DIY RF Condenser Mics

[lampas]

I am sorry, but would someone would be so kind and with simple words explain me point of and advantages or disadvantages of the RF condenser mics ?

 

[RuudNL]

The most important advantage is that an RF microphone has no problems with high humidity. (Condensation)
That is because the capsule has a very low impedance at the (RF) working frequency.
Also the voltage on the capsule can be much lower than with a LF condenser microphone.

 

[abbey road d enfer]

An added advantage is that AC bias, being less than DC bias, does not exert as much pull on the diaphragm. It does give designers some freedom as to  increasing sensitivity (by reducing the gap) or increasing max SPL

 

[lampas]

Thank you very much!

 

[emrr]

https://assets.sennheiser.com/global-downloads/file/11061/MKH-Story_WhitePaper_en.pdf

 

[Tubetec]

Thanks for that link Doug ,
brings us right up to date with Sennheisers state of the art.

 

[rogs]

There is quite a range of capacitive values for  different LDC capsules, and I've been experimenting with a couple of different types  - a C12  type and a K67 ...  The K67 measures as c.55pF, and the C12 as c.90pF.
Clearly these are different enough in value to look to see if certain component changes might improve performance.

In addition, I've been looking at the oscillator again, and have come to an experimental conclusion that I know Ruud won't like! 
The Spectrum 5.3uH coils are not matched exactly to what is really needed  - but they're close...  I decided to reduce the value of R5 from 470R to 47R, and to change R6 from 100R to 47R . In addition I linked out R2 .
This has the effect of increasing the current drawn by the oscillator slightly (to around 9mA  - or 4.5mA from each leg- so still within the 48V phantom power spec limits).

However the extra volts presented to the primary of T1 - in conjunction with a couple of 'capsule specific'  component changes - make for a quite dramatic further improvement in sensitivity, and with no measurable increase in the noise floor.

This improvement of the sensitivity is in the order of a further 10dB, taking the noise floor down to around -70dB ...quite respectable for such a simple design, and on a par with many quite high quality Schoeps type Hi-Z designs.

I've attached a  'capsule specific' schematic for the C12, detailing the changes. (The K67 schematic is attached to the next post. )

Edit:  The value of R6 as 47R may be too low for some phantom power supplies .
A value of R6 as 1K is probably a more useful value to start sensitivity experiments with.

 

[rogs]

And here's the schematic for the K67 type capsule

Edit:  The value of R6 as 47R may be too low for some phantom power supplies .
A value of R6 as 1K is probably a more useful value to start sensitivity experiments with.

 

[Gerard]

I am impressed with your success - and frustrated. My capsules have not arrived yet, nor the PCBs. I hope they get here before the Easter break; I am eager to try making these mics and trying them out.

Well done, Rogs. A very impressive performance from a basic, fairly unrefined (undeveloped?) design. You have put a lot of work into it and got the result.

 

[Gerard]

The Spectrum 5.3uH coils are not matched exactly to what is really needed 

What inductance values do you think would be better? Is it worth opening up and trying to rewind the coils for a better match?

 

[emrr]

I am glad to see this explored. 

 

[rogs]

What inductance values do you think would be better? Is it worth opening up and trying to rewind the coils for a better match?

I think the inductance value is pretty much spot on  for this task ... the tuning range of 3 to 7.5uH specified for these Spectrum coils allow for resonant capacitor values  to fall within our requirements here.

The actual loads  within the inductor configuration we are using here are complex - way beyond my maths -  but there is still useful experimentation  to be done here, I feel.

I'm looking now at the balance between the phantom power feed impedance, and the actual oscillator current requirements...
Perhaps increasing the value of  R6, while keeping R5 value low, for example? 
It looks as if it may be possible to reduce the oscillator current to around 4 or 5mA, without losing too much 'gain' ..

It would be useful to get the oscillator current down, if only because one of the main uses of an RF mic is outdoors -- where the phantom power is likely to be battery powered!

Experiments are continuing...

EDIT: A change in value for R6 from 47R to 4k7  seems to be a suitable compromise.  Each leg of the phantom power supply is then only required to supply around 3mA - so only a power dissipation of around 70mW  in each 6k8 resistor.

