Questions about current cancellation from coaxial cable.

Discussion in 'General Electronics Chat' started by Dong-gyu Jang, Nov 23, 2015.

1. Dong-gyu Jang Thread Starter Member

Jun 26, 2015
100
4
Hello.

I'm firstly asking you that you may want to see the attached image Recently we have used coaxial cable to transport power to the load (plasma) in ICP (Inductively Coupled Plasma) experiment. We had chosen coaxial cable with expectation that it minimize radiation from the cable. The basic assumption of such a choice was that current of central conductor of the cable at certain position z is equal to the current of ground at that position (mesh shield) except for direction. (in other word, out of phase). Mathematically it means Δφ is zero in the attached picture. (Z_eff represents equivalent impedance covering the load and wires between the coaxial cable and the load)

However, my colleague said radiation from the cable was measured with significant strength such that nearby instrument became malfunction. I said radiation source should not be the cable itself but the other becuase of current cancellation within the cable but it looks he didn't believe me.

So now, I'm also confused about my original concept. Does the coaxial cable have current cancellation within itself no matter what external connection is done to it or is case dependent? (for example, even for the case that two wires connecting the cable and the load have different length and composition?)

If it is latter, what factors should I take into account?

2. KL7AJ AAC Fanatic!

Nov 4, 2008
2,043
292
Hi Dong-gyu:
I first want to say how nice it is to meet someone who even knows what an ICP is! At UCLA we created the world's first stable ICP at atmospheric pressure! Major technological hurdle. I wish I could talk to you about this project for a few hours. (We measured 55,000 degrees C in the device!)

Now, to answer your question, the current on the INSIDE surface of the outer conductor should balance the currents on the center conductor. However, current can also flow on the outside surface of the coax outer conductor...but this is more of a problem when working with antennas. Unless your transmission line is a significant fraction of a wavelength long, you SHOULDN'T see much of a problem with your setup.

I need to ask, what frequency are you operating at? At HIPAS Observatory, we ran ours at 4.53 MHz.....just because we had a few 100KW transmitters already on the frequency!

Also, the induction coil itself was unbalanced (grounded on the cold end), but it looks like you're running a balanced coil.

Let me know a little more information. I would LOVE to work on one of these again! Where are you located? I have TONS of ICP experience.

Talk to you soon!
Eric

Dong-gyu Jang likes this.
3. GopherT AAC Fanatic!

Nov 23, 2012
6,304
4,027

Eric, nice to know - are you an analytical chemist or pure EE working in an instrument lab? I have been out of the hands-on ICP game for some time. The technology has come a long way since I stopped using one. I just heard the phrase PPQ (1/1,000,000 of a ppb) for the first time. I don't know if that is a goal or an every-day thing. Still amazing.

I might drop you a PM if I get far enough along with my hobby project.

Cheers.

4. GopherT AAC Fanatic!

Nov 23, 2012
6,304
4,027
@Dong-gyu Jang

Not all coaxial cable is the same. Look for the specs on your cable - voltage and power and frequency should match what you are doing.

Also, if you have an outer conducting layer and some has a third shield layer intended to be connected to ground. Only connect the outer shield layer to ground at one end - not both.

Connectors are also important - what connectors are you using?

The boards also have to be designed correctly - where the connectors attach.

5. KL7AJ AAC Fanatic!

Nov 4, 2008
2,043
292
http://www.w8ji.com/transmission_line_theory.htm
Here is a good overview of the "balun" which is used when interfacing coaxial cable with a balanced load. Again, I don't think this is your problem.
Eric

6. KL7AJ AAC Fanatic!

Nov 4, 2008
2,043
292
Hi Gopher:
Actually, I'm just a high power R.F. guy! I ended up working for the UCLA plasma physics department for no other reason than that. I learned all the plasma physics and chemistry by osmosis!

7. GopherT AAC Fanatic!

Nov 23, 2012
6,304
4,027
Ah, osmosis. Now you're talking real chemistry - Physical chemistry.

8. KeepItSimpleStupid Well-Known Member

Mar 4, 2014
1,265
222
If it counts - RF/DC sputtering and CVD systems.

9. GopherT AAC Fanatic!

Nov 23, 2012
6,304
4,027
If your doing CVD, then the purity issues can be in the ranges I was describing - and measured by ICP. What area are you in? LEDs, ICs or solar (other? Surface modifications?).

10. Dong-gyu Jang Thread Starter Member

Jun 26, 2015
100
4
It's also nice to see guys in the similar field. Our ICP is not powerful compared to your thing Its maximum operating power is 600 W and used only as test bed for some plasma diagnostics tool development. It's on 13.56 MHz and the cable length is ~1.3 M.

I guess our coil is unbalanced as one end of the coil is connected to ground of the coaxial cable. (I guess it is 50 Ohm coaxial cable with 7/16 DIN connectors on both ends)

Actually I would like to ask following questions for you to clear this issue.

