I did a series of runs today at various power supply voltages and gas pressures to try to build up a nice scatter plot of fusor operation versus the parameters. It was largely sucessful, and for what it's worth, also validates a theory Richard has mentioned in the past. This is that having some D implanted in the walls helps things - he thought it might be because it allows ions to be knocked out at the right place in the field, I think it's more like collisions between d-'s and neutrals in the wall - the data support either, as the Q does go down quickly as the walls heat up and the D is released.
There's more here:
http://www.coultersmithing.com/forums/v ... 3929#p3929
As this site makes it a little troublesome to embed multiple youtube videos, and you really do need video to see a 4d plot. I'll be adding to the thread on my site, including a setup that shows Q vs time within a run, which kind of shows the above - the fact that D in the tank walls is good - pretty clearly. But here's a teaser to get you to go see the real data at my site.
Why has no one done this before? This is how you can explain to the newbies, nope, that cranking up the pressure, or the power isn't the way - we all know we've tried it of course, it's an obvious thing to try, but here's real, honest, reduced to real numbers data that makes this unmistakable to anyone who can fog a mirror.
Plotdat.png shows the setup and axis mapping I used to make the video of the 4d plot.
The plot video is here:
http://www.youtube.com/watch?v=WJe0YBAX ... e=youtu.be
As well as on my site. Note, I mis-spoke on the video - Q is normalized to power, not just current in this plot, as you can see from the setup jpg I include here.
This and other data vindicate Richard's theory about putting some D into the tank walls. I see Q about double after doing that under all conditions, until the tank heats up and the Pd I depo'd on there lets all the D go - next time I use Ti which will hold it till higher temperatures, and maybe do more cooling than the fans I have on there now, which are pitiful in terms of my hundreds of watts input.
Here's a teaser plot showing *not* Q, but just sheer output vs the other conditions, no ion source for this one - that's yet to come (took all day to get this much!). With an ion source I can more reliably run at the edge of stability, and also at lower gas pressures, which other data I've collected says helps a lot - but I'll soon have it in pretty form like this so there is no doubt I've interpreted my logs correctly or not. I did this plot with a different setup that just shows neutron cpm, not my arbitrary-unit Q.
Parametric sweep of fusor parameters plotted in 4 dimensions
-
Doug Coulter
- Posts: 1312
- Joined: Sun May 27, 2007 3:18 pm
- Real name: Doug Coulter
- Location: Floyd, VA, USA
- Contact:
Parametric sweep of fusor parameters plotted in 4 dimensions
- Attachments
-
-
Why guess when you can know? Measure!
-
Jim Kovalchick
- Posts: 717
- Joined: Wed Apr 13, 2011 8:00 pm
- Real name:
Re: Parametric sweep of fusor parameters plotted in 4 dimensions
Thanks for sharing Doug. Your plot looks very similar to one generated by U of Wisconsin I was just looking at. I wish I could find the link again, but suffice to say that theirs expanded the surface significantly with much higher voltages.
The video was great for explaining your data collection. I noticed that occasionally your plasma extinguishes for an instant. Was that from arcing in your power supply or the grid?
Jim K
The video was great for explaining your data collection. I noticed that occasionally your plasma extinguishes for an instant. Was that from arcing in your power supply or the grid?
Jim K
-
Doug Coulter
- Posts: 1312
- Joined: Sun May 27, 2007 3:18 pm
- Real name: Doug Coulter
- Location: Floyd, VA, USA
- Contact:
Re: Parametric sweep of fusor parameters plotted in 4 dimensions
Thanks for the nice words, Jim.
That just happens sometimes for a number of causes, usually but not always when running at the lower end of pressures that will sustain current draw. Once in awhile, there will be a little arc down my feedthrough (a homebrew thing), but that's fairly rare.
This is a lot less common when I'm running an ion source of some sort (even a crummy one), I get much more stable operation at any pressure doing that. I wanted these runs to show a more-average hobbyist type fusor, though. I will also do some runs with an ion source that won't have that as much.
Here's (a?) the rub. While my data aq can be fooled by a sudden burst of neutrons before the current monitor can respond, and thus report higher Q than is "real", I have independently found out that indeed, in a pulsed mode, the Q is much higher (10-25x) on average. I have an ion source I can pulse at various rep rates and using total neutrons out vs total joules in, it's much better, but more difficult to explain.
In these runs, I edited out 5 "outliers" that showed ridiculous Q because of that data aq artifact, so it wouldn't look "better than the truth". That's out of a heck of a lotta points (you can see which if you download the logs from my site - my format allows a # to comment out a data line). But I can actually do that well with a pulsed ion source and intermittent draw, using the averages, and things do work better.
That will be its own report, though, and not the next one.
I still have some data mining to do on this one - I can easily make a plot for Q, with the color mapped to time within each run...that should show the effect of heat liberating the D from the tank coating, and make the point in an unmistakable way. We'll see, the data will tell.
The next sweep I plan to do is with a plain old DC ion source so I can better sweep the pressure parameter space and see what lower pressures (thus fewer unintended low energy scattering collisions with charge exchanges etc) and see what that does - it's the way I normally run anyway.
My own theory, which differs a little from Richards, is that we are getting some charge exchange and therefore some negative D ions, that are then accelerated into the walls at around double the power supply voltage, which is (I think) why the Q about doubles if you coat the chamber walls with a thin layer of Pd, as I have. You can see Q going down vs time (which equates roughly to chamber temperature) which says, yes, something is going on when you have D in the walls - could be Richard's theory that this makes more ions in the right place in the field, or mine, which involves charge exchange - I don't have enough data yet to say either way - but believe me, I'm looking at this.
I think the time I spent building this cool tool-set wasn't wasted at all.
I may even hang a thermocouple on the tank walls to kinda prove what I know already about that, and try using Ti coating instead, which holds D better up to higher temps, and produces fewer, less energetic, X rays to boot than the higher Z Pd does - it about doubled my X ray output over plain stainless when I added the Pd, along with about doubling the neutrons and Q.
