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

[jp8]

Refresh my memory Jan … in your prototype PCB where you’ve made room for the Takstar CM-60 rolloff switch, you incorporated the LC filter on the XLR output, right? If that ends up being effective, I’ll be curious to know what values you used.

If I were to order a bead that’d slide over the JFET gate in my build, which one should I use? I’m still a bit confused how the ferrite values and impedance works. On the impedance graphs should I be looking for one where the values are high for the frequencies I want to attenuate?

I have made 4 different PCBs, for 5 different circuits. These include:

1. For CM-60: KM84 circuit, SMT version, HPF switch, with ground-to-Pin 1 ferrite bead.
2. For CM-63: KM84 JFET circuit, CMOS polarization voltage generator, HPF and Pad switch, with ground-to-Pin 1 ferrite bead.
3. For CM-63: My own design, with transformer output ("colored" sound with T-8 transformer or "transparent" low distortion circuit with Lundahl), HPF and Pad switch, with ground-to-Pin 1 ferrite bead.
4. For CM-63: My own design, transformerless output (but not the ordinary Schoeps), HPF and Pad switch, EMC filter on output with Johanson X2Y capacitor and Common Mode Choke in the signal lines, and Ferrite bead in ground to pin 1.

So none of them have the classic LC output filter. The last design should be the best, maybe even overkill with both the X2Y cap and CM Choke. We'll see.

The output of CMOS oscillator of designs 2, 3 and 4 are adjustable. Most passive SMT components are of size 0805, so they should be easy to build (imho).

Once tested, I will share the schematics and BOM of the first two designs and make PCBs available through PCBWay and maybe other channels.

If I'm satisfied with the designs, I'll port them to other form factors, e.g. for MXL603/991, Alctron T02A and others.

Jan

 

[Cqwet Dbdfte]

I have made 4 different PCBs, for 5 different circuits. These include:

1. For CM-60: KM84 circuit, SMT version, HPF switch, with ground-to-Pin 1 ferrite bead.
2. For CM-63: KM84 JFET circuit, CMOS polarization voltage generator, HPF and Pad switch, with ground-to-Pin 1 ferrite bead.
3. For CM-63: My own design, with transformer output ("colored" sound with T-8 transformer or "transparent" low distortion circuit with Lundahl), HPF and Pad switch, with ground-to-Pin 1 ferrite bead.
4. For CM-63: My own design, transformerless output (but not the ordinary Schoeps), HPF and Pad switch, EMC filter on output with Johanson X2Y capacitor and Common Mode Choke in the signal lines, and Ferrite bead in ground to pin 1.

So none of them have the classic LC output filter. The last design should be the best, maybe even overkill with both the X2Y cap and CM Choke. We'll see.

The output of CMOS oscillator of designs 2, 3 and 4 are adjustable. Most passive SMT components are of size 0805, so they should be easy to build (imho).

Once tested, I will share the schematics and BOM of the first two designs and make PCBs available through PCBWay and maybe other channels.

If I'm satisfied with the designs, I'll port them to other form factors, e.g. for MXL603/991, Alctron T02A and others.

Jan

How do you test RFI susceptibility on those designs?

 

[joulupukki]

I have made 4 different PCBs, for 5 different circuits. These include:

1. For CM-60: KM84 circuit, SMT version, HPF switch, with ground-to-Pin 1 ferrite bead.
2. For CM-63: KM84 JFET circuit, CMOS polarization voltage generator, HPF and Pad switch, with ground-to-Pin 1 ferrite bead.
3. For CM-63: My own design, with transformer output ("colored" sound with T-8 transformer or "transparent" low distortion circuit with Lundahl), HPF and Pad switch, with ground-to-Pin 1 ferrite bead.
4. For CM-63: My own design, transformerless output (but not the ordinary Schoeps), HPF and Pad switch, EMC filter on output with Johanson X2Y capacitor and Common Mode Choke in the signal lines, and Ferrite bead in ground to pin 1.

So none of them have the classic LC output filter. The last design should be the best, maybe even overkill with both the X2Y cap and CM Choke. We'll see.

The output of CMOS oscillator of designs 2, 3 and 4 are adjustable. Most passive SMT components are of size 0805, so they should be easy to build (imho).