You could increase the value of R6  further, but the sensitivity seems to drop off more significantly if the DC supply to the oscillator falls below c. 3V

Changing R6 from 47R to 4k7  will drop the sensitivity by a little over 6dB  -  but it does reduce the current drain by around 40%....

 

[Khron]

Not that i know anything about oscillators, but i was starting to wonder - would it help anything, switching (no pun intended) to a JFET or a MOSFET for the oscillator itself (Q1)?

 

[RuudNL]

This improvement of the sensitivity is in the order of a further 10dB, taking the noise floor down to around -70dB ...quite respectable for such a simple design, and on a par with many quite high quality Schoeps type Hi-Z designs.

That is a significant improvement!

 

[Gerard]

Changing R6 from 47R to 4k7  will drop the sensitivity by around 6dB ... but it does reduce the current drain by around 40%....

What effect does this change have on the noise level?

I suspect anyone using this outdoors might prefer to choose the higher sensitivity and low noise over longer battery life. But maybe there's compromise in between that maintains very low noise with acceptable battery life.

 

[rogs]

Not that i know anything about oscillators, but i was starting to wonder - would it help anything, switching (no pun intended) to a JFET or a MOSFET for the oscillator itself (Q1)?

I did try using a JFET, but  had more success with a bipolar.  The 'Q' components of the oscillator - the crystal and T1 -  are the ones which are the most important parameters, and I found it easier to  configure for a more linear sine wave using a BC549 than with an FET....

 

[rogs]

That is a significant improvement!

Yes -- there was a lot of  'wasted' voltage with R5 being 470R.  It does need to exist to help linearise the emitter load, but 47R is a lot less wasteful of 'precious' T1 voltage swing.
Downside is more oscillator current - I'm currently looking at the best compromise between sensitivity and oscillator current, as you can read in my post #111 above....

 

[rogs]

What effect does this change have on the noise level?

I suspect anyone using this outdoors might prefer to choose the higher sensitivity and low noise over longer battery life. But maybe there's compromise in between that maintains very low noise with acceptable battery life.

The noise level will be affected by however much is introduced by  turning up the preamp to compensate for the loss of sensitivity - so in theory  around 6dB.
But if that is from a very low preamp noise floor it may not be a problem..

The increase in oscillator volts which become available as R6 is reduced creates essentially 'noise free' gain, so if the extra current drain is acceptable, the  extra sensitivity will be welcome.

The simple mod of selecting different values of R6 to determine the oscillator amplitude should make it fairly simple to  choose the  best compromise for whatever use you have planned...

 

[Gerard]

The simple mod of selecting different values of R6 to determine the oscillator amplitude should make it fairly simple to  choose the  best compromise for whatever use you have planned...

That sounds wonderfully promising! Oh, where are those PCBs and capsules? China Post, please hurry!

 

[mhelin]

I don't know anything about RF stuff, but regarding FM solution would it be too simple to use an oscillator like used on the common FM bugs (is it Hartley oscillator though in FM bugs only single coil is used) with the Foster-Seeley FM demodulator/discriminator so that the oscillator coil is actually the L1 in the demodulator:

http://aaronscher.com/Circuit_a_Day/week_by_week/August_2016_FM_Foster_Seeley_detector/FM_Foster_Seeley_Detector.html
There is also ltspice emulation asc file for the circuit:
http://aaronscher.com/Circuit_a_Day/week_by_week/August_2016_FM_Foster_Seeley_detector/documents/Foster-Seeley-Simulation.asc

L1 and L2 are "lightly coupled" which means the coils are in close proximity but are not really a single transformer (if I understood correctly).
The condenser capsule could replace C1, which is the parallel capacitor like in this FM bug:

http://www.next.gr/uploads/542-b12fce8580.gif

Also in some FM bugs the capacitor on base is replaced with a varicap diode which is then feed the audio signal to modulate the frequency, so I guess it's another spot for the capsule connection.

http://danyk.cz/stenice.gif
http://educypedia.karadimov.info/library/fm2w.gif

Wonder how Rode did it on NTG3...