Destroying USB devices

[Jimmy Tuppence]

Any help or suggestions would be greatly appreciated - and if I have posted this in the wrong forum let me know.

I have a basic setup of a laptop and a receipt printer powered by a 24v lead acid battery (2 x 12v cells in series) - crude sketch attached. The printer is 24Vdc and so is powered directly from the battery, the laptop is 19v and so is powered from the battery via a dc-dc voltage converter - in this case it is a model from Maplin. The laptop is then connected to the printer using a USB-Parallel cable/adapter such as this one.

The problem that I am experiencing is whenever the dc converter is connected/disconnected from the battery it seems to blows the chip in the USB-Parallel cable, destroying it. Having gone through quite a few cables testing out various setups I have found that the data lead wont blow if:

1. Only connected to the laptop and not printer (open end).
2. Connected to the printer and laptop but the printer is powered from a separate battery.
3. The dc-dc converter and printer are connected to the same battery but the dc converter is powering a different laptop to the one that the data lead is plugged into.

Strangely the lead will blow if the printer and laptop are connected to the same battery, but the printer is switched off. So, lead will only seem to blow when the dc-dc converter is connected/disconnected if - the printer and computer are powered from the same battery and the data lead is connected to both devices.

As a layman I would guess that connecting/disconnecting the dc-converter is creating some sort of surge or spike that is damaging the USB devices, but if anyone has any suggestions on what is causing this USB devices to blow and what the solution might be I would be very grateful.

 

[cjdelphi]

I'm confused... after switching cables everything works again?

A voltage surge could damage ICs but the cable? Do the cables melt or something? What happens

 

[cjdelphi]

I would suggest a TVS diode across the output terminals of the switching regulator, but i'm not convinced it's the problem here

 

[Jimmy Tuppence]

Crikey, thanks for the quick replies guys.

cjdelphi - The printer data cable has some sort of chip/pcb in the parallel end of the cable, presumably this communicates with the laptop when plugged in ( shows up as IEEE-1284 in windows). Connecting/disconnecting the dc-converter from the battery seems to damage this component of the cable so that it is no longer recognized - you get the standard 'USB device not recognized' message from windows when plugging the data lead back into to laptop. There is nothing as exciting as smoke or cables melting - just a printer data lead that no longer works.

 

[cjdelphi]

This puzzles me as usb will have it's own regulated 5v rail from the laptop, again there will be a regulator printer side too...

I just had a look at the maplin dc dc converter, it was designed to run on 12v and then steps it up to 19v (seems like you may be running it from 24v)

That could be the issue...

 

[Jimmy Tuppence]

Appreciate you spending time on this.

A puzzler - it's not just me then! On the packaging of the dc-converter it states that it will take 10-30v as input. I think that most of these things are designed with car electrics in mind hence 12v, but there are some, like this one that will work this a trucks 24v system too. I've attached a pic of the back of the device showing the acceptable input voltage.

Is there something we could be/should be measuring that might shine some light on what is happening?

 

[cjdelphi]

Ahh, fair enough, not that then! (Well, it shouldn't be!)

monitoring the USB voltage might be the next step, any old usb cable, strip it down and measure the voltage across it while plugging whatever in/out and look for a voltage spike, might be tricky without a scope... a regular multi Meter might not be quick enough but i'd still give it a try to see if it flashes over 5v

 

[Gryd3]

Let's clarify some things here...
Please provide the exact model number of the DC-DC converter you are using to power the laptop, I have a feeling the 'ground' or negative provided to the laptop and printer are actually at different levels.
This could cause all sorts of problems.
As you stated, this is only a problem with the same battery back is being used for both devices correct?

 

[cjdelphi]

But both grounds would be tied to each other? But i understand where you're coming from

 

[Jimmy Tuppence]

cjdelphi - will give what you suggested a try. If the multi meter (or my eyes) isn't quick enough to catch it, I'll look at getting a scope. You think a usb/pc scope would be good enough for capturing such a spike?

Gryd3 - Thanks for the reply. The dc-dc converter is a Maplin model, stock code: A83LA. The link for the device is http://www.maplin.co.uk/p/12v-auto-...ter-with-1a-usb-socket-12-charging-tips-a83la
The printer that I am using is an Epson TM-T88V thermal receipt printer - http://www.epson.co.uk/gb/en/viewcon/corporatesite/products/mainunits/overview/8396

You are correct, we can only get a data cable to fail when both the laptop and printer are powered from the same battery.

 

[Gryd3]

But both grounds would be tied to each other? But i understand where you're coming from

They 'should' but my concern is with DC-DC converter switching on the positive rail instead of the negative rail.

Now that the OP has told his his device... it looks like another problem may arise.

Jimmy, can you please tell us 'how' you connect the adaptor for the laptop? It looks to be suited for a 12V source... not the 24V source you are using with the printer.

 

[Jimmy Tuppence]

The laptop adapter is connected across the battery (24vdc) the same as the printer power. Let me know if you need more detail on this (type of connection etc).

On the device packaging and on the device itself it states that the input voltage can range from 10-30v (see pic in post #6). I assume that the Maplin website only refers to the device as a '12v ... laptop adapter', as they are aiming at the far more common car user (12v) rather than truck or marine electrics (24v, I think!).

 

[(*steve*)]

with the dc to dc converter disconnected from everything, measure the resistance between the negative input and the negative output.

 

[Jimmy Tuppence]

*steve* thanks for the suggestion.

It measures around 0.1Ω - I say around, it did flicker about a little from 0-0.4Ω but that may have been due to the quality of the multi meter I was using (uni-t ut60e) or my shakey hand!

 

[(*steve*)]

OK, that means the regulator has the topology expected.

I'll think some more on this and I'll let you know if I come up with anything.

 

[(*steve*)]

one thing... Do the usb devices fail if they're only connected at one end when the power is switched?

 

[Jimmy Tuppence]

*steve* the opposite - they only fail if they are connected both ends (USB and parallel) AND both devices are powered from the same battery. I have tried with either only-USB or only-parallel ends connected (the other end left open), and both printer and laptop powered from the same battery and cannot get the data cable to fail by switch power on/off. This only happens when both devices are connected to same battery and data cable connected between them.

Appreciate you giving this some thought.

 

[(*steve*)]

That's kind of what I expected.

Let's see if the sleep fairies bring any insight...

 

[(*steve*)]

When you disconnect the dc-dc converter, how do you do it?

Do you disconnect the +ve lead, the -ve lead, or both?

And how do you do it? With a switch, or some other method? Does it spark?

 

[Jimmy Tuppence]

Originally the dc-converter comes with a cigarette lighter socket type connector. This was swapped to a 3-pin XLR plug/socket - so both the +ve and -ve leads are disconnected and there is no visible spark.