1 ohm transmission line

feet

Talk to these guys:

I saw the movie about them, "Other Peoples's Money" with Danny DeVito playing hostile takeover guy but then falling in love with one of their family members turned attorney. A very good movie.

Anyhow, ask for micro coax. You didn't say how many amps, duty cycle and so on. But typical micro coax would be less than 15ohms so you don't have to use too many in parallel. And it'll be small.

Another option would be to talk to Parlex in Taxachussetts, see if they make FPC cable with a ground plane. I think single-ended that's usually around 50ohms if there is a plane, so if you have two 25ers all in parallel it should get you there. IIRC they can make almost any length you want.

--
Regards, Joerg 

http://www.analogconsultants.com/

for

the

feet

Getting down to one Ohm might be a bit tricky, but, in the MHD work I did at MIT Building 20 I made transmission lines from 4" wide x 1/4" thick Aluminum strips with Teflon dielectric. 50kV rated and 30,000 Amps.

Scale as desired... wonder if the stripline equations work for such a situation ;-) ...Jim Thompson

--
| James E.Thompson                                 |    mens     | 
| Analog Innovations                               |     et      | 
| Analog/Mixed-Signal ASIC's and Discrete Systems  |    manus    | 
| Phoenix, Arizona  85048    Skype: Contacts Only  |             | 
| Voice:(480)460-2350  Fax: Available upon request |  Brass Rat  | 
| E-mail Icon at http://www.analog-innovations.com |    1962     | 
              
I love to cook with wine.     Sometimes I even put it in the food.

feet

There was a thread here not long ago that discussed the ribbon cable.

Can you really roll up an unshielded transmission line? My gut tells me that you end up with two transmission lines offset by the length of the spiral, which changes over the length. Suppose you could use three layers of conductor..but even that makes me nervous...

Then there's the resistance of the foil over 50 feet.

Shouldn't be too hard to test with a couple of rolls of aluminum foil and some dielectric sheet.

Might be interesting to try to cut the ends off a foil cap to sever the parallel connections and bring out the first wrap for connection.

If you care about flatness of the pulse, I think you're gonna end up with a bunch of higher impedance stuff in parallel.

Sometimes, you can use a more conventional source with a series transformer. The transformer handles the corners and fixes up the droop. The devil is in the details.

Assume you've already solved the problem of getting that 1 ohm all the way to the place the electrons are needed.

Ribbon is ok but considering that it's 100ohms diff and John needs 1ohm that can get old quite quickly :-)

You'll normally have to wrap some stuff in there, depends on how clean the pulse is supposed to be.

[...]

--
Regards, Joerg 

http://www.analogconsultants.com/

for

the

feet

If it were to be rolled, I'd probably use a 3-layer (stripline) thing, outer grounds and hot inner conductor. That shouldn't care if it was rolled or not, or how tightly. I guess.

OK 1 oz copper is about 500 uohms per square. If it were 1" wide and 50 feet long, each strip is 0.3 ohms, which is pretty bad for a 1 ohm line. As I discharge it into the load, the current will droop.

I ordered some ebay stuff to play with. They have lots of copperclad FR4 and kapton.

As long as I can keep the parasitic inductance down around, say, 2 nH. Ugh.

--

John Larkin                  Highland Technology Inc 
www.highlandtechnology.com   jlarkin at highlandtechnology dot com    

Precision electronic instrumentation 
Picosecond-resolution Digital Delay and Pulse generators 
Custom timing and laser controllers 
Photonics and fiberoptic TTL data links 
VME  analog, thermocouple, LVDT, synchro, tachometer 
Multichannel arbitrary waveform generators

feet

These guys seem to have done both flex cable and parallel coax to get to 1 ohm Z0:

their older construction:

Looks pretty tractable- some wide polyimide film (tape?) and copper strip. 75 ns is a bit inconvenient if the stuff comes in 36 yard rolls-- probably end up with almost 50% wastage.

