Is there a way to find out if I fried the PMT, the GS-PRO, or both?

[valentin]

Yes, I am an idiot…

My daughter needed a mains adaptor for one of her toys. After her begging for a while I took the universal adaptor that powers my GS-PRO ( see viewtopic.php?f=15&t=89 ), exchanged the USB plug for a barrel connector and switched it to 12V.

Tonight I wanted to record a new spectrum, so i put back the USB plug again, even double checked for the correct polarity. Just in the second after I switched on the power, I became aware that I forgot to set it back to 5V again. Call me an idiot for overlooking this. I am annoyed with myself, because usually I double-check a dozen things before powering the probe, even use an ohmmeter to check for possible shortcuts in the probe every time I change something with the setup. And then such a stupid mistake.

Of course, I switched the power off immediately, but there were already some crackling noises coming from the electronics :-( Lesson learned: never share your most valuable scientific equipment with your kids...

Now, when I switched on the power again (5V this time, for heavens sake), the GS displayed voltage that was about +50V higher than usual. Also, there is a bit of baseline noise in the audio signal, but no pulses are visible. I verified the sound card still works normally.

So most likely this means that either the PMT is broken or some electronics inside the GS has been damaged. The dilemma is how to find out which. The problem is, that I neither have a second probe, nor an alternative driver. Of course, one option is to send the GS back to Steven for inspection or even pay for a new one, but before doing that I would like to exclude the other possible causes.

Ideas and suggestions welcome.

Valentin

[Sesselmann]

Sorry to hear that your GS has gone down, but it's hopefully easy enough to fix it.

First of all it is very unlikely that you have damaged the PMT, those things are pretty robust for a little bit of higher bias voltage.

So the problem is with the GS, and I am 99% certain you have killed either one or both of the opamps, this is really the only fragile component, and the good news is you can change them for around $1 each without any soldering required.

The opamp is called LM358 and I use the same one for the preamp and for the comparator to control the voltage, so you may as well change both.

http://www.ebay.com/sch/i.html?_odkw=LM ... P&_sacat=0

NOTE: Pay attention to which way you plug them in, the notch should match up with the notch in the socket.

I would be happy to repair the GS for nothing, but the shipping would cost way more than the part.

[rodorr]

I agree that the problem is most likely in the GS-Pro but the LM-358 should be able to handle +12 volts because they are rated for +/- 16 volts or 32 Volts single rail. Steve, do you use a regulator and/or zeners in your circuit design? Maybe an over voltage in the high voltage circuit area?

Bob
WA7ZQR

[Sesselmann]

Bob, the GS is designed to be powered from a USB port, so there are no zeners to protect the LM358 only a 22U capacitor to remove the ripple from the USB 5V source. I agree it should be able to handle more than 5V but occasionally they die for whatever reason. The LM358 has +5V on the positive power supply and +2.5V on the inverting input.

[valentin]

Thanks Steven. I replaced one of the LM358AP, and now there are visible pulses again. There is strange noise in the output, and I ordered another IC to replace second opamp as well just to be sure.

Bob: It souded like there were some spark discharges from the HV. So even if the LM358AP survived the 12V, maybe it got killed by some electrical breakdown in the input.

I will keep you updated.

Valentin

[valentin]

Actually, the output voltage is not stable and drifts up and down, so probably the second operational amplifier got damaged as well. Eagerly awaiting the parcel with the replacement circuit...

[rodorr]

Valentin,

Definitely keep us posted with your repair updates. I'll have an idea where to look when I fry mine ( I do things like that occasionally ).

Bob
WA7ZQR

[Geoff]

I feel for you. I just fried a PMT today. My home made detector wasn't fully built yet, just using heat shrink and black vinyl tape to cover it. At some point during a 48 hour spectrum the heat shrink slipped and the PMT burned up.

[valentin]

Yeah, my system is working again, even though it took me quite a while to find out how to fix it.

After replacing both opamps, the spectrometer seemed to work in principle, but overall performance had degraded. Pulses were very low, and there was lots of noise. I assumed the tube was damaged, but eventually I noticed that the background noise was even present when just connecting the coaxial cable with nothing on the other end. My GS-PRO had turned into some kind of radio receiver…

After some "debugging" with a multimeter and with the help from Steven, I found that a 1k Ohm resistor between BNC ground and GS-PRO ground had died. Presumably there was a high current discharging over the detector due to the incident with the over voltage. Fortunately, the resistor was easy to replace.

By now, audio output looks normal again. A first spectrum was acquired and displayed peaks at expected locations. I am feeling happy...

@Geoff: Sorry to hear about your PMT. Guess finding a good one for a low price is a bit harder than getting an 1k resistor.

Valentin