1. You said some portion of current flowing within outer layer of shield of the coaxial cable can be source of radiation. If this is true, I have to say that ever coaxial cable has natural radiation emitting due to this. Do you know some way to minimize this?

2. For full assurance, I really want to see some proof or logic which lead the conclusion that current of a conductor of COAX always have equal amplitude and opposite phase to the other no matter what external connection is done to it. Could you provide this or say it is only special case? If it it latter, what kinds of factor should be taken into account to make such a case?

3. Our ICP is several turns solenoid. Actually wire from the center conductor of COAX to one end of coil is not the same to the wire between other end of coil and ground of the cable. One of them looks longer by few centimeters. Do this destroy current cancellation within the cable or is something I have to correct?

p.s: It's happy to see some paper or document of your impressively powerful ICP

Last edited: Nov 23, 2015
11. alfacliff Well-Known Member

Dec 13, 2013
2,449
428
a ballanced load on coaxial cable will result in rf radiation from the shield. if you check out a dipole antenna driven with coax, they usually use a ballun at the antenna to reduce or eliminate current on the shield. some use a current balun, a bunch of ferite beads on the outside of the shield, some use a 1:1 rf transformer to isolate the current on the outside of the shield. look up transmission line theory and antenna theory. or ask a ham.

MikeML likes this.
12. MikeML AAC Fanatic!

Oct 2, 2009
5,451
1,066
Since the load is not grounded to the machine (floating), but the capacitance to the machine ground are likely a bit asymmetrical, the current on the coax shield will be quite small.

I would just get this current balun kit and install it on the feedline...

Last edited: Nov 23, 2015
Dong-gyu Jang likes this.
13. KL7AJ AAC Fanatic!

Nov 4, 2008
2,043
292
Hi again!
Alas, I regret that all the documentation for the UCLA ICP was licensed to a private entity, which unfortunately never did anything with it beyond proof of principle. Quite a shame, since it has so much potential. But we learned a lot anyway.

As far as coax currents being balanced, Kirchhoff's Current Law absolutely guarantees that the currents have to be balanced between the inside and outside conductors. unless there is a path back to the source external to the transmission line...such as through a ground plane. But you can empirically verify this by placing a current transformer around the outside of the coax. You should read no current. (Ion Physics and Pearson are of the major manufacturers of these; not sure who manufactures these in your corner of the planet).

Let me know how it goes! By the way...it could be that even if there IS some radiation, you can choke off the currents in your instrumentation cables using ferrite beads or such.

Eric

Dong-gyu Jang likes this.
14. KL7AJ AAC Fanatic!

Nov 4, 2008
2,043
292
By the way, there is a photo of a large Pearson transformer in the "pulse dipole" photo here;

http://www.thelivingmoon.com/45jack...gh_Power_Auroral_Stimulation_Observatory.html

Dong-gyu Jang likes this.
15. KL7AJ AAC Fanatic!

Nov 4, 2008
2,043
292
How are you coming along on your project? Keep me posted on your project. I would like to be there with you, but since that's a bit difficult at this time, I'd like to keep tabs on this. This is physics at its finest. )

16. Dong-gyu Jang Thread Starter Member

Jun 26, 2015
100
4
Hello.

As your suggested idea, we measured current flowing through the cable by using our current transformer (100 A corresponds to 10 V without 1/10 attenuator). Then the results is quite surprising that about 1.7A is constantly measured. As I said, I thought current cancellation within coaxial cable should give me about zero signal in current measurement. I also suspected that the signal is influence by radiation but that effect is measured about 0.37 V. It means actually measured current may be about 1.33 A.

As power used in this measurement was 400 W, I guess 1.7 A is significant portion of the whole current.

Could you imagine or guess what happens on our setup? According to your comments, it can happens if there is another path returning to the source other than the cable such as via ground plane. Yeah, maybe it is true that the metal frame closely surrounding ICP coil (coil: ~10 cm length and 7.5 turn, 1 inch diameter, spacing between the coil and the frame is about few centimeters) can receive significant portion of the current in form of EM wave (maybe called displaced current?) as the frame is grounded. (the coil itself is only grounded to the cable which is eventually connected to the grounded box. The cable length is 1.3 M) Do you think this is problematic setup? Then should I take the frame ground out thus whole frame is floated?

17. Dong-gyu Jang Thread Starter Member

Jun 26, 2015
100
4
Anyway I found mathematical proof of current cancellation in the attached image.

Maybe it will be referred to other people who might have similar question.

File size:
202.4 KB
Views:
1
18. GopherT AAC Fanatic!

Nov 23, 2012
6,304
4,027
I was thinking that 600W at 13MHz can radiate some EM energy away from the board. Anything that can receive/absorb the radiated energy and return the signal back to the source (without traveling through the coaxial cable) will cause an imbalance in the coax. Iron frames that can induce current flow, lack of ground planes, and anything else that can act as an antenna or transformer will cause issues.

Dong-gyu Jang likes this.