Then on to pulse mode, perhaps with a resonant circuit in the ballast with things timed so that the fusor re-lights when that voltage on the grid has peaked from the ringing from the previous pulse. I've done that and gotten Q's about 400-500 times what I get in the nice "stable" modes, which turn out to be about the worst place to actually run for Q. However, the way I got the pulsing before turns out to be moderately difficult to replicate in other labs, only one guy has managed so far. I will set up an easier to replicate situation, which will also be easier to fiddle with and find sweet spots in.
And, a bit of resonant ringing is an easy way to get a higher grid voltage at just the right instant, perhaps double what the power supply will provide...
You are right - my supply tops out at 50 or so KV (Spellman SL2KW), and as this shows, more current might make more neutrons, but does so at lower Q. Whether this is due to higher gas pressure needed to get to higher current, which then also creates more unintended collisions - well, I'll know soon enough - I can run the same currents at much lower pressures with an ion source. More volts would be better, but that would mean going to something like Carl showed with the feedthrough in oil on the air side - a hassle to say the least, and the X rays are already getting pretty fierce. I have supplies up to 180kv, but dealing with that in air is a real problem.
What I find interesting is that no one seems to have paid one bit of attention to the conservation laws, like spin - stuff known since Pauli, and definitely shown true in other experiments. You could easily assume that the particle orientations in a fusor are not only non-random (the field only pulls on the proton in the D) but *exactly wrong* to promote a good cross section - this is like shooting humans at each other from cannons head on and expecting to get babies - it's not an efficient way, the orientation is all wrong according to well known laws. So that's the next thrust, if you will, I'm working on a linac for D right now that will let me have control over some of that - a more complex apparatus with simpler reaction conditions, more control, and we shall see. I might be onto something head-slappingly obvious (but only in hindsight, no one else has mentioned this in the fusion lit that I'm aware of). Since I know a little of the politics of science, I'm not expecting any acclaim if it works - just "sure, we knew that" except for billions of bucks to PhD's that ignored it for all these decades...
If you check, you find that there are more possible ways to conserve spin in a DT reaction than in DD, and strangely, the cross section is just about that much larger in DT. Hmmm.
I think just making the orientations more random - via pulsing - will improve Q a lot, and have some data to that effect. But if random is better than non-random-dead-wrong, then non-random-just-right ought to be tons better yet - eh?
This is the kind of opportunity we amateurs can pursue, since we're not burdened with bureaucracy and stiff mindsets (at least not unless we choose to be), so I'm gonna do it!
I still don't think I or anyone else yet has the key to net power - a Q improvement of 500 sounds pretty good - but now I get 500 uW out for 300w in? More than that needs to be done! A couple orders of magnitude is nice, but...we need a lot more than that. On the other hand, if we can get that much improvement in reaction rate, using reactors to make medical isotopes might become obsolete pretty quickly - power isn't the only useful output here.
For good or ill, I'll post that plot of Q vs time-in-run soon, and we'll see what it shows. Just a matter of changing some parameters in the plotter software. I've noticed that it tends to go down with time, along with net neutron outputs in my lashup, it should show in the plot.
That just happens sometimes for a number of causes, usually but not always when running at the lower end of pressures that will sustain current draw. Once in awhile, there will be a little arc down my feedthrough (a homebrew thing), but that's fairly rare.
This is a lot less common when I'm running an ion source of some sort (even a crummy one), I get much more stable operation at any pressure doing that. I wanted these runs to show a more-average hobbyist type fusor, though. I will also do some runs with an ion source that won't have that as much.
Here's (a?) the rub. While my data aq can be fooled by a sudden burst of neutrons before the current monitor can respond, and thus report higher Q than is "real", I have independently found out that indeed, in a pulsed mode, the Q is much higher (10-25x) on average. I have an ion source I can pulse at various rep rates and using total neutrons out vs total joules in, it's much better, but more difficult to explain.
In these runs, I edited out 5 "outliers" that showed ridiculous Q because of that data aq artifact, so it wouldn't look "better than the truth". That's out of a heck of a lotta points (you can see which if you download the logs from my site - my format allows a # to comment out a data line). But I can actually do that well with a pulsed ion source and intermittent draw, using the averages, and things do work better.
That will be its own report, though, and not the next one.
I still have some data mining to do on this one - I can easily make a plot for Q, with the color mapped to time within each run...that should show the effect of heat liberating the D from the tank coating, and make the point in an unmistakable way. We'll see, the data will tell.
The next sweep I plan to do is with a plain old DC ion source so I can better sweep the pressure parameter space and see what lower pressures (thus fewer unintended low energy scattering collisions with charge exchanges etc) and see what that does - it's the way I normally run anyway.
My own theory, which differs a little from Richards, is that we are getting some charge exchange and therefore some negative D ions, that are then accelerated into the walls at around double the power supply voltage, which is (I think) why the Q about doubles if you coat the chamber walls with a thin layer of Pd, as I have. You can see Q going down vs time (which equates roughly to chamber temperature) which says, yes, something is going on when you have D in the walls - could be Richard's theory that this makes more ions in the right place in the field, or mine, which involves charge exchange - I don't have enough data yet to say either way - but believe me, I'm looking at this.
I think the time I spent building this cool tool-set wasn't wasted at all.
I may even hang a thermocouple on the tank walls to kinda prove what I know already about that, and try using Ti coating instead, which holds D better up to higher temps, and produces fewer, less energetic, X rays to boot than the higher Z Pd does - it about doubled my X ray output over plain stainless when I added the Pd, along with about doubling the neutrons and Q.
Then on to pulse mode, perhaps with a resonant circuit in the ballast with things timed so that the fusor re-lights when that voltage on the grid has peaked from the ringing from the previous pulse. I've done that and gotten Q's about 400-500 times what I get in the nice "stable" modes, which turn out to be about the worst place to actually run for Q. However, the way I got the pulsing before turns out to be moderately difficult to replicate in other labs, only one guy has managed so far. I will set up an easier to replicate situation, which will also be easier to fiddle with and find sweet spots in.
