Archived - Dave Xanatos - Fusor Foundations

Current images of fusor efforts, components, etc. Try to continuously update from your name, a current photo using edit function. Title post with your name once only. Change image and text as needed. See first posting for details.
User avatar
Dave Xanatos
Posts: 111
Joined: Tue Apr 22, 2014 5:59 pm
Real name: Dave Xanatos
Location: Western Massachusetts, USA
Contact:

Re: Archived - Dave Xanatos - Fusor Foundations

Post by Dave Xanatos » Tue Oct 11, 2016 11:33 pm

Neutron Detector build underway.

Using Richard Hull's Last Two PMTs!

Burn-in testing of a few different HV Supplies for the PMTs. These guys run up to around 1800Vdc, not bad for $8 eBay Hong Kong specials. Pictures show voltage running around 1500Vdc through a 10M/1k voltage divider.

I'll be using BC720, I know it's got a low efficiency... but I like the concept. And I get to play :)

More pics to follow of the nicely machined-to-match-the-PMT scintillator plastic, etc. My machinist is fabricating a really nice housing.
NDHVPS1.jpg
NDHVPS2.jpg
It would take decades of work, by thousands of scientists, in a particle accelerator powered by dump trucks of flaming grant money! - Professor Farnsworth/FUTURAMA

User avatar
Rich Feldman
Posts: 1142
Joined: Mon Dec 21, 2009 11:59 pm
Real name: Rich Feldman
Location: Santa Clara County, CA, USA

Re: Archived - Dave Xanatos - Fusor Foundations

Post by Rich Feldman » Wed Oct 12, 2016 1:00 am

Dave, looks like you are ahead of me on biasing PMT's.

You are probably over-volting your 10 megohm metering resistor. I bet its rated working voltage is 350 V (see https://www.seielect.com/catalog/SEI-CF_CFM.pdf).

Each resistor series is rated for max power and max voltage. For very high resistance values (and/or low duty cycle pulse applications), the voltage limit gets you first. If you want your voltage divider to have positive design margins, you could use five or six of the high-value resistors in series, or get HV-rated resistors. Ohmite makes some similar-looking resistors (Mini-Mox) whose 1/4 and 1/2 watt sizes are rated at 500 and 1000 volts. http://www.ohmite.com/cat/res_minimox_200_300.pdf At 1 watt size you can find resistors rated for 10,000 volts, as presented here by John Futter: viewtopic.php?f=11&t=10332&p=70119#p70119
Mike echo oscar whisky! I repeat! Mike echo oscar whisky, how do you copy? Over.

User avatar
Dave Xanatos
Posts: 111
Joined: Tue Apr 22, 2014 5:59 pm
Real name: Dave Xanatos
Location: Western Massachusetts, USA
Contact:

Re: Archived - Dave Xanatos - Fusor Foundations

Post by Dave Xanatos » Wed Oct 12, 2016 2:17 am

Hi Rich,

I bet you're right!

This was just a test setup for a burn-in. Are you suggesting that the measurements may be inaccurate? I've not usually had issues with mA level measurements with standard resistors below 2kV. I am very aware of the resistor issues/requirements when measuring my fusor grid voltage/current, but for testing my PMT supply (an order of magnitude or more lower than fusor supply) for drift over time, I am admittedly being a little less than impeccable :)

Since ultimately my serious testing will be mid-range biasing the PMT and watching the output on my scope with a low-level fast neutron source irradiating the BC720 bonded to the PMT, I only need to be in the ballpark initially, and be able to distinguish neutrons from background reliably. I have no plans to continuously meter my PMT supply once I dial the voltage in to get best neutron response from the PMT. After that, amplification, pulse shaping and counters will be designed in.

All that said, am I missing something important? Are you thinking that my voltage measurements are off by such a significant factor by using that 1/4 W 10M that it is creating a false reading? It seems to be correct based on the specified range of the supply and the position of the pot relative to min/max, so at the moment I feel confident that my readings are within at least 10%... but I'm certainly open to the possibility that I'm very wrong... please advise.

Thanks,

Dave
It would take decades of work, by thousands of scientists, in a particle accelerator powered by dump trucks of flaming grant money! - Professor Farnsworth/FUTURAMA

User avatar
Rich Feldman
Posts: 1142
Joined: Mon Dec 21, 2009 11:59 pm
Real name: Rich Feldman
Location: Santa Clara County, CA, USA

Re: Archived - Dave Xanatos - Fusor Foundations

Post by Rich Feldman » Wed Oct 12, 2016 3:53 am

Sorry to be cluttering your post, Dave. I bet the resistor in your picture will be OK for short-term use, and for bottom-line accuracy on the order of 5% as indicated by resistor stripes.

