Quantcast
What should follow an electret mic capsule? - Gearspace.com
The No.1 Website for Pro Audio
What should follow an electret mic capsule?
Old 15th September 2012
  #1
Lives for gear
 
🎧 10 years
What should follow an electret mic capsule?

Hey all,

I've been playing around with some 6mm electret mic capsules for a small, body-mounted mic for various purposes (hoping it might work for some live stuff, though I know feedback might become the death of that).

I've tried a couple of very simple transistor circuits (all bipolar) along with some basic circuits mentioned in the datasheets (1 resistor, 1 cap, that sorta thing). But I don't really know what should follow the capsule. As long as I power the capsule, does it matter what kind of preamp follows? Will some things work better, and is there a non-trial-and-error way to know? (i.e. JFET, bipolar, some type of IC, etc). I'd like to have it unbalanced so I can run it through some guitar pedals I have, and possibly build some custom effects for each instrument.

Before anybody says: Yes, there are many things out there that would work very well for my purpose. But, as you can most definitely tell, I've got no idea what I'm doing and I want to learn. So please help, not heckle unless the heckle is really funny

Thanks!
Old 15th September 2012
  #2
Lives for gear
 
🎧 10 years
Related to the mount:
If I have a capsule mounted on the instrument, how bad would it be to have the preamp/power (battery) in a box on the floor with a 1/8" cable between the two? I'm trying to reduce size/weight of the mount on the instrument. A lot of the designs I've seen/used need a 9V and those take up a ton of space. Would I need some high quality shielded wire? I doubt it'd go more than 8 feet since I don't move much in performance (usually in the back or off to the side).
Old 15th September 2012 | Show parent
  #3
Lives for gear
 
Lotus 7's Avatar
Quote:
Originally Posted by Torea ➑️
Related to the mount:
If I have a capsule mounted on the instrument, how bad would it be to have the preamp/power (battery) in a box on the floor with a 1/8" cable between the two? I'm trying to reduce size/weight of the mount on the instrument. A lot of the designs I've seen/used need a 9V and those take up a ton of space. Would I need some high quality shielded wire? I doubt it'd go more than 8 feet since I don't move much in performance (usually in the back or off to the side).
The usual small "powered" electret mics have built-in FET impedance converters that lower the output impedance to (typically) between 1k and 4.4k ohms.

The "one resistor - one capacitor" circuit on the spec sheet is just a power supply noise filter for the 3 to 10 volt (depends on the specific mic) power source for the built-in impedance converter. Try to keep that circuit close to the capsule. You still will need a mic preamplifier with 30 to 60 dB gain How much gain again depends on the mic capsule.

Although the impedance is moderate (1k to 4.4k ohms) The "single-ended" output from the mic capsule itself can still easily drive 8 or 10 feet of 1/8-inch shielded cable. It's best to use cable with fairly low capacitance (try for less then 25 pF/ft.) You can then follow up with a small single IC mic pre amp IC powered by a couple of 9 volt batteries (or a line power supply if you don't need complete portability. Many such IC's are available with both single-ended and balanced outputs possible. Input impedance for those IC mic pre designs (see the IC spec sheet) is usually between 2k and 10 K ohms. For use with a small FET electret capsule, try to stay toward the higher end of that range (5k to 10k ohms).
Old 15th September 2012
  #4
Lives for gear
 
Richard Crowley's Avatar
 
🎧 10 years
If you want to use your electret capsule like a "conventional microphone", you will need additional circuitry to further lower the output impedance, balance the output signal, and to convert the 12~48V phantom power to the voltage needed for the mic capsule. I highly recommend Rick Chinn's circuit: Pressure Zone Microphones - History and other information
Yes, it says "PZM" but that is a mechanical configuration that has nothing to do with the electronic circuit.
Old 15th September 2012 | Show parent
  #5
Lives for gear
 
🎧 10 years
Quote:
Originally Posted by Lotus 7 ➑️
The usual small "powered" electret mics have built-in FET impedance converters that lower the output impedance to (typically) between 1k and 4.4k ohms.

The "one resistor - one capacitor" circuit on the spec sheet is just a power supply noise filter for the 3 to 10 volt (depends on the specific mic) power source for the built-in impedance converter. Try to keep that circuit close to the capsule. You still will need a mic preamplifier with 30 to 60 dB gain How much gain again depends on the mic capsule.

