ICP ion source
-
steve_rb
- Posts: 145
- Joined: Sun Oct 18, 2009 4:18 am
- Real name: Steve Robinson
- Location: Tehran University
ICP ion source
I am trying to do ICP (inductively coupled plasma) simulation. Only software I saw can do the job is Plasmator from Kinema (a software jointly with Fluent). They used to offer demo version but I missed it and no longe demo is available. Any one has demo or knows where to find the software Plasmator or any other similar software for ICP simulation please let me know.
Re: ICP ion source
I have a fair bit of experience with argon ICP and am unaware of any simulation tools for it (thats not to say that they don't exist). I have a simple electrical model for it for matching to the RF generator but I suspect this is not what you are after. I have read a few documents on the processes and physics of ICP and simulating such a dynamic beast would be daunting to say the least. I would be interested if you find anything but would be sceptical of finding any acurate model.
-
steve_rb
- Posts: 145
- Joined: Sun Oct 18, 2009 4:18 am
- Real name: Steve Robinson
- Location: Tehran University
Re: ICP ion source
Yes as you said I also found it very difficult to do plasma simulation. Only software I am aware of is Plasmator from Kinema which I am looking for it and I will post results if I found it and did good simulations. Model you mentioned may help a bit and there is no harm having a look at it. Also your experinece in ICP is very valuable and I will need it . I am trying to build a simple working plasma system as the one Carl posted. If you have any part of your experiences in written form that will be very helpfull if you care to send me some important points.
-
John Futter
- Posts: 1850
- Joined: Wed Apr 21, 2004 10:29 pm
- Real name: John Futter
- Contact:
Re: ICP ion source
Steve
Forget the modelling
You will need an experimental approach.
The conditions before the plasma is struck and the conditions with the plasma running are almost exactly 180 degrees apart on the Smith chart.
As is Dustin, I also look after a cantankerous ICP with a stable plasma @ 1.8kW (max 2.5kW) used for Be dating work. This unit runs @ 11.56MHz and just polishing the inductive coil around the quartz plasma tube shifts the tuning point one eighth to one quarter way around the smith chart. This unit uses a Henry Radio RF deck with a custom autotuner on the output, and even that has to be fooled into tuning the right way when the plasma strikes, by hand starting the tune point low ( in C in this case).
Looking at the auto tuner balance points ie 1.5kW forward 5-10 watts reflected all depends on the gas pressure, sample makeup, day of the month, likelyhood of getting a payrise, nebulizer gas load, blah, Blah.
Thank god for a forgiving tetrode in the output while the autotuner hunts for a hopefully a global maxima in forward power transfer.
I've been following Carls efforts very closely and I still can not decide whether to go with RF excitation or the more traditional DC excitation for my ion sources.
FWIW
Forget the modelling
You will need an experimental approach.
The conditions before the plasma is struck and the conditions with the plasma running are almost exactly 180 degrees apart on the Smith chart.
As is Dustin, I also look after a cantankerous ICP with a stable plasma @ 1.8kW (max 2.5kW) used for Be dating work. This unit runs @ 11.56MHz and just polishing the inductive coil around the quartz plasma tube shifts the tuning point one eighth to one quarter way around the smith chart. This unit uses a Henry Radio RF deck with a custom autotuner on the output, and even that has to be fooled into tuning the right way when the plasma strikes, by hand starting the tune point low ( in C in this case).
Looking at the auto tuner balance points ie 1.5kW forward 5-10 watts reflected all depends on the gas pressure, sample makeup, day of the month, likelyhood of getting a payrise, nebulizer gas load, blah, Blah.
Thank god for a forgiving tetrode in the output while the autotuner hunts for a hopefully a global maxima in forward power transfer.
I've been following Carls efforts very closely and I still can not decide whether to go with RF excitation or the more traditional DC excitation for my ion sources.
FWIW
-
Chris Bradley
- Posts: 2930
- Joined: Fri May 02, 2008 7:05 am
- Real name:
Re: ICP ion source
The only software I think might have a shot at it is CST microwave studio, but I've not used it for years so don't know if they've finished integrating all the particle simulation modules together with the RF simulations yet. Of course, and unfortunately for you, they will want some money out of you for the effort of making such software! It is a commercial product. Maybe you could ask them for a trail period for the software - I always found them to be 'responsive' and interested in 'new challenges' for their software.
