Why are current probes for scopes so expensive

[wholder]

I've wanted to purchase a current probe for my scope for some time, but the things are really very expensive.  Even used ones, such as the Tek A6302, often sell for hundreds of dollars on eBay.  Does anyone know why these things are so damn expensive?

Wayne

[dfmischler]

Same reason my friend says his divorce was so expensive: "Because it's worth it."  Build one for yourself and you'll know why.

[Lightages]

I am close to doing a review of a cheap current probe, the Hantek CC-65 AC/DC current clamp probe. I can tell you right now that it isn't that great but it isn't all bad neither. For the $70 or so it is worth buying but I will give a more detailed review on my youtube channel.

[alm]

There are various kinds of current probes. AC only, with a bandwidth that might range from a few Hz up to 100 MHz or so, are relatively simple (a current transformer) and affordable. Available new for under $1000 and used for around $100 if you're patient. Here is a DIY design. Note that most commercial probes would be more complex because they feature a split core and higher bandwidth. Polishing the surfaces of the split core to optical flatness so they meet without gaps is already quite tricky, for example. AC/DC probes with bandwidth up to audio frequencies, like the one lightages refers to, are also fairly affordable. These just feature a Hall sensor.

It's the DC to tens of MHz bandwidth probes, like the A6302, that are complex beasts. They combine a Hall sensor and current transformer in one core, which is extremely tricky to pull off with good bandwidth and noise performance. Tektronix pioneered this process. The core and sensor in the old P6302 is essentially identical to current products like the TCP202A. These are popular, unique Tektronix products. These days the patents have expired and there are two other manufacturers (LEM and Hioki?) of similar (but slightly inferior) designs. The Lecroy probes are made by one of those companies, for example.

There are just no cheap manufacturers like Rigol and Atten making these probes. They don't have the technology or R&D to pull this off. These probes also don't sell in the same volumes as say a basic 200 MHz passive probe. So you have a product with an extremely complex production process, a limited sales volume, and a few high-end brands selling them. Note that a basic 1x/10x switchable 200 MHz passive probe from Textronix will already cost in excess of $100. It's not unusual to pay more for a high-end probe than for a high-end scope. Probes are not just pieces of wire with BNC connectors.

New products are pretty much identical to the old A6302. Compare the specs. They extended the bandwidth to a useless 100 MHz (good luck making use of that) and changed the amplifier. That's about it. So working used probes still command high prices.

[Lightages]

Yes, the Hantek I am talking about is only "good" up to 20kHz, so if you are looking for a probe with the full bandwidth of any scope, look elsewhere. If you are looking for any kind of accuracy, again look elsewhere.

[tszaboo]

i have some DC- some Khz current probe at work from Fluke. It comes with a banana plug, not BNC. Not sure about the type. One day, I was measuring a step response with it with a scope, and I saw, that the beginning of the step response is going THE OPPOSITE way than the current. It looked awful.
I was playing around if I can come up with an inverse characteristics of the probe, it was hopeless, I gave up after some hours. So we bought a faster one (same company). This is usable at least. So the bottomline is that even expensive current probes can be awful.

[Alexei.Polkhanov]

All commercial current probes I have seen use hall sensor or some sort of flux gate devices and they allow current to be measured without breaking the connection, non-intrusive so to speak. Then I realized that solution with differential op-amp and a current shunt soldered in or connected with banana plugs works even better in my case anyway. Here is the nice Application note on subject -

http://cds.linear.com/docs/en/application-note/an105.pdf.

All you need is few resistors and one of OPAMPs listed in the note. I used LT6200 and LT1806. You can purchased 0.1ohm and 0.01ohm 1% shunt resistors:

http://www.digikey.ca/product-detail/en/LVR03R1000FE70/LVRB-.10RCT-ND/1166532

Almost certainly you will burn few OPAMPS if you are not careful and put too much current through wrong type of circuit (high sense/low sense), but they are $3/piece so just solder in another one. Perhaps it make sense to add some diodes at the input to protect OPAMP from excessive swings, but I found that they limit bandwidth. Maybe I needed different diodes?

Also you have to calibrate it with trimmer resistor if you want to see accurate trace on scope and there is some phase shit to be dealt with - I used skew function on my scope.

Someone suggested me to use Current Transformer - there are transformers capable of few MHz bandwidth but I think transformers are more appropriate for higher currents.

[alm]

All you need is few resistors and one of OPAMPs listed in the note. I used LT6200 and LT1806. You can purchased 0.1ohm and 0.01ohm 1% shunt resistors:

http://www.digikey.ca/product-detail/en/LVR03R1000FE70/LVRB-.10RCT-ND/1166532

Now you've replaced the problem of measuring current with characterizing your shunt. What's the inductance of that 10 mOhm shunt? At 10 MHz, even 1 nH of inductance will swamp that resistance.

Someone suggested me to use Current Transformer - there are transformers capable of few MHz bandwidth but I think transformers are more appropriate for higher currents.

That just depends on the turns ratio. The more turns you add, the higher the sensitivity. Commercial current probes with current transformers go down to 1 mA/mV without amplification.

[Alexei.Polkhanov]

Now you've replaced the problem of measuring current with characterizing your shunt. What's the inductance of that 10 mOhm shunt? At 10 MHz, even 1 nH of inductance will swamp that resistance.