Although the impedance is moderate (1k to 4.4k ohms) The "single-ended" output from the mic capsule itself can still easily drive 8 or 10 feet of 1/8-inch shielded cable. It's best to use cable with fairly low capacitance (try for less then 25 pF/ft.) You can then follow up with a small single IC mic pre amp IC powered by a couple of 9 volt batteries (or a line power supply if you don't need complete portability. Many such IC's are available with both single-ended and balanced outputs possible. Input impedance for those IC mic pre designs (see the IC spec sheet) is usually between 2k and 10 K ohms. For use with a small FET electret capsule, try to stay toward the higher end of that range (5k to 10k ohms).

Yep, the one I've been playing with most says 2.2k. So if I attach the resistor and the cap to the mic before the 1/8" jack, I can have the battery/PS further away?


Quote:
Originally Posted by rcrowley ➑️
If you want to use your electret capsule like a "conventional microphone", you will need additional circuitry to further lower the output impedance, balance the output signal, and to convert the 12~48V phantom power to the voltage needed for the mic capsule. I highly recommend Rick Chinn's circuit: Pressure Zone Microphones - History and other information
Yes, it says "PZM" but that is a mechanical configuration that has nothing to do with the electronic circuit.

Interesting link. But it looks like it'd be hard to do without the PCB. And buying the PCB kinda takes the fun out of it. Also, I often play with a group that likes to pick inconvenient places to play, which means phantom isn't guaranteed, so I'd like to keep it to battery use if possible. And balanced out isn't entirely necessary since the places that don't have phantom usually want us to bring amps so I end up using a guitar amp.

In the case that I plug into a guitar amp, should my circuit's output impedance be higher?
Old 15th September 2012
  #6
Lives for gear
 
Richard Crowley's Avatar
 
🎧 10 years
There is nothing magic about the PC board. This circuit lends itself quite well to hacking the parts you need and building on a generic breadboard.

There is no need to raise the output impedance just because the load impedance is high. A lower source impedance is always beneficial because it removes cable effects from the circuit. Operating into a guitar amp input, the signal level would be my primary concern. Even the raw output from the electret capsule may exceed what a typical passive guitar pickup produces.
Old 15th September 2012
  #7
Lives for gear
 
🎧 10 years
Quote:
Originally Posted by rcrowley ➑️
There is nothing magic about the PC board. This circuit lends itself quite well to hacking the parts you need and building on a generic breadboard.

There is no need to raise the output impedance just because the load impedance is high. A lower source impedance is always beneficial because it removes cable effects from the circuit. Operating into a guitar amp input, the signal level would be my primary concern. Even the raw output from the electret capsule may exceed what a typical passive guitar pickup produces.

Ah, well I'll give it a better look. Thanks again for the link.

So I might need to dampen the signal for a guitar amp? Is it typically closer to a line level signal?
Old 16th September 2012 | Show parent
  #8
Lives for gear
 
Richard Crowley's Avatar
 
🎧 10 years
Quote:
Originally Posted by Torea ➑️
So I might need to dampen the signal for a guitar amp? Is it typically closer to a line level signal?
Yes, attenuation may be necessary.

No nowhere near line-level. But guitar-level is even lower than mic-level and the output of the electret capsule likely exceeds "guitar-level".
Old 16th September 2012 | Show parent
  #9
Lives for gear
 
🎧 10 years
Quote:
Originally Posted by rcrowley ➑️
Yes, attenuation may be necessary.

No nowhere near line-level. But guitar-level is even lower than mic-level and the output of the electret capsule likely exceeds "guitar-level".


Big thanks! I got another demo mic hooked up, changed it around so I could get that 2.2k resistor and the cap inside the body. Should be fun to play with =)
Old 16th September 2012
  #10
Lives for gear
 
🎧 5 years
I have an electret condenser mounted inside a sink plunger that I use for trumpet. It's just a cheap one from the Radio Shack. Electronics are just a junction box with a 9 volt and a pot it. From there I plug into a multi effect which takes instrument level and has outputs for both instrument and mic level. The pot is usually set from 10 o'clock to noon, and I think that it's 100k. I use this either as a plunger mute or push it over a Harmon mute. It gives a pretty lo fi sound but very useable for a lot of things.
Old 17th September 2012
  #11
Gear Addict
 
Minion's Avatar
 
🎧 15 years
Here is a simple curcuit i have used before for a electret capsule that has a low output impedance and runs of of +48v phantom power .....



This one is simpler but with a higher output impedance ....



They aren"t my designs .....

Cheers
Old 17th September 2012
  #12
Lives for gear
 
🎧 10 years
Interesting. Another one to consider if I end up making a phantom power version. Thanks!
Old 17th September 2012
  #13
Lives for gear
 
Lotus 7's Avatar
If you do use one of the phantom powered circuits posted by Minion, you may need to adjust the Zener diode voltage and the mic series resistors depending on the mic voltage rating. Most capsules are designed for a specific voltage: Many run at 1.5 volts, and but others require specific voltages such as 5, 6, 9 and even 10 as common values. Some will work over a small range of voltages such as 5 to 10v.

