XLR to TS grounding/shock concerns

Tutorials, tools, techniques, methodologies......

Moderator: Modulators

User avatar
FAP
Merzwow
Posts: 633
Joined: Wed Jan 08, 2020 11:50 am
Has thanked: 129 times
Been thanked: 278 times

Re: XLR to TS grounding/shock concerns

Post by FAP »

An update:

1.JPG
Viewed 588 times
2.JPG
Viewed 588 times
3.JPG
Viewed 588 times

I do plan on making more of these things eventually, hopefully improving on each earlier iteration: as of now, only one exits, and it's in the hands of a happy customer.

Features, in brief:
- Handset w/ toggle between earpiece (passive) and mouthpiece (active via 9v battery) microphones
- Base unit w/ rotary dial killswitch/tremolo effect
- Bi-directional input/output setup i.e. you can plug whatever instrument you want into either jack
- Optional hook switch bypass: hang-up the phone to cut/mute the signal, or turn the bypass on to disable that functionality entirely
User avatar
FAP
Merzwow
Posts: 633
Joined: Wed Jan 08, 2020 11:50 am
Has thanked: 129 times
Been thanked: 278 times

XLR in conjunction with rotary dial and mic

Post by FAP »

Alright, guys, riddle me this:

I'm trying to make this this thing work with XLR; the reason for XLR is to use phantom power to activate the carbon mic (mouthpiece).
If you all may recall, I tried this schematic which uses a 600:600 transformer to no avail: the only way I could get the carbon mic to be activated [with the involvement of a transformer] was if the mic element was on the same side as the power source, contrary to the aforementioned schematic.
I achieved this with 1/4" by placing a 9v battery in series with the mic element: obviously this would be undesirable if using phantom power.

Anyways, the other main component here is the rotary dial, which essentially acts as 1 to 10 n.c. momentary switches in a row (depending on what number is dialed).
With 1/4", simply placing the rotary dial in series with the + lead worked fine.
With XLR, placing the rotary dial in series with the + lead (pin 2) yielded... mixed results.
On one hand, if I ran the carbon mic element through the rotary dial, every momentary disconnect yielded a sharp power spike that peaks levels on a mixer; I want to clarify this is definitely not supposed to happen and is not ideal, since this never happens with 1/4".
However, I also tried running a "pro" XLR mic through the exact same setup and it worked as expected, so what gives?

I looked up specs for the "pro" mic I was using, and based on what I've found I've come up with two possible theories:
1. The mic element is at fault. The "pro" mic element is [according to what I've read] cardiod, whereas the rotary phone mouthpiece element is carbon: perhaps carbon mics just can't be used for balanced mics?
2. The wiring is at fault. The "pro" mic uses white wiring for + (pin 2) and red wiring for - (pin 3), which is the reverse of what I've been lead to believe: why? Furthermore, there's no wiring for signal GND (pin 1) anywhere in the mic: it's just connected internally to the mic's metal chassis. In other words, the mic's chassis GND is connected to the signal GND of the mixer or whatever else it's ultimately being run into. Is this common practice?

Please see this image for reference:
help-me-infographic-for-xlr-rotary-hpone-112922.jpg
I also tried placing the rotary dial on other lines (e.g. in series with pin 1 or 3 instead of 2) and got similar results.
Interestingly, while using either of the two "ground" planes on the carbon mic as GND activates the mic, it appears this was never the intent: the phone handset only connects to the center and ring on the carbon mic i.e. the carbon mic was never [meant to be?] grounded when used for its original purpose as a phone handset mouthpiece.

Help me: what's actually going on here?
User avatar
crochambeau
Merzwow
Posts: 599
Joined: Fri Nov 22, 2019 11:21 am
Location: Cascadia
Has thanked: 241 times
Been thanked: 184 times
Contact:

Re: XLR to TS grounding/shock concerns

Post by crochambeau »

Is DC on/passing through the rotary switch stack when the noisy configuration is used?
When in doubt, add resistance.

http://www.rochambeau.net/
User avatar
FAP
Merzwow
Posts: 633
Joined: Wed Jan 08, 2020 11:50 am
Has thanked: 129 times
Been thanked: 278 times

Re: XLR to TS grounding/shock concerns

Post by FAP »

Yep, phantom power is on in both instances.
User avatar
crochambeau
Merzwow
Posts: 599
Joined: Fri Nov 22, 2019 11:21 am
Location: Cascadia
Has thanked: 241 times
Been thanked: 184 times
Contact:

Re: XLR to TS grounding/shock concerns

Post by crochambeau »

FAP wrote: Sun Nov 20, 2022 4:35 am Yep, phantom power is on in both instances.
The Gemini microphone you have listed above is a dynamic though, what are you using phantom power for in that instance? Is the DC getting blocked by a matching transformer? Dynamic moving coil microphones do not need a DC charge/bias in order to operate, in fact, DC is bad when mixed with moving coil microphones - unless it is 100% common mode across the + & - legs of the coil (meaning no DC across the coil, even if you measure DC in reference to ground).

