Is there a way to reduce cell phone interference in DIY KM-84?

[joulupukki]

I didn't follow the discussion in detail; just one question: Ike Zimbel recommended using the Neutrik EMC connector. Have you tried that? It is the easiest way to make a product with a Pin1 problem functional.

https://www.neutrik.com/en/product/nc3fxx-emc-b

Nick Salis

Haven't yet tried it, but willing to give it a shot. It certainly couldn't hurt. I am curious though if distortion from cell phone frequencies is able to get into the audio circuit's audible range before it reaches the XLR connector wouldn't the problem be elsewhere?

Does the MP body tube have a better connection to the bottom bell than the DIY?

Or do they both have poor bonding and that's why the mics need to have a short gate lead?

I don't know that one is better electrically. For serviceability I very much prefer the way that Takstar's bottom bell works ... except I'm not a huge fan how the board connects to the XLR connector. I guess it's a toss-up either way. Not sure if both suffer from poor bonding. I did clean the bottom bell and inside of the body where the bell slides into with some sandpaper to get it nice and shiny. It maybe helped just a tiny bit.

This RF sensitivity was a known problem for older neumann mics,
I suggested to check the u87 model which has a RLC filter at the input and less sensitive to mobile devices. It is an easy mod to adapt this filter.

I'll try to study up on it and see if there might be a way to adapt it to this project without having to rebuild a PCB.

That's at least worth a try - both outputs are FAR lower-impedance nodes than the gate connection.

I'm not really a fan of how this particular PCB makes it difficult to replace the FET. The back side pads are covered by the trimmer resistor. I guess I could heat it from the top and yank them out that way ... or at least attempt that.

 

[Matt Nolan]

Haven't yet tried it, but willing to give it a shot. It certainly couldn't hurt. I am curious though if distortion from cell phone frequencies is able to get into the audio circuit's audible range before it reaches the XLR connector wouldn't the problem be elsewhere?

The point is that the RF may be coming in through the XLR connector. An RF filter there isn't necessarily just to stop stuff getting out.

But I thought you already said you had tried caps at the XLR pins and it hadn't helped? The Neumann connector may be more effective than caps inside the mic body as they may have less parasitic inductance, but if you saw no improvement at all with internal caps, then it suggests that the problem is elsewhere.

Maybe try powering up the mic without a capsule on it but with a tinfoil cover instead - tight around the mic body but not touching the capsule connection pin. That might help rule out the grounding / screening of the capsule as having a problem.

 

[joulupukki]

The point is that the RF may be coming in through the XLR connector. An RF filter there isn't necessarily just to stop stuff getting out.

But I thought you already said you had tried caps at the XLR pins and it hadn't helped? The Neumann connector may be more effective than caps inside the mic body as they may have less parasitic inductance, but if you saw no improvement at all with internal caps, then it suggests that the problem is elsewhere.

Right. And also ... the MP mic, using the same exact XLR cable doesn't have this issue (as much). The problem is, there are multiple differences between the two builds. I suspect, at this point, that the most likely candidate is the longer grid wire from the JFET. It's just not the easiest to quickly test, especially if I want to try to preserve the existing JFET that I'm using.

Maybe try powering up the mic without a capsule on it but with a tinfoil cover instead - tight around the mic body but not touching the capsule connection pin. That might help rule out the grounding / screening of the capsule as having a problem.

This is what I'm seeing with tinfoil hats on the DIY and the MP mics with capsules removed. DIY is on the left (Analog 1) and MP on the right (Analog 2) with my cell phone using LTE data sitting right on top of the mics:

 

[Khron]

What does that huge signal look like on a spectrum analyzer?

 

[joulupukki]

What does that huge signal look like on a spectrum analyzer?

How exactly would I set that up to capture the full spectrum? This is a peak output with the tinfoil hat of the DIY mic and the cell phone right on top of it (showing the peaks) but this is only the audible spectrum as captured by REW.

