] PHP Warning: in file [ROOT]/vendor/twig/twig/lib/Twig/Extension/Core.php on line 1236: count(): Parameter must be an array or an object that implements Countable

Re: VESC from Enertion: DRV8302 spectacularly blown, what went wrong?

Postby rew » 06 Sep 2016, 15:01

Yes, the DRV is one of the devices exposed to the motor voltage. As are the mosfets. Both rated for 60V, but I have the feeling that the DRV is quite sensitive to short transients. Partly because of course, the DRV is built to handle say "1.5W" at most, and the capacitors and battery are usually quite capable of handling power levels on the order of 1.5-2kW.... So when things go, there IS 2kW available (probably 10x more in practise) to make the DRV go up in smoke. The mosfets, are built to handle up to 370W disspation so might survive for a few microseconds longer.

During regen, the motor is still driven to both rails, so no diode drop.

What some people fail to realize is that when your powersupply is say 55V, the mosfets will have to edure > 55.6 due to KNOWN transients on the phase lines.

No, the breakdown voltage of mosfets goes UP with temperature. I saw a datasheet once that claimed "30V" as the breakdown voltage in the "title" of the mosfet, but if you read the datasheet carfully, the 30V was guaranteed at 25C, and you had to derate the voltage in the -40 to 25 degrees temperature range....

pf26
Posts: 310
Joined: 28 Mar 2016, 14:37
Location: FR Valence
[phpBB Debug] PHP Warning: in file [ROOT]/vendor/twig/twig/lib/Twig/Extension/Core.php on line 1236: count(): Parameter must be an array or an object that implements Countable

Re: VESC from Enertion: DRV8302 spectacularly blown, what went wrong?

Postby pf26 » 07 Sep 2016, 08:07

I was refering to the likely "diode drop" in the power switch during regen, not in the VESC.

As for breakdown voltages, I think only avalanche devices (like MOSFETs) have a positive temperature coefficient. (Zener) diodes have a negative tempco. I don't know for sure if the DRV8032 breakdown voltage is avalanche based, but I doubt it.
The FETs are probably less sensitive to breakdown voltage failure, because they are avalanche rated (they can absorb some energy when the voltage reaches their breakdown voltage).
If the FETs have a positive tempco, and the DRV breakdown voltage a negative tempco, then this would mean the FETs would somewhat protect the DRV (by absorbing voltage spikes) much more at low temperature than at higher temps. Less potential issues in Sweeden than in Spain ?

Anyway, when playing so near to the limits, I think we need find solutions to mitigate voltage spikes reaching the DRV power supply. (or get rid of DRV).


[phpBB Debug] PHP Warning: in file [ROOT]/vendor/twig/twig/lib/Twig/Extension/Core.php on line 1236: count(): Parameter must be an array or an object that implements Countable

Return to “Hardware Problems”

Who is online

Users browsing this forum: No registered users and 1 guest