RF 'echoes'

For the design and construction details of ion guns, necessary for more advanced designs and lower vacuums.
User avatar
Doug Coulter
Posts: 1312
Joined: Sun May 27, 2007 7:18 pm
Real name: Doug Coulter
Location: Floyd, VA, USA
Contact:

Re: RF 'echoes'

Post by Doug Coulter » Sun Jan 24, 2010 11:44 pm

You will be happy to know that the uWave source does indeed handle crazy-fast gas flows fine. It's just loafing at normal flow rates and pressures. I started with mid range rates, and have been going down from there ever since, and getting better results as I do. Some of the byproducts are best kept in the tank, it seems. I don't think gas flow in general is a big limit with any of the ion sources I've yet seen, anyway. Some I've tried wouldn't go low enough for me, this one will, and also too fast, just turn the knob for instant response, very nice indeed.

And it's not like D is real cheap to just blow through the system fast or that the approach is "efficient" by any definition at all. I'd concentrate on keeping them from recombining instead, if possible, and keep purity high via good vacuum technique, which is only a one time cost, not a flow rate. if you look at some ion traps, or even a quadrapole mass spectrometer, they can keep ions happy for long periods without losing them. So it's possible in practice at least under some conditions.

I would very much deny that most fusors don't need or couldn't be improved by an ion source -- just give both ways an honest try and you'll know that too. Here that's not what I see experimentally at all, even running the "conventional" modes. At the point a fusor is its own best ion source, the gas pressure is often too high, the current rises without apparent limit (some of this perhaps due to a lot of wasted tank volume in my setup that provides a lot of surface for recombinations and a lot of spare gas to reionize over and over), and the Q is terrible compared to other parameters I run much more successfully with here, all the time. I can reproduce all the results reported here by others on this system though, and have already as my baseline for improvements. It's just a scramble to reduce the gas after the thing lights off to a point where even a 2kw supply can get the volts up high enough for fusion again, without letting it go out and having to start over.

My point was simply, why go after better efficiency in the *lowest* power draw part of the system before you even get anything working? Doesn't make sense to me as an engineer or as a scientist.

It's too soon in the scientific process to deliberately limit your ability to search the multidimensional parameter space by limiting this and that too much in a misguided search for 50% better efficiency when many more orders of magnitude are what's needed to get to gain. All that stuff is simple (re)engineering once you have nailed down what you really need. For now, you can't say you know that. No one can. Got net gain, anybody? I don't see any raised hands, yet. No one is within a factor of magnitudes you can count on a hand even.

Believe it or not, btw, the hot rod is actually far more controllable, due to vastly better traction and an anti-everything-bad computer it contains (among other things like very nice tires and suspension), which neither my truck or Buick do -- I got some pleasant surprises in the recent ice storms here when I went out and deliberately tested that. I just keep it in first.

I do walk or take the Buick or kart when it's nasty as there is also a cliff to fall over and replacement cost for the old Buick is a lot less if the worst happens. Where I live, btw, no one walks to church, or nearly so (in my one case, it's practically in my front yard). Too far, all our land holdings dictate that one - we're all way too far apart for walking there in farm country, particularly the old and frail.

This ion source is also very controllable. Due to magnetic field line curvature outside it from the ECR magnets, you can have ion rates of very near zero if the extraction field isn't strong enough to pull the particles past that. It is extremely controllable even though it's also powerful. True, it's not efficient if you only want 100ua of ions (eg camaro at idle). I don't yet care about that, and neither should you.
And BTW, I'm off the grid and on solar or gas generator power, so in general, I DO care about what power things need to run. Compared to all the other things, this just isn't on the radar as yet.

I too think while in my armchair that pure monatomic ions would be best, it makes sense. Get rid of all electrons, and many (but not all) of the loss mechanisms go with them. for example, and those who think they don't have to deal with space charge regardless are very provably in utter error with the results I now have here. And the old math that also backs that statement up.

However, this new mode I'm running in with super high Q compared to all else here, about ~300x, there's only enough ions at the start to get a pulse triggered, rest are neutrals (most of them, in other words) and at gas pressures so low a fusor won't quite trigger itself without an ion source. So much for even trained intuition with a ton of real experience behind it (eg mine) -- the armchair fails again as a source of progress -- hindsight and standard model backed up by real experiment wins again.