Once tested, I will share the schematics and BOM of the first two designs and make PCBs available through PCBWay and maybe other channels.

If I'm satisfied with the designs, I'll port them to other form factors, e.g. for MXL603/991, Alctron T02A and others.

Jan

Thanks for that update Jan! I'm excited to hear how those turn out.

While I had one of the mics open (the one I've been experimenting with) I switched the main three tantalum 4.7uF caps to 22uF electrolytics just to see if there'd be any benefit (MP uses 22uF). These ones are a little fatter than the 5x11mm used in the MP. I had to leave the C3 cap leads a little long so the C4 tantalum cap could bend into the groove and still fit inside the mic body. Result: no audible difference.




After spending a little more time with these two new cables, one of them is still more susceptible to cell noise than the other. At first, I thought it could have been my mics, but definitely there's some sort of difference between the Neutrik EMC connectors or maybe in how I assembled them or something. Thought that was interesting.

 

[ricardo]

My sole purpose of the test was to see if having the ceramics on the PCB or on the XLR would make a difference as that is what many people think they do.

What I and Dip Ing. Wuttke found was that you need the ceramics on the pins as well as the bead/inductor connecting p2 & 3 to the PCB.

Btw, this ground bead is also in the Neutrik EMC plugs to prevent RF CM currents flowing through your PCB. So it's not only relying on the circumferential connection of the cable shield to the XLR connector housing.

Yup. I have a document in MicBuilders on the subject including stuff from the Neutrik inventor. Some people have replaced the inductors/beads from XLR p2 & 3 to PCB with a 22R resistor but I have not done any tests on this.

If you look at the Schoeps CMC5 schematic, you'll immediately see that they took EMC measures seriously: a bead in the JFET gate (Dr5 = Drossel 5 = Inductor 5). And it has the LC filter on the XLR output. But those parts are wired components not suitable to suppress UHF frequencies as cqwet-dbdfte already pointed out. So whether it will hold up against cell phone RFI...?

Yes to Cqwet-Dbdfte and to SMD, ground planes etc for a new design (which hopefully gets the 'ground tag' to merge with the mike body properly too).

I don't have a Calrec CB Radio anymore but I do have the first production Calrec stick mikes with these mods. (measurements on MicBuilders to see how they have fared after 45yrs)

They seem OK with my Nokia / Microsoft Lumia phone which probably is doing its best to contact the nearest phone tower from Cooktown, Far North Queensland

 

[klem]

@joulupukki
- I was unclear if you built 1 or 2 cables w/ the Neutrik EMC connectors.

- Did you make just one end of the cable an EMC type, e.g., the XLR-F, or did you do both M and F sides of cable?

If you built 2 cables (with either one or both ends with the EMCs), could you put them in series and see if there's any further improvement over your first measurement a couple pages back? I'm of course curious whether there's a 'dose-response' curve to these thingamabobs...

Glad to see you're getting some benefit with them!

 

[joulupukki]

@joulupukki
- I was unclear if you built 1 or 2 cables w/ the Neutrik EMC connectors.

I build two mics and two cables with the Neutrik EMC connectors.

- Did you make just one end of the cable an EMC type, e.g., the XLR-F, or did you do both M and F sides of cable?

Yep. Just the female end that plugs into the mics.

If you built 2 cables (with either one or both ends with the EMCs), could you put them in series and see if there's any further improvement over your first measurement a couple pages back? I'm of course curious whether there's a 'dose-response' curve to these thingamabobs...

I’ll have to give that a shot tomorrow just to see. One of them does a better job than the other. I’ve swapped mics and the one cable connector just isn’t as good for some reason. I put them together in the same way.

Glad to see you're getting some benefit with them!

Me too! Thanks to you all for the help on this!

 

[dai h.]

hi,

re: Neutrik EMC XLR connectors, I thought this was an interesting post on another forum:

https://gearspace.com/board/connect...ries/1131104-neutrik-emc-xlrs-real-world.html

also, FWIW here is a picture of a more recent Schoeps pin 1 to chassis connection (not my pic but reproduced for educational purposes) :



edit: (adding this because I think it makes it easier to understand what the deal is with the Schoeps pin 1 to chassis end connector thing. Source: www.smar.com)

 

[joulupukki]

If you built 2 cables (with either one or both ends with the EMCs), could you put them in series and see if there's any further improvement over your first measurement a couple pages back? I'm of course curious whether there's a 'dose-response' curve to these thingamabobs...

