Hello,

I have a system where I have a device with a USB port connected to a docking station for a tablet. This docking station uses the same power supply as the device itself (Ground loop?).

In normal operation this system works as intended, but to configure the device I connect the device to a Laptop for configuration. Afterwards I plug the device back in the docking and thats when the trouble starts. The device is no longer recognised (Device descriptor request failed). It also no longers works on the Laptop. The USB seems completely dead.

I have already had this happen two times and I can't seem to find the problem. The manufacturer suggested to use an USB isolator but I don't really want to use this solution (extra cost). Besides the laptop is not earthed. Running completely on thr battery.

Please help!?

Regards,

Vytas

 

Hello,

I have a system where I have a device with a USB port connected to a docking station for a tablet. This docking station uses the same power supply as the device itself (Ground loop?).

In normal operation this system works as intended, but to configure the device I connect the device to a Laptop for configuration. Afterwards I plug the device back in the docking and thats when the trouble starts. The device is no longer recognised (Device descriptor request failed). It also no longers works on the Laptop. The USB seems completely dead.

I have already had this happen two times and I can't seem to find the problem. The manufacturer suggested to use an USB isolator but I don't really want to use this solution (extra cost). Besides the laptop is not earthed. Running completely on thr battery.

Please help!?

Regards,

Vytas

Hot plugging is an important part of the original design brief.

Polyfuses (PTC thermistors) are a common type of protection, but there could be a regular PCB fuse if you're unlucky. I've broken one or two - but never fried any.

 

The device itself is not broken. There is a CAN interface on the device which seems to be working fine. The damage seems to be limited to the USB interface.

 

If I understand correctly; the USB on your laptop and on the docking station is fine, it's your other device device that actually has the problem.
When the problem occurs, how do you get the device working again? For example is the USB part of the device actually damaged, requiring parts replacement to get it working again? Or does power cycling it get it going again?

 

That's correct, the device is broken and stays broken. I suspect the FTDI chip is failing.
The mechanism behind it is however a complete mistery to me.
I suspect something like a charged capacitance to chassis of the device but I'm totally clueless

 

The USB standard specifies self resetting fuses. Hot swapping is a given.

If you won't heed manufacturer recommendations, what do you want from us? Doing the same thing repeatedly and expecting different results doesn't make much sense.

Do something to isolate the problem.

 

The second testing was not intentionally. And as testing is expensive because it sort of destroys a 500 euro unit I am reluctant to do more testing before I have a firm understanding of what is possibly happening.

 

The USB standard specifies self resetting fuses. Hot swapping is a given.

If you won't heed manufacturer recommendations, what do you want from us? Doing the same thing repeatedly and expecting different results doesn't make much sense.

Do something to isolate the problem.

Well resourced - it takes determined ########### to wreck a USB port.

 

Obviously the amount of determination to wreck a USB port is less than you image as I have accomplished it.
I'm just trying to understand how to mitigate this risk.

 

If you won't heed manufacturer recommendations, what do you want from us? Doing the same thing repeatedly and expecting different results doesn't make much sense.

Do something to isolate the problem.

Perhaps the manufacturer doesn't comply with the standard and the unit is bad design?
Also the manufacturers recommendation is probably from someone who is not the engineer of the unit. I'm hoping to get a better solution than just using an isolator from someone who has the knowledge to answer such a question.

 

Obviously the amount of determination to wreck a USB port is less than you image as I have accomplished it.
I'm just trying to understand how to mitigate this risk.

Never managed it so far, its uncharted territory for me.

Let me know when you figure it out - I'll add it to the list of "no - no's".

 

Never managed it so far, its uncharted territory for me.

Let me know when you figure it out - I'll add it to the list of "no - no's".

I haven't either untill now. My guess at the moment is that the unit's design is faulty. Hopefully I find out what this unit does differently then say a keyboard.
The only thing that I can think off is that the docking station shares a power supply with the device, something a keyboard or a mouse does differently.

 

Perhaps the manufacturer doesn't comply with the standard and the unit is bad design?

That's possible, or the device may be hard on USB ports. You have little chance of finding out unless you make the effort to eliminate things as potential problems.

Also the manufacturers recommendation is probably from someone who is not the engineer of the unit.

