I am at AES in Europe (which?is really fun) and a vendor approached me with a concept?of their?custom chip that supplies a constant current instead of using 48V and 6.81K resistors. It is here?

Based on both Homeros and my own measurements of a FET based mic circuit and my OPA circuit I don't?see an issue that needs resolving. I kind of feel this is a solution looking for a problem.?

Thoughts?

Jules

Homero - Thank you for the updates!??

1. Is the charge amp being accomplished through the Op Amps???

2. Is it possible to use a Jfet with a built in diode for the triode section?? Maybe something like the K596 Jfet?

Jerry - Thank you for the input.? My train of thought was that a Jfet Op Amps may be used to replace the discrete Jfet in the triode mode.? My understanding is that Jfet Op Amps essentially have multiple Jfets?structures on one IC.? Is it not possible to use them in this way?? ?

Thanks in advance!??

On Friday, May 6, 2022, 08:50:19 AM PDT, Homero Leal <homero.leal@...> wrote:

Thank you Jerry Lee and Henry,

@Jerry Lee Marcel?I reduced the value of C8 to 1uf. Also there was an error on the C15 value, it should be 1000p instead of 0.1uf.

@William Spragens?The additional RC filter after VR was added to a new PCB design and sent to PCBWay. Thank you!

Here is version 1.1:

/g/MicBuilders/files/Homero%20Leal%27s%20Files/Pat%27s%20Mic/Pat%27s%20Mic%20V1.1.png

Also a quick update, replacing 10V zener with two 5.1V zeners in series, seems to fix the problem with the LF noise, but VCC is reduced from 9.4V to 8.5V aprox. This should not be a problem, since AFAIK...? JFET drain voltage is not critical to get it working in triode mode.

Waiting for the new PCB version with the additional RC filter. Will let you know how it?goes in a?few days.

Kind regards!

HL

El vie, 6 may 2022 a la(s) 09:42, henryspragens@... (henryspragens@...) escribió:

An additional RC filter would work, or simply adding a resistor between the BC549 and C14, and increasing C14 to push the RC time constant below audibility. The op-amps have good power supply noise rejection, but the FET has none. It needs pure DC.

The advantage of the zener plus transistor style VR in China mics is operation at low voltage, like 3V PIP for use with PC sound cards. The disadvantage is it's noisy.

An additional RC filter would work, or simply adding a resistor between the BC549 and C14, and increasing C14 to push the RC time constant below audibility. The op-amps have good power supply noise rejection, but the FET has none. It needs pure DC.

The advantage of the zener plus transistor style VR in China mics is operation at low voltage, like 3V PIP for use with PC sound cards. The disadvantage is it's noisy.

Homero,

That's a really cool project and a nice tribute to Pat.?

Would love to hear a comparison to a regular mic.

Thanks,
-Scott

On 4/12/22 20:22, Homero Leal wrote:

HI folks!

As some of you may know, one of the dreams of our friend Patrick Davis (aka. P@T, Enjoybiking) was the idea of designing a mic that could have the "tube" sound, but that could also be very cheap to build.

We discussed the topic many times, and the idea of having a solid state tube emulation was a really fascinating option that was always on the table.

Some time ago I found a paper from Dimitri Danyuk called "Triode Emulator", and in this paper he proposes very a simple way to "emulate" triode behavior using a JFET with very specific values for drain and source resistors.

Later, I found this page:



Which provides a way to calculate (look at the end of the page) the required drain and source resistor values using some JFET parameters like:

• VP
• IDSS

And also the input voltage of the circuit (VCC).

So, I decided to try a design using the Dimitri Danyuk approach, with the values calculated from the Fetzer Valve page.

You could find the schematic here:

/g/MicBuilders/files/Homero%20Leal%27s%20Files/Pat%27s%20Mic/Pat%27s%20Mic.png

Circuit design uses a Charge Amplifier input stage, a JFET middle stage (providing the "valve" tone"), and finally a impedance balanced output stage with a Sallen Key HPF at about 19.XHz.

Also, a custom voltage regulator is used, using a BC549C NPN and a zener, very similar to the one used on some very known chinese mics. This VR provides about 9.4V for VCC.

I used two drain resistors and two source resistors, so it could be easier to approach the right values calculated by the Fetzer Valve page, by using the combination of two resistors.

As the JFET stage may have an amplification level on the signal, I also used two capacitors for the charge amplifier, this way you could reduce output level by placing an additional capacitor (some math required).

I know the design may have some room for improvement, but I can confirm I have working mic with this design, and it seems to my untrained ear, that it has a very sleek and warm sound, very similar to the sound of the valve mics.

For the build, I used a cheap Zramo donor mic from Amazon, which that very nice 26mm electret capsule we have talked about before.

I decided to name this schematic? "Pat's Mic"... in memory of our friend and former member of this group, P@T?(Patrick Davis, R.I.P.)

This post is for your review and consideration, and please if you have any ideas for improvement, don't hesitate to let me know.

Thank you, and kind regards!

HL

-- 
---- Scott Helmke ---- scott@... ---- (734) 604-9340 ----
"I have ceased distinguishing between the religious and the secular,
for everything is holy"  - Joe Henry