Blown talon srx modules

[Mike Copioli]

This is an ESD issue, not related to thermal.

Originally Posted by Michael Hill

Talons are being blown; Not by one team, but by several

So far only two teams have reported this problem. Not several.

Originally Posted by Michael Hill

ESD should not kill products that are used in a manner in which they are intended to be used, if not also have tolerance for wiring mistakes (this is a competition full of people who make mistakes).

This is far from the truth. No amount of ESD precautions can guarantee protection from an ESD strike.

[philso]

Originally Posted by Michael Hill

I'm not saying the temperature rose due to ESD. My guess is that the robot had been running prior to the event and heated up.

It doesn't matter why the transorb got that hot. The point in my last paragraph was that these devices continue to work up to the point where we melted the solder, way beyond 125 C.

Why do you think that the Talon had heated up enough that the transorb was at or near 125 C?

[philso]

Double post

[GregGarner]

During all the previous failures the Talon SRX modules were not warm at all. They were essentially at room temperature. We haven't noticed the Talon SRX modules getting warm even under fairly heavy loading. I attribute this to the extremely low on resistance of the H-Bridge inside the Talon, just not much power dissipation there. We also have the Talon's mounted to aluminum to keep them cool.

We are quite sure that the cables have no internal shorts or other faults, because we haven't replaced the cables and the replacement Talons connected to those cable are working properly.

I did hook up the scope last night, and I found that there is a large transient on the 3.3V wire at the Talon SRX. In particular, I do see large transients on the 3.3V line when running the motor, but I see no such transients when moving the motor manually. I think this is strong evidence that the 3.3V wire in our cable is coupling in noise from the motor wires. Again, we are possibly the only team making our own custom ribbon cables, so it is likely that most other teams aren't seeing this problem because they didn't run the 3.3V signal out to the end of a 8 foot cable in parallel with the motor wires.

I briefly looked at the 5V line and the transients seemed much lower, perhaps because there is a load on the 5V line in the form of the encoder.

Therefore we will now proceed to reworking the cables as previously described. I am hopeful that this will solve the problem. The breakout boards from CrossTheRoadElectronics just arrived (Thanks Mike!), so I should be able to get the team to make progress on this change tonight.

Thanks to everyone who is trying to help us solve this problem!

Here is the highest transient I saw on the 3.3V line at the Talon connector while running the motor. I think the peak voltage was probably around +-7V:


GregGarner
FRC GearHogs 3612

[Michael Hill]

Originally Posted by Mike Copioli

This is far from the truth. No amount of ESD precautions can guarantee protection from an ESD strike.

Are you saying it's acceptable for your cell phone or your laptop to completely die from an ESD strike?

[philso]

Originally Posted by GregGarner

During all the previous failures the Talon SRX modules were not warm at all. They were essentially at room temperature. We haven't noticed the Talon SRX modules getting warm even under fairly heavy loading. I attribute this to the extremely low on resistance of the H-Bridge inside the Talon, just not much power dissipation there. We also have the Talon's mounted to aluminum to keep them cool.

We are quite sure that the cables have no internal shorts or other faults, because we haven't replaced the cables and the replacement Talons connected to those cable are working properly.

I did hook up the scope last night, and I found that there is a large transient on the 3.3V wire at the Talon SRX. In particular, I do see large transients on the 3.3V line when running the motor, but I see no such transients when moving the motor manually. I think this is strong evidence that the 3.3V wire in our cable is coupling in noise from the motor wires. Again, we are possibly the only team making our own custom ribbon cables, so it is likely that most other teams aren't seeing this problem because they didn't run the 3.3V signal out to the end of a 8 foot cable in parallel with the motor wires.

I briefly looked at the 5V line and the transients seemed much lower, perhaps because there is a load on the 5V line in the form of the encoder.

Therefore we will now proceed to reworking the cables as previously described. I am hopeful that this will solve the problem. The breakout boards from CrossTheRoadElectronics just arrived (Thanks Mike!), so I should be able to get the team to make progress on this change tonight.