Hopefully, it feeds the armchair part of me and now I make better predictions, and we wash, rinse, repeat. In some active drive mode with pure ions, yes, that may be better-- hope so, there's a lotta orders of magnitude left to go, so more better is needed. But I found the mode without that. In fact, with nearly pure ions, I can't get this pulse mode that has the high Q so far at all, but I'm getting up to active pulse drives rather than ballasted DC to continue to try that, as my armchair does think that will be best in the end.

I do now have a theory of all this that's passed the Feynman test a few times, so I will share that on the appropriate thread soon - I am in running fusion and gathering data to back it up further mode just now.

Easier to resist the firestorm of flames I know I'll get from the skeptics when I am utterly sure I'm right (already, 20 runs replicated with changes matching predicted results thereof) and can by golly prove it in any way I'm asked to, which is almost "right now". Just a little more data collection to go at this point.
Why guess when you can know? Measure!

Linda Haile
Posts: 230
Joined: Mon Jan 04, 2010 9:28 pm
Real name:

Re: RF 'echoes'

Post by Linda Haile » Mon Jan 25, 2010 12:48 pm

So SWR or echoes is not something we need to consider at this stage? I'll start another thread soon on inductively coupled plasma. I'm considering a transistor driven HF tank circuit rather than a magnetron to start with. I've achieved CCP before using a similar circuit with argon at STP, so I think I have a starting point now, especially if I don't need to worry about SWR.

User avatar
Doug Coulter
Posts: 1312
Joined: Sun May 27, 2007 7:18 pm
Real name: Doug Coulter
Location: Floyd, VA, USA
Contact:

Re: RF 'echoes'

Post by Doug Coulter » Mon Jan 25, 2010 8:41 pm

Not with a magnetron, no, no worries. With solid state, yes, definately! SS ability to handle any peaks of any sort is insanely terrible compared to any tube thing. Much solid state homebrew class gear will go up in smoke in a second or two with SWR at 2::1, so you'd need to test and match at lower power, or provide a lot more available power in the SS system than you need so you can still light it off at something under half power -- opps, that's not efficient! Sometimes what is best isn't as obvious as you'd think.

Though why you'd use an expensive SS thing compared to a free dumpster magnetron I haven't a clue unless you're running it on batteries or something up on the HV terminal. From reports I've seen, though it doesn't make a heck of a lot of sense, people are using much higher powers at the lower frequencies (13.59 mhz) than at the Ghz, maybe it takes more there, but it's not obvious to me why that would be. I chose the 2.5 ghz as that makes the magnetic field for ECR (which is a huge improvement in efficiency and running range of gas pressure) a decent number -- not too big to make easily, and not so small that fields in the lab would affect it. When you get down to the few gauss range, that becomes a problem itself. ANd if you're not doing ECR, you're not caring about efficiency -- that's a factor of about 3-5 better with than without in testing here, not to mention running a lot bigger range of gas pressures with no other changes, which is merely handy. With the magnets on my thing it will stay lit to below e-6 mbar, which is kind of nice sometimes.
Why guess when you can know? Measure!

Linda Haile
Posts: 230
Joined: Mon Jan 04, 2010 9:28 pm
Real name:

Re: RF 'echoes'

Post by Linda Haile » Mon Jan 25, 2010 9:12 pm

Very interesting, Doug. You ask why transistors? I wouldn't say I'm out to re-invent the wheel, Maybe the wheelwright's apprentice, though. I want to start from first principles, or at least with what I already know, and work up from there. I'm doing this to learn about ICP from the experimental viewpoint.

User avatar
Doug Coulter
Posts: 1312
Joined: Sun May 27, 2007 7:18 pm
Real name: Doug Coulter
Location: Floyd, VA, USA
Contact:

Re: RF 'echoes'

Post by Doug Coulter » Mon Jan 25, 2010 10:01 pm

In that case, you should look into some of the designs the hams are using for solid state linear amps, many are fairly extra rugged and moderately cheap as well. Now, a linear amp won't be very efficient (about 50% if lucky) but it would be a good thing to have in the lab anyway, and those only take a couple watts to drive. Many of the ones I've seen run fine on 12 volt dc, making it possible to do battery operation should you need to go off ground as well. Then get some ideas from the design (many are written up in detail in ham publications) and make your own if you like. You'll find the reasons this or that pcb track gets wide here and narrow there for close control over capacity to ground at some point and a lot of other cool and necessary tweaks that make them rugged and broadband. Many will cover the range of interest here.
Why guess when you can know? Measure!

