Microphone FET bias and noise

GroupDIY Audio Forum

Help Support GroupDIY Audio Forum:

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

RuudNL

Well-known member
Joined
Apr 26, 2009
Messages
3,106
Location
Haule / The Netherlands
Some time ago, I bought a pair of McCrypt BM-700 microphones. (Also sold as ISK BM-700) They were cheap, but sounded amazingly good!
I saw that Kirstein had the same type of microphone on sale for an even lower price than I bought them for, so I bought two more of them.
Now the type number was Pronomic CM-22. At first sight they looked exactly the same as the BM-700's I already had.
But to my surprise these CM-22 microphones were very noisy. The BM-700's were extremely quiet, so there must be something wrong, I thought.
Here is a handdrawn schematic PRR found for me on the internet:

BM700.GIF


At first, I suspected the FET after the capsule. The microphones use a no-brand 2SK30A, so I replaced it by a 2SK170BL.
While I was busy, I also replaced the 0.1 mfd capacitors by 1 mfd, to get a better LF response.
I readjusted the bias for the same voltage on the drain as I had originally measured (= +5.6V, voltage over the FET stage = 7.5V, Uds=3.7V), but I was surprised that he noise level was still high, you could still "hear the sea"...
When I changed the bias on the FET, I reached a point were the noise seemed to 'fade out'.
Microphone sensitivity remained the same, and no distortion was heard.
To check this, I injected a signal from a generator though a small capacitor (6.8 pF) to the gate and measured the distortion after the microphone amplifier. Maximum output suffered slightly, but distortion figures were still very low.
I ended up with +4.6V on the drain.

What could be the explanation that the noise depends so much on the FET bias? I think the difference in s/n is in the order of 15 dB, compared to the state they were in when first delivered! Now they sound very good, but I am just curious...
 
A Fet is like a Highway to a working site : if there is too many car passing trough it becomes noisier from the auditor , so Biasing is like just let the right amount of car go trough te working site,  too much is noisier and the job can't get done because some of them have nothing to do,  Just an analogy.  ;)

  Best,
DAn,

 
For a JFET, Johnson noise current is proportional to the square root of the drain current. Flicker noise current is directly proportional to the drain current.
 
I reached a point were the noise seemed to 'fade out'

I just experieced this myself ...... I a while back I made a little Oktava MK012 circuit on a vero board with a few mods I had instaled a bias trimmer.... I was just blindly tweaking the trimmer while speaking near the mic & found the same , I didnt notice the noise much until it went away the things in totally silent now same nice sound level about the same

cheers
 
Back
Top