The two posted circuits are clearly designed to work with 1.5 or 2 v. mics, not mics that require more voltage for the internal impedance converter.
Old 8th July 2013
  #14
Here for the gear
 
🎧 5 years
Hi Minion,

On the schematic you show in post 11 above, how far will that circuit push a decent signal down mic cable? Will it go 20 feet? How about 40 feet?

What is the circuit's gain?

What is the best way to hook it up for an unbalanced amp downstream?

Thanks
Old 9th July 2013 | Show parent
  #15
Lives for gear
 
Richard Crowley's Avatar
 
🎧 10 years
Quote:
Originally Posted by oneoldude ➑️
On the schematic you show in post 11 above, how far will that circuit push a decent signal down mic cable? Will it go 20 feet? How about 40 feet?
Depends on the cable. Depends on how much high-frequency you need to keep.

Quote:
What is the circuit's gain?
Looks like "unity" (i.e. no voltage gain).

Quote:
What is the best way to hook it up for an unbalanced amp downstream?
First answer the question how are you powering it? If you are using an "unbalanced amp upstream" does that mean you are not using the conventional phantom powering scheme?
Old 10th July 2013
  #16
Here for the gear
 
🎧 5 years
Hi Richard,

No, I am not using conventional phantom power.

I found this schematic on the Naint website. It is very similar to the circuit posted by Minion above.



On the website they said the front end of this design floats the capsule and gives most all the benefits of the Linkwitz mod without cutting the capsule. That is nice.

But it is a phantom power design for a balanced system. I use the mic capsules to measure loudspeakers. And my system is unbalanced but I can provide whatever voltage is needed for this circuit.

So I came up with the following modification of the balanced circuit to use in my unbalanced system. I am taking the output from C1 referenced to ground. Will it work as he advertised, but in an unbalanced mode?

Should I load the output with a resistor to V+ and still take the signal out referenced to ground?

I presume in my version it is a buffer with about unity gain. Right?

Any suggestions?

Old 10th July 2013 | Show parent
  #17
Lives for gear
 
Richard Crowley's Avatar
 
🎧 10 years
Quote:
Originally Posted by oneoldude ➑️
On the website they said the front end of this design floats the capsule and gives most all the benefits of the Linkwitz mod without cutting the capsule. That is nice.
Cutting the capsule is not exactly microneurosurgery. The capsules are cheap enough that you could get a couple of spares in case you do damage the first one.

Quote:
But it is a phantom power design for a balanced system. I use the mic capsules to measure loudspeakers at home as a hobby. And my system is unbalanced but I can provide whatever voltage is needed for this circuit.
What system is it? What is it expecting as a mic input?
Does it already provide power for an electret capsule?
Why do you think you need an additional buffer when the capsule already has an identical FET source-follower circuit inside?

Quote:
So I came up with the following modification of the balanced circuit to use in my unbalanced system. I am taking the output from C1 referenced to ground. Will it work as he advertised, but in an unbalanced mode?
Yes, that looks OK to me(*). Five AA cells would power it essentially for the shelf-life of the batteries.

(*) I'm not wild about that "non-Linkwitz mod".
The shell of the electret capsule is typically used as circuit ground and so the shell provides essential shielding properties.
Using the low-side of the electret capsule as the INPUT NODE is the worst-possible case and makes it MUCH MORE sensitive to ambient EMF noise.
It will likely require you to construct an acoustically-transparent, grounded Faraday shield around the capsule and all the circuitry to keep it quiet.

Quote:
I presume in my version it is a buffer with about unity gain. Right?
Correct.

Quote:
Any suggestions?
We are presuming that you know what is the nature of the input to your system, and that you really need an additional buffer stage.
Old 10th July 2013
  #18
Here for the gear
 
🎧 5 years
Thanks for the quick reply. I will try to respond your points in order.

My mic capsules are 6mm dia Panasonic WM60 and WM61 and are installed in brass tubing only a hair larger than the capsules. I have some capsules on order (WM64 series) and am willing to cut them. But I do not want to cut my calibrated capsule. So a circuit that will handle both versions would be nice. I understand that with a Linkwitz cut capsule setup for two wire operation you simply invert the capsule connections and you are off and running. Is that right?

