home-brewed current probe

[jack11]

I have a need to look at some current waveforms with my Tek scope, and haven't been able to find much to my liking on ebay, etc for brand-name current probes. Some don't do what I want, some cost as much as the scope itself.

Has anyone built a simple current probe with the following rough specs:

measures currents for DC, AC, and switched DC waveforms
about 100A or so range, with decent resoluton compatible with range, for taking measurements on RE systems and devices
about 100KHz bandwidth
plugs into BNC connectors on an older scope
able to clamp around cables up to about 1/2" diameter

I have several voltage probes, so maybe some calibrated current-to-voltage converter would make sense.

Has anyone home-built anything like this, that would be fairly easy to build, use easy to get parts, and have reasonable accuracy?

[joestue]

http://www.ebay.com/itm/NEW-LEM-LA125-P-Closed-Loop-Hall-Effect-125A-DC-Current-Transducer-/321088956622

for the price you might as well buy one, though its not a clamp meter..

if you can make a precision cut in the iron core, there are about 12 laminations, core cross section is .2" by .15" you can make your own clamp meter, this is a project that's on my list of things to do. i've got two of them, if you want i'll take a photo of the inside.
-the hall effect sensor has .05" pin spacing, and there's a single coil of iirc, 2000 turns. if you want to preserve the 100Khz bandwidth i'd say don't mess with the bobin, just split the core and fashion it into a clamp meter. you will need to keep the air gap around the hall effect sensor the same, so when you cut the core, you'll have to cut off the other end as well, by the same amount.
i'd dremel it with one of those .020" thick cut off disks, then polish both sides square with a file.

or just buy a clamp meter for 100$...

[dnix71]

http://www.ebay.com/itm/LEM-Current-Transformers-LA-200-S-SP34-200-Amps-Modules-/190595408739?pt=LH_DefaultDomain_0&hash=item2c605f2363

$5 plus shipping. You will need to add cable with a BNC. The isolation is 5KV for 60 seconds. That's higher than the BNC could handle.

I have 2 of these. They take a fair amount of power to run, but being Swiss made are accurate and simple.

[jack11]

It looks like I am limited to Hall-effect current sensors if I want to measure AC, DC, pulsed-DC currents.

So, these Hall effect-based LEM modules seem to be in-circuit current-measuring devices (the circuit needs to be broken and the module inserted in series, just like a regular Amp-meter). Then, the Hall effect is sensed inside the module and amplified.

I am looking for external out-of-circuit measurements, just like using a voltage probe. Perhaps a Hall effect sensor integrated with some ferrite ring that I could place around the wire with current to be measured. There are BNC-ended probes like that on ebay, but they run several hundred $$$.

Joe, can you post here some pics of your project disecting your LEM module to make it work as a clamp?
This way we can get a better idea of what's involved, also if I am capable of doing this work.

[Flux]

Not sure about that LEM unit, but they make many versions with split cores that can be clamped round a cable like an ac clamp meter.

Don't under estimate the complexity of building something to cover the bandwidth you want, even voltage probes are a challenge. None of the Tektronics current probes I met were capable of handling low frequencies, just useful for high frequency pulses.

You can make split core current transformers for operation at the lower frequencies but I suspect you will not get their bandwidth up to the region you are looking at.

A passive probe will be a severe challenge, I suspect you will fall back on the variious form of Hall sensors, either voltage out or current balancing. Their only downside is that they require significanpt power, sometimes an issue with continuous monitoring but no issue with a scope surely.

Flux

[madlabs]

A long time ago I cut a ferrite ring just wide enough to stick in a Hall sensor. Back then current sensors were real expensive. After calibrating it VS a friends nice clamp on, I made a probe OK for DC, so-so for 60hZ AC and crappy for anything else. Still, it was cheap and fun!

Jonathan

[joestue]

http://www.johansense.com/bulk/LEM125-p.JPG

the circuit works as follows: the opamp amplifies any voltage out of the hall effect sensor, and the coil is connected from the opamp to the burden resistor, thus the current out of the opamp through the coil and the burden resistor opposes the magnetic field, and so the net magnetic field is always close to zero. you then read the voltage across the burden resistor.

it would also be possible to take a standard ac only current clamp probe, cut a piece of the core out, stick the hall effect sensor in there and epoxy it together, but the bandwidth and the offset current would probably be an order of magnitude worse, as i suspect LEM does not use the cheapest m-6 non grain aligned steel available.

btw: there is only one coil. why its split into two bobbins... i have some suspicions.. and it has to do with LEM's recommendation that if you wrap more than one turn through the core, you do so across the top, which would be right over the top of the wide portion of that bobbin.
i suspect there are an equal number of turns on both sides of the hall effect sensor.

[jack11]

that's strange, because their data sheet seems to show two separate coils/windings. One to run the primary current Ip thru, and the other one for the sense current Is. The rest is pretty much as I thought, closed-loop current-balance way as you describe.

I also found some open-loop, split-core sensors on the LEM website, the HTR series. Looks like the core is hinged, so it can be placed around the wire with no cuts/disconnects in the wiring.
There is only one coil shown in the data sheet, and no current-balance scheme. So, they may have to deal with the nonlinearity in  the Hall effect response, and I hope there is compensation for that somewhere in the output signal conditioning. Also, the bandwidth is only 10kHz.

I haven't got the tech details or prices yet from LEM, but if I end up getting one of these then I'll post here what I found.

[joestue]

here's the datasheet for what i have.
http://www.lem.com/docs/products/la%20125-p%20sp4%20e.pdf

single coil, closed loop. flux in the core remains at nearly zero.
also, taking a closer look at the circuit, there appears to be some kind of discrete bipolar amplifer on the output of the opamp, which is an LM201

[jack11]

I think you are right, this primary Ip circuit on the left must be their way of showing the wire with current running thru the aperture, and not wired in-circuit. There seem to be only 4 pins they define, and not 6.
But for repetitive measurements it would be difficult to use, because not a split core.

I called them about the price on the open-loop split-core HTR 100 SB, it's about $125 delivered by mail, with a long lead time to make and ship (they come from China). Maybe there are some on ebay, etc for cheaper, like the other LEM devices above.