Current Viewing Resistor for kA pulses / General Science and Electronics / Forums

Forums

4hv.org :: Forums :: General Science and Electronics

« Previous topic | Next topic »

Current Viewing Resistor for kA pulses

Move Thread

LAN_403

klugesmith

Thu Mar 03 2016, 10:37PM

Registered Member #2099 Joined: Wed Apr 29 2009, 12:22AM
Location: Los Altos, California
Posts: 1716

A contemporary thread about what's inside a broadband current transformer
prompted me to order a 2.5 mÎ© Current Viewing Resistor, the best specimen of three on ebay.
I'd had my eye on them for literally years, and the seller had just reduced the asking price.

No galvanic isolation, of course. The instrument output shares a ground with high current circuit.
But it's supposed to be accurate from DC to 48 MHz. Rated risetime is 8 ns. What does that say about magnitude of parasitic inductance?

[edit] The CVR arrived today, in a package whose label says 19 pounds!

DerAlbi

Thu Mar 03 2016, 11:19PM

Registered Member #2906 Joined: Sun Jun 06 2010, 02:20AM
Location: Dresden, Germany
Posts: 727

But it's supposed to be accurate from DC to 48 MHz. What does that say about magnitude of parasitic inductance?

Nothing

It can also be that at 48MHz the capacitive part dominates.. one can only measure what the problem is. What i can imagine is a problem for a current shunt is skin depth at this point. You need a quite large hollow cylinder to provide a constant and really low resistance with a sheet thicknes so thin that skindepth is irrelevant.
Maybe its a ceramic substrate with some deposited silver coating... just a guess.

Silver at 48MHz has 9.1Âµm skin depth. Meaning the sheet thickness cant be much more than 3..5um. If you would know at which distance the sense wires are placed on that resistor, one could estimate a diameter of the conductor. I would predict that its quite large, as mentioned before. What a high diameter dictates is actually a ultra low inductance - as intrinsic inductance is proportional to the shortest field line around the conductor. (Thats why thinner wires have higher inductance)

klugesmith

Sat Mar 05 2016, 07:18PM

Registered Member #2099 Joined: Wed Apr 29 2009, 12:22AM
Location: Los Altos, California
Posts: 1716

Here is my new toy. Rather bigger than I had imagined, even though its output BNC connector is visible at small end in OP image, and its datasheet is online.

Insulation on the input side suggests that it could drop thousands of volts. Not for long, given 225 watt dissipation rating and 2800 joule impulse rating. That's 2800 joules lost in the current sense resistor. And what about mechanical forces from six-figure currents?

The hex nut came off after some filing on dented 3/4"-16 threads, which are silver-plated brass like the other external surfaces.

An electrical test with about 5 amperes DC showed a voltage drop consistent with 2.5 mÎ© resistance. I saw no difference in voltage between the large stud and the BNC center terminal. A primitive ohmeter measurement looking into the BNC showed approximately 0 ohms. 50 ohms would have been unsurprising, and might have made it easier for designers to include compensation to flatten the frequency response.

In response to Der Albi, I had not considered the skin effect issue. Having thought about making a CVR from scratch, I bet it has a large diameter, thin wall cylinder of resistive material with enough mass to handle the energy rating. Probably very close to a coaxial "ground" cylinder that carries the return current. That would form a transmission line with extremely low inductance and Z0.

My comment about inductance was because that often bites in low-ohm current sense applications.
Typical board-mounted inductance of small surface-mount R's and C's is on the order of 1 nH. That's 6.28 milliohms at 1 MHz, and 6.28 ohms at 1 GHz.

It's easy to download and print impedance-chart graph paper. Not so easy to find blank charts that go below 0.01 ohms, so I made one.

Horizontal axis is frequency in Hz. Vertical axis is |Z| in ohms. Solid black line is Z of an ideal 2.5 mÎ© resistor. Red lines are inductance in decade steps from 1 nH (top) to 1 pH (bottom). Green lines are capacitance in decade steps from 1 nF (top) to 1 uF (bottom). The 1 nH and 1 nF cross at 1 ohm and 1/2pi GHz.

Signification

Sun Mar 06 2016, 12:09AM

Registered Member #54278 Joined: Sat Jan 17 2015, 04:42AM
Location: Amite, La.
Posts: 367

Ain't it great getting this type thing 'working'--any more left?

klugesmith

Sun Mar 06 2016, 04:02PM

Registered Member #2099 Joined: Wed Apr 29 2009, 12:22AM
Location: Los Altos, California
Posts: 1716

>>any more left?

Yes, Coin Shrinker Club Guy.

Often there are no Current Viewing Resistors on ebay, but the last time I looked there were several. "Mavin" had two more of the same model as mine. One was missing some screws at the small end, and the terminal stud on the other had more damaged threads than mine.

Be careful with the connection between CVR ground and capacitor ground. Its inductance will develop a voltage during high di/dt events, that could be bad for instruments and people connected to the BNC.

[edit] Took me a while to understand why on my CVR, the big nut jams about 1/4" above the stud bottom. The threads there look perfect, even under magnification. Turns out the whole stud is very slightly bent, and visibly leans off-axis. Not bad for US $40 + shipping.

Moderator(s): Chris Russell, Noelle, Alex, Tesladownunder, Dave Marshall, Dave Billington, Bjørn, Steve Conner, Wolfram, Kizmo, Mads Barnkob