I'm no expert with high voltage resistors. Somebody in another thread did some Googling, and said that excess voltage (without excess power, or flashover between the leads) affects linearity and long-term stability of resistance. I would question a commercial design that exposed any resistor to 5x its rated voltage for extended periods, unless its designer knew how to safely venture far away from the road.
Mike echo oscar whisky! I repeat! Mike echo oscar whisky, how do you copy? Over.

User avatar
Richard Hull
Moderator
Posts: 11605
Joined: Fri Jun 15, 2001 1:44 pm
Real name: Richard Hull

Re: Archived - Dave Xanatos - Fusor Foundations

Post by Richard Hull » Wed Oct 12, 2016 4:36 pm

Note***** Few if any PMT's ever need much over 1200 volts for proper operation and most work just under 1KV!! The higher the voltage, the more noise pours into your amplifier system. No one needs super amplified noise!

I am somewhat of an expert on HV resistors. There are high ohm resistors (very special above 1 gigohm); then there are high voltage resistors (also very special above 10kv). The two types are often very separate in construction and very expensive, though today's technology has made both easier to fabricate and this would reduce their price, save for demand for them being near zero as few applications demand either in the modern world of solid state electronics.

Precison is extremely expensive in either type. This is due almost solely due to temperature variability of their environment, (temperature coefficient) which is often a function of the current forced through them.

Common, modern film reisistors are cheap, sturdy and accurate. For voltages under about 5kv I tend to prefer these 3 cent each resistors, (in bulk pricing). For my most common range of 1kv-2kv made up dividers using these types, I tend to prefer a string of 9, one megohm 1/4 w 2% resistors. I measure across a special 10th resistor setup placed at the bottom of this string that is a special set, "net", 1 megohm resistor that connects to ground, (1/10 HV reading). I am careful to use a 100 meg input impedance, digital panel meter set into this lower most, one meg calibration resistor. 2% or better accuracy is easily obtained....Provided.... you tweek the meter leg's potentiometer against a good standard across the HV input to match. I have such a high voltage standard calibrated system.

Such "lower voltage", high voltage metering is found in testing and using, PMT's, Neutron proportional tubes, GM tubes, etc. Yes, 1,000 to 3,000 volts in our biz is a "lower" high voltage!!

All of the good digital panel meters are 0.2 volt meters and it can be a daunting task to put them in a series-parallel situation in this special lower 10th, bottom one meg, "net" resistor. It is left as an exercise to the student to noodle-out this electronic-hack to avoid blowing up the 200mv meter as part of a 100 volt potential placed across this net 1meg reistance metering rig with 1,000 volts across the entire divider.

The result can be a highly accurate and stable 1-2kv meter that will place no more than a 100ua load on a 1kv power supply. Supplies for the above nuclear instrumentation are nefariously low current supplies and will not suffer even a weak loading by normal electronic standards before buckling.

At voltages up to about 20kv, A pack of 100 of these cheap resistors, placed in separated series strings of 10 within a divider ladder is still acceptable, using good practices. Above 20kv special resistors, (expensive), are demanded. Sometimes these very special, HV rated, resistors are found surplus at only $2.00 each. Only a good scrounger can win without spending much money.

The quickie little pigtal divider seen in this post is fine for Q&D, (Quick and Dirty), order of magnitude measurements. Where they fail or lie to such Q&D users is their failure to consider the load on their supply or the impedance of their meter. Remember, accuracy is no better that the distributed net division of final impedances in the divider and the meter is a big, big part of it. In the case of this post, Dave did good for a Q&D setup

I do not like to stress a modern, high ohm, film resistor beyond 200 volts across it or 100ua through it when in a high voltage circuit.

Richard Hull
Progress may have been a good thing once, but it just went on too long. - Yogi Berra
Fusion is the energy of the future....and it always will be
Retired now...Doing only what I want and not what I should...every day is a saturday.

User avatar
Dave Xanatos
Posts: 111
Joined: Tue Apr 22, 2014 5:59 pm
Real name: Dave Xanatos
Location: Western Massachusetts, USA
Contact:

Re: Archived - Dave Xanatos - Fusor Foundations

Post by Dave Xanatos » Thu Oct 13, 2016 3:25 am

Very good info Richard, thanks. The pictured resistor setup is indeed just a Q&D, running near the top end of the unit's maximum rated output. Honestly, these things are from Hong Kong and only $8.00 each, and I just wanted to run them for an hour or so because I fully expected them to crap out! But they're little troopers, handle being shorted out for up to 5 seconds (as far as I was willing to torture them), and dealt with rapid power cycling with ease, returning to the set voltage in under a second after each sequence.

For actual builds of anything, I'd never employ the setup you see, I think since I posted the images, I should state that explicitly for anyone in the future reading this thread and trying to learn for the first time. But for my purposes in that moment, it did the trick.

I have the data sheet you sent me for the PMTs I bought, and I'm designing for the values specified therein (actual running voltage between 900 and 1250 if I recall correctly.) Ultimately, though, all actual "run voltages" will be dialed in real-time while watching the output on my scope. I also have gamma scintillator plastic (BC420 I believe) and several good gamma sources, so I can develop the electronics with an easier-to-excite media and then go for the not-so-easy neutrons.