Thanks to everyone who is trying to help us solve this problem!

Here is the highest transient I saw on the 3.3V line at the Talon connector while running the motor. I think the peak voltage was probably around +-7V:


GregGarner
FRC GearHogs 3612

Greg

Please post a photo showing how you have the scope probe connected to the Talon. The Ground clip of the probe should be connected to pin 10 of the Talon Analog Input connector. The wire for Ground clip of the probe forms a loop and will cause the probe to pick up noise from magnetic fields around the probe. Minimizing the area of that loop will minimize the spurious signals. The wire can be wrapped around the probe leaving the minimum length require to reach the Ground pin on the Talon Analog Input connector.

You can also connect the probe to the breakout board with no cables connected. If you still get noise, it is coming from inside the Talon or is being picked up "from the air" by your scope probe.

Please run the Talon with the same commands as used to get the scope shot but with the motor connected then disconnected.

You may want to twist the motor wires (about 2 twists per inch) and see if that reduces the noise.

[Mike Copioli]

Originally Posted by Michael Hill

Are you saying it's acceptable for your cell phone or your laptop to completely die from an ESD strike?

Yes.

If the strike's potential is higher than what the device is designed around, you should expect the device to fail.

Would you think your television design was unacceptable if it fried due to a lightning strike? I can almost guarantee it will die if this happens.

BTW lightning is ESD.

There are boundaries to every design. Just because you hit one of these boundaries does not make a product crap or unacceptable.

[JamesBrown]

Originally Posted by Michael Hill

Are you saying it's acceptable for your cell phone or your laptop to completely die from an ESD strike?

You aren't comparing apples to apples here, the talon is not a consumer electronics product. It is a component in an electrical system. It is not unreasonable to expect the customer who purchased the product to take appropriate precautions during installation and in their application.

[GregGarner]

Note that the wiring in this robot has been torn down, so I am unable to take any more pictures or scope shots until we get it rebuilt.

I can probably take some more scope shots when we get it rebuilt using the breakout boards in a day or two. I expect that we will so little to no noise on 3.3V rail since it will only have a short ribbon cable connected to it.

We are changing over to shielded wiring to go from the breakout board out to the encoders, so that should greatly help the 5V line to not pick up any noise, and of course the encoder cables will be run separately from the motor wires now.

I would like to re-emphasize that our original implementation was flawed, it is not good design practice to run signal wires in close proximity to high current motor wires for any significant distance. I also think that the idea of using this super tiny ribbon cable to go all the way out to the encoders was flawed, since this ribbon is just not very robust.

It is much better and more reliable to run the ribbon a short distance out of the talons and then use the breakout board to switch over to a more robust shielded cable to go the long distance to the encoder.

Hindsight is 20/20, but for any other teams out there that see this thread please think about this when designing and building your robot.

GregGarner
FRC3612 GearHogs

[philso]

Originally Posted by Mike Copioli

The Talon SRX is tested to both EN 61000-6-1 and EN 6100-4-2, 8KV air gap and 4 kV direct contact. This is the standard that most consumer electronics are tested to.

We used the attached TVS diode in several places in the design. It is rated to >+/-30kV.

Originally Posted by Mike Copioli

Yes.

If the strike's potential is higher than what the device is designed around, you should expect the device to fail.

Would you think your television design was unacceptable if it fried due to a lightning strike? I can almost guarantee it will die if this happens.

BTW lightning is ESD.

There are boundaries to every design. Just because you hit one of these boundaries does not make a product crap or unacceptable.

Michael

It is expected by those who wrote the standards and those who follow them (customers and manufacturers), that applying levels beyond those called out in the standard can lead to failure of the product. I am not aware of any standards that require a product not fail with no limit on the level of the stimulus.

[GregGarner]

Good news from Gearhogs Team 3612:

The shotgun approach worked, as our Talon SRX modules are all functioning nicely.

We bagged up our competition bot last night, fully functioning. We ran it quite a bit, so I am pretty confident that we solved the problem.

I haven't had a chance to look at any more waveforms.

Greg Garner