And, a bit of resonant ringing is an easy way to get a higher grid voltage at just the right instant, perhaps double what the power supply will provide...
You are right - my supply tops out at 50 or so KV (Spellman SL2KW), and as this shows, more current might make more neutrons, but does so at lower Q. Whether this is due to higher gas pressure needed to get to higher current, which then also creates more unintended collisions - well, I'll know soon enough - I can run the same currents at much lower pressures with an ion source. More volts would be better, but that would mean going to something like Carl showed with the feedthrough in oil on the air side - a hassle to say the least, and the X rays are already getting pretty fierce. I have supplies up to 180kv, but dealing with that in air is a real problem.
What I find interesting is that no one seems to have paid one bit of attention to the conservation laws, like spin - stuff known since Pauli, and definitely shown true in other experiments. You could easily assume that the particle orientations in a fusor are not only non-random (the field only pulls on the proton in the D) but *exactly wrong* to promote a good cross section - this is like shooting humans at each other from cannons head on and expecting to get babies - it's not an efficient way, the orientation is all wrong according to well known laws. So that's the next thrust, if you will, I'm working on a linac for D right now that will let me have control over some of that - a more complex apparatus with simpler reaction conditions, more control, and we shall see. I might be onto something head-slappingly obvious (but only in hindsight, no one else has mentioned this in the fusion lit that I'm aware of). Since I know a little of the politics of science, I'm not expecting any acclaim if it works - just "sure, we knew that" except for billions of bucks to PhD's that ignored it for all these decades...
If you check, you find that there are more possible ways to conserve spin in a DT reaction than in DD, and strangely, the cross section is just about that much larger in DT. Hmmm.
I think just making the orientations more random - via pulsing - will improve Q a lot, and have some data to that effect. But if random is better than non-random-dead-wrong, then non-random-just-right ought to be tons better yet - eh?
This is the kind of opportunity we amateurs can pursue, since we're not burdened with bureaucracy and stiff mindsets (at least not unless we choose to be), so I'm gonna do it!
I still don't think I or anyone else yet has the key to net power - a Q improvement of 500 sounds pretty good - but now I get 500 uW out for 300w in? More than that needs to be done! A couple orders of magnitude is nice, but...we need a lot more than that. On the other hand, if we can get that much improvement in reaction rate, using reactors to make medical isotopes might become obsolete pretty quickly - power isn't the only useful output here.
For good or ill, I'll post that plot of Q vs time-in-run soon, and we'll see what it shows. Just a matter of changing some parameters in the plotter software. I've noticed that it tends to go down with time, along with net neutron outputs in my lashup, it should show in the plot.
Why guess when you can know? Measure!
-
Doug Coulter
- Posts: 1312
- Joined: Sun May 27, 2007 3:18 pm
- Real name: Doug Coulter
- Location: Floyd, VA, USA
- Contact:
Q vs time plotted in 4 dimensions
Is here with some words:
http://www.coultersmithing.com/forums/v ... 3935#p3935
Or here on youtube.
http://www.youtube.com/watch?v=qgczy369 ... e=youtu.be
Note, somehow my tools got the audio WAY out of sync with the picture and added some hum not present in thee hardware here, no idea why, but as to the sync, the computer was heavily loaded with another app (stock trading) at the time.
I'm thinking more and more that adding the Pd so there's more D in the chamber walls improves things due to fusion right at the walls, not because the ions are produced in a better place. This is because both myself and Tyler have seen location-dependent neutron output in other runs, he with a bubble detector, me with my small hornyak button. This implies either neutron beaming, which no theory I'm aware of supports, or fusion right at the tank walls which can look like a beam due to the nearness of the detector - you'd only ever see that with a non moderated type fast neutron detector, as a moderator would smear out all the spatial resolution.
At any rate, as the tank walls heat up and the D is driven out, the Q goes down...it's pretty obvious on this plot, even though there are about 3 points that look like outliers (out of a few thousand points).
Such is life - a few outliers will fall into any data aq, the main question is - were they real, or just errors.
http://www.coultersmithing.com/forums/v ... 3935#p3935
Or here on youtube.
http://www.youtube.com/watch?v=qgczy369 ... e=youtu.be
Note, somehow my tools got the audio WAY out of sync with the picture and added some hum not present in thee hardware here, no idea why, but as to the sync, the computer was heavily loaded with another app (stock trading) at the time.
I'm thinking more and more that adding the Pd so there's more D in the chamber walls improves things due to fusion right at the walls, not because the ions are produced in a better place. This is because both myself and Tyler have seen location-dependent neutron output in other runs, he with a bubble detector, me with my small hornyak button. This implies either neutron beaming, which no theory I'm aware of supports, or fusion right at the tank walls which can look like a beam due to the nearness of the detector - you'd only ever see that with a non moderated type fast neutron detector, as a moderator would smear out all the spatial resolution.
At any rate, as the tank walls heat up and the D is driven out, the Q goes down...it's pretty obvious on this plot, even though there are about 3 points that look like outliers (out of a few thousand points).
Such is life - a few outliers will fall into any data aq, the main question is - were they real, or just errors.
Why guess when you can know? Measure!
-
Carl Willis
- Posts: 2841
- Joined: Thu Jul 26, 2001 7:33 pm
- Real name: Carl Willis
- Location: Albuquerque, New Mexico, USA
- Contact:
Re: Parametric sweep of fusor parameters plotted in 4 dimensions
Nice collection of data, Doug, and you've mustered together an excellent data-logging capability that will be widely useful in future fusion experiments.
Do you automate this process? It looks to me like you set a constant voltage on the power supply and sweep through pressure either by manual means or automatically. How long do you count at each point?
-Carl
Do you automate this process? It looks to me like you set a constant voltage on the power supply and sweep through pressure either by manual means or automatically. How long do you count at each point?