Let me rephrase the line of inquiry: when your rotary dial exhibits unwanted popping: is there ANY difference in signal (AC and or DC) at the actual switching mechanism that is causing the popping than when you can use the switching mechanism WITHOUT the unwanted popping sound?

The switches are doing what they are designed to do, produce heavy enough spikes on the signal line to communicate with a telephone exchange miles away from the handset location. In order to bend the application of the switches to your desires you need to hinder their reach on DC power rail (if you're dead set on using phantom and a charged carbon mouthpiece) and isolate the switch function on the signal AC only.
When in doubt, add resistance.

http://www.rochambeau.net/
User avatar
FAP
Merzwow
Posts: 633
Joined: Wed Jan 08, 2020 11:50 am
Has thanked: 129 times
Been thanked: 278 times

Re: XLR to TS grounding/shock concerns

Post by FAP »

crochambeau wrote: The Gemini microphone you have listed above is a dynamic though, what are you using phantom power for in that instance?
Now that you mention it, I just sort of assumed it needed it; I didn't notice any difference with phantom power on vs. off during testing.
crochambeau wrote: Is the DC getting blocked by a matching transformer?
No clue. Do you mean is there another transformer in the Gemini mic? If so, then also no clue: I'd have to crack open the mic element to find out and I don't want to do that.
crochambeau wrote: Dynamic moving coil microphones do not need a DC charge/bias in order to operate, in fact, DC is bad when mixed with moving coil microphones - unless it is 100% common mode across the + & - legs of the coil (meaning no DC across the coil, even if you measure DC in reference to ground).
Didn't know this: the Gemini mic acting the same regardless of phantom power status makes a lot more sense now.
crochambeau wrote: When your rotary dial exhibits unwanted popping: is there ANY difference in signal (AC and or DC) at the actual switching mechanism that is causing the popping than when you can use the switching mechanism WITHOUT the unwanted popping sound?
Not 100% sure what you're getting at but I think the answer is no: aside from the mics (Gemini vs. mouthpiece) and the wiring thereof, all else is equal.
crochambeau wrote: In order to bend the application of the switches to your desires you need to hinder their reach on DC power rail (if you're dead set on using phantom and a charged carbon mouthpiece) and isolate the switch function on the signal AC only.
Yep: the only problem, then, is figuring out how to do it.
I thought maybe I could use a relay in some way to artificially "clone" the rotary dial terminals (sort of like transforming a SPST switch into a DPST switch) but I feel like that may require a separate power source, which would defeat the purpose of using phantom power as the sole power source.

Honestly, the more I look at this, the less reason I see to use XLR or phantom power for this project: it's not powering the carbon mic like I'd expected it to (regardless if the rotary dial element is included or not), and it's not like the handset is made of metal, so there's no reason to convert the handset into a passive balanced mic, either.

What I really don't understand and what's so frustrating to me is how I've built the same initial schematic (see attached) over and over again and it's never worked.
I've only ever gotten the carbon mic to activate if its power source is on the same side of the transformer (the original schematic shows the power source on the opposite side).
This initial schematic is derived from one used by Godfried-Willem Raes: he's not the pinnacle of truth in human form or anything, but I'd think a guy who's been around that long in the world of experimental audio production techniques wouldn't waster his time throwing up something he knew wouldn't work.
Attachments
og_circuit.jpeg
User avatar
FAP
Merzwow
Posts: 633
Joined: Wed Jan 08, 2020 11:50 am
Has thanked: 129 times
Been thanked: 278 times

Re: XLR to TS grounding/shock concerns

Post by FAP »

For reference, here's the schematic I used for the 1/4" version of the build: notice how the 9v battery is in line with the mouthpiece i.e. the mouthpiece is powered directly by the battery.
This is what I mean when I say I've only ever gotten the mouthpiece to work if its power source is on the same "side" of the transformer as the mouthpiece.
quarter-inch-phone-handset-schematic_090322.jpg
User avatar
crochambeau
Merzwow
Posts: 599
Joined: Fri Nov 22, 2019 11:21 am
Location: Cascadia
Has thanked: 241 times
Been thanked: 184 times
Contact:

Re: XLR to TS grounding/shock concerns

Post by crochambeau »

FAP wrote: Mon Nov 21, 2022 3:31 pm
crochambeau wrote: When your rotary dial exhibits unwanted popping: is there ANY difference in signal (AC and or DC) at the actual switching mechanism that is causing the popping than when you can use the switching mechanism WITHOUT the unwanted popping sound?
Not 100% sure what you're getting at but I think the answer is no: aside from the mics (Gemini vs. mouthpiece) and the wiring thereof, all else is equal.
You do have tools with which to measure voltages at various points in a circuit; do you make it a habit to use them?

I'm going to pound on this nail a bit, so bear with me and please understand I mean mean no offence.