No improvement. It’s as good as the worst one (my “B” cable). Not sure why, but it is what it is.

 

[jp8]

How do you test RFI susceptibility on those designs?

I have described my RF Jammer on my website. But as the website is not yet online, I have made a pdf of the page involved and attached it to this post. An RF expert like you will either shake his head or roll on the floor laughing. I don't care. It works for me when doing comparative tests.

I have a LiteVNA 64 available, but I haven't done any measurements on mic filter circuits yet with it.

Jan

 

[jp8]

Thanks for that update Jan! I'm excited to hear how those turn out.

While I had one of the mics open (the one I've been experimenting with) I switched the main three tantalum 4.7uF caps to 22uF electrolytics just to see if there'd be any benefit (MP uses 22uF). These ones are a little fatter than the 5x11mm used in the MP. I had to leave the C3 cap leads a little long so the C4 tantalum cap could bend into the groove and still fit inside the mic body. Result: no audible difference.


After spending a little more time with these two new cables, one of them is still more susceptible to cell noise than the other. At first, I thought it could have been my mics, but definitely there's some sort of difference between the Neutrik EMC connectors or maybe in how I assembled them or something. Thought that was interesting.

You're welcome! I'm excited too, but I'll have the PCB's just before my holidays. Don't expect any results before the end of April.

Tantalum and electrolytic caps contain diode like structures that cause nonlinearities, which could perhaps cause AM demodulation. But I left the idea because I could not find any reference to it on the web and the diode in the tantalum cap is reverse biased and and will not rectify the signal. The elcap has two diode structures, in both electrodes one, but it was exactly that mic having the elcaps which had the highest RF immunity.

Did you use the same cable brand/type/length for both cables? Shielding efficiency differs from brand to brand and shield type. I still have some Mogami Quad Core cable which I'd like to test and compare to good and bad regular microphone cable. That cable should be really good.

Jan

 

[jp8]

What I and Dip Ing. Wuttke found was that you need the ceramics on the pins as well as the bead/inductor connecting p2 & 3 to the PCB.

Sure, beads in the outputs will help. Similarly as a ground bead, it would reduce RF CM voltages on the cable to cause RF currents to run through your circuit.

But I found the order in which bead and caps were place somewhat confusing. Maybe I'm missing something, but in order to work as an LC filter for RF sigals entering from the cable, I would have expected the caps to sit on the PCB. I see that also being done in other mics, e.g. Gefell MT71S comes to mind. In the Schoeps circuit the beads and caps would act as an LC filter for outgoing RF signals. So I assumed they were (also) meant to filter the outgoing oscillator noise and prevent the mic from becoming a shortwave transmitter. Don't know about the Schoeps, but at least the MXL770 needs the output caps to filter the 2MHz oscillator.

When designing the output filter circuit of the Calric mics, did you also consider the LC circuit as used in a.o. the Gefell M295?

Jan

 

[Cqwet Dbdfte]

I'm curious how well a shielded CAT6A etc etc cable with shielded connectors would fare.
The system is mostly concerned with preserving digital information, but may pan out OK for low level audio.
It would be inexpensive.
The XLR connectors and cables were evidently not designed for a high RFI environment in the pre wifi era.
.

 

[joulupukki]

You're welcome! I'm excited too, but I'll have the PCB's just before my holidays. Don't expect any results before the end of April.

Darn. Well, okay. Enjoy the holdays. We'll still be here when you return. Hehe.

Did you use the same cable brand/type/length for both cables? Shielding efficiency differs from brand to brand and shield type. I still have some Mogami Quad Core cable which I'd like to test and compare to good and bad regular microphone cable. That cable should be really good.

I had two inexpensive cables I've been using. No idea where I got them from originally but I've had them for years. I had to cut off about 4 inches on one of them because of a strange short near the end, but other than that they are identical.

 

[Cqwet Dbdfte]

The shielding on various mike cables differ. Gotham sells some special "Reussen" shielded cables. Better?

 

[jp8]

At line level, this is already done. For analog signals from mics, I don't know.

