Considering giving up Electronics as a hobby - Geiger counter problems / General Science and Electronics / Forums

Forums

4hv.org :: Forums :: General Science and Electronics

« Previous topic | Next topic »

Considering giving up Electronics as a hobby - Geiger counter problems

1 2 next

Move Thread

LAN_403

Inducktion

Sat Nov 19 2016, 07:32AM

Registered Member #3637 Joined: Fri Jan 21 2011, 11:07PM
Location: Buffalo, NY
Posts: 1068

I've had a rough past few projects, least to say.

Recently, I've been working on a Geiger counter project. Past few days, quite a few hours each day. Uses a TracerLab mica window geiger tube, which requires 1.3 kV to operate properly.

Figured, I could try using a single li-ion battery (18650), use a boost converter, a ZVS driver, and then a voltage multiplier to get the 1.3kV as needed, with active regulation, by feedback from the voltage multiplier back to the boost converter to control the voltage to the ZVS driver.

Got everything built, and put together. Tested out each component, separately, then together, then figured out the necessary resistor values and potentiometer value to get the best "range" of voltage to figure out what the geiger tube needs.

And.... it doesn't work.

Nothing. No clicks, no beeps no anything from the tube's circuit, just, a lot of "sine wave" noise from the ZVS driver.

And then from my poking around, trying to diagnose the issue, I think I somehow killed the boost converter, rendering that portion of the circuit useless.

Then I try just powering the ZVS driver from my power supply, to see if I can get anything.

Still, nothing. Just in line noise from the ZVS driver. I can't even fucking tell if it's because of the tube or if it's my circuitry, or what. And that's the irritating part. I don't know what's wrong because my circuitry could have very easily been working fine but the tube could be broken, or vice versa and now because the boost converter is toast, I won't be able to try it again as I envisioned.

This seems to be a reoccurring theme with me, and when I build something. I make it, spend a lot of time on it and then it just doesn't end up working, some way or another either because of faulty components, or human error, or whatever, and all the time I spent building it ends up going out the window. It's getting really, REALLY frustrating...

Does anyone have any words of wisdom... because I feel like i'm starting to reach my point of, screw this, because it's not fun anymore, not when everything you build fails to work.

radiotech

Sat Nov 19 2016, 07:51AM

Registered Member #2463 Joined: Wed Nov 11 2009, 03:49AM
Location:
Posts: 1546

How did you test the GM tube?

Electronics is not my hobby.

Few perfectly working projects teach you anything.

Sweating the details paid off for me. To keep

my job, for the first half of my career I did that

chore better than anyone, who tried to oust me from

my work.

After you posting all your ideas in this peer reviewed snake pit,

this long you wont go anywhere.

So fix the damn thing.

Carbon_Rod

Sat Nov 19 2016, 08:16AM

Registered Member #65 Joined: Thu Feb 09 2006, 06:43AM
Location:
Posts: 1155

1.
After basic principles of analog design:

Read the entire series of: Encyclopedia of Electronic Circuits
...And try to and figure out how each circuit works.
you will only need to do this once if you know basic physics.

2.
Read everything the ARRL has ever published at your local library.

3.
Ask Google how common bad tubes with delicate mica windows leak.

Sulaiman

Sat Nov 19 2016, 10:11AM

Registered Member #162 Joined: Mon Feb 13 2006, 10:25AM
Location: United Kingdom
Posts: 3140

If you just want to follow 'recipies' and build well tried circuits then electronics is a fairly simple hobby
If you want to combine ideas and circuits to come up with an original design, then it gets more interesting :)

If you do not create failures then you are not trying !

I have had projects that ended up flying towards the shed/lab walls at great velocity

it can be a very 'trying' hobby :)

In this case I would look at point #3 in carbon rods post first as I have zero experience with mica window GM tubes.
My hobby is amateur radio but who has time to read the entire published works, or an encyclopaedia ? not me.

IF you had some high value resistance between your voltage multiplier and G-M tube then your experiments should not have damaged the tube.
If not then WHAT WERE YOU THINKING ?
(an ionisation event causes the tube to look like a near short circuit)

I do not use zvs inverters as I find them too fragile,
have a look at ccfl inverter circuits that use a positive feedback winding and saturating core - reliable.
I would have opted for a small flyback inverter.
and because I'm lazy I would also consider;

If you show your circuit diagrams I'm sure that the combined brains of 4HV.org can help.

hen918

Sat Nov 19 2016, 12:37PM

Registered Member #11591 Joined: Wed Mar 20 2013, 08:20PM
Location: UK
Posts: 556

I hate Mazzilli ZVS drivers. They are the most temperamental power electronic circuits on the planet. (In my experience) The likelihood of the circuit stalling and becoming a dead short is increased with fast changing input voltages which makes input voltage regulation a tricky thing to implement.
I have had partial success with pulse skipping, shorting the gates to GND with a couple of diodes to prevent further power being delivered to the LC tank, but now I avoid Mazzilli drivers like the plague.