My current mic system is all DIY. I use the Panasonic WM6x series of capsules and at my typical measurement settings I am at around 90 dB SPL. That translates to an ECM output of about 6mv. That is fed into a mic pre that has a gain of about 30 dB and is based on the 2N3906 transistor (I will change that soon to about 40 dB and use the LM833N) . That feeds line-in of an M-Audio Transit Audio Interface which feeds my computer. From there I use ARTA as measurement software but am considering giving REW a shot. I would like to get the signal to line-in up to about 0 dB (1V for consumer apps).

My current setup is most inconvenient. The circuit is not very good and I would rather have a circuit that I can put into an XLR so I don't need to have the extra box. I can provide power to the circuit and the mic by using XLR connectors in an unbalanced mode with one pin for signal one pin for V+ and one pin for ground. I can do that without mucking up the standard cables. Only my mic and preamp need to know.

The reason I need a buffer is that I am pushing the mic signal through 40 feet or more of cheap mic cable. Certainly more when I do outdoor measurements. Total capacitance is probably on the order of 1800P or more and since I am using the mic for measurement I would like to see all the high end that is actually there. Even Linkwitz admonishes you to use cables less than 30 feet long on his schematics. Heck, if the Linkwitz circuit could accept a single supply and fit into an XLR connector I would be done.

EMF can certainly be a problem. I shield everything as well as I can and have never had a problem that I know of.

I hope I have answered your points.

Surely you can tell I am not a circuit designer and need all the help I can get. I am not in love with the circuit I posted. So if there is a better one out there that will do what I need and that will fit into an XLR connector, please show it to me. In any event, any help on this circuit will be greatly appreciated.

Thank you
Old 11th July 2013 | Show parent
  #19
Lives for gear
 
Richard Crowley's Avatar
 
🎧 10 years
Quote:
Originally Posted by oneoldude ➑️
My mic capsules are 6mm dia Panasonic WM60 and WM61 and are installed in brass tubing only a hair larger than the capsules.
That puts the mic capsule and your brass tube at the input node (the very most sensitive part of the circuit). Don't be surprised if it is unusable because of picking up environmental hum.

Quote:
I have some capsules on order (WM64 series) and am willing to cut them. But I do not want to cut my calibrated capsule.
Can't you calibrate one (or all) of your other capsules against your "reference" capsule?

Quote:
So a circuit that will handle both versions would be nice. I understand that with a Linkwitz cut capsule setup for two wire operation you simply invert the capsule connections and you are off and running. Is that right?
There are several ways of using a modified capsule. To wit...



Quote:
My current setup is most inconvenient. The circuit is not very good and I would rather have a circuit that I can put into an XLR so I don't need to have the extra box. I can provide power to the circuit and the mic by using XLR connectors in an unbalanced mode with one pin for signal one pin for V+ and one pin for ground. I can do that without mucking up the standard cables. Only my mic and preamp need to know.
I would use the reference circuit that Rick Chinn publishes for using those mic capsules for balanced XLR
Pressure Zone Microphones - History and other information
You can ignore all the talk about "PZM" as all capsules are basically the same, regardless of they physical configuration.

Quote:
The reason I need a buffer is that I am pushing the mic signal through 40 feet or more of cheap mic cable. Certainly more when I do outdoor measurements. Total capacitance is probably on the order of 1800P or more and since I am using the mic for measurement I would like to see all the high end that is actually there. Even Linkwitz admonishes you to use cables less than 30 feet long on his schematics. Heck, if the Linkwitz circuit could accept a single supply and fit into an XLR connector I would be done.
Using industry-standard, low-impedance, balanced protocol for your measurement microphone would solve your cable length and interchangability problems.
Old 11th July 2013 | Show parent
  #20
Here for the gear
 
🎧 5 years
Thanks for the comments Richard,

Quote:
Originally Posted by Richard Crowley ➑️
That puts the mic capsule and your brass tube at the input node (the very most sensitive part of the circuit). Don't be surprised if it is unusable because of picking up environmental hum.
Yes, that issue raised its ugly head in your last post. I think I may have the problem licked though.


Quote:
Originally Posted by Richard Crowley ➑️
Can't you calibrate one (or all) of your other capsules against your "reference" capsule?
That is exactly why I am going through this. I would like a good standard with which to calibrate all my other mic capsules. Just in case.


Quote:
Originally Posted by Richard Crowley ➑️
There are several ways of using a modified capsule. To wit...
Thank you for the images.

Quote:
Originally Posted by Richard Crowley ➑️
I would use the reference circuit that Rick Chinn publishes for using those mic capsules for balanced XLR
Yes, I have read that a couple times before. It is a bit over my head and quite a bit more complex than what I want to get into.