I may still wind up with an He3 detector ultimately, but I am fascinated by PMTs and scintillation methods, and I'd really enjoy taking a shot at making a good scintillating neutron detector. Even if I fail miserably, I'll learn a ton and have fun.

Besides, 10 years from now we'll all be using small 3.3Vdc solid state high-sensitivity fast neutron counters anyway :)

Dave
It would take decades of work, by thousands of scientists, in a particle accelerator powered by dump trucks of flaming grant money! - Professor Farnsworth/FUTURAMA

User avatar
Dave Xanatos
Posts: 111
Joined: Tue Apr 22, 2014 5:59 pm
Real name: Dave Xanatos
Location: Western Massachusetts, USA
Contact:

Re: Archived - Dave Xanatos - Fusor Foundations

Post by Dave Xanatos » Mon Oct 17, 2016 1:02 am

Last MIT Flea Market of 2016... Picked up a few goodies, including this fat 14M resistor. Any idea what the wattage on this puppy might be?

It measures 7.5" overall length by 2" diameter, and is hollow. Looks really nice, measures 13.40M. Should find a home in some future HV project! Got it for $1.00.

Dave
14M1.jpg
14M2.jpg
It would take decades of work, by thousands of scientists, in a particle accelerator powered by dump trucks of flaming grant money! - Professor Farnsworth/FUTURAMA

User avatar
Richard Hull
Moderator
Posts: 11605
Joined: Fri Jun 15, 2001 1:44 pm
Real name: Richard Hull

Re: Archived - Dave Xanatos - Fusor Foundations

Post by Richard Hull » Mon Oct 17, 2016 6:01 am

I think 10 watts at least. Be aware that those white nicks may have cut across one of the black bands of the resistor. Meaure it with a good 20 meg DVM if accurate and the band is only partially cut, then your wattage is reduced. These are special carbon spirals usually on a white ceramic base, thus a severe nick or blow can compromise the resistance and wattage.

Richard Hull
Progress may have been a good thing once, but it just went on too long. - Yogi Berra
Fusion is the energy of the future....and it always will be
Retired now...Doing only what I want and not what I should...every day is a saturday.

User avatar
Dave Xanatos
Posts: 111
Joined: Tue Apr 22, 2014 5:59 pm
Real name: Dave Xanatos
Location: Western Massachusetts, USA
Contact:

Re: Archived - Dave Xanatos - Fusor Foundations

Post by Dave Xanatos » Mon Oct 17, 2016 3:23 pm

Hi Richard - thanks for the info - I would have thought more than 10 Watts, the thing is almost as big as a beer bottle!

As for what look like scratches - that is actually reflection - see the photos below with different lighting. This unit was in its original foam packaging and labels - it's pristine, and shiny like a mirror. And as I noted, I measured it on a 20M DVM at 13.40 Meg. It's a really nice unit... even if I never use it in a HV application, I'll use it for something - might make a desk lamp of it for a while! :)

Dave
14m3b.jpg
14m2b.jpg
14m1b.jpg
It would take decades of work, by thousands of scientists, in a particle accelerator powered by dump trucks of flaming grant money! - Professor Farnsworth/FUTURAMA

User avatar
Richard Hull
Moderator
Posts: 11605
Joined: Fri Jun 15, 2001 1:44 pm
Real name: Richard Hull

Re: Archived - Dave Xanatos - Fusor Foundations

Post by Richard Hull » Tue Oct 18, 2016 7:57 am

Glad they were not scratches. I have three 1 teraohm resistors of this ilk and many electrometer grade precision resistors in the 1 gig ohm to 750 gigohm range. Considering that the 10% tolerance on the 1T ohm resistor means that it can be off its 1T ohm rating by 100,000 million ohms and still be in perfect spec; me and my great friend and fellow engineer, Bob Deaton, always called these items, highly accurate, calibrated and qualified open circuits. At 1 volt across these resisitors, electrons dribble and drool through like baby slobber. Based on a draft across these puppies you just aren't sure if schrodingers cat or the electrons are dead or alive.

A 1farad cap in series would form an RC time constant of a terasecond, (30,000 years). Give or take 100,000 million seconds,(3000 years). Of course in most any measuring scenario, checking state of charge would wreck the time constant via loading. You could never know the state of charge, but you could know the circuit ought to be there at any given moment.(quantum stuff, LOL).

Of course, the equivalent series resistance of the cap would foil this effort dragging it back out of the jokingly quantum situation.

I have never seen a manufactured resistor with greater than a teraohm marking.

Richard Hull
Progress may have been a good thing once, but it just went on too long. - Yogi Berra
Fusion is the energy of the future....and it always will be
Retired now...Doing only what I want and not what I should...every day is a saturday.

Post Reply