Jan

 

[jp8]

I'm curious how well a shielded CAT6A etc etc cable with shielded connectors would fare.
The system is mostly concerned with preserving digital information, but may pan out OK for low level audio.
It would be inexpensive.
The XLR connectors and cables were evidently not designed for a high RFI environment in the pre wifi era.
.

At line level, this is already done. I don't know whether it's being used for analog mic level signals.

Jan

 

[klem]

Darn. Well, okay. Enjoy the holdays. We'll still be here when you return. Hehe.

I had two inexpensive cables I've been using. No idea where I got them from originally but I've had them for years. I had to cut off about 4 inches on one of them because of a strange short near the end, but other than that they are identical.

​

I'd highly consider some Canare L-4E6S Star Quad by the foot. Cheap and robust, though the braided shield is always a bit irksome to work with. It sounds like the current cable may have unidentified faults, no?

I'm curious if the new cables in series also were negligible in further attenuating the remaining RFI. Maybe the cable is compromising the EMC capabilities, or maybe the EMC design doesn't prove additional benefit when using more than a single connector.

That gearspace thread raised some interesting questions with no real good response as well. Ultimately, I imagine many people here, myself included, wish there were some in-production and field-proven, affordable, and clear-cut solutions for both RF and EMI. Location recording feels like a roll of the dice these days...

In trying to sum up the proposed solutions so far, I'm seeing:
1) ensure grounding best practices, pin 1 to shell then to PCB (?) and / or
2) a cap (gearspace OP mentioned 0.1uF 100v ceramic Z5U axial cap soldered between pin 1 and cable shell)
3) FET - specific ferrite beads for the leg to capsule
4) carbon mesh stuffed against either side of PCB
5) copper mesh surrounding mic shell
?

So much of this is a bit over my head, but it is interesting how many different theories and suggestions there are in this thread among people who clearly have experience with this stuff, but few if any clear cut suggestions for part #s etc. Maybe that's reflective of the recalcitrant / intractable nature of the issue, but the absence of at least some agreed upon best practices seem conspicuously absent. I'm sorry you're having to wade through it all!

 

[ccaudle]

maybe the EMC design doesn't prove additional benefit when using more than a single connector.

It is clear when you look at the construction that there is only a benefit when installed at a chassis connector on a conductive chassis. A Neutrik EMC connector installed in the middle of a cable run will have zero benefit.

 

[Ike Zimbel]

Darn. Well, okay. Enjoy the holdays. We'll still be here when you return. Hehe.

I had two inexpensive cables I've been using. No idea where I got them from originally but I've had them for years. I had to cut off about 4 inches on one of them because of a strange short near the end, but other than that they are identical.

The EMC connectors can only be as good as the cable they are mounted on. In low-cost manufacturing, even two "identical" cables can have different properties. Don't use crap cable.

 

[Ike Zimbel]

​

I'd highly consider some Canare L-4E6S Star Quad by the foot. Cheap and robust, though the braided shield is always a bit irksome to work with. It sounds like the current cable may have unidentified faults, no?

I'm curious if the new cables in series also were negligible in further attenuating the remaining RFI. Maybe the cable is compromising the EMC capabilities, or maybe the EMC design doesn't prove additional benefit when using more than a single connector.

That gearspace thread raised some interesting questions with no real good response as well. Ultimately, I imagine many people here, myself included, wish there were some in-production and field-proven, affordable, and clear-cut solutions for both RF and EMI. Location recording feels like a roll of the dice these days...

In trying to sum up the proposed solutions so far, I'm seeing:
1) ensure grounding best practices, pin 1 to shell then to PCB (?) and / or
2) a cap (gearspace OP mentioned 0.1uF 100v ceramic Z5U axial cap soldered between pin 1 and cable shell)
3) FET - specific ferrite beads for the leg to capsule
4) carbon mesh stuffed against either side of PCB
5) copper mesh surrounding mic shell
?

So much of this is a bit over my head, but it is interesting how many different theories and suggestions there are in this thread among people who clearly have experience with this stuff, but few if any clear cut suggestions for part #s etc. Maybe that's reflective of the recalcitrant / intractable nature of the issue, but the absence of at least some agreed upon best practices seem conspicuously absent. I'm sorry you're having to wade through it all!

Actually, the Neutrik EMC cables have shown consistent, desired results. So I think they could be considered a solution to this issue over a wide range of manufactured and DIY microphones.