Quote:
Originally Posted by Richard Crowley ➑️
Using industry-standard, low-impedance, balanced protocol for your measurement microphone would solve your cable length and interchangability problems.
Indeed, I have a preamp mixer for my Behringer mic. But it is a pain to drag out and arrange all the connections with the rest of my stuff. I am building a test box/jig with a chip amp and mic preamp in it as well as the M-Audio Transit. Minimalist switching and a couple patch cords and a USB to the computer. Done! At least that is what I am shooting for. Pro audio industry standard has its place. But it is problematical in a consumer system like mine.

Since I do not understand FETs at all, I have decided to go another way. The high input impedance in FETs is not always necessary and buying a bunch to select one is not a treat.

The ST AN1534 app note shows how to adapt one of their chips to the WM-60A capsule. It is essentially the identical capsule that I am using. In that app note they point out that an 18k ohm input impedance is fine for interfacing with the WM-60A. So I designed a BJT CC Buffer using the 2N5210. It is a relatively quiet transistor and since I am not recording it should do just fine. I used a program to design the buffer with an input impedance of 20k ohm, output impedance of 100 ohm and an Ic of 20uA. That is where the 2N5210 is at its lowest noise contribution. The schematic is below. All parts numbered in the 100s will be in the test box/jig and "Cable" represents the cable capacitance. Yes there is a cap after the Pot. Note that now the MIC case is grounded and no longer at the input node. See, I listened.

So, would you please be so kind as to look it over and see what glaring mistakes I have made. If it looks like it will work, I will build it in an XLR connector. Hoo Haw, what fun.

Thanks for your comments.

Old 11th July 2013
  #21
Lives for gear
 
Lotus 7's Avatar
R4 is labeled wrong. With a 162K resistor as the emitter resistor the output impedance is higher then the input impedance. Looks like the "K" a typo? 162 ohms is probably lower than you need even to drive 1800pF of cable. With a 162 ohm emitter resistor plus the series resistor, the output impedance will be over 250 ohms, not 100 as stated.

If the voltage on the collector was actually 7.5 volts and the emitter resistor was 162 ohms the emitter current would be around 21 mA. Of course, your power source has a 2K series resistor, so it can't supply anything like that current. You apparently used a 162 K ohm resistor in your design. As drawn using a 162k resistor for R4 it won't work because of the very high output impedance. If the resistor is changed to 162 ohms to lower the output impedance to around 250 ohms, it won't work because of the low current power source (limited by R101).

Using a resistor (for R101) that can provide 25 or so mA and a 7.5 v Zener in place of R102 to stabilize the voltage feeding the capsule and the impedance converter may allow it to work and will lower the distortion at low frequencies.
Old 11th July 2013
  #22
Here for the gear
 
🎧 5 years
As I said earlier, I'm no designer. And I am no engineer either. I used a some software to do the design of the CC Buffer. It gave me the very numbers you see. I input them into LTSpice and the curve you see is what I got. The voltage you see is also what I got in the sim. Of interest is that the App Note I mentioned used the same sort of divider to supply V+ to the capsule. No zener and no regulator. I just used it for the whole front end circuit because my V+ is very stable and has very low ripple (.02 % line and load regulation and 0.5mv noise +ripple). I figured a bit of additional filtering and all would be well. Hmmm Maybe not.

I was and am concerned about using 20uA of Ic simply to get noise down. Is that really enough current? I am also concerned whether the divider I have implemented will supply sufficient current to the circuit to get it to run properly. Also, I do not know how to model an ECM in LTS and am only using a voltage source to mimic one.

I used the divider because I do not have any zeners in the 7 to 9 volt range nor do I have fixed regulators in that range. I mean, how bad can it be when the higher voltage source is pretty good to start with?

Well, I checked again and found some 5.1V zeners. I guess I could give it a go with them. The ECM spec sheet says the operating voltage range is 2-10V. 5.1V is right in there. What do you think?

Oops, the program I used only designs for output impedance in CE amps. I have no clue what the output impedance of this amp is and have no idea how to calculate it either.

Assuming a 5.1V zener based voltage source, what input resistance, and Ic would you suggest for a CC design? Also which transistor would be better, the 2N3904 or the 2N5210? For a CE design what Gain, Input impedance and output impedance would you suggest and which of the two transistors above? Those are the inputs I need for the software I have. Indeed, which design is better suited to my needs, CC or CE?


Thanks
Old 11th July 2013
  #23
Lives for gear
 
Lotus 7's Avatar
Sorry, clearly you don't understand how your circuit actually works and how the source impedance of such a circuit is determined.

Your buffer is a simple emitter follower operating in Class-A. The output impedance is set primarily by the emitter resistor. The DC bias on the transistor sets the operating point of the follower. It has to be high enough to allow the maximum peak to peak voltage swing of the signal without the transistor saturating or the voltage across the emitter resistor reaching "0". The DC voltage across the resistor determines the static current.

