If you're having problems, you should look at these areas things first:
- Soldering - missed joints, too much solder.
- Make sure you have added all components to the board and all joints are soldered.
- Look closely at the back of the board for problem joints.
- Review the Soldering section in the Build Instructions.
- Solder Dust and Flux -
problems - especially HV problems - are sometimes caused by flux, solder dust, or some other film on the board.
- Look closely at the back of the board for bits of solder - especially between the pad and backplane. Brush the back of board with an old toothbrush.
- For cleaning off flux, a safe way is to use alcohol (from hardware store) and a tooth brush. Afterwards, blow off the water formed. There are other options you can Google for, but be careful with what solvents you use.
This board is an extreme example of these problems.
- Part orientation wrong or wrong part installed.
- Compare your board with the pictures in the Build Instructions. Check orientation on all polarized parts.
- Go through the "Build Sequence" again in the Build Instructions. Be sure to read info in the right hand column.
- If you find that you must remove a part, read the Unsoldering Parts section below.
- Other things you can do:
- Look for "simple problems" - battery OK, LCD contrast pot set, etc.
- Use the schematic and your meter to troubleshoot.
Do continuity checks with the power off. Connect one lead to Ground, and check for continuity to all places that should be ground. Do the same with Vcc (+).
- The schematics can also help you narrow down the problem area.
If you discover that you soldered in the right part the wrong way, or the wrong part, you'll need to unsolder the part. Removing some parts "in tact" without damaging the part or the board can be difficult - especially if a pad is connected to the ground plane. The usual methods apply - solder sucker or solder wick. I've found that a flux pen
really helps. Adding fresh flux right on the solder wick, and also the part, improves the flow. However, if you have a replacement part, it's generally easier to destroy the bad part and replace it. There are 2 options for doing this:
One option is to simply cut out the part and solder the new part to the old leads or pads on the top of the boards. If you'd rather do a clean removal, I use the method described in the following paragraph:
Cut the part off leaving as much lead on the board as possible. Heat the pad, and pull out the remaining lead. Then use solder wick or sucker to clear the hole. If that isn't working, rather than struggling and possibly lifting the pad, remove enough solder to make a "dent" with the solder being further down the hole. Then drill out the hole. This is easier then it sounds if you get a pin vise with a very small drill bit - smaller than the hole in the pad. You should be able to find a cheap pin vise
, with tiny bits included, at a hobby center or hardware store. Turning the pin vise will drill through the solder in the hole pretty quickly. Be sure to brush the solder shavings afterwards.
No High Voltage
This is the most common problem when the Geiger board is not working. The usual troubleshooting applies as a first step; the right parts are installed in the right way, none missing, the soldering is good (no bridges), a clean board with no solder dust or film, the battery voltage is good, and you are measuring for the HV in the right spot. Once you've checked all of these things try the following:
- Double check the orientations of D1, D2, and IC1 using the pictures in the Build Instructions.
- Check that you installed the correct part for Q1, Q2. Q1 is a 2N4401, Q2 is a STX13005 (MPSA44 on older kits). If Q1 & Q2 reversed max HV is only 100V.
- On v5.x boards, make sure R2 is 1kΩ (BR, BL, RD) and not 10kΩ (BR, BL, OR).
- (If you have the 4.0.2 PCB make sure Q2 is soldered into the "B" pad as described in the Build Instructions.)
- It's best not to mess with R5 if you are not getting any HV. It should be pre-adjusted to give you a workable voltage. Adjusting it can just add another variable to the problem.
- If you do need to adjust R5, note that it is a 25 turn pot. It should have been preset to ~26Ω which will give you HV. Turning it clockwise will increase the HV (by lowering it's resistance). Note that if it's turned too far clockwise (below 12
Ω) the HV drops to battery voltage. If you don't know where the pot is set, do this:
- Power off the board.
Set your meter to "Ohms".
- Connect one probe of the meter to Ground (negative side of battery, "-" side of power header, etc.)
- On the bottom side of the board - touch the other probe to the solder pad of the pot that is closest to the middle of the board.
(Or from the top of the board - the emitter of Q2 - which is the left side with flat side facing you.)
- Adjust the pot to 30Ω.
- I've seen a board where the fine wires on the bottom of L1 were broken. Check resistance across L1 - you should get about 20-30Ω.
- If you have a scope, or your meter measures frequency, you should see about 3kHz at pin 3 of IC1. See the traces shown on the Circuit Description page.
