Ground current through scope - USB connection

[pigtwo]

Hello all,

I'm working on a board that uses USB to send data to and from a computer. Power is supplied not through USB but an external isolated power supply. I'm having some weird problems with my scope while measuring on the board. The scope will randomly turn off and then won't turn back on unless I switch the power cord(makes no sense to me). This doesn't happen immediately and the scope can be running for hours before it will turn off.

So I thought maybe there was a problem with the grounding. I'm pretty sure the USB isn't isolated so the ground on my board is earth ground. I also know that the scope ground is earth ground so I can't just put my ground clip where ever I want. I must connect it to ground and this is what I've been doing the whole time. So I measured the current going through the scope probe ground and it is about 10mA. Is this a reasonable amount? I'm not sure why there would be a difference between these grounds such that it would draw that much current.

I have verified that the power supply is isolated and with just the scope and power supply plugged in there is no ground current.

Hopefully someone has seen something like this and can give me some insight.

Thank you!

 

[andre_luis]

It is easy to know the amount of consumed current on each USB port of PC, at least on Windows. If I remember correctly, it is 500mA and once you reach this, a protection scheme turns off the output of the chipset for that specific slot and you can restore its working just with a hardware restart. Perhaps you are draining something too close from that so that the minimal change ( spurious, overheating, etc... ) could be triggering that shut off.

 

[BradtheRad]

Power is supplied not through USB but an external isolated power supply.

It sounds as though you have various return paths in your setup. Since you focus on ground connections, could the problem be in a shared current return path?

* Do you have strong rapid current switching through a power supply, or any ground wire?

* Could you have a ground loop problem?

* Do you get drooping gaps in supply V to any unit? It may help if you install a momentary supply dropout detector. (This type of circuit is easy for you to make, no doubt.)

* How do your units (scope and computer) take power from mains AC? From the same fusebox (fuse circuit)? Are appliances causing glitches on the house wiring?

* Just on the chance... Does your wall plug have 2 prongs? There is a common remedy when we get hum in a sound system. Turn the plug in the opposite direction. Glitches can appear on one wire that don't appear on the other wire.

 

[pigtwo]

Sorry for the late reply, I came down with a cold so wasn't checking the forum.

It sounds as though you have various return paths in your setup. Since you focus on ground connections, could the problem be in a shared current return path?

I think this is the core of the problem. I was thinking about it again and it seems obvious that any current coming in through the USB connection is probably using the scope to get to ground instead of the ground connection on the USB cable. I don't see anyway around this when working with USB(or any other non-isolated supply) besides isolating the scope by maybe cutting the third prong on the power cord to the scope.

* Do you have strong rapid current switching through a power supply, or any ground wire?

No, I don't think so. Everything on the board is pretty low power and the high power stuff was not connected when the problem happened.

Do you get drooping gaps in supply V to any unit? It may help if you install a momentary supply dropout detector. (This type of circuit is easy for you to make, no doubt.)

I'm not sure but I can check.

How do your units (scope and computer) take power from mains AC? From the same fusebox (fuse circuit)? Are appliances causing glitches on the house wiring?

Yes, both my computer and the scope get power from the same outlet.

* Just on the chance... Does your wall plug have 2 prongs? There is a common remedy when we get hum in a sound system. Turn the plug in the opposite direction. Glitches can appear on one wire that don't appear on the other wire.

No, I'm using three prong wall plugs for everything.

Thank you!