DIY RF Condenser Mics

[reega]

Yes, Coils are from a local parts recycler. I've measured them to be 5-13uH. On T1, near the limit but not quite at it, I do get a peak signal but I think T2 may be too far out of range. I'll try finding some smaller ones. FYI: output measurement is poor man's differential probe of two single ended probes and using scope's math function.

At the JFETs input, what should the proper signal look like? An RF modulated signal? I'm seeing a wierd mix of audio level and RF level signal?

(makes a fine radio out of shield with scope leeds attached btw)

 

[rogs]

Did the coils come with a type number and a spec ? ... Unless they are very similar to the Spectrum 5u3H coils it's going to be difficult to advise where your problems may lie.
(see this page :SPECTRUM COMMUNICATIONS about half way down for the specs on the 5u3HH and the earlier 5u3H coils )

The signal observed on the gate of the JFET will depend very much on how much the bridge is unbalanced... anything from a couple of hundred millivolts to 2 or 3 volts. You won't see the modulation affect the signal very much (if at all).
Also, the scope lead will load and detune the secondary of T2 and won't give you a meaningful objective reading, just an indication of the RF levels present.
The same is true where ever you apply the scope leads to the RF circuitry. Even with a x10 probe you can expect a 1MOhm impedance with a 10 or 20pF additional capacitive load at the measuring point.

But to be able to move forward you really need to know just how similar the coils are to the specified Spectrum coils?....

 

[Pariah Zero]

Then it's time to wait for shipping. That's not a huge loss, as it's time to learn more.

@rogs You'll be happy to hear that after switching to the new capsule, the first mic works quite well!

Now I just need to finish assembly of the other two mics (I have the boards assembled; just need to put them in the donor bodies & mount the capsules).

 

[Marik]

Just some information about microphones based on the RF principle:
In the 60's, the technical department of Dutch National Radio (NRU, Nederlandse Radio Unie) developed their own microphones based on the RF idea. The schematic they used was super simple! They work on 12 volts.
I still have one of a similar microphone (MO432) in my own collection and even after 60 years it is still functioning and sounding good.
I have the complete article about the microphone development available, but (unfortunately) it is in Dutch...

The schematics shows only 3 different core transformers/inductors, but I see on the PCB quite a few more. Does the article give any data for those? What is the FF90 on the PCB?

Best, M

 

[reega]

So, I got some spectrum coils, and they are in place. Sadly, the signal looks very much the same. After the first coil, I get a 10Mhz signal at about 1.9V amplitude at peak, The signal going to the JFET looks much like what you see above though. I checked for peak voltages around the circuit and I'm getting 33V. Could my Steinberg UR12 not be providing enough phantom voltage current?

 

[rogs]

V.5 of the schematic draws less than 3mA from the phantom power supply in total, so even a relatively 'weak' supply should function OK.

Using a DMM you should be able to measure voltages within the following values at these points, relative to gnd (pin1):
• Junction of R11 and C7 - c.35V DC
• R11 (which is a link) - c.20v DC
• Junction of R2 and Q1 collector - c.7.5v DC
• Across both R4 and R10 - between c.2v and c.4v DC, depending on the Vp cut off voltage of the specific FET used. Both values should be the same.

Using an oscilloscope with at least a 20MHz bandwidth, and a x10 probe with an impedance of c.1M and a capacitive loading of c.15pF -
(again reference to gnd) :
• Junction of R3 and T1 primary - between 4.5 and 5v AC p-p at 10MHz
• At each end of T1 secondary - between 3.5 and 4v AC p-p at 10MHz.
(those last 2 readings may be slightly different from each other, but typically less than 250mV difference )

• Measuring the signal at Q4 gate will not give a very accurate reading, because the 'scope lead will load T2 secondary, but will typically be a sine wave between 500mV and 2v AC p-p at 10MHz.

Can you confirm that you are getting readings close to those values? ...

 

[reega]

R11 - C7 = 33.8V
R12 - D1 - R11 = 18.0V
R1 - Q1 = 6.9V
Across R4 = 0.58V (something wrong here?)
Across R10 = 9.7V (something wrong here?)

