Current measurement from insulated probing of conductor I-prober 520

[ErikTheNorwegian]

Wha

[Blofeld]

This certainly looks interesting. The only drawback I can think of at the moment is position dependency - if you hold this thing at a slightly different angle, the measured value might change. So for precision measurements I would be sceptical, but for everything else it looks like a winner.

[robrenz]

I think Mikeselectricstuff either has one or has one on order

[mikeselectricstuff]

I think Mikeselectricstuff either has one or has one on order

I 've had it a while and started a review video but got distracted... must get round to finishing it soon.
My initial view is that it is of limited usefulness for DC measurements due to position dependence. Not yet sure how much better it is for absolute AC measurements, although for stuff like switching regulators a lot of the time you're looking for things like inductor saturation etc.  by looking at wave shape.
The thing that attracted me was for applications like tracing shorts on PCBs, and if you used a pulsed power source so you're only tracing AC itd oes seem pretty useful, although I do wonder if you couldn't get similar functionality for less money using something like a tape head or small inductor as an AC probe.

[DaveW]

I have been using one for quite a few months now. They are good for the money.
As for accuracy, it is very good compared to the Fluke current clamps in the closed mode (at least partially because the hole is much smaller).
Probing current through a PCB track is a little fiddly, as it has to be held upright from the board and centered over the track. As is normal with these, you also have to adjust it for the width of track you are measuring. Without taking too much care, my measurements have normally come out within 10% of a measured reading, and taking a bit of time, better than that, down to 5% or so.

I use it a lot for working out where shorts are in boards, and with 5MHz bandwidth, it is sometimes possible to see pulsed currents being drawn in operation

[T4P]

I think Mikeselectricstuff either has one or has one on order

I 've had it a while and started a review video but got distracted... must get round to finishing it soon.
My initial view is that it is of limited usefulness for DC measurements due to position dependence. Not yet sure how much better it is for absolute AC measurements, although for stuff like switching regulators a lot of the time you're looking for things like inductor saturation etc.  by looking at wave shape.
The thing that attracted me was for applications like tracing shorts on PCBs, and if you used a pulsed power source so you're only tracing AC itd oes seem pretty useful, although I do wonder if you couldn't get similar functionality for less money using something like a tape head or small inductor as an AC probe.

An inductor loop . Is a cheaper option , it is basically what they are .

[Short Circuit]

Have one too, and indeed worthless for any absolute measurements.
However very usefull for general insight in current waveform, and
as mentioned before; for finding shorts on a PCB.

[alm]

For finding shorts, you could also look for a used HP 547A current tracer. It will only respond to pulsating current, but that's usually easy to supply from a function generator. The usual used price is a lot less than $900.

[qno]

It is in Dutch!!!!

[SeanB]

I wondered why they were complaining about the language...................

I d like the contactless current probe, made a probe years ago from a ferrite bead and 10 turns of 40SWG wire, worked quite well for applications where I could lift a lead to thread the bead through it.

[alm]

Language aside, Dutch is , well surly not the most commen understood language, but HEY, we got the pictures to enjoy!

Agreed, why didn't they use a common language like German, Chinese, Spanish or Hindi? Weren't you at some point advocating the use of Norwegian?

But is the probe sett Worth the money for a regular repair workshop? Or is a tool for the more developing end of the industry?

I would consider it strictly for R&D. It's about an order of magnitude too expensive for finding shorts on PCBs, and the number of faults that would benefit from this probe would probably not justify the $850 price tag. The limited bandwidth and high noise levels would probably severely limit R&D applications 6 mA rms noise at 5 MHz... you could fly to the moon and back on 6 mA, which would all be lost in the noise. And 5 MHz isn't exactly comfortable above the harmonics frequency of modern switchers. Even with 2 Hz bandwidth, when it's essentially a DC meter, noise is still 1.5 mA rms.

For comparison with the traditional clip-on AC-DC current probe, a Tek A6302 has 50 MHz of bandwidth and has ~0.15 mA rms of noise from DC to 100 MHz, less when BW limited to 5 MHz.

[mikeselectricstuff]

Teardown video and experiments in 2D current density mapping.
Dave just tweeted that he has one on the way - I know he hates unsoldering shielding cans so I've saved him the trouble 

[free_electron]

The problem with that thing is the positional accuracy... It is good to have an 'idea' of what a current looks like, but you can't really measure with it...
Shift it or tilt it a bit and the amplitude can be way off....