The power supply must have a low enough impedance that the collector voltage (and in your circuit the voltage feeding the mic) is stable at all audio frequencies at the currents that the transistor will draw. Using a simple voltage divider with the high resistance values in your circuit, as drawn, does not meet those conditions.

The application note that used the simple resistor divider was intended to supply only the low current of the mic. A properly designed emitter follower, with an appropriately sized emitter resistor will draw much more current (10 to 20 mA, not 20 uA) which is why the simple bypassed resistor divider won't work.
Old 11th July 2013
  #24
Here for the gear
 
🎧 5 years
Louts 7,

I think this is what you had in mind. The Zener is 4.7V cause LTSpice did not have what I want. But it is illustrative for this discussion. The mic will draw .5 ma and the buffer will draw 20 ma. R101 will now provide 25ma to be on the safe side. Output impedance is now very low. And R101 is 1/2W.



My questions are:

Does Q1 really have to draw 20ma when the input signal will most likely never see .25V?

Does it make a difference at this current level whether I use the 2N5210 or the 2N3904 with a 2dB higher noise spec?

Does it make any difference that the Zener regulator is downstream rather than very close to the MIC and Q1?

Thanks
Old 11th July 2013
  #25
Lives for gear
 
Lotus 7's Avatar
Closer, but still no cigar.

I was just about to re-draw your schematic.

There is no need for R5, but you do need a coupling capacitor in its place: 200 to 250 uF will not increase the output Z by more than 40 ohms at 20 Hz. It's never a good idea to have DC on an audio output, unless you are absolutely, positively sure that the following input is AC coupled, and you don't care about loud "pops" when you connect/disconnect the signal line. The circuit has no voltage gain or negative feedback (so needs no "phase-shift protection"), and current limiting to protect the transistor from a short can be done with R-101, so all R5 does is increase the output impedance.

With a 4.7 volt Zener. the emitter should be biased to about 2 volts for the best voltage swing. If the Zener is not close to the emitter follower, at least place the bypass cap near them. You need a low AC source impedance at the transistor collector.

Even with a 1800pF load an output impedance of 220 to 250 ohms is low enough to give decent response to 50kHz (way higher then the mic will work at), so change R4 to 250 ohms.

With a 250 ohm emitter resistor, the DC current will be 9.1 mA. Set R-101 to 500 ohms for a current of about 10 MA into the emitter follower, mic and bias resistors, and 10 mA into the Zener.

I'd stick with the 2N5210. Since the transistors Hfe (forward current gain) can vary from less than 200 to as much as 600 you will probably have to adjust R2 to obtain the optimum bias on the transistor (2.0 volts on the emitter). The 2N5210 transistor noise factor will still be significantly lower than the mic, especially with it connected with the Linkwitz mod.

I have no idea of what the DC bias on the MIC output will be so you will have to check it to properly determine the correct polarity of C1. If the voltage on the mic at the R-1 connection point is more then 2.8 volts DC then the cap's (+) lead goes to the mic. If the voltage is less than 2.8 volts, the (-) lead goes to the mic.

Assuming that you use a WM-61A, and assuming that the "Linkwitz" mode of operation reduces the output by 10 dB (a total guess) the output of the mic might be -45 dBv at 1Pa (94 dB SPL). At a loud 114 dB SPL the output signal may be around -25 dBv or 56 mV RMS. for a sine wave the P to P voltage swing will be 158 mV. Of course, musical waveforms are rarely pure sine waves and may easily have Peak to RMS values that exceed 5:1. That means you would see a voltage swing of as much as 1.57 volts (P to P) at a mic average SPL of 114 dB. That's about the point that the buffer will clip (if the mic will actually produce that voltage swing). Who knows?
Old 12th July 2013 | Show parent
  #26
Here for the gear
 
🎧 5 years
Lotus7, your are extremely gracious and giving of your time. This has been a real pleasure.

Quote:
Originally Posted by Lotus 7 ➑️
It's never a good idea to have DC on an audio output, unless you are absolutely, positively sure that the following input is AC coupled, and you don't care about loud "pops" when you connect/disconnect the signal line.
This is a limited purpose item and I am absolutely sure the preamp is AC coupled (after the pot) and I do not mind the pops.

Quote:
Originally Posted by Lotus 7 ➑️
With a 4.7 volt Zener. the emitter should be biased to about 2 volts for the best voltage swing.
Here I may have unintentionally thrown you a curve. I used a 4.7V zener cause that was the closest LTS had in its database. I will actually be using a 5.1V zener. How much of a difference will that make?

