PI coil cable - triboelectric noise - problems - solutions?

[Polymer]

I believe I have ran into a problem with triboelectric noise when testing my PI Pinpointer.

I have not found anything about triboelectric noise in cables on this forum. Odd.

Still in my ears: "make sure the cable to the coil doesnt move" - it will cause falsing.
Sure enough this is true. There was never really an in depth explanation that I am aware of.

This problem occurs with my switchable coil/probe setup on the MPP E.

The pinpointer probe has a cable to the detector housing and obviously is moved quite a bit when pinpointing.
(yeah - talk about bench testing vs. real life testing - you have my full attention)

This triboelectric noise in cables can run into the 100s of millivolts. Gosh, we are playing with microvolts in a PI.

It would be nice to keep this switcheable setup between coil & PP on the MPP, rather than build an extra PI pinpointer in a rigid pipe without a cable.

I am wondering how these pinpointers dealt with it though:

Any ideas or experiences with this?

[waltr]

Interesting and caused me to google and read about this effect.

Maybe this is also why different results from different cables has been an issue and much discussed on this forum.

On my PI (hammerHead2) detector, I used Cat5, unshielded cables with twisted pairs and have not had any noise or cable
movement issues that is so commonly complained about.

On paper I was reading on "triboelectric noise in cables" indicated it is the center to the shield movement causing the problem.
https://www.pcb.com/contentstore/Mkt...er-Signals.pdf

Solution is better cables.

Did you use coax for the pin-pointer connection?
Have you tried other types of cables?

Try Unshielded twisted pair, microphone cable and other types of coax.

[Ferric Toes]

I believe I have ran into a problem with triboelectric noise when testing my PI Pinpointer.

I have not found anything about triboelectric noise in cables on this forum. Odd.

Still in my ears: "make sure the cable to the coil doesnt move" - it will cause falsing.
Sure enough this is true. There was never really an in depth explanation that I am aware of.

This problem occurs with my switchable coil/probe setup on the MPP E.

The pinpointer probe has a cable to the detector housing and obviously is moved quite a bit when pinpointing.
(yeah - talk about bench testing vs. real life testing - you have my full attention)

This triboelectric noise in cables can run into the 100s of millivolts. Gosh, we are playing with microvolts in a PI.

It would be nice to keep this switcheable setup between coil & PP on the MPP, rather than build an extra PI pinpointer in a rigid pipe without a cable.

I am wondering how these pinpointers dealt with it though: [ATTACH]49657[/ATTACH]

Any ideas or experiences with this?

The detectors that the divers have are Aquapulse AQ1B's which I am very familiar with. The orange cable is RG58 coax but with a polyurethane outer jacket rather than pvc. For use in seawater, the minimum delay is 30uS. I would guess they are divers working the Spanish wrecks off Florida and also they are using 8in search coils, which are the popular size when looking for gold and silver coins. A ferrite cored probe is available, but I haven't seen them used much.

What can happen with coax, is that where the braid end is twisted together and soldered to the coil wire, it becomes a substantial conductor, and if close to the coil winding gives a detectable signal, particularly at shorter sample delays than 30uS. The problem arises where the cable flexes just up from the entry point in the coil shell. This flexing results in the braid conductors tightening and loosening, thereby varying the conductivity of the braid in that location. For the 8in coil, I put in a spacer that lifts the coax joint away from the coil winding by about 1". For a 10" coil with spokes and a centre swivel, the coax entry is near the centre and well away from the coil winding.

The problem is worse for a ferrite cored probe as the coax and joints to the coil wires are right on the axis of the core. In making such a probe, I always have the outer tube long enough so that the coil wires to the coax joint are several inches away from the back face of the core. Always put a drain wire to the core as they are slightly conductive. One thing I have noticed with some cores is that waving in the earth's field generates a signal almost as though it were trying to act as a fluxgate element.

Eric.

[6666]

Hi Eric, interesting post, when you added the 1" spacer was there a really noticeable difference in just such a short distance ?

[Ferric Toes]

Hi Eric, interesting post, when you added the 1" spacer was there a really noticeable difference in just such a short distance ?

Yes, it does make a big difference. As you can seen in the picture, if the coax were brought into the shell without a spacer, the braid and soldered joint to the coil wire would be right up against the coil. It's like a small nugget would respond; you can detect it on the coil edge, but nowhere else. The coil wires go down the short tube of the spacer and join the coax just by the bottom of the strain relief.



Eric.

[6666]

Yes, it does make a big difference. As you can seen in the picture, if the coax were brought into the shell without a spacer, the braid and soldered joint to the coil wire would be right up against the coil. It's like a small nugget would respond; you can detect it on the coil edge, but nowhere else. The coil wires go down the short tube of the spacer and join the coax just by the bottom of the strain relief.

[ATTACH]49668[/ATTACH] [ATTACH]49669[/ATTACH]

Eric.

Thank you Eric for pictures and info, much appreciated.

[Polymer]

Interesting and caused me to google and read about this effect.