Junction R3-T1 primary measures between 4.48 and 4.98V p-p at 10Mhz.
Each end of T1 secondary is between 3.5 and 4V p-p at 10Mhz, it fluctuates a bit.

Signal at Q4 gate is about 1.2V p-p and generally noisy.

Things are going sideways at Q4? I'll try a different Q4. I don't have a transistor characterizing devices, often thought about rigging one up between sig. gen and scope but never got there.

Any other thoughts?

 

[soliloqueen]

returning to this project, could someone give me a quick rundown of exactly what kind of capsule works best for a project like this? i'm seeing insanely high impedance, which can be done, what about tension? what's the current state of knowledge?

 

[Khron]

To date, I have had my best results with capsules above 65pF in value.
Edge terminated capsules tend to have a higher sensitivity, which is also useful for this type of project.
Centre terminated 'K.67' style capsules tend to be more effective than 'K.47' types - again because of the higher sensitivity.
It does seem to be a bit of a lottery, when it comes to buying these cheap Chinese capsules though......

returning to this project, could someone give me a quick rundown of exactly what kind of capsule works best for a project like this? i'm seeing insanely high impedance, which can be done, what about tension? what's the current state of knowledge?

There's at least part of your answer. And since virtually no DC bias is involved, low(er) tension should be fine, as i understand.

 

[rogs]

returning to this project, could someone give me a quick rundown of exactly what kind of capsule works best for a project like this? i'm seeing insanely high impedance, which can be done, what about tension? what's the current state of knowledge?

Capsules with the parameters I suggested in the post that Khron has linked to seem to work best - at least when using these specific inductors.

On of the main reasons for using RF bias is the massive reduction in operating impedance that the concept brings with it.
The impedance is reduced from the many Mohms that an AF condenser capsule normally operates at, down to a few hundred ohms.
( More details on the first page of Manfred Hibbing's notes here: https://assets.sennheiser.com/global-downloads/file/11061/MKH-Story_WhitePaper_en.pdf )

 

[rogs]

R11 - C7 = 33.8V
R12 - D1 - R11 = 18.0V
R1 - Q1 = 6.9V
Across R4 = 0.58V (something wrong here)
Across R10 = 9.7V (something wrong here)

Thanks for the reference points! Investigating ...

Looks as if the FET may be faulty? ..... The gate is effectively grounded from a DC point of view, so in this configuration the FET is self biased around Vp.
The voltage cross the drain and source resistors (R10 and R4 respectively) should be the same.

 

[reega]

Looks as if the FET may be faulty? ..... The gate is effectively grounded from a DC point of view, so in this configuration the FET is self biased around Vp.
The voltage cross the drain and source resistors (R10 and R4 respectively) should be the same.

Wow, seems to be a lot of variation. First one showed .5V and 10V, second showed 0V and 21V, third showed .5V and 7V ... wish I had a tool to characterize these FETs.
I tried a fourth one and again, I'd gotten 0.5V on one resistor and 10V on the other. That seems to be the most common. Could I put a resistor on either the source or the drain to balance it? That will reduce gain won't it. Or, maybe I could adjust R10 or R4.

Here is the signal I'm getting to Gate now ..

 

[rogs]

Could I put a resistor on either the source or the drain to balance it? That will reduce gain won't it. Or, maybe I could adjust R10 or R4.

There should be no need to adjust the value of R10 or R4 to obtain the same voltage across each resistor. In this configuration the FET will sit naturally at the cut off point . The current flowing in the source resistor will be the same as that flowing in the drain. The voltage across both should be identical.
If this is not the case, there is something else confusing the issue.
If you temporarily unsolder one end of R3 to stop the oscillator, what DC voltages do you measure across R4 and R10 in that situation?.....
(The DC supply voltages will rise a little overall, as there is no oscillator current flowing, but the voltages measured across R4 and R10 should still remain the same as each other)
EDIT: Note that in the case of measuring equal voltages across R4 and R10 we are talking about measuring across the resistors, not simply with respect to ground.
In the case of R4 it's the same thing, but in the case of R10 you need to actually measure the volts across the resistor itself.

 

[rogs]

Here is the signal I'm getting to Gate now ..