Quote:
Originally Posted by Lotus 7 ➑️
If the Zener is not close to the emitter follower, at least place the bypass cap near them. You need a low AC source impedance at the transistor collector.
How small can I go on that cap? I want to put this circuit in an XLR connector that serves as the MIC wand base. It will allow me to swap out the wands as I choose when I choose. Very similar to the Naint drum mics they show on their website. Theirs are short and stubby and mine have a wand sticking out of the RCA jack. I thot I was clever on that design until I saw theirs and discovered they beat me to it.

Quote:
Originally Posted by Lotus 7 ➑️
.... so change R4 to 250 ohms.
Done

Quote:
Originally Posted by Lotus 7 ➑️
Set R-101 to 500 ohms for a current of about 10 MA into the emitter follower, mic and bias resistors, and 10 mA into the Zener.
Is 500 close enough at 5.1V?

Quote:
Originally Posted by Lotus 7 ➑️
I'd stick with the 2N5210.
Done.

Quote:
Originally Posted by Lotus 7 ➑️
....you will probably have to adjust R2 to obtain the optimum bias on the transistor (2.0 volts on the emitter).
Will be done.

Quote:
Originally Posted by Lotus 7 ➑️
If the voltage on the mic at the R-1 connection point is more then 2.8 volts DC then the cap's (+) lead goes to the mic. If the voltage is less than 2.8 volts, the (-) lead goes to the mic.
Will be done.

Quote:
Originally Posted by Lotus 7 ➑️
Assuming that you use a WM-61A, and assuming that the "Linkwitz" mode of operation reduces the output by 10 dB (a total guess) the output of the mic might be -45 dBv at 1Pa (94 dB SPL). At a loud 114 dB SPL the output signal may be around -25 dBv or 56 mV RMS. for a sine wave the P to P voltage swing will be 158 mV. Of course, musical waveforms are rarely pure sine waves and may easily have Peak to RMS values that exceed 5:1. That means you would see a voltage swing of as much as 1.57 volts (P to P) at a mic average SPL of 114 dB. That's about the point that the buffer will clip (if the mic will actually produce that voltage swing). Who knows?
For now there will be no capsule mods. I got ten WM64Ps in the mail today and they are not conducive to the mod. Maybe on another batch. But in my use, measurement not music , the capsule will rarely if ever see more than 100 dB SPL. Might happen on an MLS signal though.

Here is the schematic. You do the thinking I'll do the drawing. It will save you time better spent using your brain. I am retired, all I have is time.

Everything on the right side is in the preamp. Everything on the left is in an XLR connector.

Is it done?

Thanks

Old 12th July 2013
  #27
Lives for gear
 
Lotus 7's Avatar
Allowing DC on the 10K AC-coupled input is OK as long as you are sure that this is never going to be used with a transformer or DC-coupled input. With a 250 ohm output impedance and a cable capacitance of 1800 pF the HF cutoff frequency (3 dB drop) will be over 350 kHz.

The low frequency cut-of is set with the approximate 9k input impedance of the emitter-follower circuit and the 33 uF coupler. It works out to about 0.6 Hz which is really lower than necessary. You may want to drop C1 to 6 to 10 uF to reduce the physical size and to reduce the turn-on delay as the cap charges.

Using a 5.1 volt Zener (instead of a 4.7 V) will make no significant difference. The emitter voltage should be set to: (the supply (collector) voltage) - (0.8 volts) (the base-emitter drop for the 2N5210) /2.

(5.1-0.8)/2=2.15

Ideally, you should adjust the value of R2 for +2.15 volts on the emitter. But a few tenths either way won't significantly affect the peak-to-peak output swing. The exact value will be affected somewhat by the Hfe of the specific transistor you install.

The 500 ohm resistor is fine. The Zener will draw the appropriate amount of current as long as your supply is actually a stable +15 volts. With a 5.1 volt Zener, the current through the resistor will be 19.8 mA, so the resistor will dissipate 196 mW. Use a 1/2 watt 5% resistor minimum, or a 1 watt if you have the room. Using a "standard" value of 510 ohms is also fine.

Again, be sure to actually check the DC output voltage point of the mic. Connected conventionally, I'd expect it to be low so the polarity of C-1 appears to be correct, but do check it anyway.

250 uF total is plenty for the +5.1V. bypass. Using a 100uf cap near the mic capsule and another 150-250 at the Zener is fine.
Old 12th July 2013
  #28
Here for the gear
 
🎧 5 years
Lotus7,

Thanks loads.