Maybe this is also why different results from different cables has been an issue and much discussed on this forum.

On my PI (hammerHead2) detector, I used Cat5, unshielded cables with twisted pairs and have not had any noise or cable
movement issues that is so commonly complained about.

On paper I was reading on "triboelectric noise in cables" indicated it is the center to the shield movement causing the problem.
https://www.pcb.com/contentstore/Mkt...er-Signals.pdf

Solution is better cables.

Did you use coax for the pin-pointer connection?
Have you tried other types of cables?

Try Unshielded twisted pair, microphone cable and other types of coax.

Thank you waltr!

Yes, this is certainly an interesting topic, worth looking further into. I agree that this may very well be the cause of some cable issues here.

I do not have this problem or do not perceive it with a "fixed" cable, snaking down the shaft to a normal search coil.
The problem is very pronounced when the the cable to the pinpointer is flexed whilst moving it about.

Yes, I have been using coax. I have implemented "better" cables.

I have tested three cables so far:
The worst offender is a "low noise" microphone cable - ??? WTF
The next best one is a "normal" microphone cable, but still not suitable for the pinpointer
The one that works with my pinpointer is also a microphone cable.

Talk about confusing things - aaargh.

I will sift through the technical specs. if available and try to figure out if there is some logic to it.

I did find a triboelectric series chart, cannot find it right now.

[Polymer]

The detectors that the divers have are Aquapulse AQ1B's which I am very familiar with. The orange cable is RG58 coax but with a polyurethane outer jacket rather than pvc. For use in seawater, the minimum delay is 30uS. I would guess they are divers working the Spanish wrecks off Florida and also they are using 8in search coils, which are the popular size when looking for gold and silver coins. A ferrite cored probe is available, but I haven't seen them used much.

What can happen with coax, is that where the braid end is twisted together and soldered to the coil wire, it becomes a substantial conductor, and if close to the coil winding gives a detectable signal, particularly at shorter sample delays than 30uS. The problem arises where the cable flexes just up from the entry point in the coil shell. This flexing results in the braid conductors tightening and loosening, thereby varying the conductivity of the braid in that location. For the 8in coil, I put in a spacer that lifts the coax joint away from the coil winding by about 1". For a 10" coil with spokes and a centre swivel, the coax entry is near the centre and well away from the coil winding.

The problem is worse for a ferrite cored probe as the coax and joints to the coil wires are right on the axis of the core. In making such a probe, I always have the outer tube long enough so that the coil wires to the coax joint are several inches away from the back face of the core. Always put a drain wire to the core as they are slightly conductive. One thing I have noticed with some cores is that waving in the earth's field generates a signal almost as though it were trying to act as a fluxgate element.

Eric.

Thank you for the very interesting information Eric and also the photos. This is definitely noteworthy and you have created a really nice solution.

Alas, it does not apply to my pinpointer setup. Otherwise the problem would now be clearly solved.
The probe I made is very long and the solder/braid points are about 16 inches away from the ferrite core.

My testing so far are pointing to the cable itself being the problem.

Ah yes, the problem with the drain wire to the ferrite.
I would like to fully agree, but my setup appears to work better without a grounded ferrite. They are conductive as you have stated.
The how and why is still a mystery to me. I would like to delve further into this.

How do problems with an ungrounded ferrite manifest in your books?

I have had one core so far that showed the "fluxgate effect" you mention. Talk about being puzzled. Carl I believe made me aware of it.
Also an interesting topic to go down the rabbit hole with.

Cheers, Polymer

[Ferric Toes]

Hi to All,
This is a subject worth pursuing further, both for cables and for ferrite rods.

As for cables, I have been testing some guitar cables, particularly Sommer Cable XXL and LLX. Both of these are very rugged and extremely flexible. Also, I am looking for low capacitance which the LLX in particular has. Both have a carbon conductive layer between the braid and the inner insulating core similar to microphone cables. XXL is 88pF/m and LLX 52pF/m. Both have pro's and con's with the LLX winning electrically and the XXL mechanically. Both are greater in diameter than standard RG58 with XXL at 6.8mm diameter and LLX 7.5mm. XXL goes snuggly into a PG7 strain relief but for LLX a larger PG9 is required. Full specs are on the Sommer Cable website.

I have only tested these cables in the workshop but have not noticed any effect that could be called triboelectric by knocking and flexing the cables. I remember some years ago, trying a microphone coax that also had the conducting layer and that cable acted like a microphone when knocked, so was discarded. I have yet to test the above cables outdoors with it wrapped around a shaft.

With regard to ferrite rods there are several interesting points. 1) My testing indicates that long rods have more range than short rods of the same diameter, which seems logical as per the field from bar magnets. 2) Some rods ring mechanically if driven too hard; presumably due to magnetostriction. You can sometimes hear them singing. 3) Some rods also show sensitivity to the earth's magnetic field. Swinging North - South gives a signal while East - West doesn't. Perhaps the susceptibilty changes. 4) How best to shield a ferrite core.