The photo you have attached to this post (post #512) doesn't seem to make much sense? .... it appears shows a waveform with a frequency of 141.8KHz?

Here are a couple of screenshots showing typical waveforms you might expect to see.
The first is from the junction of R3 and T1 primary. The second is from the gate of the JFET.
Note the frequency (10MHz) and the time base setting for measuring that (20nS).
As I mentioned earlier, the gate measurement is not likely to be very accurate, amplitude wise.
The scope lead itself will detune the loading on T2 and reduce the 'Q'.....

 

[reega]

characterize

There should be no need to adjust the value of R10 or R4 to obtain the same voltage across each resistor. In this configuration the FET will sit naturally at the cut off point . The current flowing in the source resistor will be the same as that flowing in the drain. The voltage across both should be identical.
If this is not the case, there is something else confusing the issue.
If you temporarily unsolder one end of R3 to stop the oscillator, what DC voltages do you measure across R4 and R10 in that situation?.....
(The DC supply voltages will rise a little overall, as there is no oscillator current flowing, but the voltages measured across R4 and R10 should still remain the same as each other)
EDIT: Note that in the case of measuring equal voltages across R4 and R10 we are talking about measuring across the resistors, not simply with respect to ground.
In the case of R4 it's the same thing, but in the case of R10 you need to actually measure the volts across the resistor itself.

Right. Yes. Disconnected R3, and measured voltages. R10 showed 21V in the sense of pic showed below and R4 showed -0.6V. Hmm. Looking around for shorts or diodes/transistors in backwards by they look OK.

It is definitely the JFET. Think I have to find a better source. It appears to be pulling low between the gate and the drain/source. When I remove it, and short source and drain, you get the expected 16V of a voltage divider where the high voltage is 32V.

By the test method here .. https://www.allaboutcircuits.com/textbook/semiconductors/chpt-5/meter-check-transistor-jfet/
This transistor is a fake J113. Bought from Ali Express. Hasn't happened to me before! Measuring resistance with Drain positive, get Open Circuit, Source positive, get 6MOhm (should be Open Circuit). Other way, get Open Circuit on one and again 6MOhm on the other.

 

[Jen]

Capsules with the parameters I suggested in the post that Khron has linked to seem to work best - at least when using these specific inductors.

On of the main reasons for using RF bias is the massive reduction in operating impedance that the concept brings with it.
The impedance is reduced from the many Mohms that an AF condenser capsule normally operates at, down to a few hundred ohms.
( More details on the first page of Manfred Hibbing's notes here: https://assets.sennheiser.com/global-downloads/file/11061/MKH-Story_WhitePaper_en.pdf )

 

[Jen]

Hey thanks for posting the technical paper on the MKH series of mics. I own a pair of MKH-30 figure of 8 capsule mics and just love them.

 

[Khron]

By the test method here .. https://www.allaboutcircuits.com/textbook/semiconductors/chpt-5/meter-check-transistor-jfet/
This transistor is a fake J113. Bought from Ali Express. Hasn't happened to me before! Measuring resistance with Drain positive, get Open Circuit, Source positive, get 6MOhm (should be Open Circuit). Other way, get Open Circuit on one and again 6MOhm on the other.

Mouser carries legit J113's...

 

[rogs]

Hey thanks for posting the technical paper on the MKH series of mics. I own a pair of MKH-30 figure of 8 capsule mics and just love them.

One of the things it's almost impossible to copy with RF mic experiments is that extra quality that comes with the MKH series capsules......
High sensitivity , low tension, ultra linear, 'push-pull' construction. ..... Designed for use with RF mics of course!
You can occasionally get lucky on some of those qualities with random 'cheapo' Chinese LDCs, but not often -- and certainly not the advantages from the 'push-pull' characteristics.. .

 

[khalidgul]

So I want to add replace sdc with ldc(cardioid capsule with 2 cables)on my Sennheiser mkh 416 rf condenser shotgun microphone. Please let me know how to do this, biasing and modification. Happy to pay if somebody is interested. by the way my first post on the group. Just joined it. Thanks everybody. You are lovely people. With love from pakistan