GS-1100A2 teardown and review

The Australian company Bee Research offers the Gamma Spectacular USB powered scintillation probe bias supplies with integrated preamplifier and audio output. With this box, a scintillation detector and the excellent free PRA software it is possible to perform some basic gamma spectrometry far cheaper than obtaining a multichannel analyser, which usually cost thousands of dollars/euros.

I bought one of these from a  Yahoo groups Gamma spectrometry member out of curiosity. What’s in this little box?

Basically it’s a CCFL inverter driven by 5V USB power. The high voltage from the inverter is rectified, filtered and regulated by a string of zener diodes. The big knob selects how much zeners are put in series parallel to the high voltage, regulating the voltage and probably removing some ripple, anything above the zener voltage will be connected to ground.
This more or less works, but the output voltage sags by about 100V when connecting my probe, a load of about 5.6 megaohms. At 900V this means 160 µA of current which really is not a lot. In my opinion, a scintillation power supply should at least be able to supply up to 500 µA (0.5 mA) with ease. I tried different power supplies, a USB charger with plenty of current and my desktop PC but the voltage drop persisted.

Additional information added 29/12/2012:

The voltage problem seems to be related to the Gamma Spectacular expecting higher resistance than the 5.6 M Ohms of my detector. I built the voltage divider according to specs from Scionix  but their voltage divider designs are optimized for stationary setups which often have more than enough power to drive low resistance voltage dividers. “Standard” scintillation detectors often have a load resistance of 60 M Ohms and up.

According to a Scionix employee, a higher total resistance only influences the maximum count rate, spectroscopy should be possible with any voltage divider.

 

I have made power supplies for scintillation probes myself and also encountered this problem. Usually it is better to regulate the input voltage based on the output, using a feedback mechanism.

It is possible to compensate for the voltage drop, just turn up the knob, all the way to eleven if it’s needed. Of course, the highest voltage of 1100V can not be reached, but with my probe it’s no a problem, I usually run it at 850V which the unit can supply.

Unloaded power supply ripple is slight, didn’t see anything significant on my scope.

Pulses from the detector are decoupled using a HV capacitor, preamplified with a transistor and supplied to the output jack. With the A2 model you can either do the decoupling in the unit itself, or use a separate signal cable if your detector has a separate output, just turn the rocker switch on the back of the unit. Apart from positive or negative pulse polarity, it seems to output the same signal to my computer’s soundcard.

Does this thing work? Yes, a Cs-137 spectrum was easily obtained within minutes. Of course, more testing is needed to compare it to my current setup, a power supply using a professional PMT power supply and a DIY audio interface.

So, in short, this box works, but the power supply does not seem completely up to its job. If you want to experiment with scintillation detectors and you don’t have a lot of electronics experience it could be a nice solution. However, if you know how to solder circuits the same or higher performance can be had from DIY circuits like the earlier mentioned PRA_amplifier and CCFL power supplies like these. This is also far cheaper than this unit, as it costs about about $250 but basically consists of parts costing some tens of dollars.

Addition: I made a spectrum of a radium clock after I wrote this article:

Radium clock spectrum


11 Responses to GS-1100A2 teardown and review

  1. Wonderful explanation! I hope everyone gets into Gamma Spectroscopy! We need more isotope detection and less speculation!

    Good job!

  2. Avatar TxeScientist
    TxeScientist says:

    Thank you very much for teardown and review! I was wondering for a long time what is inside 🙂
    Best regards and keep up the good work!

  3. Avatar TxeScientist
    TxeScientist says:

    Are those 3 HV capacitors voltage multiplier x3? CCFL module works from 12V, so try to hook it to separate power supply. HV multiplier limits output current. To increase current and voltage output> add separate 12V power supply to CCFL module and get rid of voltage multiplier. CCFL modules can give up to 1600 – 1800V. Are low bias current zeners used? Some require several mA to work, so they can drain quite power. Hope this helped. Also see this link for regulated PSU> http://charliethompson.50megs.com/LowRippleHVsupply.html

  4. Txe, I did open it to get some idea of the circuit workings. The HV capacitors seem to be used to form a basic RC filter. Of course, the zeners also reduce ripple on the output voltage.

    Modifications of the circuit to provide more output current should be possible but I have to look at the circuit more closely to see what can be improved.

    I have also thought about circuits like these in general as stable, low ripple PSU’s are really useful for the gammaspectroscopy hobbyist. The best of course are those NIM bin HV PSU’s available on eBay at price. Those are similarly priced to the Gamma Spectacular units and you only have to build a simple preamplifier to hook everything up to a computer.
    Maybe I’ll just buy one of those supplies to see what is inside. I guess some HV generator with a feedback loop, some very accurate reference voltages and temperature compensation applied to all sensitive parts of the circuit.

    The power supply in the URL also seems interesting, quite a chance that I will build it one of these days in the search for clean HV for a decent price 🙂

  5. Avatar Roberto Cena
    Roberto Cena says:

    Theremino Geiger & MCA kits

    You can build or buy a sensible and cheap #theremino Geiger or MCA hw kit. Look at http://www.theremino.com/contacts/price-list

    The #theremino #radioactivity modules have millivolt ripple. Look at the elusive Potassium K40 in the gamma spectrum http://www.theremino.com/blog/gamma-spectrometry/images

    Thanks for your attention,
    Roberto Cena

    The Theremino team

  6. Any chance you developed a schematic off this? I’ve been trying to run across one. Or would you have a schematic diagram of any similar devices? I’ve wanted one of these for some time but don’t really have the cash for it, so was doing research tonight on how I might build one… starting, of course, by looking for the circuit diagram. But the other ones you mention sound good too. Any ideas would be helpful. Of course, then I have to find a good scintillation crystal too… 🙂 Maddening hobby, this, eh? Picked up a Gamma Scout because of Bionerd23’s videos on youtube and recently got a real good deal on an Eberline ESP-1 (without scintillation crystal). But it seems to work! Anyway, great site. Enjoyed this article! All the best!

  7. Take a look at the Theremino PMT adapter “spammed” in the comments here. It’s far cheaper than the Gamma Spectacular.
    These devices are nothing more than a adjustable, stabilised HV supply and an amplifier/filter for the pulses. My first spectrum was with a self built power supply and a small amplifier/filter schematic:

    https://secure.flickr.com/photos/julietmikebravo/6932000074/in/set-72157625600003560

    You could also mod your Eberline, if the HV is stable enough you can decouple the signal with a capacitor and a 100K resistor, run it through a lowpass filter and directly connect it to a PC for analysis.

    If you need more info just ask me.

  8. WILL THIS WORK WITH A CdZnTe detector?

    • As far as I know diode radiation detectors require bias voltages up to tens of volts, not the hundreds of volts of photomultiplier and geiger tubes.

  9. Nice review..

    Everyone will be pleased to know that the new GS-1100-PRO and GS-2000-PRO will soon be hitting the shelves. These new models have been completely redesigned, and use an internal precision voltage reference source, to stabilise the output and are variable from 0 to 2000V, with virtually unmeasurable ripple.

    See the preview here

    Steven

  10. Steve, normally I would consider this spam, but since I do a lot of sound card spectrometry 😛

    I wouldn’t mind also reviewing the 0-2000V model, is it also available with BNC or is this undesirable at higher voltages? I am also in need of a better sound card PMT driver which has higher voltage and lower ripple. I currently have the GS-1100A2 which does have some ripple and noise, signal to PC is also a bit low.