Wednesday, March 23, 2011

Geiger Counter Kit - Part 3

Please visit this web site if you want to purchase or learn more about the Geiger kit I am offering. 

I'll leave the pics and video here, but you'll find all the details about the kit through the link above. However, you can leave a comment here if you like.

Geiger Kit PCB  . . .

The video gives a pretty good introduction to the GK-Plus


  1. I will buy one of your boards, I ordered a SBM-20 from eBay a week back, still waiting.

  2. good work.... myself also working on same project i use pic18f2550 and i am struggling with 500v power supply for the gm tube. i have used 4khz PWM (50%)... :-|

  3. Depending on the current drain, which I expect is very low, I'd prefer that you design a unit with the option of being powered by a single 9V battery. One of your earlier schematics had that, the one without the microcontroller. I assume you just wanted to avoid the extra components for regulation down to 5V for the microcontroller, but the 9V option is more convenient for battery changes because there are cases with snap on/off 9V battery compartment lids. It would be nice if at least the option was given on the PCB for 9V power with the locations for regulator parts left unfilled in not desired.

  4. Seeed Studios is an excellent, extremely cheap source for PCBs. I've used them once and the turnaround time was a month from Gerbers emailed to receipt in the US with majority of the delay being postal transit time.

  5. Thank you all for your comments and ideas. The kit is moving along nicely.

    Re the 9V option, you make a good point about cases that include a 9V battery. I think your suggesting:

    9V to HV ckt, optional regulator to ATmega. I'll look into doing that. I should have room for a TO92 style regulator on the right side of the board.

    I'm thinking about using IteadStudio for the PCB. I'd like to go with Yellow boards. (I hope the silk screen is black.)

    I'll also have a small quantity of SMB-20 tubes available.

  6. "Re the 9V option, you make a good point about cases that include a 9V battery. I think your suggesting:

    9V to HV ckt, optional regulator to ATmega. I'll look into doing that. I should have room for a TO92 style regulator on the right side of the board."

    Yes, that's exactly what I was talking about. Thanks for considering it. I think it would be a great option.

  7. OK, I added a 78L05 but I couldn't add it for the ATmega only. It had to be for the whole circuit because the hex inverter is 6V max and I'd have to level shift for the Arduino.

    Since regulators burn mA, I'll recommend replacing it with a jumper if your using less than 6V input. I think that should work.

  8. I found that using 1/4" fuse clips worked well for mounting a SBM-20 on a PCB. Something like:

    In my case I had to file the leads down a bit.

    Thanks for the circuit, I've just finished a PIC variant on veroboard, with a MAX232 and bluetooth. Your schematic was a great starting point.

    My understanding of 555's is a bit limited, but to get my charge pump going I had to drop the 3.9KOhm resitor to 1.5KOhm and kill the link between pins 3 and 7. Not sure if that's something to do with the MPSA42 I'm using. It is however producing 200Vdc at the test point, so *shrug*.

    Finished build:



  9. "and kill the link between pins 3 and 7"

    Note sure what you mean by that. Pins 3 and 5 of the 555 are not connected to each other.

    I had some ideas on the clips, but those look better - thanks.

    Nice job on your build!

  10. CuddlyDragon - Thanks for the specific AG3 fuse holder part number. I've read that soldering to the tube is not a good idea because it could break the vacuum seal if you apply heat for too long. A quick tacking can work because I've seen it successfully done, but it's better to use clips.

    I've seen a SBM-20 tube diagram elsewhere that indicates that the contact at the end marked with + is the positive terminal (of course) and the tube body is ground. Is the contact on the opposite end from the positive contact also a ground?

  11. Yes, the other end of the tube is negative. (See pic in "Part 2" below.) You definitely shouldn't solder to the tube. (I wrapped the thin wire around the terminals, and soldered the wraps together. I am planning to include the AG3 fuse clips in the kit.

  12. I know I could just pop your .brd file into Eagle, but I'm not on my machine with Eagle installed so I'll ask for the dimensions of the board so I can have a case ready for it. Don't see any mounting holes, so what's the plan for mounting?

  13. Nice work BroHogan,
    Could you post a link to your brd/sch files like you did in Part1.

  14. The dimensions of the board is 1.6" x 3.6" (4.06cm x 9.14cm). The #4 screw holes are centered .10" from each edge - see updated board image.
    I am only posting images of the board and schematic at this time.

  15. Great! Thank you very much.

  16. Hi! I live next to Fukushima.

    We're still OK right now but the situation is still serious.
    Now I am trying to build geiger counter but your circuit seems to be much better than mine.

    I am trying to gather some parts to make your circuit but I hope to purchase your kit when it will be completed!

    Anyway, Thanks a lot for your giving precious imformation!

  17. 'Eddy-san,
    Thank you for your comments, and I'm sorry for what your country is going through.

    I already have all of the parts. In your situation, I would be happy to send you a set. If that would help, then send me an email at my contact on this page.

  18. Will the ATmega come preloaded with software and is the LCD included in the kit? If not, which LCD from Sure Electronics should be used?