Linda Haile
Posts: 230
Joined: Mon Jan 04, 2010 9:28 pm
Real name:

Re: RF 'echoes'

Post by Linda Haile » Mon Jan 25, 2010 10:45 pm

Thanks for the advice Doug. I plan to do some experimenting soon starting with my CCP circuit but modified for ICP. As I blow things up I'll modify it with an aim to increase frequency. I'll be using argon to start with as it seems easy to ionize and it is cheap.You can't beat hands on experimentation for learning rather than just copying another design. I find your posts and Chris Bradley's quite interesting but I'm not sure where Chris's experiment is going.

User avatar
Doug Coulter
Posts: 1312
Joined: Sun May 27, 2007 7:18 pm
Real name: Doug Coulter
Location: Floyd, VA, USA
Contact:

Re: RF 'echoes'

Post by Doug Coulter » Mon Jan 25, 2010 11:13 pm

You are most welcome, thanks for putting up with my sometimes excess verbiage.
Hard to beat hands on, and yes, I use argon too. I found N2 does some interesting but mainly nasty extra strange effect things -- I will be putting up some movies about weird sudden ion recombinations/large N polymer decompositions soon on you tube. Till then, Ar or He are good, as Ne is expensive. Though, I must say, really impressive in an ionizer, blinding bright when you get it right.
You could also fool with H2 from electrolyzed water, ala Seltzman's thing. But be careful with that so an accident doesn't pull water into your system.

I do know where Chris is headed, but I don't know if I'm supposed to tell. His theory is fantastic and a real breakthrough, but it seems he got caught up some in some pretty glows short of pushing to make it real according to plan. I think this is fine, as everyone does that at some point before moving on the the real deal Basically, he's trying to make a novel cyclotron, wherein more than one set of beam going roundy round in the same H field "rub" at the edges, and has a way to re-focus with low energy loss those particles that don't fuse but do scatter, as well as handling the monkey-motion along the H field axis, the idea is pretty slick, actually (and patent pending IIRC). Now if we can just get him to start running real conditions and trying to make his idea into actual reality.....My analysis (under NDA) shows it will actually make fusion practical, or a heck of a lot closer than we are now (even including my recent 300x improvements). Could be good, stay tuned. If I could talk him across the pond to use my lab etc, I would.
Why guess when you can know? Measure!

Linda Haile
Posts: 230
Joined: Mon Jan 04, 2010 9:28 pm
Real name:

Re: RF 'echoes'

Post by Linda Haile » Mon Jan 25, 2010 11:32 pm

Maybe he has limited resources, but I think we're going off topic here, although this thread has served it's purpose as far as I'm concerned.

John Futter
Posts: 1460
Joined: Thu Apr 22, 2004 2:29 am
Real name: John Futter
Contact:

Re: RF 'echoes'

Post by John Futter » Thu Jan 28, 2010 10:15 pm

Lyn

Please drop the term echos it grates the nerves to an Rf professional
The correct term is return loss or mismatch loss. A slang of this used commonly by amatuer radio enthusiates is SWR. shortened from VSWR or Voltage Standing Wave Ratio.

My companies website has a little discussion on why bipolar and mosfet RF amps are susceptable to imperfect loads and getting more power to the load by using attenuators (Iknow this sounds counbterintuative)
This probably is the best way for you to use a solid stae PA to drive your ICP or CCP without damaging your PA. Using an SWR meter between the attenuator and your ICP load will then allow you to tune without damaging the PA or have it rapidly turn down the power

Link
http://www.iel-rf.com/AttenuatorsDescription.html

Linda Haile
Posts: 230
Joined: Mon Jan 04, 2010 9:28 pm
Real name:

VSWR...return loss.....mismatch loss....

Post by Linda Haile » Thu Jan 28, 2010 10:36 pm

John, Thankyou for your input, I appreciate it. 'echoes' is a term I came accross while researching RF sources. Doug explained that this is the same as SWR, which I have come accross before. I also came across the term 'auto-match unit', as used in commercial plasma sources. I agree that I should have dropped the term 'echoes' earlier in this thread because, as you point out, it is a slang term. Thanks for the link, I will explore it in greater depth in the near future. I'm still learning the nuances of RF sources and any useful pointers are most appreciated. I believe I now have a rudimentary understanding of what was meant by the term. I admit I do find VSWR a lot easier to comprehend than 'echoes'.

Post Reply