More later, as have to deal with reorganising life in view of corona virus situation.

Eric.

[surfdetector]

Interesting and caused me to google and read about this effect.

Maybe this is also why different results from different cables has been an issue and much discussed on this forum.

On my PI (hammerHead2) detector, I used Cat5, unshielded cables with twisted pairs and have not had any noise or cable
movement issues that is so commonly complained about.

On paper I was reading on "triboelectric noise in cables" indicated it is the center to the shield movement causing the problem.
https://www.pcb.com/contentstore/Mkt...er-Signals.pdf

The accelerometer paper was a very interesting read. I was somewhat pleased to see that the first solution listed was (1) in vibratory environments minimize cable ?whip? ..... a solution that I could comprehend

Cheers

[bklein]

Triboelectricity is caused by two materials rubbing causing electricity. You can also get voltage created by changes in capacitance due to changes in wire spacing due to movement. Another possibility is that the wire insulation itself is piezoelectric. Some of the low capacitance insulation’s have this behavior. Go play with some Kynar wire wrap wire and a scope. :-)

[Polymer]

Here is a link to a triboelectric series with good info.
Especially noteworthy the last paragraph right at the bottom of the page regarding air and dust.

https://www.alphalabinc.com/triboelectric-series/

[Polymer]

Triboelectricity is caused by two materials rubbing causing electricity. You can also get voltage created by changes in capacitance due to changes in wire spacing due to movement. Another possibility is that the wire insulation itself is piezoelectric. Some of the low capacitance insulation?s have this behavior. Go play with some Kynar wire wrap wire and a scope. :-)

Yep, the "fluor" plastics are shown to be very electronegative ...

Changes in capacitance during movement. Yikes, why are we still using cables?!

I remember seeing the concept of having a preamp etc. in the coil, somewhere here in the forum.
That makes much more sense now.

[Polymer]

Hi to All,
This is a subject worth pursuing further, both for cables and for ferrite rods.

As for cables, I have been testing some guitar cables, particularly Sommer Cable XXL and LLX. Both of these are very rugged and extremely flexible. Also, I am looking for low capacitance which the LLX in particular has. Both have a carbon conductive layer between the braid and the inner insulating core similar to microphone cables. XXL is 88pF/m and LLX 52pF/m. Both have pro's and con's with the LLX winning electrically and the XXL mechanically. Both are greater in diameter than standard RG58 with XXL at 6.8mm diameter and LLX 7.5mm. XXL goes snuggly into a PG7 strain relief but for LLX a larger PG9 is required. Full specs are on the Sommer Cable website.

I have only tested these cables in the workshop but have not noticed any effect that could be called triboelectric by knocking and flexing the cables. I remember some years ago, trying a microphone coax that also had the conducting layer and that cable acted like a microphone when knocked, so was discarded. I have yet to test the above cables outdoors with it wrapped around a shaft.

With regard to ferrite rods there are several interesting points. 1) My testing indicates that long rods have more range than short rods of the same diameter, which seems logical as per the field from bar magnets. 2) Some rods ring mechanically if driven too hard; presumably due to magnetostriction. You can sometimes hear them singing. 3) Some rods also show sensitivity to the earth's magnetic field. Swinging North - South gives a signal while East - West doesn't. Perhaps the susceptibilty changes. 4) How best to shield a ferrite core.

More later, as have to deal with reorganising life in view of corona virus situation.

Eric.

You mention one thing that has been difficult for me to bring together as one.
That would be the electrical & mechanical properties.
The cables I have tried so far suck at either one or the other. This perspective is for much flexed pinpointer cables.

As to the ferrite rods:
1.) Yep, my testing indicates that too. There's a nice chart somewhere with diameter/length ratios etc. showing this
2.) This has me laughing -memories- I once had a singing ferrite - the look on my face. Sounded quite good actually.
2a.) One ferrite I drove way to hard, forgetting the limiting resistor, violent ferrite noise, panic switch off.
This ferrite hasn't been "good" to use ever since. Something must have changed in the structure or so.
3.) May have to do with saturation and/or biasing as far as I've understood so far.
4.) Work in progress ...

Looking forward to further posts. May your re-organisation of life flow with ease.

[waltr]

Here is a link to a triboelectric series with good info.
Especially noteworthy the last paragraph right at the bottom of the page regarding air and dust.

https://www.alphalabinc.com/triboelectric-series/

good read, thanks

Yep, the "fluor" plastics are shown to be very electronegative ...

Changes in capacitance during movement. Yikes, why are we still using cables?!

I remember seeing the concept of having a preamp etc. in the coil, somewhere here in the forum.
That makes much more sense now.

Yes, a pre-amp at the RX coil makes good sense. It is true anytime you have a very weak signals (radio receive antennas or metal detector receive coils).
Is is always better to boost the signal before sending down a cable. This helps keep the SNR high and also help keeps other EM from badly affecting the signal of interest.
It is even better to send a differential signal onto twisted pairs (Ethernet, etc). This helps to greatly reduce CM from affecting the signal and receiver.