  19. For SW, I'm planning to pre-load a simple sketch that outputs CPM to a 2 line LCD and also to the serial port.
    I'll use the LiquidCrystal library, so any HD44780 compatible LCD will work. LCDs like Sure's 1602, SeeedStudio LCD104B6B are just examples. In short, any LCD you can run from the Arduino will work.
    Hope that helps.

    I did not include an LCD with the kit. I figured everyone has different needs for display - size, # of lines, etc. - (some don't even need a display) so your better off selecting your own.

  20. Great kit!
    I am thinking that the code from libelium should work with minimal changes (pin numbers, basically) on your Geiger counter.

  21. Thanks Florin,
    Actually, I think the libelium code will even work as is - we both use the same INT and LCD library. But I have certain features in my sample code I'd rather keep. One is that the sample time is not fixed, but varies based on if your are getting just background counts, or high counts. That way, changes in high counts display faster.
    Definitely want to look into the CMP to Sv conversion they came up with though.
    I posted a rough version of the sample code at:

  22. Just mentioned your kit and the Cooking Hacks one in the Yahoo GeigerCounterEnthusiasts group and provided links, so your hits should go up.

    Please don't sell them all before this wbblair3 person gets one. ;-)

  23. On the 14th, you said you expected to have kits available "by this weekend" which was the 16th-17th. It's now a week later than that and no word. Problems?

  24. Sorry to have kept you in the dark. No "problems" other than overestimating the mail system and underestimating the demand.

    I will post an update here later today, but briefly, the status is this . . .
    I received a small batch of PCBs this weekend. I built one out and it tested fine. (see updated video) The rest were already spoken for (some to Japan), so I am out of the first batch of PCBs already. I "hope" (same word I used last time) to receive more PCBs by 4 May. More info will be in the update.

    Please bear with me. It's not easy getting something like this started.

  25. How does one reserve boards from the next batch? I don't see an email link to you. Perhaps the best way to get an idea of how many PCBs you need to order would be by taking prepaid kit orders?

  26. My email address is under "Contact" in the upper right side of the page underneath "Labels". If you contact me I will reserve a kit for you.
    There will be more about reserving and prepaying when I update this post. I was hesitant to take prepayments until I had walked through the whole process - from supplying the components to printing the postage. I'm finishing that up today.

    So please contact me to reserve a kit or get instructions on how to prepay one. I thank everyone for their patience.

  27. Hi Bro,
    congratulation for the integration of the counter and thanks for sharing the project.
    I have 2 question: One regarding how to get 1300V by HV supply for my photomultiplier. 2nd how to set the efficency of different probe for 12000CPM/1mR/h (RAM63-3) to read correct uSievert value. Many thanks. Daniele

  28. Got the full kit and it works well , no problems at all.

    Now onward to the "Experimenting" fun ... want to see if I can get it to work with a LND 7616 GM tube , just need to get the HV up to about 760V ... then its up to 900V for the old CDV-700 tubes.

    Now if the board only had the USB parts to code the Arduino in circuit my lazy butt would be overjoyed.

    Good job on the design , I give it 5 out of 5 stars.

    Think I'm a Guest/Anonymous but the real name is Lew Morris ... I'll have to make an account here now.

  29. Daniele,
    I apologize for missing your post - especially since I clicked on your name and discovered that you work with extremely beautiful women. Can I hang around with you? Probably not because I have no good answers for you. :-(

    Lately I tried a little bigger inductor (15mH instead of 10mH) and saw a good increase in HV - maybe 800V, but even going bigger I don't think it will reach the 1300V you need for that tube. I think a HV circuit based on a "voltage multiplier" is what you need.

    Your second question is also difficult. I use a conversion from CPM to uSv that is popular now for the SBM-20 tube. Exactly how you convert the specs on the tube to this factor I am not too clear on. Maybe you can find someone who used a standardized radiation source with that tube.
    Sorry not much help, but if I was taking the photographs that you take, I'd never leave the studio!

  30. Lew,
    I am glad you liked the kit. Thank you!
    As I mentioned in the comment above I think going to a 22mH inductor might work for those tubes. (I am going to try it too.) I will PM you with more details.
    I wish there was an easy way to add USB without a FTDI board. There are no through hole FTDI chips and the SMT chips available are really hard to solder. So the FTDI port on the board was as far as I could go.

  31. John,
    I received the full kit and it is working great!! Thank you for your hard work on this project and for making the kit available. Building the kit turned out to be quick and easy. You can see my implementation at

    The only problem I encountered during assembly was the 555 IC did not oscillate properly. This was easily identified using an oscilloscope combined with the fine documentation provided on the blog. A replacement 555LP I had on hand quickly solved the problem.

    As suggested, I obtained the FTDI Breakout Board from Sparkfun to use with this kit. Adding the FTDI port really makes programming changes a cinch - What a great feature! Plus this makes a great battery saver when using the counter at home or in a car with a USB charger cable.

    Congratulations and thanks for a job well done!

  32. Thanks very much Tom. That's great to hear. I'm glad you find the FTDI connector handy. If I have the right guy, it was you who suggested it many moons ago!

    Unfortunately, I couldn't see your pics on Picasa - maybe they're not shared?

    If you want you can eamil me some and I'll put them in the Gallery on the Geiger Kit site.

