Electron gun- what's wrong here?

For the design and construction details of ion guns, necessary for more advanced designs and lower vacuums.
David Kunkle
Posts: 285
Joined: Thu Apr 17, 2014 4:43 pm
Real name: David Kunkle

Re: Electron gun- what's wrong here?

Post by David Kunkle » Wed Jul 27, 2016 12:10 am

Rex Allers wrote:Since, in my last post, I talked about removing or bypassing the caps in the output of the inverter, I thought I would comment about Steven's circuit. His configuration, with two diode strings, is a voltage doubler. In this case the two caps on the inverter are put in parallel as the charge storage for the first stage of the doubler. That's a valid approach and doesn't require any hacking on the inverter board.

Steven also mentions using fast diodes, but the 1N4007's in his schematic aren't particularly fast; about 2 uS trr, I think. For this kind of application I ordered some surface mount BYG23M diodes from Mouser, a while back. They are spec'ed at 1 KV, 1.5A, 75 nS trr. Overkill on the current for this but pretty cheap, if I remember right. Maybe others have recommended part numbers to look for.
Rex,

The 5V will be coming from an adjustable supply. Be interesting to see how much I can vary the output. Good tip.

I have found 1KV, 1A fast switch diodes on ebay with trr of 500ns. 4 for about $15 incl. shipping.

You're right about the grounds being connected. I found the jumper between the 2 grounds. What would be the advantage to disconnecting these?

Thanks.
If your experiment needs statistics, you ought to have done a better experiment.

Ernest Rutherford

User avatar
Steven Sesselmann
Posts: 2096
Joined: Thu Aug 11, 2005 1:50 am
Real name: Steven Sesselmann
Location: Sydney - Australia
Contact:

Re: Electron gun- what's wrong here?

Post by Steven Sesselmann » Wed Jul 27, 2016 12:11 am

David,

The capacitor should be around 10 nF and yes another one will smooth the ripple, but for your purpose ripple is not an issue.

One more thing I should have mentioned above is that you need to turn the diodes the other way, because you obviously want negative bias on your filament.

These inverters are not going to supply much current, but since you are heating the filament you just need some negative bias on the filament to make it become an electron emitter, this rather weak electron emitter should generate enough ions to keep the fusor lit up at low pressure.

Good suggestion from Rex to put a potentiometer on the input, so you can control the bias on the filament.

Steven
http://www.gammaspectacular.com - Gamma Spectrometry Systems
https://www.researchgate.net/profile/Steven_Sesselmann - Various papers and patents on RG

David Kunkle
Posts: 285
Joined: Thu Apr 17, 2014 4:43 pm
Real name: David Kunkle

Re: Electron gun- what's wrong here?

Post by David Kunkle » Wed Jul 27, 2016 12:48 am

So you can make it + or - bias just by turning the diodes around?

According to this: http://dkc3.digikey.com/PDF/CA2011/P2792.pdf , pink column, digi-key part # 445-1621-ND , the output is about 10mA if both outputs are wired together?

On that same pdf, same page, I see the largest (V-wise) inverter outputs 2100V. If my 900V inverter will hit 1400V, would that 2100V inverter wind up putting out about 3200V if I needed more electrons or would it just run out of mA being used up through beam current?

Thanks again.
If your experiment needs statistics, you ought to have done a better experiment.

Ernest Rutherford

Rex Allers
Posts: 236
Joined: Sun Dec 30, 2012 8:39 am
Real name:
Location: San Jose CA

Re: Electron gun- what's wrong here?

Post by Rex Allers » Wed Jul 27, 2016 1:49 am

David, there's no reason in this application to float the HV ground. I just mentioned it to be complete. I think I had a reason once, but I can't remember why now. Maybe monitoring current out.

Those diodes you mentioned seem very expensive. Also, 500 nS isn't real fast as "fast diodes" go.

You can do much better buying new direct from Mouser.com, probably Digikey too but I didn't check there. I looked at mouser for those BYG23M that I bought earlier. I bought 100 when I ordered. Looks like the price for 100 now would be $16.90. Those are smd package. A standard fast axial diode is UF4007; 100 of them is less, $10.20 for the first Fairchild listing.

Both of these diodes are 1KV, 1A, 75nS. You can order smaller quantities but the price goes up a little. Still way better than the ones you found. If you have time to kill, you could try the parametric search at mouser or digikey. There may be something similar with 2KV rating or etc.

Edit: I forgot to mention, yes, you can switch all the rectifier diodes around to change the polarity of the output.
Rex Allers

David Kunkle
Posts: 285
Joined: Thu Apr 17, 2014 4:43 pm
Real name: David Kunkle

Re: Electron gun- what's wrong here?

Post by David Kunkle » Wed Jul 27, 2016 3:15 am

Just got the proper diodes from Mouser. Got 6 for 45 cents each- then there's the $8 for shipping.
If your experiment needs statistics, you ought to have done a better experiment.