Yes you are right, wire-wound resistor is a very bad choice for this application. Probably this is better choice: http://www.digikey.ca/product-detail/en/SR10-0.10-1%25/SR10-0.10-FDKR-ND/2182131.
I used shunt resistor similar to this:
http://www.digikey.ca/product-detail/en/CSR1206-0R001F1/696-1369-1-ND/2813302 I pulled it out from some sort of metering device.

If shunt resistance is 1 ohm, then at 10MHz 1nH inductance will add 0.06ohm to total impedance at 3MHz it will be 0.02ohm.
3db bandwidth with LT6200 was only about 3.5MHz not 10MHZ. I needed only 3MHz.

I think that with LT1806 20 MHZ 3db bandwidth is possible. Perhaps by adding 4 resistors in parallel in feed-back loop and only 2 resistors in parallel to ground would balance out the inductive component in shunt resistor.

[alm]

Increasing the shunt resistance will indeed reduce the effect of parasitic inductance, and increase the effect of parasitic capacitance . I'm sure there's an optimum somewhere. 1 Ohm is probably much better than 10 mOhm.

[wholder]

Thanks for all the replies.  I guess I'll just have to wait for a deal to pop up on eBay for a Tek probe.

Wayne

[V3r50n]

I've wanted to purchase a current probe for my scope for some time, but the things are really very expensive.  Even used ones, such as the Tek A6302, often sell for hundreds of dollars on eBay.  Does anyone know why these things are so damn expensive?

Wayne

I was able to get a pair of them really cheap on ebay and they seem to work as advertised.  http://www.ebay.com/itm/120957416259?ssPageName=STRK:MEWNX:IT&_trksid=p3984.m1497.l2649

[mtbolha]

Same reason my friend says his divorce was so expensive: "Because it's worth it."  Build one for yourself and you'll know why.

Nice one

[chlan]

The Labdevice current probes are not expensive.  Around 300USD. They measure AC and DC from the low uA range up to 1A. Perfect for electronic development.
www.labdevice.ch

[D3f1ant]

http://www.keysight.com/en/pd-2229629-pn-N2820A/3-mhz-50ua-high-sensitivity-ac-dc-current-probe-2-ch?cc=US&lc=eng

Speaking of expensive current probesb I just need a 90% discount

[djQUAN]

I bought mine cheaper than the Hantek and claims a little better frequency response of DC-100kHz but I have not tested it yet (no circuit to test on). I did manage to take it apart once I got a hold of it.

http://quan-diy.com/misc/multimeter/etcr.htm

[HighVoltage]

Speaking of expensive current probesb I just need a 90% discount

If you look around, you can find these deals 90% discount...
I just got lucky with a N2893A for $ 300 +$49 in express shipping (€ 4413 List price in Germany)
But I had to repair it:
https://www.eevblog.com/forum/testgear/agilentkeysight-n2893a-current-probe-%28teardown-and-repair%29/

[Macbeth]

Forgive my ignorance, but (other than earth isolation) why not measure the current by sticking your probe across a shunt or something like Daves uCurrent?

[Mechatrommer]

Forgive my ignorance, but (other than earth isolation) why not measure the current by sticking your probe across a shunt or something like Daves uCurrent?

sometime shunt doesnt exist, sometime a good differential CMRR is not around, and sometime the circuit is not easily available or we are not into the hesitation, only 2 wires poking out is what needed...

[HighVoltage]

Forgive my ignorance, but (other than earth isolation) why not measure the current by sticking your probe across a shunt or something like Daves uCurrent?

Many times I have to test currents on running systems that can not be shut down to add a resistor and on many applications it is not allowed to open a circuit at all. And in addition, I have to use a probe that is certified.
 

[woodchips]

Has anyone tried to use these clip on EMC suppression ferrite cores as the basis for a current probe?

They are certainly cheap enough, available in a range of sizes and grades, split, can't be too hard to try.

[_Wim_]

The Labdevice current probes are not expensive.  Around 300USD. They measure AC and DC from the low uA range up to 1A. Perfect for electronic development.
www.labdevice.ch

That seems more like a device with an internal shunt and a differential amplifier. A bit of a different beast then the ones discussed above...

[joeqsmith]

Pictures showing my Steward 28A2024-0A2 current probe from Kenneth Wyatt's article.   Towards the back is my Pearson 410 and in the center is my LEM  LA55P.     The LEM is a 50A RMS sensor that is good for +/-70 Amps.  DC-200KHz.    The LEM is also 50A RMS and has a 3dB cutoff of 120Hz to 20MHz.   

The video will show all three along with my antique, modified Tektronix P6042.   This video shows my  Tektronix probe after I modified it. 


Upfront, let me just say this is going to be one crappy setup but I hope it will provide some idea the limitations and strengths of these different probes. 

[joeqsmith]

Video showing the cheap and dirty current transformer from the article.

[CustomEngineerer]

There are just no cheap manufacturers like Rigol and Atten making these probes. They don't have the technology or R&D to pull this off.

While true just imagine if Rigol could pull it off. You could buy the cheap, AC only couple of Hz probes, hack them easily to the MHz bandwidth and to enable the DC option. While the hardware would be fairly solid, the probe's firmware would be buggy as hell and take many updates to fix and add new bugs. They would become wildly popular because of the fairly cheap price tag and easy hackability, and would force other low end current probe makers to compete. Of course all the competitors on the low end would put out even buggier, slightly cheaper probes, but we would at least have some new toys to play with and not mind if we accidentally blow them up.