    Thanks again for your great review.

  33. Hi John - Yes, nice memory on your part, you have the right guy.

    I did correct the sharing setting on Picasa, so my build photo album of your DIYGeigerCounter kit can now be viewed:

    Please feel free to use any of the pictures for your gallery if you like.

    I continue to enjoy my kit. This quality kit deserves the best review possible IMHO! - Tom

  34. Got the kit and built it, works fine. This is one of the best geiger circuit boards around I think. For the price it delivers like no other: you can use different tubes, connect an LCD easily, reprogram it to suit you specific needs, get into datalogging, etc.
    I am now in the process of putting it in a nice box with some buttons. I will later modify or write a new program for adding additional funtionality ( different time scales for detecting low levels just above background).

    I do have one question however: I prefer "clicks" over beeps, is it possible to modify the circuit to achieve this?

  35. Johan,
    That is nice to hear.

    For the clicks, I think a speaker might make better clicks than the piezo. Perhaps something like this:
    It's 45ohms, so 110mA @ 5V. IC2 max is 50mA, so a resistor in series.

    However it may be easier and better to drive it with a transistor from an Arduino pin with a small change in the SW.

    Just some ideas.

  36. What I meant was not using a tone but a simple pulse for a classic geiger counter sound.

    It's possible to pulse a pin with each event but that involves putting commands in the interrupt routine which could interfere with counting large amounts of interrupts per second.

    Anyway, I can always look into this later, first I need to connect buttons and get some basic interface going.

    I want to implement storing custom CPM/uSv/hr rates and longer integration times with rolling averages.

  37. In case someone wants inspiration for the enclosure, here is some:

  38. My inductor in kit broke due to overheat.
    Is it the same as in kit ? 13R156C

  39. Daniel,
    That's amazing that the inductor would overheat!
    In any case, yes. That is the inductor I use in my kits.

  40. I used solder paste and heat gun. Powder core cracked. Plastic coating was untouched. Crazy thing. Probably due to high dT/dt.

  41. Hello,

    I'm thinking on using a SBT-10 GM tube ( witch seems to be an alpha capable tube. Do you think I should change in any way the values for Resistors R6, R7, R8 or R9?.
    Pedro Moreno

  42. Wow - interesting tube. It's made as a "platform" for analyzing foods.

    Per section, it has similar specs to a SBT-9 I have, and it runs fine with the stock values.

    The anode resistance is not specified. But a working current is.

    Does working voltage / working current = anode resistance? Hmm that would be ~400M, but it assumes the tube had 0 resistance. I don't think that's right.

    I'd start with an additional 10M right at each tubes anode just to be safe. R8 & R9 should be OK as is.

    I will be interesting to hear how you make out.

  43. I'll post my findings as progress in made

  44. Eventually my SBT-10A arrived and after connecting each of the ten individual counter section to a 10Mb resistor, the counter went nuts as the digital counter was showing more than 10000 counts per minute. So disconnected all 10 counter sections and reconnected one by one to verify witch one was the offending section. Some counter sections had a background between 150 to 300 CPM but five of them had a whopping background of 5000 CPM or more no matter what the working voltage was used (from 100 to 430v).
    I have contacted the seller to request a replacement or alternatively my money back ($118).

  45. Humph, that's too bad. Maybe the part is contaminated. Or else it's the reason it was removed from the original equipment in the first place. Personally, I have a good opinion of that seller, so hopefully he will make good on it.

    Since it's an alpha tube, I wonder how consistent each segment is with alpha. If you have an Am241 pellet from a smoke detector you should try that.
    Good luck with it, and let us know how it turns out. It's an interesting part.

  46. No radioactive contamination detected in the GM tube. Seems to be a faulty part.

  47. Ok but make sure you keep your cathode lead short just to make sure you are not picking up and counting noise.

  48. I thought you should know...
    The 'Overview of the Kit' page on your site has missing images.
    The pics that are missing follow the following statements... "The kit looks like this . . ." "The current revision of the board looks like this . . ." "This is what the assembled board looks like . ."
    I am very interested in this kit and would like to see these images. Thanks.

  49. Hmmm, I cleared my cache in both Chrome and IE, and still both get the pictures.
    I wonder if others are seeing this?
    Can't think of what the problem could be.
    Thanks for letting me know

  50. Finally the replacement for the faulty SBT10-A tube arrived and after connecting it to the arduino the counter shows a 160/190 CPM with all the segments connected to a 10Mohm resistor. The anode voltage seems to be critical for this tube as it creates avalanches if anode voltage is over 370v. Care must be taken in the voltage calibration as most voltmeters will show erroneous reading due to intrinsecal impedance, so in my case the adjustement was made to 330v (digital readout) but the real voltage without voltmeter load was 370v.
    In my case this tube detects 2400 CPM for a lantern mantle at 10cm from the mica window without alpha blocked and 690 CPM with alpha blocked.
    I have also a SBT11 GM tube and will test this as well when have some spare time.

  51. Pedro, That's great to hear. Thanks for getting back to us on this interesting tube. I've had several people ask if it would work with Geiger Kit, and now I can say someone has it working.
    I posted your comment on the GM Tube Info page
    Let me know if that's not OK.
    Thanks again for the info on this tube.

