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 (v.3) . . .
.JPG)
what the assembled kit looks like . . .
.JPG)
the assembled kit running the default sketch with an LCD display added . . .
I will buy one of your boards, I ordered a SBM-20 from eBay a week back, still waiting.
ReplyDeletegood 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%)... :-|
ReplyDeleteDepending 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.
ReplyDeleteSeeed 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.
ReplyDeletehttp://www.seeedstudio.com/depot/services-c-185.html?zenid=53eea25d5949339657c5b5a69fd15963
Thank you all for your comments and ideas. The kit is moving along nicely.
ReplyDeleteRe 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.
"Re the 9V option, you make a good point about cases that include a 9V battery. I think your suggesting:
ReplyDelete9V 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.
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.
ReplyDeleteSince regulators burn mA, I'll recommend replacing it with a jumper if your using less than 6V input. I think that should work.
I found that using 1/4" fuse clips worked well for mounting a SBM-20 on a PCB. Something like:
ReplyDeletehttp://onecall.farnell.com/littelfuse/102-074/fuse-clips-pcb/dp/1596547?Ntt=159-6547
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:
http://www.flickr.com/photos/cuddlydragon/5575506250/
Cheers,
CD.
"and kill the link between pins 3 and 7"
ReplyDeleteNote 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!
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.
ReplyDeleteI'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?
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.
ReplyDeleteI 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?
ReplyDeleteNice work BroHogan,
ReplyDeleteCould you post a link to your brd/sch files like you did in Part1.
Thanks
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.
ReplyDeleteI am only posting images of the board and schematic at this time.
Great! Thank you very much.
ReplyDeleteHi! I live next to Fukushima.
ReplyDeleteWe'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!
'Eddy-san,
ReplyDeleteThank 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.
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?
ReplyDeleteFor SW, I'm planning to pre-load a simple sketch that outputs CPM to a 2 line LCD and also to the serial port.
ReplyDeleteI'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.
Great kit!
ReplyDeleteI am thinking that the code from libelium should work with minimal changes (pin numbers, basically) on your Geiger counter.
Thanks Florin,
ReplyDeleteActually, 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:
https://sites.google.com/site/diygeigercounter/software
Just mentioned your kit and the Cooking Hacks one in the Yahoo GeigerCounterEnthusiasts group and provided links, so your hits should go up.
ReplyDeletePlease don't sell them all before this wbblair3 person gets one. ;-)
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?
ReplyDeleteSorry to have kept you in the dark. No "problems" other than overestimating the mail system and underestimating the demand.
ReplyDeleteI 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.
John
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?
ReplyDeleteMy 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.
ReplyDeleteThere 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.
John
Hi Bro,
ReplyDeletecongratulation 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
Got the full kit and it works well , no problems at all.
ReplyDeleteNow 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.
Daniele,
ReplyDeleteI 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!
Ciao
Lew,
ReplyDeleteI 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.
John
John,
ReplyDeleteI 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 https://picasaweb.google.com/tking2097/DIYGeigerCounter.
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!
Tom
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!
ReplyDeleteUnfortunately, 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.
https://sites.google.com/site/diygeigercounter/gallery
Thanks again for your great review.
John
Hi John - Yes, nice memory on your part, you have the right guy.
ReplyDeleteI did correct the sharing setting on Picasa, so my build photo album of your DIYGeigerCounter kit can now be viewed:
https://picasaweb.google.com/tking2097/DIYGeigerCounter
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
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.
ReplyDeleteI 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?
Johan,
ReplyDeleteThat is nice to hear.
For the clicks, I think a speaker might make better clicks than the piezo. Perhaps something like this:
http://www.goldmine-elec-products.com/prodinfo.asp?number=G16840
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.
John
What I meant was not using a tone but a simple pulse for a classic geiger counter sound.
ReplyDeleteIt'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.
In case someone wants inspiration for the enclosure, here is some:
ReplyDeletehttp://timewitharduino.blogspot.com/2011/08/diygeigercounter-now-completed.html
My inductor in kit broke due to overheat.
ReplyDeleteIs it the same as in kit ? 13R156C
http://search.digikey.com/scripts/DkSearch/dksus.dll?lang=en&site=US&KeyWords=+13R156&x=27&y=16
Daniel,
ReplyDeleteThat's amazing that the inductor would overheat!