-Carl
-
Doug Coulter
- Posts: 1312
- Joined: Sun May 27, 2007 3:18 pm
- Real name: Doug Coulter
- Location: Floyd, VA, USA
- Contact:
Re: Parametric sweep of fusor parameters plotted in 4 dimensions
FWIW, I've open sourced all the code and hardware design on this setup. You can also get us to make you one - we're not in it for the money, so they will be pretty cheap as these things go.
I think it will be worth it to a lot of people - this has told me just what direction to go in, which will save me money on false starts.
But there's no need if you can make your own - all the info to do it is free and already up on my site. I may extend the software in the uP to add more a/d channels, time will tell - it's not too hard and they are there in the hardware. Just that the cost of all those BNC's starts to add up if you're thinking product for this. I've not yet found a cheaper connector that does as good a job of keeping the EMI out.
The process isn't automated other than the data taking, once begun. I'm lucky to have this killer-good Spellman supply, so for each run I simply set the voltage and current limits, then fiddled with the gas, or let the system drift - which it does with temperature and "phase of the moon" seemingly sometimes.
Each point is whatever the fusor was doing on the logged parameters for one second. The number of points is just how many seconds I ran the data aq before stopping it. In this case, to get rid of outliers that make the plot autoscale poorly, I started and stopped the data aq inside a fusor run - no points taken at zero input, at least none on purpose - else a cosmic ray hitting the neutron detector of course reports infinite Q! Each run at different settings went into its own file, and I simply concatenated all those files to make this plot. This is so each run could start clean from identical conditions otherwise.
All the PC side software is in perl, which means that if you install the right stuff, it will run on linux (easy), mac (also easy) and windows (more stuff to install there, like the gtk runtime and perl itself).
If anyone needs a binary of the uP code (or the source) all they have to do is go to my site, or ask me for the latest - which I will keep updating up there in the thread for each part.
I'm using gnuplot to make the plots. Most of the work was figuring out the documentation for that. *After* you get something working, you can kind of see what the doc meant, but not necessarily before, there are some serious quirks there (you set line type and color to set point type and color for example, and they don't tell you that the set point style and color commands don't work!). Most of the rest of the challenge was figuring out how to make it simple enough for a scientist to use, without him having to become a gnuplot or perl expert. I'm not sure I really succeeded there, but it works for me and a couple other people so far. It's not perfect by any means, and for gosh sake, don't type 'rm -rf'; into one of those axis mapping code windows - it'll *work*!
Or you can roll your own, the log file format is pure ascii (but with lthe inux line terminator of \\n, not \\r\\n for windows or \\r for macs - everyone went out of their way to be incompatible there). Gnuplot will eat it directly, but doing any programming on the axis mapping means learning its quirky math stuff, something I was unwilling to do myself.
Here's a couple sample lines of a log file:
# name:/home/doug/bin/sdaq/8_5_2012_16_32_37.log
# 50kv20ma no ion source, unstable pressure
0:00:01 C0 10 TC0 10 C1 17 TC1 17 AD0 29530 AD1 17812 AD2 38854 AD3 33084
0:00:02 C0 14 TC0 24 C1 20 TC1 37 AD0 29952 AD1 17862 AD2 38902 AD3 33116
0:00:03 C0 12 TC0 36 C1 22 TC1 59 AD0 30179 AD1 17811 AD2 38856 AD3 33087
0:00:04 C0 11 TC0 47 C1 29 TC1 88 AD0 28086 AD1 17831 AD2 38873 AD3 33067
...
0:07:36 C0 42 TC0 16409 C1 19 TC1 12164 AD0 34995 AD1 17877 AD2 38939 AD3 33139
0:07:37 C0 30 TC0 16439 C1 18 TC1 12182 AD0 35206 AD1 17853 AD2 38904 AD3 33083
0:07:38 C0 37 TC0 16476 C1 13 TC1 12195 AD0 35131 AD1 17840 AD2 38887 AD3 33086
0:07:39 C0 27 TC0 16503 C1 19 TC1 12214 AD0 34995 AD1 17859 AD2 38906 AD3 33113
As you can see, parsing this, or just reading it is fairly trivial (on purpose) - I'd bet things like Excel would eat this format fine too. The device itself just looks like a serial port on USB, so you don't need any of my (free) software anyway - you can use a terminal program to collect and log it. Bonus - most serial terminal programs will eat any line terminator and supply the right one for your system instead.
For display, I scale the counts per second up to counts per minute, as that's what I think in.
During the acquisition, I have real time updating 2-d plots of counts and a/d inputs so I can be assured I'm not driving something to clipping or getting some sort of really strange noise in my signals. It's what I actually watch when tweaking the fusor parameters during a run, as it's all in one place on a big screen TV monitor and easier than looking at all the meters all over the rig. It also lets me stay behind all that nice shielding I have so I don't get fried!
Obviously, I'm a little proud of this one - I hopefully earned it. Here's a picture of the device. The PIC uP used can also be had in other packages that are easier to DIY with. Not all the parts on this board are required, like the max-232 chip - that's my debug port. The gray connector is for in-circuit programming. The connectors cost more than the rest! The main additions to the real thing are a box to put it in, and some RC low pass filters on the a/d inputs, and all those BNCs. It actually takes and sums 64 samples/channel/second to get a little better resolution than the 10 bit a/d has - it's actually good to 12-13 bits out of the 16 produced, with noise providing the dithering. Old DSP trick.
Props to Joe Jarski who did the PCB layout so nicely that I've been able to use the same board in multiple projects. There's quite a lot of PIC I/O we haven't tapped into yet as well - I2C, PWM and so on. I may use one to automate my gas flow control for example.
I think it will be worth it to a lot of people - this has told me just what direction to go in, which will save me money on false starts.