Ernest Rutherford

User avatar
Steven Sesselmann
Posts: 2096
Joined: Thu Aug 11, 2005 1:50 am
Real name: Steven Sesselmann
Location: Sydney - Australia
Contact:

Re: Electron gun- what's wrong here?

Post by Steven Sesselmann » Wed Jul 27, 2016 4:37 am

David,

Plenty on ebay if you dont mind waiting for them, just search for "8kv rectifier diode"

http://www.ebay.com/itm/20-x-30mA-8kV-1 ... SwcF9UXGpR

Steven
http://www.gammaspectacular.com - Gamma Spectrometry Systems
https://www.researchgate.net/profile/Steven_Sesselmann - Various papers and patents on RG

Rex Allers
Posts: 236
Joined: Sun Dec 30, 2012 8:39 am
Real name:
Location: San Jose CA

Re: Electron gun- what's wrong here?

Post by Rex Allers » Wed Jul 27, 2016 6:30 am

Steven and Dave,

I've had good results with that ebay seller 'high-voltage-hv' too. I've bought a number of hv diodes, those blue ceramic hv caps, and some hv wire from them. No complaints so far. Check out their 'store' or 'see other items' on ebay.

Outside of ebay, the site hvstuff.com seems pretty good too.
Rex Allers

Jeroen Vriesman
Posts: 278
Joined: Mon Feb 07, 2011 9:08 pm
Real name: Jeroen Vriesman
Location: Netherlands
Contact:

Re: Electron gun- what's wrong here?

Post by Jeroen Vriesman » Wed Jul 27, 2016 12:05 pm

I don't really get the idea of a high negative bias on the filament.

Indeed, it would accelerate the electrons more, but imho this might be the problem, not the solution, and the negative power supply would just be used for running current through the diode (filament + anode ring = diode).

If you take a look at the field strength between the filament and the anode ring, and compare that with the field strength between the anode ring and the final target:

if the distance between the filament and the anode ring is about 4mm (estimated from the photo), the field strength will be 320/4E-3 = 80 kV/m.
That means, if you want any substantial electron yield, the field strength between the anode ring and the target must be stronger than 80 kV/m near the center of the ring!

What I would do:

increase the distance between the filament and the anode ring a bit, give the target a high positive potential (e.g. 10 kV), make the voltage on the anode ring variable.

Now start with 0V (with respect to the filament) on the anode ring, measure current on the target, you should be able to make the current on the target zero by making the voltage on the ring a bit negative (a few volts), just like a triode.

Then slowly increase the voltage on the ring, a current on the target will flow, you will see that there is a maximum met the voltage on the ring is a bit positive, but when it gets too positive all electrons will go directly from the filament to the ring and not to the target.

User avatar
Steven Sesselmann
Posts: 2096
Joined: Thu Aug 11, 2005 1:50 am
Real name: Steven Sesselmann
Location: Sydney - Australia
Contact:

Re: Electron gun- what's wrong here?

Post by Steven Sesselmann » Wed Jul 27, 2016 12:45 pm

Jeroen Vriesman wrote:I don't really get the idea of a high negative bias on the filament.

Indeed, it would accelerate the electrons more, but imho this might be the problem, not the solution, and the negative power supply would just be used for running current through the diode (filament + anode ring = diode).!
.
Jeroen, the idea behind a negatively biased ion source is to accelerate the electron more which means accelerating the positive ion less, Preferably we want the deuterium nucleus to stand still, and we can in fact achieve that by ionising it at -62 kV, but the world of fusioneers still believe in building atom smashers to overcome the imaginary Coulomb force, which is why they are still trying after more than 60 years.

Unfortunately the little CCFL inverter won't be able to deliver much current, so the negative bias will fall off as the current goes up, I used a rectified MOT in my ion sources, but that becomes a bit more dangerous.

Steven
http://www.gammaspectacular.com - Gamma Spectrometry Systems
https://www.researchgate.net/profile/Steven_Sesselmann - Various papers and patents on RG

Jeroen Vriesman
Posts: 278
Joined: Mon Feb 07, 2011 9:08 pm
Real name: Jeroen Vriesman
Location: Netherlands
Contact:

Re: Electron gun- what's wrong here?

Post by Jeroen Vriesman » Wed Jul 27, 2016 1:17 pm

Hi Steven,

I just tried to answer the original question about an electron source, not the ion source related stuff.

The filament-anode ring-target setup is comparable to a ordinary triode (cathode-grid-anode) with well-known behaviour.
Making the grid in a triode positive a little bit gives more anode current, but make it too positive and the current will only flow between the cathode and the grid (and it will burn the grid in a regular triode).

The absolute potential of the entire setup doesn't make any difference, and having a large potential difference between the filament and the anode ring just draws maximum current between them (limited by emission, space charge or supply current) . Having the filament and the anode ring both biased with an negative potential is the same as have a higher positive potential on the target, but with the side effect that the chamber walls are positive with respect to the electron source, this will pull the electrons to the walls of the chamber.

I've seen these effects with the emission measurements of my thoriated tungsten wire activation experiments, something with the coulomb force I guess.

Post Reply