Royer mod MXL R2 value question

GroupDIY Audio Forum

Help Support GroupDIY Audio Forum:

This site may earn a commission from merchant affiliate links, including eBay, Amazon, and others.

soapfoot

Well-known member
Joined
Dec 27, 2010
Messages
1,068
Location
LA + Brooklyn
2A864837-1E0D-4C2F-B010-6433E5A6F66C.gif

Since deciding to try soliloqueen’s “flat k47” in these mics, I pulled them back out after about 12 years of non-use.

One thing that stands out to me is the 22M value for R2.

Many other mics seem to use a substantially higher value—100M up to 1G—for capsule polarization. Of course many also use a smaller value for C1.

This project was clearly intended to be a relatively easy and inexpensive DIY entry point—and IIRC 20 years ago 22M was much cheaper and easier to come by in a small 1/4w metal film form factor than the higher values.

My question is—if I were to swap this with a 250M metal film I’ve got lying around, what sort of change in operation (if any) might I expect?

Is the smaller value compensated by the relatively-large C1 (meaning I could expect effectively no significant change)?

What if I used a nicer polystyrene cap in a more-manageable .01 value and tried a 250M resistor instead?

Since it appears to be for capsule polarization (not grid leak or anything tube-related), there should be no danger to the 5840, correct?

And since there’s no current being drawn through it, it also shouldn’t materially impact polarization voltage—is that a fair assumption?

Just want to double check that it’s safe to try!

While we’re at it… any other optimizations/modifications to suggest?
 
Last edited:
There's pretty much no reason to up the value of that resistor. There might be some merit to that idea IF the "bottom end" of the capsule (on the schematic) was grounded, and the bias voltage was applied to the front diaphragm (which would then also require a DC-blocking cap ahead of the tube's grid connection.

As it stands, that 22M resistor is simply the series resistor in that RC low-pass / high-cut filter. Increasing that value would only lengthen the RC time-constant - the time it takes the capsule bias voltage to reach its full and final value (as it stands, 22M & 0.1u gives 2.2 seconds to reach 63%, and 10-11 seconds to reach "full" value).

Look carefully how the capsule is connected here:
16909-52f9855e13ef3a86c69dc4e26bd4d66c.data
 
Back
Top