Soldering iron makes USB devices drop out

[MBaadsgaard]

Maybe an odd place to ask about this, but I'm stumped and can't think of a better solution.

I got my computer and my hobby electronics at the same desk.
Under the desk are 3 extension cords going to 3 different outlets.
One powers my oscillloscope/soldering iron/PSU/3d Printer.
One powers my computers, monitors and peripherals and KVM.
One powers my european odd bits and bobs.

When I turn off my soldering iron, USB devices disconnect, then reconnect.
Didn't think much of it, until tonight where it led to my KVM's USB ports stopping working.

I have had something similar with the 3d printer, too, with an old monitor.

The soldering iron is a Weller WD 1M. A hefty size so I assume there's a big transformer inside.

Any suggestions to what could cause this and what I could try to alleviate it?

[wizard69]

I'm not familiar with British electrical systems so this may be garbage, but make sure if the system uses neutrals that they are properly installed.   This really sounds like a wiring problem so maybe an electrician is in order, especially one that really understands your local codes and practices.

[Kim Christensen]

The right sized MOV or a RC snubber network across the primary winding of the transformer should help. (Assuming this is a traditional 50/60Hz transformer and not a switching power supply)

[MBaadsgaard]

I'm not familiar with British electrical systems so this may be garbage, but make sure if the system uses neutrals that they are properly installed.   This really sounds like a wiring problem so maybe an electrician is in order, especially one that really understands your local codes and practices.

It's a newly-built property, so you might very well be right. They're notoriously dodgy here in the UK.
I'll get a professional on it if I don't find another solution. Though paying for potentially nothing is always a hard pill to swallow.
Thanks for giving me something to go by.

The right sized MOV or a RC snubber network across the primary winding of the transformer should help. (Assuming this is a traditional 50/60Hz transformer and not a switching power supply)

Given it's a Weller soldering iron, I can only imagine they already took care of this, but maybe some components inside have expired.
But it's pretty old. I've heard of snubber capacitors and I've heard of capacitors drying out. Would this be a thing to open it up and look for?

[wraper]

Most likely it means USB cable (including in enclosure if front USB port is used) or KVM is a substandard garbage. Soldering station is not at fault here, it just something that breaks what's already marginally working.

[MBaadsgaard]

Most likely it means USB cable (including in enclosure if front USB port is used) or KVM is a substandard garbage. Soldering station is not at fault here, it just something that breaks what's already marginally working.

Well the soldering iron causes the transient, so can't say it's not part of the problem.

[wraper]

Most likely it means USB cable (including in enclosure if front USB port is used) or KVM is a substandard garbage. Soldering station is not at fault here, it just something that breaks what's already marginally working.

Well the soldering iron causes the transient, so can't say it's not part of the problem.

It's not. Soldering station is working as intended and device under question must tolerate it. If the thing fails from turning off the soldering station (usual transformer), device is broken (likely by design) and it will fail from other normal things too. I would first check if shields on USB cables are actually connected on both ends. If not, throw away the trash. Also in cheap cases USB cables to front panels often are just flat cables with no shield whatsoever.

[MBaadsgaard]

Most likely it means USB cable (including in enclosure if front USB port is used) or KVM is a substandard garbage. Soldering station is not at fault here, it just something that breaks what's already marginally working.

Well the soldering iron causes the transient, so can't say it's not part of the problem.

It's not. Soldering station is working as intended and device under question must tolerate it. If the thing fails from turning off the soldering station (usual transformer), device is broken (likely by design) and it will fail from other normal things too. I would first check if shields on USB cables are actually connected on both ends. If not, throw away the trash. Also in cheap cases USB cables to front panels often are just flat cables with no shield whatsoever.

So you're thinking poorly shielded USB cable got current induced in it causing device to fail?
I found the USB ports for devices still work, it's the host USB port for one PC that died. This cable happens to cross the power cable to the soldering iron.

So I think you're spot on. Thanks for the insight.
I have a higher quality cable somewhere I'll dig out and will try to route things differently.

[Ian.M]

Does your 3D printer have a grounded PSU + a USB cable to the PC?  If so you've got a ground loop, which can cause these sorts of issues and even much worse!   To break the loop, you can add a USB isolater between the PC and the printer.

[MBaadsgaard]

Does your 3D printer have a grounded PSU + a USB cable to the PC?  If so you've got a ground loop, which can cause these sorts of issues and even much worse!   To break the loop, you can add a USB isolater between the PC and the printer.

No, it doesn't have a USB cable, I just print from an SD card every time.
It is grounded.
It wasn't on when I had a USB device failure, was more mentioned because it previously made a monitor turn off briefly when I turned it off.
Printer and soldering iron are on a different outlet than monitors, usb devices, computer etc.
However, the monitor that the printer made flicker had its power, USB and signal cables come down along where the printer's power cable is.

[Kim Christensen]

The right sized MOV or a RC snubber network across the primary winding of the transformer should help. (Assuming this is a traditional 50/60Hz transformer and not a switching power supply)

Given it's a Weller soldering iron, I can only imagine they already took care of this, but maybe some components inside have expired.
But it's pretty old. I've heard of snubber capacitors and I've heard of capacitors drying out. Would this be a thing to open it up and look for?

Well, if you find that replacing the USB cables with good quality ones doesn't solve your problem, I'd definitely open it up and have a look. No harm in having a peak for a burnt out MOV, etc...

[MBaadsgaard]

The right sized MOV or a RC snubber network across the primary winding of the transformer should help. (Assuming this is a traditional 50/60Hz transformer and not a switching power supply)

Given it's a Weller soldering iron, I can only imagine they already took care of this, but maybe some components inside have expired.
But it's pretty old. I've heard of snubber capacitors and I've heard of capacitors drying out. Would this be a thing to open it up and look for?

Well, if you find that replacing the USB cables with good quality ones doesn't solve your problem, I'd definitely open it up and have a look. No harm in having a peak for a burnt out MOV, etc...

I had a look. There's nothing on the primary side between live and neutral except the switch and the transformer.
I could potentially do something about that, I might even have the needed bits around, but I know nothing about this and it involves mains voltage.
I think I'll go look for an alternative. I have another soldering iron, which isn't great, but will hold up while I find an alternative.
Along with the other USB cable that'll hopefully solve it.

Would have been nice to know what caused it, to know for the future.
While the unshielded cable is a good guess, I can't really be 100% sure that was it.

But I think it's better to not experiment and potentially fry another KVM or USB hub.
If the issue crops up in some other form, I'll come and revive the post or make a new one,
but for now, thanks for the help guys.

[wraper]

You can measure resistance between outer shells on both ends. While connection does not guarantee there is a shield (they may be connected just to GND wire) and that cable is good, lack of connection guarantees that cable is trash.

[MBaadsgaard]

You can measure resistance between outer shells on both ends. While connection does not guarantee there is a shield (they may be connected just to GND wire) and that cable is good, lack of connection guarantees that cable is trash.

Good point, though I checked and there's definitely a connection.