[SOLVED] Power Supply For Cold Cathode Crookes Tube

[resurgence2012]

I am trying to build a power supply for a cold cathode crookes tube. I am
told the tube needs between 10 and 15KV. Does anyone have any information about appropriate power supplies circuits that you can share? The simpler, the better.

I have been trying to use a flyback transformer. Using a simple single transistor driver, and 30 volts going into the primary, I am guessing that I have more than 20kv coming from the secondary.

The output from the secondary is high frequency, but I believe it is DC because the flyback is modern. But the tube only goes on for about a second and then goes out. After that I
have to switch everything off and wait about ten minutes before trying again. I think I have probably done something basically wrong but I can't figure out what it is.

Thanks in advance for any help.

Pete

 

[ginasterafan]

I am trying to build a power supply for a cold cathode crookes tube. I am
told the tube needs between 10 and 15KV. Does anyone have any information about appropriate power supplies circuits that you can share? The simpler, the better.

I have been trying to use a flyback transformer. Using a simple single transistor driver, and 30 volts going into the primary, I am guessing that I have more than 20kv coming from the secondary.

The output from the secondary is high frequency, but I believe it is DC because the flyback is modern. But the tube only goes on for about a second and then goes out. After that I
have to switch everything off and wait about ten minutes before trying again. I think I have probably done something basically wrong but I can't figure out what it is.

Thanks in advance for any help.

Pete

>a flyback transformer
This is just the approach I would have chosen; I'm all for recycling any kind of kit.

>tube only goes on for about a second and then goes out
Why do you have to wait ten minutes to try again I wonder ?
Is that because the tube current requirement is overloading your psu causing it to trip? Do you have any way of measuring the potential across the tube ?

Also, must admit I'm not familiar with the Crookes tube - do you know yours to be good ?

 

[resurgence2012]

Measuring the voltage is another challenge. I am not sure how to go about measuring 20, 000 Volts. I would like to be able to 'see' what is happening.

The alternative of course is to buy a power supply ready made, but they tend to
be over 100 dollars.

From what I can make out the comercially available power supplies are mostly based on induction coils and have a wide spark gap on the top.

I saw one yesterday advertised as a resonant tesla coil with a variable voltage from 10 - 100 KV.

 

[ginasterafan]

Measuring the voltage is another challenge. I am not sure how to go about measuring 20, 000 Volts. I would like to be able to 'see' what is happening.

OK, so assuming you have a multimeter that will measure at 1000Vdc, then I would remove the Crookes tube from the circuit and replace it with a series resistance chain of say ten 10MΩ or 20MΩ resistors (making a load of at least 100MΩ or 200MΩ). Probably a good idea to check each resistor value first too, make sure they're similar. Then fire up your psu and measure the potential across the resistor close to earth (safest there). Ten similar resistors will give one tenth of the load voltage across each. So now you can calculate the load voltage. In fact you could use any value resistors just so long as you have a pretty high total load resistance, then just a matter of Ohms law to get the current and then add up the voltage drops.

>But the tube only goes on for about a second and then goes out.
It could be that the on load voltage with the Crookes in circuit is pulling down your psu output so it won't sustain the tube for more that a second or so. Above method should give you some indication of what's happening.

 

[resurgence2012]

Thanks. I will try what you suggest. I have a bunch of 10 M. 1/2 watt resistors somewhere close at hand. I'll try that and see how it goes. I did wonder whether I should perhaps be putting a resistance of some kind in series. I am running it off a 30volt. 5 amp DC supply so I get the feeling there is quite a bit of current there.

Cheers

Peter

 

[chuckey]

The output from a flyback transformer will be AC. A thin hi voltage pulse followed by a low voltage negative pulse. So the target will only be lit up for a certain percentage of the total time, it will be a dim image. Can you not monitor the input V and I to see how its varying when the tube goes off. it might give you a clue.
Frank

 

[ginasterafan]

The output from a flyback transformer will be AC. A thin hi voltage pulse followed by a low voltage negative pulse. So the target will only be lit up for a certain percentage of the total time, it will be a dim image. Can you not monitor the input V and I to see how its varying when the tube goes off. it might give you a clue.
Frank

Hi
Doesn't a flyback (though I'd normally say line output) transformer normally include an integral diode to provide d.c. output ?

 

[resurgence2012]

Hi Frank

I believe that the flyback due to being modern has an integral
diode. That's what I have read everywhere. However, it
is obviously time varying.

What I will do is set up a nice string of 10
M resistors in series. I will insulate the 9 closest to the live end. Then I
will start by looking at the voltage and current with my cheapest
DM.

 

[resurgence2012]

Hi. *I thought I would give you an update. **I decided to try setting up the flyback so there was a spark gap between the high tension output and the earth. ***With this in parallel with the CRT it started to flicker on and off. *So am wondering whether the output of the flyback needs some kind continuos load.

I will try that but then I went out at the weekend and bought a 10 dollar crackle disk (you know like a plasma globe, but one of those flat discs)

Took it apart in two minutes and had the output of the high voltage connected to the CRT. **It turns out that the high voltage terminal is actually minus a few thousand volts. *So with only this one wire attached to the
cathode and absolutely nothing connected to the anode - hey presto! *- a nice clean green line showing where the electron beam is hitting the fluorescent screen. *




So it works and I kind of understand why it works but still don't really understand why my own ridiculously overkill and potentially lethal flyback circuit doesn't work, unless it is true of these devices in general that what they won't is a high negative potential rather than a high positive one.

I still need to take the time to study the output from the
flyback. I have it set up now with 10 x 10m ohm resistors in series with the positive end,
so I just need to get the meters out and have a look at what is going on.

Cheers

 

[betwixt]

Beware, most resistors, unless you have special high voltage types are only rated to about 250V so if you have more than 2.5KV you risk flashover and damaging your meter. Don't measure the voltage across the resistors, measure the current through them and calculate the voltage using Ohms law.

In a Crookes tube, it's possible the voltage is too high and you are drawing the electron field to the target rather than letting some of them overshoot and hit the phosphors.

Brian.