 

[Chrisfromthepast]

Wow. I was not expecting this issue.
How are you addressing the hole for the high pass switch?

 

[joulupukki]

Wow. I was not expecting this issue.
How are you addressing the hole for the high pass switch?

Simple piece of tape labeling one of them mic "A" and the other one "B". They don't need to be beauty queens and it seems pretty effective so far.

 

[Chrisfromthepast]

Im no RF expert, but the aperture size seems fine for the many cell phone bands. Try some fine mesh under the tape, or maybe even aluminum tape would work.
The paint and adhesive are insulators, so you will have a hole in your faraday cage unless you overlap, and ideally bond the conductors.

 

[joulupukki]

Im no RF expert, but the aperture size seems fine for the many cell phone bands. Try some fine mesh under the tape, or maybe even aluminum tape would work.
The paint and adhesive are insulators, so you will have a hole in your faraday cage unless you overlap, and ideally bond the conductors.

This aluminum tape didn’t improve it at all…and it definitely has electrical contact with the mic body (tested it with a DMM):

 

[Chrisfromthepast]

So, the E field isn’t the problem, its M field?
That aperture is right over the primary winding.
(Very sorry to those who are cringing about my misuse of terms) is it B field

 

[joulupukki]

So, the E field isn’t the problem, its M field?
That aperture is right over the primary winding.

I’m not sure I follow. To rule out the mic body being the issue I temporarily inserted the DIY board into the MP body (couldn’t screw it in because the screws won’t align). Same problem exists - it picked up lots of cell phone interference.

 

[Chrisfromthepast]

I feel bad pushing this recipe on you. I have not had this problem.
Engineers in the orchestra world have told me that km84’s could expose a musician who brought a cell phone onto the stage against union rules. I haven’t run into this issue with one of my builds, but I haven’t explicitly tested for it either.
Thanks for the learning experience. I appreciate that you are trying 1 variable at a time. We’ll figure it out.

I had a 105 open the other day that made an interesting connection to the shell of the mic.

 

[joulupukki]

I feel bad pushing this recipe on you. I have not had this problem.

Don’t feel bad at all. The MP build obviously has something about it that better controls this situation so it’s gotta be possible. I just hope it doesn’t mean re-designing a PCB. Although I won’t rule that out.

Engineers in the orchestra world have told me that km84’s could expose a musician who brought a cell phone onto the stage against union rules. I haven’t run into this issue with one of my builds, but I haven’t explicitly tested for it either.
Thanks for the learning experience. I appreciate that you are trying 1 variable at a time. We’ll figure it out.

I had a 105 open the other day that made an interesting connection to the shell of the mic. View attachment 123568

Wow, that’s an interesting connector! Reminds me of a diagram of the Neutrik EMC connector.

 

[joulupukki]

This morning I shortened as much as I could (without removing it) the existing JFET's gate lead. But, I probably need to entirely replace it because the source and drain legs just aren't enough to where I can get the gate lead to hug the isolated terminal. Zero change in resilience to cell phone interference, but I'd say I only was able to shorten it by 3-4 mm in total. I kinda don't want to have to replace the JFET because I landed on two that are pretty similarly matched ... and I'd have to probably destroy this one in order to get it out since I don't easily have access to the underside of the board so I'd have to heat these leads up from the top side in order to get them to come out.


Edit: And maybe it's also the 1GOhm resistor leads now that I look at it? Those leads in the MP build are also shorter. More and more this is feeling like a new PCB design is in order.

 

[Cqwet Dbdfte]

All that plastic insulation at the end of the PCB looks like it is ready to handle 40kV!
Certainly a PCB can be laid out with SMT parts on a 4-layer stack using some better dielectric, teflon, RO4350, etc, including slits for HV insulation. Better RF centric parts can be used, and the whole thing RF qualified before assembly into the mike resonant cavity ehhh .. body.
Circuits can be placed under RF snap-on shields, like those from Masach, placed on a continuous copper plane.
The word "ground" in RF has limited use, the term "reference plane" may be more suitable.Any ground impedance could make circuit signalling differential.
The board can be assembled by JLC, Gogo, etc.
The spectral plots shown has very even 100 Hz harmonics. Is this in Europe?
If this is power line harmonics the problem would be easier to solve.