I will now explain what I meant by the inquiry above.

"When your rotary dial exhibits unwanted popping"

I think this part of the query acts as a suitable stand alone opening statement, it makes sense to me that when a circuit is acting up, it will tend to get measured in hopes of discovering what ails it.

"is there ANY difference in signal (AC and or DC) at the actual switching mechanism that is causing the popping than when you can use the switching mechanism WITHOUT the unwanted popping sound?"

I'm realizing that this sentence may have been unclear, I will now try to rectify that.

ASSUMING as I do that voltage state measurements would be undertaken when a build is acting up - since you have apparently enjoyed the build WITHOUT the problematic aspects, it stands to reason that it is possible to configure the build in a way that does not introduce the problematic popping, yes?

....and being able to configure a build in a way that is not objectionable (albeit not ideal for whatever design criteria drives you), you can then MEASURE THE PROBLEMATIC SECTION OF THE CIRCUIT WHEN IT IS NOT MALFUNCTIONING, and then compare your readings of the two states to determine what (if anything) has changed.

Then from there you can leverage those findings towards more hypothesis, one of which hopefully ringing true.

Anyway, on to your schematic related questions: When you hang a 9 volt battery across the microphone capsule there is 9 volts DC across the capsule and it works. Did you measure across the capsule in the phantom power configuration? It looks to me like the +48 common mode voltage is supposed to enter in on pin 2 & 3, which ALSO *accepts* an AC differential voltage coming from the capsule/transformer that will be amplified at preamp while the common mode (+48) is ignored.

Following me so far?

That diagram shows the phantom DC power being fed equally to the outer legs of a center tapped winding of a transformer - the center tap of which is connected to a resistor that then feeds DC to one side of the capsule.

That side sees the DC voltage BLOCKED by a capacitor before entering the other winding of the above transformer.

Now, there is another line from the XLR jack, tied to pin 1. Pin 1 SHOULD be ground/common/shield/drain. But on the schematic you have posted above, that leg has a +48V note on it, even though it connects to an XLR jack that has clearly got a ground symbol tied to it.

So - Did you measure DC voltages at the failed implementation of the phantom powered microphone at the actual microphone capsule?

If the capsule has +48 volts on both sides, as far as the capsule is concerned, it is a a zero or near zero voltage potential - uncharged.

If the "low" side of the microphone transformer is actually connected to ground, then you should read +48 volts *across* the capsule, which would power it.

If you want the switch to work at interrupting your signal, without massive spikes, you need to interrupt your signal at a point in which it is NOT riding in tandem with +48V phantom power.

There is a spot on that schematic that meets that criteria: not in the DC charged lines, AC signal only. I have a hunch you can find it.
When in doubt, add resistance.

http://www.rochambeau.net/
User avatar
FAP
Merzwow
Posts: 633
Joined: Wed Jan 08, 2020 11:50 am
Has thanked: 129 times
Been thanked: 278 times

Re: XLR to TS grounding/shock concerns

Post by FAP »

Alright, so I did some initial measurements this morning & here's what I found:
- Voltage across mouthpiece capsule is 0v
- Voltage from center tap (before 6.8kΩ resistor) to GND is 0v
- Voltage from end of 6.8kΩ resistor to GND is also 0v
- Voltage across XLR pins is as expected (1 + 2 = 48v, 1 + 3 = 48v, 2 + 3 = 0v)
- Tried changing value of cap from 1µf to 10µf: no difference
- Tried flipping polarity of the cap[s] (polarity isn't specified in the original schematic): no difference
- Tried shorting 6.8kΩ resistor: no difference
- Tried shorting pin 2 lead to center tap: this activates the mic (makes sense since there's now +48v flowing through the capsule to GND, completely bypassing the transformer)
- Same applies if pin 2 is swapped out for pin 3 in above scenario

Note that I've removed the rotary dial element entirely for these measurements.

It's to my understanding that the center tap on one side of the transformer feeds DC voltage to the other side, then is blocked by the DC blocking cap, right?
So then if I'm not reading anything from the center tap, does that mean I got a dud transformer?
User avatar
crochambeau
Merzwow
Posts: 599
Joined: Fri Nov 22, 2019 11:21 am
Location: Cascadia
Has thanked: 241 times
Been thanked: 184 times
Contact:

Re: XLR to TS grounding/shock concerns

Post by crochambeau »

FAP wrote: Tue Nov 22, 2022 6:07 am So then if I'm not reading anything from the center tap, does that mean I got a dud transformer?
With your common multimeter lead connected to whatever is tied to pin 1/ground of the mixer (ie: system ground), measure DC voltage at both outer legs of the transformer secondary - where secondary is the winding that physically connects to the mixer side of your build.

Do you measure +48V at both locations there?
Do I need to produce an illustration to accompany this or is my instruction clear?
When in doubt, add resistance.

http://www.rochambeau.net/
Post Reply