I hope you used a current limiting resistor, but even if you didn't, get a new tube and NEVER GIVE UP! (on electronics; it's OK to give up on individual projects occasionally)

DerAlbi

Sat Nov 19 2016, 05:53PM

Registered Member #2906 Joined: Sun Jun 06 2010, 02:20AM
Location: Dresden, Germany
Posts: 727

Hmmh. I like hen918s post and the first 3 lines of sulaimans post very much.
Considering the thread title i just would like to summarize it a little different: Dont give up. But manage your expectations.
Electronics is a very complicated hobby and achieving something in this line of timespending is expensive. Electronics suffers from a rather large division in education level.
There is a difference between hobbyist and professionals. End even many professionals are more related to the hobyist level, to be honest. But thats a problem of the education system and the internet.
There are the hobyists which keep repeating to build the same circuits over and over again and the most climactic experience is when a circuit finally does not blow up, even if they dont know why it actually works. Then the experience gets shared and repeated and "explainations" are published without regard of ther factual correctnes or pedagogic value. Any progress from that level is extremely hard, because the circuits are not based on a scientific method and people who learned with those resourc es are stuck in a dead end with bad or at least incomplete thinking models.

If this is what happended to you is yours to decide. I am not sure.
Every circuit works on his own - no problems. But do the circuits actually work under any circumstance? Are they truely designed to be blackboxes where you only care about input and output or do they have more specific needs at their interfaces?
Thats the thing. Once electronics reach a higher abstraction level every building block must be designed to be a bulletproof thing. This requires engineering not imitation or adaption.

So to give up or not.. complicated. It basically all comes down if you want to continue so suffer from that frustration but thats the only way to a hobbyist can get beyond the dead end. Its the only way to understand what most of the people actually do wrong.
But if you continue here is one advice: move on to a different topic. Do something else. I mean a completely different. Microcontrollers, PCB design software, or the best: circuit simulation. You have to evolve and learn other topologies.

To be honest... to every hour i spend on the soldering iron i spent 60-100h on the design. 40% of this time i spend in simulation and modeling, the other 40% i spend in research and datasheet study and maybe 20% i spend on actual schematic and PCB design. That does not mean that this is the only right thing to do of course.

Proud Mary

Sat Nov 19 2016, 08:03PM

Registered Member #543 Joined: Tue Feb 20 2007, 04:26PM
Location: UK
Posts: 4992

Inducktion, remove the GM tube from the circuit and replace it with a 100MÎ© resistor to simulate a healthy GM tube when it is not conducting.

What is the voltage across this resistor? (Remember that you are shunting this resistor with the input impedance of your meter, very often 10MÎ©)

If you have no high voltage across the test resistor, it means that your HV power suppy is defective.

When you have corrected the HV power supply fault so that HV now appears across the resistor, you can replace the GM tube and see if it all now works.

Is your GM tube one of TracerLab's famous TGC end window series? Some of these early tubes from the late 1940s/early 1950s used neon in the counting gas, so that a small reddish flash can be seen with each avalanche event when the tube is viewed in a darkened room.

2Spoons

Sat Nov 19 2016, 08:52PM

Registered Member #2939 Joined: Fri Jun 25 2010, 04:25AM
Location:
Posts: 615

Well, I've been a professional electronics engineer since 1989. And a hobbyist for about 12 years prior to that.

And I still make mistakes - though I make far fewer these days, having learned many lessons from past mistakes.

Don't give up - if something doesn't work its really important to figure out why. You'll be all the better for it.

At a company I worked at a couple of years ago someone had tried to make a high-power PCB inductor. It was a good idea, badly executed - the coils overheated and the idea was abandoned without further investigation. When I heard about it I chose to figure out what had gone wrong. From there I worked out how to fix the issue, and the result ended up in a patent for wireless power transfer.

Failures are important, not giving up is even more important.

Inducktion

Sun Nov 20 2016, 01:52AM

Registered Member #3637 Joined: Fri Jan 21 2011, 11:07PM
Location: Buffalo, NY
Posts: 1068

Thank you everyone for your replies and advice, I really appreciate it.
I was in the heat of the moment, and honestly was severely disappointed in that it didn't work properly.

RadioTech, I have no real method of testing the tube outside of the circuit. That does mean there's some possibility of the tube being defective, which it might very well be.

Carbon, I already have a pretty decent understanding of most basic circuitry, or at least, I like to think I do. I learn most of the things I do through application, and watching other people's attempts, and then figuring out what I could do differently to make it better.

Sulaiman, and to everyone else who asked or said, yes, I had a 5.6 meg ohm resistor in series with the tube to prevent current overloads. In order to test the output voltage of the supply I was using a 50 uA analog meter and a 250 meg ohm resistor in series. I too think ZVS drivers aren't the most reliable things in the world, however they are relatively simple, and with the way I had mine goin' it only drew around 200 mA from my supply (including the boost converter too.). It was something I was familiar with as well, so I figured it would make the most sense. I was getting around 5 uA ish, give or take 1 uA meaning I was getting the right voltage from my supply. I also checked it using my oscilloscope, to see the ripple output (which, strangely, was a sine wave?... this doesn't seem quite right, but I'm curious if it was interference from the ZVS driver or if perhaps my diodes had failed...)