That isn't a valid prejudice. It's very likely that none of us had anything to do with the design of the unit. That doesn't mean we can't apply common sense and trouble shooting skills to try to isolate the problem.

I'm hoping to get a better solution than just using an isolator from someone who has the knowledge to answer such a question.

Why won't you do this? It's a simple step to eliminate the USB client as being the problem. Get a powered USB hub, power it, and connect/disconnect the client as you've been doing. If it doesn't damage the host, that may eliminate the client as the problem.

Plug the client into other hosts from other manufacturers (at the risk of said hosts) to see if the problem still occurs.

 

That's possible, or the device may be hard on USB ports. You have little chance of finding out unless you make the effort to eliminate things as potential problems.
That isn't a valid prejudice. It's very likely that none of us had anything to do with the design of the unit. That doesn't mean we can't apply common sense and trouble shooting skills to try to isolate the problem.
Why won't you do this? It's a simple step to eliminate the USB client as being the problem. Get a powered USB hub, power it, and connect/disconnect the client as you've been doing. If it doesn't damage the host, that may eliminate the client as the problem.

Plug the client into other hosts from other manufacturers (at the risk of said hosts) to see if the problem still occurs.

How would I tackle a potential problem?

I think a regular person working there doesn't have the combined knowledge to solve this problem if the only suggestion is an usb isolator.

I have done used this setup several times in the past with no issues.

I don't have the luxury of randomly breaking these units.

 

I think a regular person working there doesn't have the combined knowledge to solve this problem if the only suggestion is an usb isolator.

You have no control over the qualifications of any support person you interact with. The same is true on this forum. You have no way of knowing that the advice we're offering is more or less sound than the advice from the manufacturer that you're choosing to ignore.

Use a powered hub to see if the problem changes. If you don't have one, spend $20 and get one.

 

What is your device (that is failing) called, so that we have a name to make discussions easier?

I'm assuming it's a device you have designed yourself or at your company? The first step is figure out what part inside the device is failing. Then the next thing I would do is get the datasheet for that device and check the wiring of the device against the datasheet. Maybe a ground is missing, or maybe a TVS diode is missing, etc..

 

Hello,

I have a system where I have a device with a USB port connected to a docking station for a tablet. This docking station uses the same power supply as the device itself (Ground loop?).

In normal operation this system works as intended, but to configure the device I connect the device to a Laptop for configuration. Afterwards I plug the device back in the docking and thats when the trouble starts. The device is no longer recognised (Device descriptor request failed). It also no longers works on the Laptop. The USB seems completely dead.

I have already had this happen two times and I can't seem to find the problem. The manufacturer suggested to use an USB isolator but I don't really want to use this solution (extra cost). Besides the laptop is not earthed. Running completely on thr battery.

Please help!?

Regards,

Vytas

When your failing usb device is a FTDI chip then the following could help you;
Download the FTDI tool 'mprog' or latest version including the drivers.
This tool allows you to detect, test and modify chips internal settings.
The older drivers do require the same usb port this goes sometimes wrong result operating system can't find usb chip.
I did had that war several years ago ending up with 'I am not mad I am an apple tree'.
But survived and won the battle.

Picbuster

 

Hello I'm sorry for the delay.

The device is actually a motor controller for low voltage brush less motors and is not designed by us.

I inspected the PCB of the motor controller and found that a small 6 pin IC has a bulge. The IC is marked KA2 but haven't found the item yet. I can't see any other damage on the PCB except for that little IC. You can see it pretty clearly on the images I attached.

You have no control over the qualifications of any support person you interact with. The same is true on this forum. You have no way of knowing that the advice we're offering is more or less sound than the advice from the manufacturer that you're choosing to ignore.

Use a powered hub to see if the problem changes. If you don't have one, spend $20 and get one.

I think I'm able to filter out useful and non-useful advice, but thanks.

 

I think I have found the IC with the marking KA2:

https://www.diodes.com/assets/Datasheets/ds30154.pdf

The datasheet says it's a fast switching diode array. I don't have a clue yet why this part would be destroyed.

 

Assuming there's no warranty, I wonder if the manufacturer would give you any help determining why this part may have failed, or at least help you determine what its purpose is, so that maybe you can figure out if its failure is the cause or just a symptom of the real cause. Perhaps the board needs to be in a specific state before it's unplugged? Worst case, swap that part and see if it comes back to life.