Best regards, Spehro Pefhany

--
"it's the network..."                          "The Journey is the reward" 
speff@interlog.com             Info for manufacturers: http://www.trexon.com 
Embedded software/hardware/analog  Info for designers:  http://www.speff.com

.highlandtechnology.com  jlarkin at highlandtechnology dot com

Reynolds Industries used to make some high voltage cable similar to the flexible cables used in cars to wire up the dashboard. They had a ribbon center conductor and two ribbon conductors on each side. That was about 50 years ago. It is now Teledyne Reynolds. As I remember the impedance was not much more than 1 ohm. They were used in EBW systems.

Dan

"John Larkin" wrote in message news: snipped-for-privacy@4ax.com...

Make it fatter at the far end.

Tim

--
Deep Friar: a very philosophical monk. 
Website: http://seventransistorlabs.com

On Sun, 14 Apr 2013 14:00:22 -0700, John Larkin wrote:

feet

This kind of works, but I don't know how lossy the Ls might be in real life.

Version 4 SHEET 1 2104 680 WIRE 96 80 16 80 WIRE 176 80 96 80 WIRE 208 80 176 80 WIRE 320 80 288 80 WIRE 352 80 320 80 WIRE 464 80 432 80 WIRE 496 80 464 80 WIRE 608 80 576 80 WIRE 640 80 608 80 WIRE 752 80 720 80 WIRE 784 80 752 80 WIRE 896 80 864 80 WIRE 928 80 896 80 WIRE 1040 80 1008 80 WIRE 1072 80 1040 80 WIRE 1184 80 1152 80 WIRE 1232 80 1184 80 WIRE 1408 80 1312 80 WIRE 1456 80 1408 80 WIRE 1488 80 1456 80 WIRE 16 128 16 80 WIRE 176 144 176 80 WIRE 320 144 320 80 WIRE 464 144 464 80 WIRE 608 144 608 80 WIRE 752 144 752 80 WIRE 896 144 896 80 WIRE 1040 144 1040 80 WIRE 1184 144 1184 80 WIRE 1408 144 1408 80 WIRE 1616 160 1456 160 WIRE 1488 208 1456 208 WIRE 1616 208 1616 160 WIRE 1488 224 1488 208 WIRE 16 256 16 208 WIRE 176 256 176 208 WIRE 176 256 16 256 WIRE 320 256 320 208 WIRE 320 256 176 256 WIRE 464 256 464 208 WIRE 464 256 320 256 WIRE 608 256 608 208 WIRE 608 256 464 256 WIRE 752 256 752 208 WIRE 752 256 608 256 WIRE 896 256 896 208 WIRE 896 256 752 256 WIRE 1040 256 1040 208 WIRE 1040 256 896 256 WIRE 1184 256 1184 208 WIRE 1184 256 1040 256 WIRE 1344 288 1312 288 WIRE 1408 288 1408 224 WIRE 1408 288 1344 288 WIRE 1504 288 1408 288 WIRE 752 304 752 256 WIRE 1408 320 1408 288 WIRE 1504 320 1504 288 WIRE 1616 320 1616 288 WIRE 752 416 752 384 WIRE 1408 432 1408 400 WIRE 1504 432 1504 384 FLAG 752 416 0 FLAG 1408 432 0 FLAG 1616 320 0 FLAG 1488 224 0 FLAG 1456 80 LINE FLAG 1344 288 LOAD FLAG 96 80 END FLAG 1504 432 0 SYMBOL cap 160 144 R0 SYMATTR InstName C1 SYMATTR Value 9n SYMBOL cap 304 144 R0 SYMATTR InstName C2 SYMATTR Value 9n SYMBOL ind 192 96 R270 WINDOW 0 32 56 VTop 2 WINDOW 3 5 56 VBottom 2 SYMATTR InstName