Hello Guys, that's what I assumed when I first saw the KM84 circuit in the late 70's. I have attached the KM84 schematic for reference and it will appear once approved by Taperssection.
On first blush, I presumed that C2 a 4pf capacitor was acting like the 8pf capacitors that Rupert Neve used from the base of the transistor to ground in the EQ section of the 1073 to prevent RF interference.
This assumption led me early in my career to believe that the Neumann capsules were smoother and that Neumann had somehow "mechanically" smoothed out the rise in the upper midrange attributed to the physics of capsule design.
We had AKG 451 pencil microphones during my 20 year tenure at Ocean Sound Studios in Vancouver. The original 451 were slightly brighter than the KM84 microphones which seemed as present but smoother. However, in the days of tape machines we cherished the extra rise in the AKG 451 as it was always better to cut Highs that boast them in the mix. Today with digital technology we have a much better signal to noise ratio and this is not so necessary.
Now, the AKG 451E had more headroom than the KM84 because of its class "A" emitter follower output stage so the typical 4db rise at 10khz of the capsule response would not negatively effect the headroom. The AKG 451e uses a similar fet input circuit as the Neumann KM84 with no HF de-emphasis and less gain.
Now, on closer examination of the KM84 circuit we find that C2 is placed between the output of the capsule (fet input capacitor) and the output of the 2n3819 fet.
When we analyze the circuit we find that the 2n3819 has an AV (gain) of 60. The 2N3819 has an internal reverse capacitance from Gate to Drain of about 4pf with the capacitance of C2 paralleled across it yielding 8pf.
According to the "Miller Theorem" the internal reverse capacitance is multiplied by the voltage gain of the circuit which is about 60 with a typical 2n3819.
The “Miller Effect” is a constant in tube/fet circuit design and something one usually tries to avoid. Microphone circuit designers like myself usually use circuits that reduce the “Miller Effect” like the CCDA circuit in our tube microphones.
However, it came to me when I was recently preparing a lecture on tube preamplifier design for SFU’s Acoustical Club that these very clever early 70’s Neumann engineers used this effect to their advantage in the KM84.
The Miller Effect will increase the apparent capacitance between gate and drain (CGD) of the drain loaded amplifier in direct relationship to the gain of the fet stage.
So, by selecting the value of C2 plus the internal capacitance of the fet which is multiplied by the gain of the circuit we can predict a high end roll-off at about the same rate that the capsule response is increasing.
Like Neumann the capsules must be skinned with 6 micron mylar to take advantage of the "Miller" enhanced roll-off effect falling in the right frequency range. C2 plus the internal capacitance of 4pf times 60 will certainly roll out very high frequencies.
Now, our circuit has 1/10 the gain in the first stage as the KM84 as we use a 2 stage class “A” emitter follower and drive the transformer from a much lower output impedance. There is 14db less loss in our lower ratio output transformer.
We increase the "C2" value in our circuit to compensate for the lower gain(less Miller effect) of our front end which yields 14db more headroom the the KM84.
The Neumann Cardiod capsule on average measures about 36pf while ours seem to measure on average about 39pf. So, we also take this into account.
The 36pf to 39pf of the capsule results in an output impedance of about 4 million ohms at 1khz and 400 million ohms at 100hz.
Our circuit is not a direct "copy" of the Neumann circuit but it will yield the same results with our 2 stage class "A" as the KM84 circuit by optimizing these capacitor values while retaining the increased headroom and low output impedance of our circuit.
The emitter of the output transistor in our circuit has a 50 ohm impedance compared to the 10,000 ohm output impedance of the 2N3819 fet.
We also provide the option of switching this "de-emphasis" capacitor out retaining the 3-4db rise in the 9-12khz range. Our friends who play Portuguese guitar love this rise on their instruments.
We have a perfect example of a working KM84 here in the shop and when we put it against our CM1084 in the "V" the differences are less than subtle at best.
Two CM1084 microphones are being used as we speak in conjunction with KM84 microphones by our good friends at Angel Studios in London, England to record a 40 piece orchestra.
The KM84 microphones have been workhorses of the Recording Industry since their introduction in the early 70's but now these venerable old microphones are now over 40 years old.
I have three KM84 microphones currently in my shop. One of these is in perfect condition but the other two have compromised capsules from years of use and abuse.
Cheers, Dave Thomas
www.aamicrophones.com
Topic: Advanced Audio km84 "clone" (Read 9387 times)