On the whole, though, I'd stick with John F's advise. There's always experimentation and tuning needed in such systems over a few MHz, simulation is only 'guidance' anyway! You're potentially gonna have to work out how to match your source to your load over a vast range of impedances. Just start, and see what happens!..
On the whole, though, I'd stick with John F's advise. There's always experimentation and tuning needed in such systems over a few MHz, simulation is only 'guidance' anyway! You're potentially gonna have to work out how to match your source to your load over a vast range of impedances. Just start, and see what happens!..
-
Doug Coulter
- Posts: 1312
- Joined: Sun May 27, 2007 3:18 pm
- Real name: Doug Coulter
- Location: Floyd, VA, USA
- Contact:
Re: ICP ion source
I'd have to agree with most of the above postings. The conditions to "strike" the ionization are a lot different from ideal running conditions, and all the literature mentions this, and besides, it's obvious that there will be a big difference between no load and a load. Indeed, that's what I measure here, the lit is right. My ionizer uses a 600w magnetron running at more like 50w -- plenty of fudge, and it works great, rock solid, nothing gets hot.
The best thing to do is to have some extra capability in your RF stuff to handle that, and then tweak and learn some things. Since the actual power needed delivered to a load isn't big, that's pretty easy if you get to design the RF parts rather than buying something fragile intended to match 50 ohms +/- not much at all. You may find this actually cheaper than what it takes to get really fine control of all the other conditions that affect the operation, or at least I did.
So, to loosely quote Richard (and others here) just get to the lab and DO IT. Don't use lack of a simulator as an excuse. It will work, or not, but you'll surely learn and that's gotta be the main goal of it all.
Depending on predictions and simulations would be a great way to play the stock market (what I do for fusor bucks) if there were any good ones that you fully understood the unstated assumptions of.
But there aren't. Sometimes learning by doing is just the best way (maybe the only way?).
The best thing to do is to have some extra capability in your RF stuff to handle that, and then tweak and learn some things. Since the actual power needed delivered to a load isn't big, that's pretty easy if you get to design the RF parts rather than buying something fragile intended to match 50 ohms +/- not much at all. You may find this actually cheaper than what it takes to get really fine control of all the other conditions that affect the operation, or at least I did.
So, to loosely quote Richard (and others here) just get to the lab and DO IT. Don't use lack of a simulator as an excuse. It will work, or not, but you'll surely learn and that's gotta be the main goal of it all.
Depending on predictions and simulations would be a great way to play the stock market (what I do for fusor bucks) if there were any good ones that you fully understood the unstated assumptions of.
But there aren't. Sometimes learning by doing is just the best way (maybe the only way?).
Why guess when you can know? Measure!
Re: ICP ion source
If I were to recommend a method I would say a valve oscillator would be the best approach. Once the plasma strikes the impedance change is mostly compensated for by a change in the frequency of the oscillator. Modelling the plasma itself probably won't help you very much. My model is frequency / coupling co-efficient / power dependant and at atmospheric pressure so would probably not be very relevant.
As john acurately notes, small changes make big differences so experimentation would be the best route.
Don't let me put you off, it won't be as hard as we make it sound once you have set fixed conditions (much easier at lower pressures ) and would be a fun project.
Dustin
As john acurately notes, small changes make big differences so experimentation would be the best route.
Don't let me put you off, it won't be as hard as we make it sound once you have set fixed conditions (much easier at lower pressures ) and would be a fun project.
Dustin
-
steve_rb
- Posts: 145
- Joined: Sun Oct 18, 2009 4:18 am
- Real name: Steve Robinson
- Location: Tehran University
Re: ICP ion source
Dustin wrote:
> I would be interested if you find anything but would be sceptical of finding any acurate model.
I found following links indicating CST can simulate ICP plasma media. I am new with CST and need tutorials to try these simulations. I will post more information when I find .
www.cst.com/Content/Documents/Events/UG ... -Leroy.pdf
www.cst-korea.co.kr/Application/KAIST_ICP_JunHyunSu.pdf