L2 SYMATTR Value 9n SYMBOL cap 448 144 R0 SYMATTR InstName C3 SYMATTR Value 9n SYMBOL ind 336 96 R270 WINDOW 0 32 56 VTop 2 WINDOW 3 5 56 VBottom 2 SYMATTR InstName L3 SYMATTR Value 9n SYMBOL cap 592 144 R0 SYMATTR InstName C4 SYMATTR Value 9n SYMBOL ind 480 96 R270 WINDOW 0 32 56 VTop 2 WINDOW 3 5 56 VBottom 2 SYMATTR InstName L4 SYMATTR Value 9n SYMBOL cap 736 144 R0 SYMATTR InstName C5 SYMATTR Value 9n SYMBOL ind 624 96 R270 WINDOW 0 32 56 VTop 2 WINDOW 3 5 56 VBottom 2 SYMATTR InstName L5 SYMATTR Value 9n SYMBOL cap 880 144 R0 SYMATTR InstName C6 SYMATTR Value 9n SYMBOL ind 768 96 R270 WINDOW 0 32 56 VTop 2 WINDOW 3 5 56 VBottom 2 SYMATTR InstName L6 SYMATTR Value 9n SYMBOL cap 1024 144 R0 SYMATTR InstName C7 SYMATTR Value 9n SYMBOL ind 912 96 R270 WINDOW 0 32 56 VTop 2 WINDOW 3 5 56 VBottom 2 SYMATTR InstName L7 SYMATTR Value 9n SYMBOL cap 1168 144 R0 SYMATTR InstName C8 SYMATTR Value 9n SYMBOL ind 1056 96 R270 WINDOW 0 32 56 VTop 2 WINDOW 3 5 56 VBottom 2 SYMATTR InstName L8 SYMATTR Value 9n SYMBOL ind 1216 96 R270 WINDOW 0 32 56 VTop 2 WINDOW 3 5 56 VBottom 2 SYMATTR InstName L1 SYMATTR Value 18n SYMBOL voltage 752 288 R0 WINDOW 0 73 36 Left 2 WINDOW 3 68 79 Left 2 SYMATTR InstName V1 SYMATTR Value 290 SYMBOL sw 1408 240 R180 WINDOW 0 71 37 Left 2 WINDOW 3 60 69 Left 2 SYMATTR InstName S1 SYMATTR Value SW1 SYMBOL res 1392 304 R0 WINDOW 0 -76 41 Left 2 WINDOW 3 -77 77 Left 2 SYMATTR InstName R1 SYMATTR Value 0.1 SYMBOL voltage 1616 192 R0 WINDOW 0 58 37 Left 2 WINDOW 3 -117 -63 Left 2 WINDOW 123 0 0 Left 2 WINDOW 39 0 0 Left 2 SYMATTR InstName V2 SYMATTR Value PULSE(-10 10 50n 1n 1n 250n) SYMBOL res 0 112 R0 SYMATTR InstName R2 SYMATTR Value 1K SYMBOL diode 1520 384 R180 WINDOW 0 -53 27 Left 2 WINDOW 3 -60 -4 Left 2 SYMATTR InstName D1 SYMATTR Value DID TEXT 8 400 Left 2 !.model SW1 SW(Ron=.01 Roff=1Meg Vt=0 Vh=-.5 Lser=1n Vser=0) TEXT 280 312 Left 2 !.tran 0 250n 0 1n TEXT 136 352 Left 2 !.model DID D(Vfwd=1 Ron=5m Roff=1G) TEXT 488 -24 Left 2 ;TRANSMISSION LINE PULSER JL APR 14 2013 TEXT 1136 16 Left 2 ;1 OHM LINE

--

John Larkin                  Highland Technology Inc 
www.highlandtechnology.com   jlarkin at highlandtechnology dot com    

Precision electronic instrumentation 
Picosecond-resolution Digital Delay and Pulse generators 
Custom timing and laser controllers 
Photonics and fiberoptic TTL data links 
VME  analog, thermocouple, LVDT, synchro, tachometer 
Multichannel arbitrary waveform generators

for

the

vaguely

feet

or

1" copper strip in 0.002",0.003",0.004",0.005",0.007",0.010",0.016",0.021" is easily available @ 100ft.. At the thickest DCR is only about 5% of Z0 for a strips x 40m round trip.