But there's no need if you can make your own - all the info to do it is free and already up on my site. I may extend the software in the uP to add more a/d channels, time will tell - it's not too hard and they are there in the hardware. Just that the cost of all those BNC's starts to add up if you're thinking product for this. I've not yet found a cheaper connector that does as good a job of keeping the EMI out.
The process isn't automated other than the data taking, once begun. I'm lucky to have this killer-good Spellman supply, so for each run I simply set the voltage and current limits, then fiddled with the gas, or let the system drift - which it does with temperature and "phase of the moon" seemingly sometimes.
Each point is whatever the fusor was doing on the logged parameters for one second. The number of points is just how many seconds I ran the data aq before stopping it. In this case, to get rid of outliers that make the plot autoscale poorly, I started and stopped the data aq inside a fusor run - no points taken at zero input, at least none on purpose - else a cosmic ray hitting the neutron detector of course reports infinite Q! Each run at different settings went into its own file, and I simply concatenated all those files to make this plot. This is so each run could start clean from identical conditions otherwise.
All the PC side software is in perl, which means that if you install the right stuff, it will run on linux (easy), mac (also easy) and windows (more stuff to install there, like the gtk runtime and perl itself).
If anyone needs a binary of the uP code (or the source) all they have to do is go to my site, or ask me for the latest - which I will keep updating up there in the thread for each part.
I'm using gnuplot to make the plots. Most of the work was figuring out the documentation for that. *After* you get something working, you can kind of see what the doc meant, but not necessarily before, there are some serious quirks there (you set line type and color to set point type and color for example, and they don't tell you that the set point style and color commands don't work!). Most of the rest of the challenge was figuring out how to make it simple enough for a scientist to use, without him having to become a gnuplot or perl expert. I'm not sure I really succeeded there, but it works for me and a couple other people so far. It's not perfect by any means, and for gosh sake, don't type 'rm -rf'; into one of those axis mapping code windows - it'll *work*!
Or you can roll your own, the log file format is pure ascii (but with lthe inux line terminator of \\n, not \\r\\n for windows or \\r for macs - everyone went out of their way to be incompatible there). Gnuplot will eat it directly, but doing any programming on the axis mapping means learning its quirky math stuff, something I was unwilling to do myself.
Here's a couple sample lines of a log file:
# name:/home/doug/bin/sdaq/8_5_2012_16_32_37.log
# 50kv20ma no ion source, unstable pressure
0:00:01 C0 10 TC0 10 C1 17 TC1 17 AD0 29530 AD1 17812 AD2 38854 AD3 33084
0:00:02 C0 14 TC0 24 C1 20 TC1 37 AD0 29952 AD1 17862 AD2 38902 AD3 33116
0:00:03 C0 12 TC0 36 C1 22 TC1 59 AD0 30179 AD1 17811 AD2 38856 AD3 33087
0:00:04 C0 11 TC0 47 C1 29 TC1 88 AD0 28086 AD1 17831 AD2 38873 AD3 33067
...
0:07:36 C0 42 TC0 16409 C1 19 TC1 12164 AD0 34995 AD1 17877 AD2 38939 AD3 33139
0:07:37 C0 30 TC0 16439 C1 18 TC1 12182 AD0 35206 AD1 17853 AD2 38904 AD3 33083
0:07:38 C0 37 TC0 16476 C1 13 TC1 12195 AD0 35131 AD1 17840 AD2 38887 AD3 33086
0:07:39 C0 27 TC0 16503 C1 19 TC1 12214 AD0 34995 AD1 17859 AD2 38906 AD3 33113
As you can see, parsing this, or just reading it is fairly trivial (on purpose) - I'd bet things like Excel would eat this format fine too. The device itself just looks like a serial port on USB, so you don't need any of my (free) software anyway - you can use a terminal program to collect and log it. Bonus - most serial terminal programs will eat any line terminator and supply the right one for your system instead.
For display, I scale the counts per second up to counts per minute, as that's what I think in.
During the acquisition, I have real time updating 2-d plots of counts and a/d inputs so I can be assured I'm not driving something to clipping or getting some sort of really strange noise in my signals. It's what I actually watch when tweaking the fusor parameters during a run, as it's all in one place on a big screen TV monitor and easier than looking at all the meters all over the rig. It also lets me stay behind all that nice shielding I have so I don't get fried!
Obviously, I'm a little proud of this one - I hopefully earned it. Here's a picture of the device. The PIC uP used can also be had in other packages that are easier to DIY with. Not all the parts on this board are required, like the max-232 chip - that's my debug port. The gray connector is for in-circuit programming. The connectors cost more than the rest! The main additions to the real thing are a box to put it in, and some RC low pass filters on the a/d inputs, and all those BNCs. It actually takes and sums 64 samples/channel/second to get a little better resolution than the 10 bit a/d has - it's actually good to 12-13 bits out of the 16 produced, with noise providing the dithering. Old DSP trick.
Props to Joe Jarski who did the PCB layout so nicely that I've been able to use the same board in multiple projects. There's quite a lot of PIC I/O we haven't tapped into yet as well - I2C, PWM and so on. I may use one to automate my gas flow control for example.
- Attachments
-
Why guess when you can know? Measure!
-
Doug Coulter
- Posts: 1312
- Joined: Sun May 27, 2007 3:18 pm
- Real name: Doug Coulter
- Location: Floyd, VA, USA
- Contact:
Re: Parametric sweep of fusor parameters plotted in 4 dimensions
BTW, if you need much better time resolution but not perfect data continuity, your best bet appears to be to get a GW-Instek digital 'scope and the GDSH tool (for linux only - as it's Tektronix-compatible, the lawyers have pulled it down from the net so Tek can sell their overpriced version, but you can get it from me). It will let you download the data from the scope as either text or bmp's - and they make these with multi-Ghz sample rates. When I get to the pulsing stuff, I'll be grabbing some of that type of data too, as one/second isn't going to cut it for tuning that!
Why guess when you can know? Measure!
