DIY HF current probe design..

[themaccabee]

Dear Experts,

I want to design a HF current probe,

Aim : Is to relatively measure the cable current radiation to estimate if my fix on PCB ,is working or not.(For pre compliance testing purposes)

Materials I have : Wurth Toroid Core 74270097 ,Teflon Wire, BNC connector

Equipment i have : An Old Spectrum Analyzer ESA-L1500A , 75Ohm, 1 MHz-1.5GHz

Input Specifications of Spectrum Analyzer : 75Ohm,75dBmV(0.4W) MAX/ 100v DC MAX

I was trying to follow the section DIY CURRENT PROBES in https://www.interferencetechnology.com/the-hf-current-probe-theory-and-application/

some details are not clear there , may be it is obvious,

1) Is it ok to use multi-stranded teflon wires
2)The one end of the wire after winding should be connected to BNC connector center point and the other to the BNC shield correct?
3)They are mentioning a copper tape to be put over the windings as electrostatic shield.Should it be connected to BNC shield Ground?


Since my spectrum analyser is 75 Ohm input impedance & I've no option for VNA or similar device to estimate the transfer function of the probe

4) Is it necessary to convert or put 50 to 75 ohm adapter in between? i dont actually have one and commercial ones are pretty expensive for me to afford.
5)Or if i can work in 75 Ohm impedance system, what else should i take care?
6)The BNC connector to the Toroid core be 75 Ohm or 50 Ohm?
7) Is there any chance that i will be overloading ( destructively) the spectrum analyzer input by any chance?what should i take care?
8) Is it ok to proceed without transfer function estimation of the probe, my idea is that observing the spectrum analyser with and without cable inside so that i can see the base line noise first ,and then inserting the cable and check the effect of rework on board by maintaining the setup same.

Any suggestions will be really helpful,

Thanks

 

[jiripolivka]

Many responses depend on the frequency your probe will operate at. Such probe is not a MEASURING device, so the 50-75 Ohm mismatch is not too important.
1. Teflon wire or cable can be used.
2. yes
3. yes, use a grounded narrow strip, take care it cannot make a shorting loop.
4. I think impedance is not important here as the loop will not have a constant impedance over frequency
5. Again, not important
6. as 5.
7. Depends on the RF power your loop will pick up. Take care and use a good attenuator.
8. As I explained, the probe is an INDICATOR, not a MEASURING device. You can try to calibrate its transfer function over frequency but it depends also on the structure ( stripline, etc.,) over which the probe will be located.

Narda and others make CALIBRATED current probes but they are quite expensive. Your trial probe will be useful to follow where current intensity is larger than elsewhere, and if spurious are present. Skill is needed to handle it as you will find.
YOu can also try a probe without a core, size depends on frequency. One turn can be OK, too.

 

[themaccabee]

Thank you very much for the reply

3. yes, use a grounded narrow strip, take care it cannot make a shorting loop.

I've a doubt regarding that section.
what i had in mind was that i will wind a copper tape without any insulation over the insulated teflon wire as shown below( brown color) now shall i short its two end together and connect it to the BNC shield to which one end of the Teflon wire is already connected.
Also if possible can u plz explain the exact functionality of this copper tape ?
Thanks

 

[jiripolivka]

Thank you very much for the reply

I've a doubt regarding that section.
what i had in mind was that i will wind a copper tape without any insulation over the insulated teflon wire as shown below( brown color) now shall i short its two end together and connect it to the BNC shield to which one end of the Teflon wire is already connected.
Also if possible can u plz explain the exact functionality of this copper tape ?
Thanks

The piece of copper tape is connected to ground (at the connector) and bent around the wire loop to prevent the electric component to induce RF voltage in the "live" loop. The loop should only get induced voltage due to the current in the RF line next to the loop, so only RF magnetic component generates loop output.
Make sure the copper tape is not making a shorted turn close to the loop, it should only function as a shield.

- - - Updated - - -

Thank you very much for the reply

I've a doubt regarding that section.
what i had in mind was that i will wind a copper tape without any insulation over the insulated teflon wire as shown below( brown color) now shall i short its two end together and connect it to the BNC shield to which one end of the Teflon wire is already connected.
Also if possible can u plz explain the exact functionality of this copper tape ?
Thanks

The piece of copper tape is connected to ground (at the connector) and bent around the wire loop to prevent the electric component to induce RF voltage in the "live" loop. The loop should only get induced voltage due to the current in the RF line next to the loop, so only RF magnetic component generates loop output.
Make sure the copper tape is not making a shorted turn close to the loop, it should only function as a shield.
In the picture on the left, only the left side of CU foil is connected to BNC ground, the right side is left free. Instead of wrapping the foil as shown you can only bend it around the loop or the toroid so it screens the above from RF VOLTAGE. Only RF current generates the output voltage.

 

[perky]

I'm also a little confused by this. I would think wrapping the Cu tape around the torroid as shown would represent a shorted turn and would act to counter any flux changes in the toroid yielding no voltage output on the main winding. Should the foil instead be partially wrapped so that there is a small gap all around the circumference of the outside of the ring? That to me would make more sense, it would shield the inside circumference of the ring from E fields but wouldn't represent a shorted turn. If so, should the connection to the BNC be done only at one end? (Or maybe it's best to connect both ends but have another gap diametrically opposite?)