Best regards, Spehro Pefhany

--
"it's the network..."                          "The Journey is the reward" 
speff@interlog.com             Info for manufacturers: http://www.trexon.com 
Embedded software/hardware/analog  Info for designers:  http://www.speff.com

Transmission line impedance goes as sqrt( L/C) (L being inductance per meter, C being capacitance per meter), so you would like to keep L low (wide ribbon would do that without a lot of cost) and C high (which leaves you making a sandwich of metal/high-dielectric-constant insulator/metal with considerable length). Kind of like a film capacitor, unrolled.

Or, you could try a ceramic with high dielectric constant, and find a way to make semirigid coax by swaging a fat pipe over a thinner pipe coated with ceramic. I was thinking of the way Calrod heater elements are made, a bunch of beads on a wire with the jacket shrunk over them by drawing through a die, but you'd want low thickness for the ceramic. Either that, or you'd want to store kilovolts and use a transformer.

Such low Z is tough. I have designed something in that ballpark on-chip but dimensions become ridiculous. Another construction I'm making right now is a 6 ohm line made out of nine 0.5mm OD Gore coaxes, as a part of a Guanella transformer.

Regards, Mikko

a really long film cap?

--
?? 100% natural 

--- news://freenews.netfront.net/ - complaints: news@netfront.net ---

for

the

feet

How about a piece of capacitively loaded stripline? If you dimension it for about 300nH/m and load it with about 300nF/m, it would only have to be about 25cm long. Not sure it's practical.

Your simulation shows 25V or so into a 0.1 Ohm load. Is that the actual target?

Jeroen Belleman

Its the parallel plate limit, which is easy. C/l = e0 er w/h and L/l = mu0 h/w, and Z = sqrt( L / C ). Edge fringing becomes negligible.

Regards, Mikko

for

the

feet

If you put 50mOhms in each inductor and 100mOhm into L1 it still works but has some droop.

Why do you want to do this with a delay line instead of bang-bang switches sync-buck style? 150nsec at 300V/100A is a wee bit challenging but not impossible.

[...]

--
Regards, Joerg 

http://www.analogconsultants.com/

[...]

But mind the skin effect.

--
Regards, Joerg 

http://www.analogconsultants.com/

for

the

vaguely

feet

That's in the range, details to be negotiated if/when I have a customer.

It seems to me that nature doesn't want us to make decent electrical delay lines in the 100 ns range. Something or something else makes them lossy. If you look at the signal delay lines in analog oscilloscopes, in the 50 ns or so range, they need elaborate equalization networks to get clean step response. There's no practical equivalent for an energy storage line... maybe.

The lumped line thing I sim'd has 9 nh inductors, so it's approaching your loaded stripline sort of thing. I think copper/skin loss and cap ESR are going to be problems.

--

John Larkin                  Highland Technology Inc 
www.highlandtechnology.com   jlarkin at highlandtechnology dot com    

Precision electronic instrumentation 
Picosecond-resolution Digital Delay and Pulse generators 
Custom timing and laser controllers 
Photonics and fiberoptic TTL data links 
VME  analog, thermocouple, LVDT, synchro, tachometer 
Multichannel arbitrary waveform generators

for

the

vaguely

feet

lines

no

If you make the pulse with FETs the copper losses vanish. ESR is no problem, just parallel enough Johanson caps. But make sure they can stomach the pulse load. I've once managed to grenade a (big) cap. When I cleaned the mess up I found that some of the shards showed a material that had mutated from white ceramic into green glass with tiny bubbles in there.

--
Regards, Joerg 

http://www.analogconsultants.com/