Here's a really rudimentary diagram of my circuit. It would take me quite a while to actually make a true schematic, but I hope this brings some slight insight to my methodology.

Sp7EdjT

The tube was connected to the multiplier through a 5.6 meg ohm resistor, as I said previously. The amplifier circuit was based off of the right portion of this circuit

Hen918, funnily enough I actually never had any issues with the circuit stalling no matter what he input voltage was doing, unless it was just straight under the voltage required to oscillate.

DerAlbi, believe me I understand that as a hobby electronics can be a bit... expensive, as you said, in both money and time. And I also understand that there are plenty of examples of circuits and schematics of them out there that have no real explanation of how they work, or are created by people who have no business doing so. I try to be careful about where I source them from and make sure that I understand why it works to some degree before attempting to actually build it, lest I waste my time making something faulty.

Stella, like I said a little up in this message I actually was getting the proper voltage and whatnot from my supply. My tube is actually a TGC-2. It came with a little pamphlet thing detailing the tube and how to start it working, and the construction of the tube itself. It's actually pretty nifty. The tube itself is filled with Helium, uses a stainless steel cathode and has a mica window of 2 mg/cm^2. I do have to wonder why the voltage required by the tube is so high, and two things off about the tube... The base of the tube can be moved around, like it twists almost like a screw in a way. It's still attached but...loose, ish? I'm wondering if maybe there's a loose connection under/inside the base, but I have no way of checking this to be sure.

And there was also a strange white powder substance stuck in and around the tube's protective aluminum cap. It's almost like mold, or fungus or something but I'm not totally sure. The tube was stored loosely, not in a container or anything so I'm wondering if water perhaps got to it somehow. the window itself still looks to be intact though. It looks almost identical to the picture you linked.

2spoons, thank you for your words of wisdom, I appreciate them a lot. I'm going to keep at it, I just need to try and figure out what went wrong, if anything. Lord knows my luck, it's likely the tube itself is just shot somehow. Which is great, seeing as I'm on a tight budget and buying another geiger tube is kind of difficult/near impossible. Plus it was also nice because it was from the US, not from Russia so shipping didn't take half a century.

Sulaiman

Sun Nov 20 2016, 08:33AM

Registered Member #162 Joined: Mon Feb 13 2006, 10:25AM
Location: United Kingdom
Posts: 3140

the audio/click amplifier is of poor design ... no resistance between Q3 collector and Q4 base,
each click short-circuits the power supply :(
I suggest 1 kOhm per volt of dc supply, when all parts are working, reduce this resistance for louder 'clicks'.

The oscillator in that diagram is of the type I suggested ... positive feedback until core saturation
a 'blocking oscillator'
Frequency determined by volts/turn and core saturation.
i.e. more primary turns or lower supply voltage = lower frequency
'magnetic energy' released per cycle is fixed by core saturation
so higher frequency = higher power, and faster battery drain.
construct it to just provide enough output at minimum battery voltage input for efficiency / battery life.
Adding a small airgap in the core would dramatically increase power - so try with no/minimal airgap.

Test, then re-wind the primary with a more suitable number of turns :)

because it is a flyback output,
changing the number of primary turns does not directly alter the output voltage,
it changes the output power.
So you can keep the secondary unchanged while altering the number of primary turns.

the physical size of the transformer will be significantly smaller than a zvs transformer of similar output power.
I imagine somewhere around 1 cm^3

D1 - D4 need to be fast diodes (e.g. UF4007) not slow mains rectifiers (e.g. 1N4007)

because a little energy is lost each time the core saturates, the transformer will get warm.
that is why a conventional flyback (without core saturation) could be better for battery life.
Hint: manufacturers care more about their production costs than the life of your batteries.

P.S. do not get dis-heartened, you are effectively designing and prototyping a complete product
- not many professional EE graduates achieve this !
(sub-components of sub-modules would be more common :)

P.P.S. many oscillators, zvs and 'blocking' included, often fail if the supply voltage is increased too slowly at start up.
Put a simple mechanical switch in the d.c. power line when testing,
the often slow rise and fall of an ac/dc psu output can cause confusion.
e.g. by the time a single quater of the a.c. input sinewave reaches its peak (1/4 x 1/60 = 1/240 second)
a 12 kHz oscillator would have been through 12000/240 = 50 complete oscillations.

P.P.P.S. your sketch shows the 'amplifier for clicks' across th G-M tube,
I assume that it is in series per the circuit diagram refered to.

1 2 next

Moderator(s): Chris Russell, Noelle, Alex, Tesladownunder, Dave Marshall, Dave Billington, Bjørn, Steve Conner, Wolfram, Kizmo, Mads Barnkob