  52. Just bought this kit and assembled a counter. A counter itself works pretty well, but when I plug
    Rx/Tx to COM-port (9600) I got one or two lines:
    "CPM uSv Vcc" and that's all.
    Where the problem could be?

  53. Sorry, false alarm! I was too quick :)
    Now I can see that:
    "CPM uSv Vcc
    94, 0.5358, 4.42V,
    902, 5.1417, 4.40V,"


  54. By the way, how to build the sketch? There is no LCD header ans library:

  55. @vitasam,
    The LiquidCrystal lib is included with the Arduino IDE environment. This is what most use to compile and download the sketch.
    If you want me to send it to you, contact me using the contact page on my web.

  56. @BroHogan,
    ah, it is a part of Arduino IDE, maybe I will install it. Do you know is it possible to use that IDE with Atmel STK500 development board directly?

  57. It's a C++ library that's included along with the IDE. I have no idea about using the Atmel STK500, but I bet you'll find an answer on the Arduino forums.
    The Arduino IDE is easy to install, free, and easy to change and upload programs to the chip. I have instructions on getting started on the Software page on the DIY Geiger web.

  58. Got my kit last week everything works fine other than the d2 diode on the v3.0b board is backwards compared to the picture in the pdf v3. Make sure the band is on the top not the bottom.

  59. Glad it's all working for you. Can't understand how you got that impression about the band. All pics show the band doing up (if you can see it). Only diff is that in v3b the lead comes of at 90 deg. from V3.

  60. In the "GK Build Instructions v3.pdf" the finished picture the one with the tube on the top.The d2 the band is on the bottom facing towards the circle. You can't see it on the ver2 board but the 1.4 board pdf was set corrctly. It didnt help when it said "See pictures for orientation." But still all is fine and i would recommend your kit. Thank you.

  61. Might have just been glare making it look like a band at the bottom too:)

  62. Trying Geiger sketch in Arduino 1.0. Got the message:
    "As of Arduino 1.0, the 'BYTE' keyword is no longer supported.Please use Serial.write() instead."

    How to solve it?

  63. Vitasam,
    Arduino 1.0 breaks lots of sketches. I haven't yet looked into changing the SW for it.

    You could try replacing byte with unsigned char in the definitions - that might get you further, but for now stick with REL 022 of the IDE.

    I should get a chance to look into it this weekend so I'll know more. I only hope the libraries I count on for display, etc. still work!

  64. (2 hours later ...)
    You can now download an IDE 1.0 compatible version of the default sketch from the Software page on the Geiger kit site.
    The changes were not difficult - unlike the changes needed for the Logging Shield and the other stuff I'm working on. :-!

  65. By the way, just a curiosity - why you're not using any GIT hub (e.g. Google Code) for keeping a SW development there?

  66. I expect to do that at some point.

  67. Thank you for sharing.I did your porject and I succeed to run on bread board.I have a geiger tube(CTC-5)which is similar shape and size with sbm 20.(Working Voltage 360-440V Initial Voltage 280-330V)
    Nevermind I gain 210-240 Vdc at test point and can't gain much more how can I do gain more voltage or it is enough to operate 210 vdc?.

  68. Good to hear.
    If you are using just a regular meter 240V means that you are probably actually getting ~450V. See the HV Test & Adjust page on my site.

  69. Hello,

    At first thank you for the nice projekt. Greate work.

    I build the geiger board, it works nice, but I noticed the power consuption is approximately 50mA (whole bord 555 + schmidt trigger + atmega + LCD).
    HV-Part and Schmidt Trigger needs appr. 12mA
    Atmega +LCD(without backg. ligth) needs 40 mA.
    It means 9V Battery is empty after few (3-5 hours) time.

    My questions are:

    1. What is your power consuption? Of your board? What are your values?

    2. How can I lower the power consuption of Atmega? Lower clock-frequency?


    1. The last time I checked the HV & Schmidt used 4.75mA on a 5V batt and no reg. Running on 9V with a reg is a waste.
      Not sure if your talking about a kit or you built your own.If you built your own the 555 must be CMOS and some varieties of the Schmidt use more current.

      But it's the uC and display that uses most of the current. Having a resistor for your backlight would help. For the uC you could run at a lower voltage, use sleep, etc. Best advice would be on the Arduino forum.
      Again, I don't recommend using a 9V batt if you can avoid it. They only have ~600MaH. (If yours goes dead in 5 hours you are using ~120mA)

  70. Any chance you can do expedited shipping to NYC? I just emailed you about my particular situation...Sorry to cross post here- I'm just trying to make received and build the kit before I leave on an upcoming trip- many thanks for your help!

  71. Very slowly my kit is progressing, yesterday I've connected a display:
    What does represent the vertical line above "1" in "0.21"?