- It is possible to "let out the magic smoke" in the CMOS 555 timer (IC1). It can fail if something was mis-wired. If you live near a RadioShack, they have them. Make sure you get this CMOS version.
- Check to see if something is loading down your Vcc (+ voltage). Measure at the 4 pin power header.
- If you are not using the regulator you should read the same as your battery.
- If you are using the regulator, you should see 5V (or 1.4V less than your battery voltage if the battery voltage is below 5V.)
- If you are seeing less than what is expected, remove IC1 and test again (this time to Pin 8 on the socket. If the voltage goes up, IC1 may have a problem. The same applies to IC2.
- If IC2 was ever installed backwards it could be fried. It may be warm or even hot. Remove it and see if you start to get counts on the LCD.
- Troubleshoot the board! This means using your meter and the schematic and the picture of the board traces shown in the Build Instructions.
- With power on - attach one probe to Gnd and look for 5V at the points in the schematic that say Vcc.
Then do the opposite - attach one probe to Vcc and look for 5V at the points in the schematic that say Gnd.
- With power off - check continuity between parts using the picture of board traces.
First check you are getting HV - see the Build Instructions. If you are getting HV . . .
- make sure your tube is orientated correctly - on most tubes the body is the cathode (-)
- put something across the HV terminals (I use my finger) to see if that gives counts
- try increasing the HV - turn R5 2 turns clockwise If still no counts turn it back
- if you have another tube try that one
Whine through piezo
Note: Kits v5 and above which use a speaker instead of a piezo are subject to a faint whine at the extreme HV settings. This is caused by the speaker coil picking up the oscillation from the inductor L1. Moving the speaker off the board, mounting it on the case will solve the problem. If the speaker remains on the board you can usually be avoid the whine by decreasing or in some cases increasing the HV setting. Other more general causes are:
- flux on board or board on a conductive surface (HV conducts easily wood and even black foam board will conduct.)
- polarity of the GM reversed
- poor solder connections around IC2 or a dirty board. Check that pin 7 of IC2 is at ground.
- noise being picked up by cathode (-) lead of the tube - try shortening the lead to see if this is the problem
- having your meter still connected to the HV test points
- if you have the UC_PWR jumper removed, you should also remove the INT jumper.
- (piezo not loud) - check C7
- check orientation of D3, D4, and D5
- Make sure you have set the LCD contrast pot.
- On the Geiger Shield, if the backlight isn't even on it indicates you don't have the Geiger Shield sketch loaded.
- Make sure the uC_PWR and INT jumpers are installed.
- Make sure you're running at the right voltage. Most LCDs need >4.2V.
- If you wired in your LCD make sure you have it wired right and the contrast pot is set.
- If you ever reversed polarity on the Vcc and Gnd of the LCD, there's a good chance it's fried.
No IR response on Geiger Shield or Geiger Kit
- Make sure your IR sensor is installed with the bulge facing out from the top edge of the board. (So if it was installed in a case it would face out of the case.) Note: if it was installed backwards, it will still work when it's installed the right way.
- It's best to use a "universal remote" the kind with several codes to try. TV codes work best. Keep trying the Mute and Power buttons.
- You can scope the "out" side of the sensor (rightmost lead) to test that it is picking something up.
microSD card will not write on Geiger Shield
- The current SW should support 2-8GB microSD and microSDHC cards formatted with FAT16 or FAT32.Larger cards may not work.
- After 100 files have been written you can not write more until you remove them from the card.
Contacting me for support
Clock problems on Geiger Shield
- Clock resets every 5 minutes: This can happen if the coin cell (RTC backup battery) is not installed or it's dead.
- Clock runs slow: My supplier accidently gave me 32.000kHz crystals instead of 32.768kHz and they may have made it into a few kits. "32.000" is marked on the side of the can. The 32.768kHz crystals often have no marking on the side of the can. If you have the wrong crystal, please contact me for a replacement.
If you've tried all of the above - or done all you can, write me for help. However, to help me help you, I have a few requests:
- Describe the problem is as completely as possible. (remember -
garbage in, garbage out)
- Tell me what steps you have taken to find the problem and what you have found.
- If I ask questions, no matter how stupid they may sound, please answer them - don't just pile on more information. I need to have the background so that I can relate to your problem. If the questions I ask in order to get this background are not answered, I may not go further with help.
- I have a lot of patience when I know you are trying to help yourself.
- At some point, I might ask for a good picture of your board. Use a "macro" setting if you can. The picture should be in focus.
- Give me the version number of the PCB and if it was ordered under a different email let me know that too.