Ferrite bead PCB track current probe ?

[chris_leyson]

Many years ago I read an article that described a PCB track current probe using a ferrite bead. The bead was ground down to a semicircle and a few turns of wire were placed on the core for the pickup coil. I don't remember which magazine the article was published in, it could be Electronic Design or EDN. The closest match that I could find is from Electronic Design https://www.electronicdesign.com/components/simple-homemade-sensors-solve-tough-emi-problems. Unfortunately there are no links to the pictures in the Electronic Design article but in hindsight I think it must have been Electronics Design because I don't remember any performance figures being published.

I made a few attempts grinding down ferrite beads but didn't have much success and the project was abandoned.
Has anyone on the forum tried this current measurement technique . I'm looking for 30MHz BW so the Aim TTi is not an option.

[T3sl4co1l]

I've always seen the one with a slitted metal tube and small air-core coil, but that will work nicely too.  Rather like a hard drive read head.   Should have wider bandwidth, and probably more up-close sensitivity for probing traces.

It's not exactly critical, just throw the parts together, add a termination resistor and you're good to go.

Note that it won't be a current probe as such, but more like a Rogowski.  Add an integrator to get waveforms closer to actual current (or just do it mentally, hand-wavingly).

Curiously, I'm currently working on a project that uses a similar concept to measure mains distribution lines...

Tim

[iMo]

Rather like a hard drive read head.

Could not be the heads from hdds used for that purposes, btw? Ie. those magnetoresistive ones? They work with pretty high frequencies, afaik. In past I tried to make the same with ferrite toroids (maybe 1.5mm diameter) from old computer core memories, absolutely no luck with creating the tiny slit.

[T3sl4co1l]

Rather like a hard drive read head.

Could not be the heads from hdds used for that purposes, btw? Ie. those magnetoresistive ones? They work with pretty high frequencies, afaik.

Maybe, but you have to bias it.  Sensitive region is probably way too small to be useful, too?

In past I tried to make the same with ferrite toroids (maybe 1.5mm diameter) from old computer core memories, absolutely no luck with creating the tiny slit.

Yeah, those aren't worth crap, they have an open B-H loop.

Tim

[SeanB]

I have wound current sense coils using ferrite beads, they do work to get the currrent waveform in things like power transistor emitters, and other things like that where you cannot fit a more linear current shunt. Does have some affect on the waveform, and does need calibration and termination both sides, and really needs thin coax cable to be a success. I used some scrap 110R PTFE twisted pair coax, normally used inside avionics to transfer data bus signals around, and used a 110R 1/8w resistor both sides, then used the scope to read it. Was reasonably linear up to around 5A IIRC, using a simple test jig made from a 555 timer running a power transistor, driving up to 5A into a few parallel resistors, fed from a bench power supply. wound up being reasonable up to 100kHz, using the largest random ferrite bead I found that had enough room to wind 15 turns of 48SWG wire on it, and also have a PTFE sleeve that prevented the component wire from removing the enamel. Glued the resistor on with superglue, and used it as a solder post for the wires, then blobbed more superglue on afterwards.

Was from an article in IIRC ETI or PW, and I wanted to look at the waveform in a SMPS, new fancy things at the time, at least in the consumer PC market, though they were common in TV sets as a scary thing that ate silicon when faulty.

[chris_leyson]

Thanks for the tips guys, I think I will just throw something together an see how it goes, could be a useful tool for EMC work. In the meantime I've got a Beehive 100B on order so that should help. I've got a board that uses a synchronous buck converter and it's failing on conducted emissions so it would be useful to get some idea where the currents are flowing.

[chris_leyson]

I just remembered why I wanted to try a semicircular ferrite bead in the first place. Maybe I can get better spatial resolution, will have to give it a try.

[joeqsmith]

I made the probes from this EDN article when it was released.   At that same time, I also made one using ferrite in the head.   This is the smaller tip of the three. 

[wilfred]

Could it have been this article in Electronics Australia April 1988 pp88?
https://archive.org/stream/EA1988/EA%201988-04%20April#page/n85

[dazz1]

Hi
What about using a VHS play head?  Specifically designed to pickup small magnetic fields. 
I've had one in my future projects bin for some years.  Just haven't had the need/justification to make it.

[Gyro]

Could it have been this article in Electronics Australia April 1988 pp88?
https://archive.org/stream/EA1988/EA%201988-04%20April#page/n85

OT: As that's the April 1988 issue, the article on Mercury speaker cables is worth reading. 

[SpencerTC]

Great, another project to add to the growing list of projects that I haven't had time to do.

Thanks for all of the info, the various sources for this type of probe has made me very interested.

[Ice-Tea]

You could also just buy an I-Probe 520 

[dazz1]

Hi
I know this is a really old post but:
If you want resolution much less than the size of the probe, then Google monopulse radar sum difference.
I am not proposing to use radar to find currents.  I do think that the sum-difference maths could be applied to the problem of tracing currents.

For years, I have kept an video tape head in my junk box, with the intention of making a sensitive high resolution current probe.  It is still there.