  72. Hmm, found it: "The program ("sketch") that is pre-loaded on the Atmega328 has evolved to provide good functionality for the average user. Basically this "default sketch" now provides: a readout of CPM and uSv/h every 5 seconds on the LCD display
    a bar graph on the display showing CPM level, or the voltage at the CPU when the battery is low"

    1. Thanks! LCD will be installed to the cover of the project box (do not have an idea about protection glass on top of it). TTLtoRS232 adapter is coming:
      This 'big brutal' Russian programmable calculator' is going to be connected as a field logger (with date/time stamps): :)

  73. "...a bar graph on the display showing CPM level, or the voltage at the CPU when the battery is low" - how do I know, is this bar now a CPM level or a battery level?

    1. When the voltage is low (<4.2V) the entire bar graph will be replaced by text. Something like "Vcc=4.1V"
      Feel free to email me if you have other questions.

  74. I just completed the Geiger board and data shield with one of the recommended LCD and GM tubes and every thing worked perfectly. I am not a serious electronics hobbiest (geologist/hydrogeologist) but can solder and use a multi meter and found this to be a thou rough and well presented kit.

  75. Is there any chance to obtain higher voltage from this kit? Say up to 1200V. Or it has to be completely redesigned?

  76. See the [Jan 2012] note on this page:
    I am not sure if excessive ripple is produced with this mod though. As the note says, you're on your own ;-)

  77. Just completed V4 board in a couple of hours. I had a LND-712 that I pulled out of an ancient Aware RM-60. Fired it up without any HV adjustment, and it is getting the usual ~20 CPM background that I expected. Plugged in an LCD display and only had to make an adjustment for the contrast. I do need to adjust the D/R part value that drives the LCD backlight, thats all - nothing but praise for the quality of the kit.

  78. Hi,
    after long period I came back to my assembled device (
    Also I've upgraded my PC to Win8, don't know yet, was it a good decision, but when I'm trying to upload sketch (7.1 or 8.0) from IDE 1.0.2 I have following message:
    "Binary sketch size: 10 654 bytes (of a 32 256 byte maximum)
    avrdude: stk500_getsync(): not in sync: resp=0x00"

    Any idea what is wrong? COM port in system seems to be Ok - I can see the Geiger log in a terminal.

  79. Vitasam,
    There is a lot on that error if you Google.
    You will see lots of Arduino forum threads.
    Hopefully you are just selecting the wrong processor or something like that. It's not a SW problem.
    Good luck, I'm sure you'll find it.

    1. Is it possible that I need a new bootloader in Atmega for new IDE?

    2. I don't think so.
      In your Board menu you should probably be using "Uno" or Due..w/328 depending on when you bought your kit.
      I don't want to start a big dialog here, so PM me if you have additional questions.

    3. Ok, problem was with settings. Board 'Duemilanove w/328' must have been selected - it works now from Win7 and from Win8.

  80. Hi guys,
    we are testing the radiation board by Libelium and just found that they use voltage 160 Volts to power the tube (J305ß or vintage SBM-20) although according to the tube datasheets it should be much higher.

    Do you know any information source which is able to relatively simply explain these terms:

    Operating Voltage Range (volts) 350 - 475
    Initial voltage (volts) 260 - 320
    Plateau length (volts) at least 100

    does it mean that we should try to set the power supply to something between 260 - 320?

    I know, there should be some relationship between sensitivity to low dose rates and voltage - that using lower voltage should make the tube more sensitive to low dose rate values.

    Did you make any tests in this case?


    1. If you are just measuring the HV with a regular meter you will get very low readings like that.
      You need a very high resistance on your meter. Not 10meg (typical for most meters) but at least 100meg or even 1000meg.
      See this page . . .

      About 400V is about right for that tube.

    2. Hi, thanks for reply. We discussed it with the support and the voltage of the board should be around 380V, so it would be OK for the. SBM-20

      However, we tested it with Ra-226 source and got strange readings. The chinese J305 tube was measuring values from 117 to 130 microSv/h, while the SBM-20 after few periods (had 10 second intervals) and few values like 120 microSv/h sent values like 0.05 microSv/h, while the chinese tube continued measuring OK (same distance, same source and same electronics).

      Do you think it is possible that the SBM-20 was saturated? According to the GSTube datasheets it should work up to 1400 microSv/h. Strange ...

  81. It's impossible to tell from here.
    You should contact Libelium if that is the board you are using.

  82. I've managed to connect DIYGeiger to my Russian programmable calculator MK-161 :)

    Sources are here:

  83. Hey that's cool! Nice work.
    I'd be happy to post your links in the Gallery
    if you like.
    Just PM me if you do.

  84. Hi,
    just out of curiosity - do you have any plans to migrate to low-power TI MSP430 family?

  85. No, no plans though I might switch to the ATMEGA32U4 at some point to gain real USB.

  86. Hi guys,
    just curious. Do you have any idea what detector could be used to measure gamma radiation in a tiny, flash disk sized usb detector like Polimaster RadFlash? I am just curious as I do not know such small GM tubes and the detector is in price range $768 - 900 USD / Piece so there cannot be any hi-end detector. Any ideas what could be used?

    PS: is it clear that the manufacturer would not give me such information :-)

  87. The specs do say it's a GM tube.

    and they also make a watch, but I haven't seen any GM tubes that small. I'd guess it's specially made.

    Maybe someone else knows.

  88. Hi guys,
    is it possible to either (i) switch HD47780 LCD display off or (ii) put it in to any low-power mode from SW?