-
Richard Hull
- Moderator
- Posts: 15024
- Joined: Fri Jun 15, 2001 9:44 am
- Real name: Richard Hull
Re: Parametric sweep of fusor parameters plotted in 4 dimensions
Superb work and data reporting! The kind of stuff this forum was made for, but rarely sees presented. I look forward to see if the issue of shell loaded target fusion at the wall vs. increased ion production at the shell due to pop out deuterons and deuterium atoms will be solved in a divintive and satisfactory manner.
I figure that I see the fusion increase due to loading of the 8-11% nickel in the 304 stainless shell. The fusor definitely improves over a very long operating session at high output levels.
Nickel and titanium are both good hydrogen absorbers. I would imagine a thick film of electrochemically deposited Pd, Ti or Ni would be demanded as the bombardment of failed deuterium neutrals could go through several moleular layers and succeeding neutrals that hit these would continuously pound Ds into the film. (those that were not bounced out due to the impact, of course.) Once loaded, thermal bounce out and simple kinetic ejection would proceed in keeping with a stasis "load-eject regime" that would probably develop regardless of absorber material.
Richard Hull
I figure that I see the fusion increase due to loading of the 8-11% nickel in the 304 stainless shell. The fusor definitely improves over a very long operating session at high output levels.
Nickel and titanium are both good hydrogen absorbers. I would imagine a thick film of electrochemically deposited Pd, Ti or Ni would be demanded as the bombardment of failed deuterium neutrals could go through several moleular layers and succeeding neutrals that hit these would continuously pound Ds into the film. (those that were not bounced out due to the impact, of course.) Once loaded, thermal bounce out and simple kinetic ejection would proceed in keeping with a stasis "load-eject regime" that would probably develop regardless of absorber material.
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
The more complex the idea put forward by the poor amateur, the more likely it will never see embodiment
Fusion is the energy of the future....and it always will be
The more complex the idea put forward by the poor amateur, the more likely it will never see embodiment
-
Mike Beauford
- Posts: 419
- Joined: Wed Apr 25, 2007 2:24 pm
- Real name:
- Location: Morton Grove, IL
Re: Parametric sweep of fusor parameters plotted in 4 dimensions
Hi Doug,
Very good info. Thanks and keep up the great work!
Mike Beauford
Very good info. Thanks and keep up the great work!
Mike Beauford
Mike Beauford
-
Doug Coulter
- Posts: 1312
- Joined: Sun May 27, 2007 3:18 pm
- Real name: Doug Coulter
- Location: Floyd, VA, USA
- Contact:
Re: Parametric sweep of fusor parameters plotted in 4 dimensions
Thanks for the nice words, Richard (evidently the first reply I posted got lost in the ether).
I'm just sorry it took me so long to do it - remember I mentioned it at HEAS back in '08 or so.
I want to officially thank you for the idea to coat the walls with something that holds D better, though I think we might disagree with why that's good - it's not important - (and hard to measure either way) - but it works, which IS important. In my world, giving credit where it's due is a good thing.
I have a lot more work to do gathering data here before I change the fusor innards again. When I do, I'll change that Pd into Ti instead - via evaporation, since you can't deposit Ti any other way I know of.
Next are some runs with DC and pulsed ion sources which want to have everything else the same for a good comparison. In that sense, this data is only the baseline. Now we can compare this to data taken with various ion sources and a wider range of gas pressures. Just so people know, I'm using the "straight" output of the PKR-251 pressure sensor, which is known to read off a bit on hydrogen, so those numbers will be different than what you get with a cap manometer. See cal chart included.
It might be prudent to improve the cooling there too - my fans are pitiful in the face of a few hundred watts of heat - I might need to go to fluid cooling to keep most of the D there. And yes, other plots I've put up on youtube (search for DCFusor, that's me up there) of Q vs time in run show that the Q goes down as the tank heats up and loses the D in the walls - that's a pretty good test.
I'm just sorry it took me so long to do it - remember I mentioned it at HEAS back in '08 or so.
I want to officially thank you for the idea to coat the walls with something that holds D better, though I think we might disagree with why that's good - it's not important - (and hard to measure either way) - but it works, which IS important. In my world, giving credit where it's due is a good thing.
I have a lot more work to do gathering data here before I change the fusor innards again. When I do, I'll change that Pd into Ti instead - via evaporation, since you can't deposit Ti any other way I know of.
Next are some runs with DC and pulsed ion sources which want to have everything else the same for a good comparison. In that sense, this data is only the baseline. Now we can compare this to data taken with various ion sources and a wider range of gas pressures. Just so people know, I'm using the "straight" output of the PKR-251 pressure sensor, which is known to read off a bit on hydrogen, so those numbers will be different than what you get with a cap manometer. See cal chart included.
It might be prudent to improve the cooling there too - my fans are pitiful in the face of a few hundred watts of heat - I might need to go to fluid cooling to keep most of the D there. And yes, other plots I've put up on youtube (search for DCFusor, that's me up there) of Q vs time in run show that the Q goes down as the tank heats up and loses the D in the walls - that's a pretty good test.
- Attachments
-
Why guess when you can know? Measure!
-
Chris Bradley
- Posts: 2930
- Joined: Fri May 02, 2008 7:05 am
- Real name:
Re: Parametric sweep of fusor parameters plotted in 4 dimensions
Doug Coulter wrote:
> The process isn't automated other than the data taking, once begun. I'm lucky to have this killer-good Spellman supply, so for each run I simply set the voltage and current limits, then fiddled with the gas, or let the system drift - which it does with temperature and "phase of the moon" seemingly sometimes.
Hi Doug, the data aq is absolutely fantastic. Just wondering if it may be better to use some standardised way to roam through the parameter space to collect the data, rather than trusting in the 'phases of the moon'.
If you make changes next time to the way you approached certain parameters in this run [for example if you approached a parameter point from high pressure to low rather than vice versa (or whatever) and get a significantly different answer] would you be likely to forget the different scenarios that lead to that differing result, which might mean you'd be prone to conclude it was something else you'd done differently? One second for the readings to stabilise seems quite a short time. I'd imagine that what happened in the preceding second might contribute significantly to the readings in the next second.