My power supply is an Artesian Technologies Model 505 potted analog power module spec'd at +- 15V at 100ma with line and load regulation of 0.02% and ripple + noise of 0.5mv. All it will service is the buffer and both sides of an LM833N as the preamp. Those are neat units, I bot several surplus for a few bucks apiece. Can't beat 'em for a small project.

BTW, you never posted re the Naint circuit at post 16 above that I posted after Minion posted a similar circuit at post 11. Naint says the circuit I posted is insensitive (my words) to FET parameters and floats the ECM so as to achieve the benefits of Linkwitz cutting without cutting the capsule. I guess they get the benefit of the Linkwitz source follower by the way they use the capsule in their circuit. Richard Crowley doesn't like the circuit due to noise concerns in unbalanced mode.

I have tried running the circuit in LTSpice the way they use it and with R6 and the MIC swapped with the MIC case grounded. All I get are crazy results both balanced and unbalanced (but I do not have a model for a mic, I use a voltage source). The circuit at post 16 would be a beauty as an unbalanced buffer at 5.1V. Low parts count, very small space needed and no fet matching. Sometime when you get a chance, take a look at it.

Thanks for all your help.
Old 12th July 2013 | Show parent
  #29
Lives for gear
 
Lotus 7's Avatar
Quote:
Originally Posted by oneoldude ➑️
[...] BTW, you never posted re the Naint circuit at post 16 above that I posted after Minion posted a similar circuit at post 11. Naint says the circuit I posted is insensitive (my words) to FET parameters and floats the ECM so as to achieve the benefits of Linkwitz cutting without cutting the capsule. I guess they get the benefit of the Linkwitz source follower by the way they use the capsule in their circuit. Richard Crowley doesn't like the circuit due to noise concerns in unbalanced mode. [...]
Richard's concerns are of course correct. Using the low side of the capsule as the differential signal source works for Naiant because they enclose (and float) the mic capsule in a grounded solid metal shell with a woven wire grill. Without that shell, adding a series resistor to the low side will create a potential electrostatic "antenna". In addition, the shell to ground capacitance will be different than the capacitance of the high side of the mic (to ground) creating a small imbalance in the mic's output unless more capacitance is added to the high side. The Naiant method is fine if you have a good shell (shield).

If I were designing a balanced out, phantom-powered circuit using a Panasonic ECM, I'd connect it conventionally (single-ended, grounded shell), probably use a low noise, dual, surface-mount op amp with one section connected as a unity gain voltage follower driving the the other as a unity gain inverter, providing both output polarities with excellent balance. Phantom power would be Zener regulated to around +18 to +24, and a filtered divider would set a + 9 to +12 volt virtual ground for the input (two allow for supplying the +/- power for the op amps). It would require a single input coupling cap, a pair of output coupling caps and a bypass for the virtual ground, plus a few resistors. Probably too much stuff to try to squeeze into a XLR sized housing.
Old 12th July 2013
  #30
Here for the gear
 
🎧 5 years
Lotus,

Yea, I started this project with the intention of using an OPA as a buffer amp in a mic handle feeding an OPA pre downstream. I use thru-hole parts. I can't see SMCs anymore. The project evolved a bit kludgy and would not fit in the handle. So I went on a hunt and happily ended up here.

I beg to differ with you on Niant products though. They have two mic formats, the X-O and the X-R. The X-O has a screen. The X-R does not and you can see the capsule case sticking out of its holder. Here is their pic.



The only obvious differences between my mics and theirs are that my cartridge holders are longer and their buffers are balanced. That balanced part is real important.

Clearly, if I strip down the Niant circuit and use it single ended (unoptimized for my use), like this



I will face EMI issues. But how bad will they be? I guess I can only build it and and see.

Thanks to Richard's information I guess I could use a cut capsule this way (unoptimized for my use)



That would put the case at ground and should be a whole lot better re: EMI if I have done it right. Of course, doing it this way negates the beauty of using an un-cut case and still get the benefits of a cut case. But, in engineering as in life, everything is a compromise.

Thanks for your help Lotus.
πŸ“ Reply
Topic:
Post Reply

Welcome to the Gearspace Pro Audio Community!

Registration benefits include:
  • The ability to reply to and create new discussions
  • Access to members-only giveaways & competitions
  • Interact with VIP industry experts in our guest Q&As
  • Access to members-only sub forum discussions
  • Access to members-only Chat Room
  • Get INSTANT ACCESS to the world's best private pro audio Classifieds for only USD $20/year
  • Promote your eBay auctions and Reverb.com listings for free
  • Remove this message!
You need an account to post a reply. Create a username and password below and an account will be created and your post entered.


 
 
Slide to join now Processing…

Forum Jump
Forum Jump