  89. It's not hard to add a switch to turn the backlight on and off. That uses 90% of the display power.

    How you do this depends on model of kit you have, so I can't give you specific details. You can PM me if you have additional info or questions.

  90. I built the kit some months back and it works
    flawlessly. Just one suggestion, though.
    How about a variant kit that would work with
    a Nokia PCD8544 display, rather than a 16x2 LCD?
    The Nokia displays are common and cheap (about $5)
    and display 6 lines of text and/or graphics.
    Yeah, and the sketch would, of course, have to be

    1. That's certainly worth considering. I've worked with them before. There are pros and cons. The pros are obvious. One con is I found them to be somewhat delicate - it's more the board they are attached to. It would be nice to have a running graph, and other things a graphic display would bring though.

      I'll have to dig one out and write some code and see what it looks like.

      Nice idea. Thanks!

  91. I received the full kit and it is working great!! Thank you.And I need feed data to from ethernet shield. Are you have sample code.
    Thank you.

  92. Good to hear the kit is working great for you. (I don't recall shipping one to Thailand, but . . .)

    I never got a chance to interface with but it's been on my list for a long time. If you do it please give me the details and I will post it on the web.


  93. Just finished my Ver 4.0.2 board yesterday. Besides one mistake I did which cost me a couple hours of troubleshooting, my board works great. I installed Q2 incorrectly and was getting like .8 volts at my tube. I looked at the picture and caption on top in the PDF, however I didn't notice that is was supposed to be installed straight across. I bent the center pin backwards. It took another 20 minutes to correct this mistake. Lucky the transistor just pulled straight out no problems when heated. Putting it back in the holes was the tough part :/ Pay attention to Q2! lol.

    What sucks now though, is literally like a week after I built my kit, the newest version GK-B5 came out with a speaker instead of piezo. I was upset, but I'll live with it. :(

  94. I agree that mounting Q2 was confusing on the v4.0.2 PCB. The board was made to support 2 different footprints for the HV transistor because I had problems with the availability of the STX. I did try to make it as clear as I could in the build instructions, but I know people don't always read the instructions in the order I expect.

    As far as the speaker goes, I had to make the change at some point. However, there is a mod to replace the piezo with a speaker for previous PCB versions. The mod is described on this page . . .


  95. I just received my kit today.I re-read the instructions, and then spent about 2 & 1/2 hrs building. No problems at all, except when I dropped a resistor and couldn't find it! Fortunately I had a spare. I checked it first minus the LCD and tube, I got a bleep and a flash when I switched on. HT was fine according to my 1G HV meter. I added the LCD, adjusted the contrast, all fine. Added the tube (SBM-20) and straightaway got my usual background count. Agreed well with my Nuclear Enterprises RDM1.
    A truly great kit. I have built hundreds of electronic kits over the years, and this has to be one of the best from the point of view of design, instructions and quality of components.

    1. Wow, thanks Steven. That's very nice to hear.
      I go way back to Knight and Heath kits myself and remember the magic in building them. Have fun.

  96. I finally received my SBM-20 Geiger tube today from Ukraine. It seems there are no US sellers at all. Hooked it all up, and it works great! Best kit I've seen out there. I absolutely love it. I don't know if my readings are accurate, but this baby picks up great. Extremely sensitive compared to my CDV-700's. Can anyone tell me if these readings sound about right:

    Americium Pellet
    ~3500 vs ~900 for CDV

    Thorium Mantly
    ~3600 vs ~1000 for CDV

    Fiestaware Saucer Plate
    ~12000 vs ~1300 CDV

    Is this because the SBM is nice and open? This was tested doing direct contact with the tube and my CDV with the Beta shield open. Its a smaller area to measure also on the CDV. Can anyone else confirm these readings? Thank you

  97. Daniel,
    Thanks for those kind words. Two good review in a row! :-)

    I have a chart on the readings I got with the SBM-20 on the GM Tube Info page of my site.

    The Am241 reading seems a bit high, but the mantle looks about right.

    Sometimes older tubes will avalanche. In that case it can sometimes be overcome by adjusting the voltage and/or the load resistor.

    I'm not saying this is the case - just FYI.

  98. Just an update everyone, working on my first custom case. This is a new type of project for me, but I have to say its coming along well. How does everyone hold everything in place in their cases?? I have some nylon spacers I hot glued in place for my screen, and I plan on hot gluing the board and gieger BUSS fuse holders in place.

    I ordered a second LCD screen also so I could run the headers slightly different. Is it okay to put a switch in the LED power cable to cut the back-light off if it isn't needed? (because when the processor section is off, my screen still stayed lit)

    I have also successfully modified my board to run with a speaker. Thank you!

    1. Daniel,
      I can't speak for the case (although I have used hot glue myself).
      The best way to turn the backlight on and off is by cutting that blue resister on the right side of the LCD. Cutting the left most pad will leave the resistor in the circuit. Otherwise the resistor can be put next to the switch.