> The process isn't automated other than the data taking, once begun. I'm lucky to have this killer-good Spellman supply, so for each run I simply set the voltage and current limits, then fiddled with the gas, or let the system drift - which it does with temperature and "phase of the moon" seemingly sometimes.
Hi Doug, the data aq is absolutely fantastic. Just wondering if it may be better to use some standardised way to roam through the parameter space to collect the data, rather than trusting in the 'phases of the moon'.
If you make changes next time to the way you approached certain parameters in this run [for example if you approached a parameter point from high pressure to low rather than vice versa (or whatever) and get a significantly different answer] would you be likely to forget the different scenarios that lead to that differing result, which might mean you'd be prone to conclude it was something else you'd done differently? One second for the readings to stabilise seems quite a short time. I'd imagine that what happened in the preceding second might contribute significantly to the readings in the next second.
-
Doug Coulter
- Posts: 1312
- Joined: Sun May 27, 2007 3:18 pm
- Real name: Doug Coulter
- Location: Floyd, VA, USA
- Contact:
Re: Parametric sweep of fusor parameters plotted in 4 dimensions
The data aq is kind of a compromise. The one second period is because shorter, and I might not get very reliable counter data (unless I demand high count rate counters - but my fusor swamps my 3He tube for example, and it counts wrong due to deadtime issues), and longer and the other channels (they are all produced in a batch, and timestamped together) smears out the other parameters too much.
There's no problem with approaching from either direction, in fact, this dataset did that multiple times per run going both ways, and the results at a particular set of parameters don't have all that much smear, though there is a little, mostly due to that slow second (from the a/d point of view). There's a little bleed in the a/d data from the previous second or two due to the anti-aliasing filter. There is also some scatter in the counters due to the short (for them) time interval used. Compromise is kind of like that...
However, don't forget that what is saved here is the raw timestamped data (something PCs are bad at, which is why we do it in an embedded uP - we can guarantee *all* the data for that second started and stopped right on the seconds boundary within a fraction of a microsecond), we can look at it any way we want, and one of the demos I put up on youtube indeed shows it vs time (actually, a few of them do), so you can see if there was a significant precursor situation to whatever feature you're examining.
I am still exploring what gnuplot can do for us, and what's worth doing. One thing I can do even on this multidimensional plot is to draw arrows between the points, according to their time order, for example, however - on this much data, it obfuscates rather than informs.
These parameters are NOT independent in a fusor. At X gas pressure, you cannot have arbitrary volts and current - they all interact. In fact, you can have a situation where the thing won't light off at a given set of pressure, volts - but once lit, will stay lit just fine (and it will stay lit for a lower pressure once lit) - there is a divergent feedback once there are a bunch of ions present that drives things to a limit (what we normally call stable operation).
I am of course, looking at automating my gas control stuff. (to do that well, I need finer control per pulse of the inlet and outlet solenoids than I now have) But then you'd have to track that too if you were interested in say, how fresh this batch of fuel was - obviously contamination builds up during a run due to things getting pounded off the walls, sputtering off the grid and so on...and as mentioned, temperature has an effect - which means, as you say, the the immediately previous conditions are somewhat important since it takes awhile to heat and cool.
This again, is why we do our science "right" and always, ALWAYS, keep all the raw data just as we collect it - that way, if a new idea comes up, we can tailor our display method of it to see if that new idea has merit or not. You can't do new ideas in data-mining if you don't keep the pure raw data around! The format does allow for comments in the log files, and I do put them in there. We just don't plot the comments...
There are of course some practical limits on what's worth logging. I'd like to have chamber temperature too for example. At some point though, the data rate gets out of control for the systems involved (The PC is generally the limit on bursts of data, it's not really a good real time thing), and the cost gets very high for all those channels. There will always be more we wish we could do, I suppose.
Here I'm trying to keep the line length per burst under 80 chars and still keep it human readable, since there's a lot of useful software out there that barfs on longer lines.
For high time resolution, I've been adding a GW Instek scope to this and grabbing screens from it's 4 channel, 2.5ghz sample rate data set - you can't get it all, but it lets you put a time-microscope on individual events. This can all be put into say, a MySQL database for later mining, though once you have two separate interfaces into a PC (and didn't get to write the opsys yourself), keeping them accurately time-aligned is more or less impossible. Two simultaneous inputs into a PC can be recognized in either order even down at the driver/interrupt level. If you have two programs stuffing things into a database, vagaries of the opsys time-slicing add to the uncertainty for both programs, and the database program *itself* may not put things in in the order recieved. We discovered this on a more ambitious prototype of this, where we had the scope and a 10 channel pro audio card all going at once along with this. At the speed transient events happen in a fusor, you really can't depend on sorting out which was cause, and which was effect unless for example, they are both on the same input (for example, the soundcard or the scope). If the signals come in on different interfaces - all bets are off.
So, doing that gets you into government-money-level custom built stuff, the sort of stuff I used to do for NSA intercept systems. I thought this was rather a decent compromise since I can take a family to dinner (admittedly at a good restaurant) for less than this will likely retail for.
No need to colonize mars yet - I'm just getting to earth orbit - while most people can't even fly, for now.
I'm putting all the demo movies up at: http://www.youtube.com/user/DCFusor?feature=mhee
And embedding them on my site, with discussion: http://www.coultersmithing.com/forums/v ... 34&start=0
which includes plots vs time on all this if that would answer any of your "what came first and from which direction" kinds of questions. You can see for example, the pressure varying (either due to my actions or randomly) and what happens before and after quite clearly on those, but not some other interesting facets we can see in the 4d plots better - depends on what you want to look for in the data mine.
There's no problem with approaching from either direction, in fact, this dataset did that multiple times per run going both ways, and the results at a particular set of parameters don't have all that much smear, though there is a little, mostly due to that slow second (from the a/d point of view). There's a little bleed in the a/d data from the previous second or two due to the anti-aliasing filter. There is also some scatter in the counters due to the short (for them) time interval used. Compromise is kind of like that...
