Possible failure of adum7441

[anderwm]

My company bought a printer (Airwolf) that uses RAMBo v1.2. The usb interface worked initially then failed to connect. I had them send me a new board and traced back the problem in the failed board to the adum7441 chip. The receive signal is present on pin 5 (input), but not on pin 12. If I understand your schematic, doesn't this mean this isolator has failed(for whatever reason)? Unfortunately, the replacement board's communication has also stopped working, so I am clearly missing some underlying cause. However, I thought it rather strange that adum7441 would fail, since it is designed to handle those possibly different supplies. Have you encountered this issue?

[johnnyr]

We have just identified this issue in the last few days. We have had a few kits come back that the ADUM7441 was non or partially functioning. The failure is triggered by a ESD <1kV to the USB shield with the the USB unplugged or plugged into a device that is not grounded. A greater than 1kV ESD charge exceeds the isolation voltage of the ADUM7441 and finds a path to ground through the isolation IC.

The patch can be done with a plug-in-patch, or adding a SMT component. I am working on documenting the patch at https://github.com/ultimachine/RAMBo/tree/master/docs/v1-2-ESD-patch .The patch is a varistor between the two grounds that provides a path to ground for the charge. We are rolling the patch out now and it should be available from your original vendor by the end of the week.

I'm sorry, but it sounds like we found the patch too late for your board. You will need to return your board again.

[anderwm]

Ok, that checks out. I was using a laptop that wasn't plugged in. Thanks for the reply, if you have verified the mod I might try it myself. I ordered a few of the AD isolators anyway.

[johnnyr]

We have tested it. We have not been able to cause a failure with >100
deliberate ESD events after patching.

I would be happy to send you the components if you want to do the patch
yourself.

The AD part is not too bad to rework if you have a preheater and hot air.
We remove the plastic from the pth connectors and mask around the area with
kapton tape. Please contact me if it doesn't work out.
On Feb 3, 2014 9:44 PM, "anderwm" [email protected] wrote:

Ok, that checks out. I was using a laptop that wasn't plugged in. Thanks
for the reply, if you have verified the mod I might try it myself. I
ordered a few of the AD isolators anyway.

Reply to this email directly or view it on GitHubhttps://github.com//issues/24#issuecomment-34028198
.

[anderwm]

Thanks, I can get the parts. Where have you decided to put the 0603 component (or is this not as good of a plan?), no good place seems to jump out at me. Also, just out of curiosity what's the problem with bridging the two grounds (I assume this is where you started). Does the stepper motor noise screw up the USB?

[johnnyr]

Please check https://github.com/ultimachine/RAMBo/tree/master/docs/v1-2-ESD-patch again. I pushed some more docs. Especially check SMT-patch.pdf

There is a known problem with USB and ground loops. It has damaged v1.1 RAMBos and other 3D printer electronics. It occurs in installations where the printer power supply or house wiring cause a voltage differential between the computer and printer grounds. It can cause communication issues and/or component failures.

[anderwm]

Replacing the AD part and applying your smt patch has brought my USB interface back from the dead. Since it seems I am good at causing this failure, I'll let you know if it happens again now that the patch is implemented.