  99. So if I cut the blue D/R resistor, the LED screen will still function, and then turn off if I pull the jumper to the micro controller (UC_PWR) ?
    I'm a lil confused as the how you mean this. If possible, could I just use the same switch for both the micro controller circuit and the backlight, or would this not work due to voltage differences?
    I have 4 switches at the moment ready to be wired up. (one for Power to the terminals, one for the speaker, UC_PWR jumper, and the last one is up in the air for LED backlight or the INT_Jump) I will send you pictures to your email of my build once its completed.

  100. Daniel,

    This is a hard way to chat. Please email me at the CONTACT address on the top right of this page.
    On the D/R resistor, basically you cut it and put a switch across the cut. Turning off the MCU with the UC-PWR jumper will not power off the backlight. You can power off both with one switch but it's trick to explain.

  101. Hi, I would probably buy your kit, just I want to know how to connect it to my arduino with sd card on it? It needed to storage all data their, on sd card. I am new in arduino, just how many wires and in what pins would be going to arduino? Thanks!

  102. Hi, The Geiger kit is an Arduino. There is also a Logging Shield for it described on the website. It has a SD card and RTC.
    If you want to interface to a additional Arduino with an SD card, you can use the INT on I/O pin 2 to interface to your sketch.

  103. I just built the geiger kit and used my SBM-20 GM tube. It was an easy build and worked first time. The circuit is well designed. I breadboarded a few others geiger HV/clicker circuits, this one used the least current. GREAT PRODUCT!!!! Highly recommended! Sincere thanks!

  104. Hi, I built the geiger kit (V5) and all is well (see above comment). I read in the online documentation "Regarding the conversion from CPM to uSV/hr... To handle two of the most common tubes, the SBM-20 and the LND712, I allow for 2 conversion ratios and a jumper that determines which will be used." I do not see this jumper on my board... what am I missing? Thanks for a great kit and a LOT of work!

  105. Hi
    Glad to hear the kit worked out well.

    There isn't a jumper on the board, but all you need to do is ground I/O pin 9 to switch to the other rate. (You will see that rate displayed when you restart.)

    I'll make this more clear on the Software Features page.

    Hope that helped.

    1. Thanks! Another question... Does the windows setup program designed for V9 work with V10.1?

      Thanks Again...

  106. It should work with v10 and v10.1, but I haven't tested it.

  107. Thanks, I'll give it a try.


  108. Wow, I really like the tone mode!!! Thanks for the work I'm sure it took to put it in the software.

  109. Hi, the above 5 anonymous posts are mine. I just ordered another kit. I'm using the first with an LND pancake GM tube. I have had trouble getting the infra-red mode to work. The remote is the one recommended with the "8201" code recommended and it works well on my Sony TV. I put a scope on the geiger board and can see a nice signal from the remote on pin 26 of the arduino, but it does nothing to invoke the menu. Since the signal is reaching the arduino, I'm guessing it is a software switch. IR support is true in the sketch... any chance there is a library setting I could tweak?



  110. (I am supporting this through direct email.)

  111. This an Octave visualization from 5-hours log file, recorded from my Geiger-MK-161 set:

  112. Another question: is this a good tube for the kit?

    1. Nice graph! It looks like that Octave sw can do many powerful visualizations.

      As far as the SI-39G tube goes, yes it will work,but I think you will find that they are not very sensitive. In general I stay away from glass GM tubes but there are exceptions.

    2. Yep, Octave is a very powerful opensource thing. My scripts are here:

  113. What is a maximum current from battery when only analog part of Geiger is assembled?

    1. Without the microprocessor and display the kit draws ~20mA

  114. Occurred to me this display might be an easy swap for a cool feature: the ability to have the unit indicate alarms or other situations, directly on the main display, through changing the color of the display. There are 2 versions of this display: black on RGB and RGB on black.

    Without rewriting the code, you could have the discrete led drive one of the display LEDs. I imagine using the count / click LED to drive it would be unnerving. :-)

  115. I went ahead and ordered one of the RGB on black modules ($13.45 PPD) and will let you know how it goes.


    1. OK cool!

      One think you will need to contend with if you are using the LCD header are the last 4 pins.

      Right now:
      pin 15 = bcklight Vcc with resistor
      pin 16 = backlight ground

      You need:
      pin 15 = bcklight Vcc NO resistor
      pin 16 = RED bklt gnd with rstr + control
      pin 16 = GRN bklt gnd with rstr + control
      pin 16 = BLU bklt gnd with rstr + control

      So you can do something like short out the "D/R" resistor, break the gnd on pin 16 and use a 2 pin female header to get pins 17 and 18.

      (Of course this is much easier if you are running wires to the display instead.)

      Have fun!

  116. Hi, I tried your e-mail address but google did not recognize it. I am very interested in your kit-it is in the range of what I can afford which is a huge selling point though after reading reviews of some of the cheaper geiger counters I have been looking to the 350.00 to 500.00 units. Your kit seems to offer a similar range of features when connected to a good GM tube.

    How long do you think the build will take for a person of medium level ability?

    How does your unit compare with a ready built like the Inspector or the Safecast Nano.

    Does the unit need to be calibrated by a professional shot?

    Thanks, Jim.

  117. Jim,
    The email on the upper right under "Contact" should work.

    If you have reasonably good soldering skills, work carefully, and follow the instructions, building the kit should take about 3 hours. If you mess up it will take considerable longer to troubleshoot and unsolder parts.