In any case, yes. That is the inductor I use in my kits.
John
I used solder paste and heat gun. Powder core cracked. Plastic coating was untouched. Crazy thing. Probably due to high dT/dt.
ReplyDeleteHello,
ReplyDeleteI'm thinking on using a SBT-10 GM tube (http://cgi.ebay.es/RARE-Russian-Geiger-counter-SBT-10-Oty1-NEW-/200644426517?pt=BI_Security_Fire_Protection&hash=item2eb756fb15) 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?.
Thanks
Pedro Moreno
Wow - interesting tube. It's made as a "platform" for analyzing foods.
ReplyDeletePer 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.
I'll post my findings as progress in made
ReplyDeleteEventually 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).
ReplyDeleteI have contacted the seller to request a replacement or alternatively my money back ($118).
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.
ReplyDeleteSince 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.
John
No radioactive contamination detected in the GM tube. Seems to be a faulty part.
ReplyDeleteOk but make sure you keep your cathode lead short just to make sure you are not picking up and counting noise.
ReplyDeleteI thought you should know...
ReplyDeleteThe '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.
Hmmm, I cleared my cache in both Chrome and IE, and still both get the pictures.
ReplyDeleteI wonder if others are seeing this?
Can't think of what the problem could be.
Thanks for letting me know
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.
ReplyDeleteIn 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.
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.
ReplyDeleteI posted your comment on the GM Tube Info page
https://sites.google.com/site/diygeigercounter/gm-tubes-supported
Let me know if that's not OK.
Thanks again for the info on this tube.
John
Just bought this kit and assembled a counter. A counter itself works pretty well, but when I plug
ReplyDeleteRx/Tx to COM-port (9600) I got one or two lines:
"CPM uSv Vcc" and that's all.
Where the problem could be?
Sorry, false alarm! I was too quick :)
ReplyDeleteNow I can see that:
"CPM uSv Vcc
94, 0.5358, 4.42V,
902, 5.1417, 4.40V,"
WOW!
By the way, how to build the sketch? There is no LCD header ans library:
ReplyDelete#include
@vitasam,
ReplyDeleteThe 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.
@BroHogan,
ReplyDeleteah, 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?
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.
ReplyDeleteThe 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.
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.
ReplyDeleteGlad 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.
ReplyDeleteIn 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.
ReplyDeleteMight have just been glare making it look like a band at the bottom too:)
ReplyDeleteTrying Geiger sketch in Arduino 1.0. Got the message:
ReplyDelete"As of Arduino 1.0, the 'BYTE' keyword is no longer supported.Please use Serial.write() instead."
How to solve it?
Vitasam,
ReplyDeleteArduino 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!
(2 hours later ...)
ReplyDeleteYou 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. :-!
Thanks! It works now!
ReplyDeleteBy the way, just a curiosity - why you're not using any GIT hub (e.g. Google Code) for keeping a SW development there?
ReplyDeleteI expect to do that at some point.
ReplyDeleteThank 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)
ReplyDeleteNevermind 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?.
Good to hear.
ReplyDeleteIf 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.
Hello,
ReplyDeleteAt 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?
Alex.
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.
DeleteNot 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)
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!
ReplyDeleteJust emailed you back.
ReplyDeleteVery slowly my kit is progressing, yesterday I've connected a display:
ReplyDeletehttp://dl.dropbox.com/u/7286384/Pictures_to_share/mix/geiger_kit.jpg
What does represent the vertical line above "1" in "0.21"?
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
ReplyDeletea bar graph on the display showing CPM level, or the voltage at the CPU when the battery is low"
Yup! Nice build BTW
DeleteThanks! 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: http://pics.livejournal.com/vitasam/pic/0006fh35/s640x480
DeleteThis 'big brutal' Russian programmable calculator' is going to be connected as a field logger (with date/time stamps):
http://pics.livejournal.com/vitasam/pic/0006gt24 :)
"...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?
ReplyDeleteWhen the voltage is low (<4.2V) the entire bar graph will be replaced by text. Something like "Vcc=4.1V"
DeleteFeel free to email me if you have other questions. BroHoganX10@gmail.com
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.
ReplyDeleteThat's really very nice to hear.
Delete