[qno]

The function mimics a gismo I made ages ago with a cassette players playback head and a audio amp.
This one works also with DC I presume?

[pickle9000]

There is some really good documentation here, good bedtime reading. PDF's are 12345 at the top of the page.

diy fluxgate magnetometer

http://perso.infonie.be/j.g.delannoy/BAT/PPMGuidelines.htm

...

[NiHaoMike]

If AC is all that's needed, could a cheap alternative be made using the read head out of a VCR or hard drive?

[pickle9000]

I'd be interested in knowing if the probe has two magnetometers because that seems to be what is needed to get a reading that measures magnitude. However the probe is a very different animal from what is in the documentation, very interesting in any case.   

...mike

[EEVblog]

The problem with that thing is the positional accuracy... It is good to have an 'idea' of what a current looks like, but you can't really measure with it...
Shift it or tilt it a bit and the amplitude can be way off....

I've played with a bit, and that is indeed the case. But the benefits of not having to hack your circuit to measure the current is worth the price of admission. It's a real neat bit of kit.

Dave.

[mikeselectricstuff]

If AC is all that's needed, could a cheap alternative be made using the read head out of a VCR or hard drive?

I think for some of the potential applications of this, in particular short tracing on PCBs, where you can use a pulsed signal and only care about relative measuremente, you could probably do something with comparable  performance with a small SMD inductor or tape  head as a probe.
 

[robrenz]

Does this crazy idea have any merit? 
Measure DC current by oscillating the sense coil through the lines of flux.
Take a very small flat wound sensing coil about 1mm in dia. and oscillate it with a voice coil.  The sensing coil would be bonded on the end of a stem that oscillates inside a thin wall polyimide tube.  This whole voice coil and stem with sensing coil would be the probe that you place tight to a trace. The voice coil would oscillate the sensing coil from flush with the outer probe tube to retracted some very small amplitude. The amplitude would depend on how high the lines of flux were above the trace surface.  Any travel outside the lines of flux won't contribute to generating a signal. Output the amplified signal to your scope or meter. You have a calibration board that has various trace widths copper thickness solder mask thickness etc. You run your anticipated max current through the calibration board test trace that best matches your situation and calibrate the probe. (oscillation frequency, amplitude and overall gain)

Just thinking out loud.

[mikeselectricstuff]

I think the main problems with that are defining the direction of oscillation relative to the trace, and of course your measurement bandwidth would be somewhat limited.
A GMR hard disk head might be worth experimenting with though.

[robrenz]

I was picturing the folowing. oscillation being axial in the probe tube.  You would hold the probe perpendicular to the board or conductor and tight against it.  This would allow the coil to almost touch the conductor at the one end of its oscillation and be repeatable between calibration  and measure.  The coil would be a single layer spiral of very fine magnet wire maybe 45 to 60 AWG. it would sit flat to the probe tip so all the coil windings are in the same position in the flux field at any time.  I would think the thickness of the flux field at mA currents would be very small so an extremely thin flat coil would be an advantage to cut the maximum lines of flux during oscillation.

[SeanB]

What is wrong with a fluxgate, with a coil to drive the core into saturation and a sense winding to measure the offset caused by external fields. Rugged and quite accurate ( measures the magnetic field of the earth with 1960's tech easily) and you can make it quite small, and even make it able to sense in 2 axes at the same time, summing them to get both a field magnitude and strength, so allowing you to measure both current and direction of flow.

[mikeselectricstuff]

What is wrong with a fluxgate, with a coil to drive the core into saturation and a sense winding to measure the offset caused by external fields. Rugged and quite accurate ( measures the magnetic field of the earth with 1960's tech easily) and you can make it quite small, and even make it able to sense in 2 axes at the same time, summing them to get both a field magnitude and strength, so allowing you to measure both current and direction of flow.

This is exactly how the i-prober works

[pickle9000]

Hey Mike,

Is it possible to tell if they are actually using two probes? My understanding was that with two probes one of the probes would be located a certain distance away from the probe doing the actual work (to get a base line). Maybe it's using the calibration feature to do the same job?

This is an edit: 30 seconds later, no, it must have two or it would be out of cal in no time.

...mike (a different mike)