    You will also need to spend time putting it in a case. Depending on your approach, that often takse more time than building the kit.

    I'm really not that familiar enough with other products and don't want to get into making comparisons. I'm personally happy with hardware and software, and get good feedback.

    Calibration is a big subject. Briefly . . .
    The kit will faithfully record events from the tube - CPM.
    Calibration comes into play in converting those counts into units of "Dose" - uSv/h, mR/hr, etc.
    So in basic terms you need a ratio between CPM and the dose unit you are using.

    Most GM tubes spec this and each tube type has different sensitivities for different isotopes - i.e Co60 vs Cs137.

    Most Geigers, including mine, use some kind of ratio from these specs for the type of tube. And I'd say it's ballpark at best.

    So the only true way to calibrate is with a lab sample by people who do this type of work.

    But think about the difference between "detection" and "measurement". I prefer to monitor by looking at the *difference* between what I detect on a daily bases.

    However for sites like Safecast it might be good to have some kind of measurement to compare with others. But I don't think many there are truly "calibrated".

    As I said, big subject! Hope this helps.

  118. Thanks, very good info. One other question occurred to me. Is it possible to hook up multiple gm tubes and switch between them depending on the particular sensitivity you desire? Oh, and I'll be ordering a kit a little later tonight. Thanks for answering my questions so quickly.

  119. Glad it helped.
    The kit support two "ratios" selectable by grounding a pin. Switching the HV from one tube to another might be tricky buy I think has been done. Another option is run both together.

  120. Hello, I installed sketch (v.8.0) on board lcd
    I have read the data every 30 minutes, I changed this
    # define LOGGING_PEROID 6000000 /

    I have to change anything else?

  121. Hi,
    Please email me directly about this.
    I am not sure what you are asking and v8 is an old sketch.
    Also provide the name your kit was ordered under.

  122. Just taken delivery of my GKPlus kit.
    Amazed at the sheer quality of the boards, components, packed properly, free resistors to make a HV probe, a free SD card, I mean what more could you ask for. Absolutely cracking service. Top marks john...

  123. Hello BroHogan,

    Thank you for great info about geiger devices. Did read lot of your writings very informative and must have taken a lot of time and effort. For me most needed was tube info and Geiger kit mini schema about HV + general info (sorry had no interest for kits).

    I did not go trough you latest designs as was interested just to drive tube with my existing automation thingy. I have two ideas that I try to contribute back, but ebay parts have not yet arrived and these might be silly / bogus.

    1. In HV design (geiger mini) there is "hard" to find HV transistor. Planned to use regular fast transistor and fast protective HV diode at hv side.

    2. You did write about some test about max "speed" of tube / kit. My ordered tube has 2 positive and 3 negative leads, thus was thinking that adding separate counter's for each negative lead would improve resolution and speed.

    Thanks and best regards,

  124. Hello BroHogan,
    I really like the looks of these kits! Thinking about building one of these in the very near future.

    Looking for a simple sketch based off of your GK-B5 kit using the internal hardware timers to build a scaler/ratemeter to integrate into by Bicron Surveyor M. I need it to just sample incoming pulses at high speed and be able to sample at CPM / CPS (settable timer and count rate out to serial) I am not looking for dose rate. The internal hardware timers are highly accurate and don't suffer lag time like interrupts.

  125. Hi BroHogan,
    Great kit, loved it.
    Just finished the kit and it works like a charm (v5.6). Just can't get the remote to do anything yet :-( to input some alarm-set points, etc. The remote is working though, my digital camera can see the IR flashing. Any suggestion welcome.

    1. Hi Frank
      Happy to hear you love the kit. Is the IR sensor facing the right way? (bulge out)
      If yes, it should be easy to fix. Best to email me directly for more help. (I'm traveling right now so be a little patient.)

  126. Hi,
    wonderful project, thank you for sharing it. Based on your HV design I built my own Geiger counter unit.

    I made a small addition to the HV part - added a 1GOhm/4.7MOhm (1% accuracy) voltage divider and buffered this signal through a MCP6041 OPAMP (1pA Input Bias) to the ADC input of my MCU.

    Based on this voltage divider ratio and measured voltage at the ADC input I can easily monitor the HV with excellent accuracy.


    1. Nicely done! Using the OPAMP was a good idea. Display of HV will be nice.

      Are you thinking about using a PID controller to keep the HV constant? (Some work though!)

      Thanks for posting your work.

    2. Hello again,

      No, not really PID. However what I did in my latest modification, was to replace the 100R Pot with a voltage controlled source.

      Control voltage is from the MCU DAC. So instead of using the 100R pot I can set up the HV using two buttons. Then the MCU is monitoring the ADC input every 100ms or so, calculates the HV and adjusts the DAC voltage to compensate accordingly.

      Please note that in my original version I implemented and tested only the OPAMP buffer and the HV readout. I am currently waiting the new PCBs to test the voltage controlled current source.

      With Best Wishes

    3. I understand. BTW, I also have a circuit where the 555 is replaced by PWM from the MCU. It works well and may be a simpler alternative if one is needed.
      Good luck on your project.