However, don't forget that what is saved here is the raw timestamped data (something PCs are bad at, which is why we do it in an embedded uP - we can guarantee *all* the data for that second started and stopped right on the seconds boundary within a fraction of a microsecond), we can look at it any way we want, and one of the demos I put up on youtube indeed shows it vs time (actually, a few of them do), so you can see if there was a significant precursor situation to whatever feature you're examining.
I am still exploring what gnuplot can do for us, and what's worth doing. One thing I can do even on this multidimensional plot is to draw arrows between the points, according to their time order, for example, however - on this much data, it obfuscates rather than informs.
These parameters are NOT independent in a fusor. At X gas pressure, you cannot have arbitrary volts and current - they all interact. In fact, you can have a situation where the thing won't light off at a given set of pressure, volts - but once lit, will stay lit just fine (and it will stay lit for a lower pressure once lit) - there is a divergent feedback once there are a bunch of ions present that drives things to a limit (what we normally call stable operation).
I am of course, looking at automating my gas control stuff. (to do that well, I need finer control per pulse of the inlet and outlet solenoids than I now have) But then you'd have to track that too if you were interested in say, how fresh this batch of fuel was - obviously contamination builds up during a run due to things getting pounded off the walls, sputtering off the grid and so on...and as mentioned, temperature has an effect - which means, as you say, the the immediately previous conditions are somewhat important since it takes awhile to heat and cool.
This again, is why we do our science "right" and always, ALWAYS, keep all the raw data just as we collect it - that way, if a new idea comes up, we can tailor our display method of it to see if that new idea has merit or not. You can't do new ideas in data-mining if you don't keep the pure raw data around! The format does allow for comments in the log files, and I do put them in there. We just don't plot the comments...
There are of course some practical limits on what's worth logging. I'd like to have chamber temperature too for example. At some point though, the data rate gets out of control for the systems involved (The PC is generally the limit on bursts of data, it's not really a good real time thing), and the cost gets very high for all those channels. There will always be more we wish we could do, I suppose.
Here I'm trying to keep the line length per burst under 80 chars and still keep it human readable, since there's a lot of useful software out there that barfs on longer lines.
For high time resolution, I've been adding a GW Instek scope to this and grabbing screens from it's 4 channel, 2.5ghz sample rate data set - you can't get it all, but it lets you put a time-microscope on individual events. This can all be put into say, a MySQL database for later mining, though once you have two separate interfaces into a PC (and didn't get to write the opsys yourself), keeping them accurately time-aligned is more or less impossible. Two simultaneous inputs into a PC can be recognized in either order even down at the driver/interrupt level. If you have two programs stuffing things into a database, vagaries of the opsys time-slicing add to the uncertainty for both programs, and the database program *itself* may not put things in in the order recieved. We discovered this on a more ambitious prototype of this, where we had the scope and a 10 channel pro audio card all going at once along with this. At the speed transient events happen in a fusor, you really can't depend on sorting out which was cause, and which was effect unless for example, they are both on the same input (for example, the soundcard or the scope). If the signals come in on different interfaces - all bets are off.
So, doing that gets you into government-money-level custom built stuff, the sort of stuff I used to do for NSA intercept systems. I thought this was rather a decent compromise since I can take a family to dinner (admittedly at a good restaurant) for less than this will likely retail for.
No need to colonize mars yet - I'm just getting to earth orbit - while most people can't even fly, for now.
I'm putting all the demo movies up at: http://www.youtube.com/user/DCFusor?feature=mhee
And embedding them on my site, with discussion: http://www.coultersmithing.com/forums/v ... 34&start=0
which includes plots vs time on all this if that would answer any of your "what came first and from which direction" kinds of questions. You can see for example, the pressure varying (either due to my actions or randomly) and what happens before and after quite clearly on those, but not some other interesting facets we can see in the 4d plots better - depends on what you want to look for in the data mine.
Why guess when you can know? Measure!
-
Richard Hull
- Moderator
- Posts: 15024
- Joined: Fri Jun 15, 2001 9:44 am
- Real name: Richard Hull
Re: Parametric sweep of fusor parameters plotted in 4 dimensions
The natural hysteresis in glow modes on and off points is maddening in fusor data collection, but allowed for many novel uses of the ubiquitous NE-2 and sabilized NE-77 three element lamps in the 50's and early 60's when used as Flip-flops, astables, saw tooth gens, etc. GE put out a large engineers desk volume on its line of neon lamps with many apps and circuits featured in the late 50's.
As we work near a ragged edge of glow and higher vacuums with 40kv drops across our glow instead of the normal 55-70 volts of an NE-2, the going gets really rough and often maddening.
Richard Hull
As we work near a ragged edge of glow and higher vacuums with 40kv drops across our glow instead of the normal 55-70 volts of an NE-2, the going gets really rough and often maddening.
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
The more complex the idea put forward by the poor amateur, the more likely it will never see embodiment
Fusion is the energy of the future....and it always will be
The more complex the idea put forward by the poor amateur, the more likely it will never see embodiment
-
Donald McKinley
- Posts: 185
- Joined: Tue Nov 01, 2005 10:53 pm
- Real name: Donald McKinley
Re: Parametric sweep of fusor parameters plotted in 4 dimensions
Doug,
8-11-2012
I like your idea of taking account of the spin properties of reactants. I have been thinking about this for quite some time. I have an experiment to suggest roughly along those lines. I will mention it in a separate post so as not to muddy up the waters here. It involves calculated de Broglie frequencies.
regards,
Don
8-11-2012
I like your idea of taking account of the spin properties of reactants. I have been thinking about this for quite some time. I have an experiment to suggest roughly along those lines. I will mention it in a separate post so as not to muddy up the waters here. It involves calculated de Broglie frequencies.
regards,
Don