 

[Matt Nolan]

This morning I shortened as much as I could (without removing it) the existing JFET's gate lead. But, I probably need to entirely replace it because the source and drain legs just aren't enough to where I can get the gate lead to hug the isolated terminal. Zero change in resilience to cell phone interference, but I'd say I only was able to shorten it by 3-4 mm in total. I kinda don't want to have to replace the JFET because I landed on two that are pretty similarly matched ... and I'd have to probably destroy this one in order to get it out since I don't easily have access to the underside of the board so I'd have to heat these leads up from the top side in order to get them to come out.
View attachment 123668

Edit: And maybe it's also the 1GOhm resistor leads now that I look at it? Those leads in the MP build are also shorter. More and more this is feeling like a new PCB design is in order.

Why not try the ferrite bead on the JFET gate suggestion? You can de-solder or cut that leg to get a bead on and resolder it without removing the JFET. Nothing to lose.

 

[Cqwet Dbdfte]

Ferrite chokes come in hundreds of flavors, with various resistance vs. frequency, so using a random bead may be a lottery. Using an air core inductor would yield more predictable result. I'd start with a 56nH SMT part in the gate lead, even a 1k or so, leaded resistor could work.
Is it a 2N3819? It has gain up to 700MHz.
If you want to try RF shielding the mike, put the foil over the capsule area. The mike body better be conductive.
If the RF gets into your cable, its shielding may be unsuitable for RF. XLR is not a good RF connector, but trying CAT6 with the shields reaching within a mm of the solderjoint on the pins, and the drainwires bonded to the XLR shell may help. Jumper grade CAT6 is normally stranded, could work as mike cable, drain wire solderability is hit and miss.
There are now more choices in semiconductors, another JFET could be considered. The Texas JFE150 is a low noise audio part. Maybe its input capacitance is too high for the KM84?
Anyway the pics of the mike above shows an old school PCB layout, a new one with SMT parts may be in order.

 

[joulupukki]

Why not try the ferrite bead on the JFET gate suggestion? You can de-solder or cut that leg to get a bead on and resolder it without removing the JFET. Nothing to lose.

I would happily try this … but am 100% clueless as to what to get (what value/etc).

Ferrite chokes come in hundreds of flavors, with various resistance vs. frequency, so using a random bead may be a lottery. Using an air core inductor would yield more predictable result. I'd start with a 56nH SMT part in the gate lead, even a 1k or so, leaded resistor could work.
Is it a 2N3819? It has gain up to 700MHz.

Yes, I’m using a 2N3819 JFET. The MP uses a J305. Is that a significant enough difference in the cell phone interference discussion?

If you want to try RF shielding the mike, put the foil over the capsule.
If the RF gets into your cable, its shielding may be unsuitable for RF. XLR is not a good RF connector, but trying CAT6 with the shields reaching within a mm of the solderjoint on the pins, and the drainwires bonded to the XLR shell may help. Jumper grade CAT6 is normally stranded, could work as mike cable, drain wire solderability is hit and miss.

I’m not sure that I follow all of this. I can use the same XLR cable with the MP mic and the cell phone RFI is almost a non-issue.

 

[joulupukki]

Speaking of JFETs, what brand is this and what does the “V524AB” indicate?

 

[Khron]

Vishay-Siliconix (logo is actually Siliconix's own); last row is production fab / batch / date code.

https://www.mouser.fi/ProductDetail/Vishay-Siliconix/J305-E3?qs=E5eJzWX2jIA6qyExkH0nCg==