JFET Follower Amplifier Cancels Distortion

A resource for technical documentation. Datasheets, application notes, instruction manuals, books and links to resources are found in the Document Library.
ricardo
Posts: 262
Joined: Sun Apr 10, 2011 7:24 am

Re: JFET Follower Amplifier Cancels Distortion

Post by ricardo »

mediatechnology wrote:ricardo - What is the typical value of R5 and R6 in the Schoep's circuit? Are those AOT?
They are indeed AOT in the original. T2/3, the emitter followers, are selected and matched and their base resistors R5/6 too. 75 -150k.

You can take another approach and just use high hfe transistors so there is effectively no voltage across the resistors so Vcb = 0. The emitter followers will easily swing to Vce < 0.1V so you have -0.5 Vp for each leg. The +ve swing depends on the current draw of the mike. A total current of 5mA (so as not to upset some vintage N***s) gives 2.5mA x 6k8 = 17V across the P48V resistors. But the 6k8 drives half the 2k preamp load so 17 x 1k/(6k8 + 1k) = 2.18Vp.

Though I've played a lot with Schoeps circuits, I've never designed one into a commercial product. In the early 80's, there was on old DIN P48V 0.5mA standard that only a lil' ol' mikemaker in Hebden bridge met. None of the big Germans did .. except for the KM84 which was the original P48V mike. We did it with transformers and got good overload with a sneaky Class B circuit.

There are many advantages with transformer output, the most obvious one being RFI/EMI design. A more subtle one is that you can arrange it so the mike can drive unbalanced inputs without loss of signal or noise degradation (You obviously lose the CMRR of a balanced input) I used Calrec transformer stick mikes into a souped up recorder's unbalanced input and later directly into a Sony PCM-F1. I made up a box with Sowter transformers for the F1 but it was another box to carry around.

For a deluxe mike to be hand crafted by virgins from Unobtainium (like genuine Schoeps), I've drawn up a chart for R5/6 vs hfe of individual T2/3 so the -ve swing on each leg will match the above 2.18Vp +ve swing

hfe 196 218 240 262 295 327 360 393 425 469 513 556 611 676 742
Rb 180 200 220 240 270 300 330 360 390 430 470 510 560 620 680k

Of course dis beach bum is just after level and leaves THD & other niceties to da pedants :lol:
User avatar
JR.
Posts: 3700
Joined: Sat May 24, 2008 7:21 pm
Contact:

Re: JFET Follower Amplifier Cancels Distortion

Post by JR. »

I missed the 1G resistor value so agreed, not injecting corrective AC voltage into gate to linearize.

I had to google AOT, I used to see SAT (select at test) on schematics and miss the good old days where schematics didn't always show values, until after we finished the design.

JR
Cancel the "cancel culture", do not support mob hatred.
User avatar
mediatechnology
Posts: 5444
Joined: Sat Aug 11, 2007 2:34 pm
Location: Oak Cliff, Texas
Contact:

Re: JFET Follower Amplifier Cancels Distortion

Post by mediatechnology »

Thanks ricardo and JR.

ricardo - Have you ever viewed the Schoeps circuit as a "current loop" interface? I say this because like a conventional industrial current loop, the load - in this case the emitter load - is distant from the microphone and resides in the preamp at the far end of the loop as a 6K81 resistor. The microphone electronics are not a voltage source - like a transformer output would be - but instead a current sink. The distinction may be six vs. half-a-dozen but that circuit has always piqued my curiosity because the emitter load(s) is (are) at the far end.
I had to google AOT
I knew someone was going to have to look that up.
User avatar
JR.
Posts: 3700
Joined: Sat May 24, 2008 7:21 pm
Contact:

Re: JFET Follower Amplifier Cancels Distortion

Post by JR. »

The transistor with resistor base emitter does look mostly like a current source "at it's base", because most of the current there is constant with only a small variable fraction from the base current. Phantom power interfaces are always a bit of neither-nor, for the emitter side to look like a current source the base needs to be pretty high Z . So perhaps at DC but not high enough Z at AC...

Unless you are talking about something else with current loop. It sure looks like a voltage interface to me.

JR
Cancel the "cancel culture", do not support mob hatred.
ricardo
Posts: 262
Joined: Sun Apr 10, 2011 7:24 am

Re: JFET Follower Amplifier Cancels Distortion

Post by ricardo »

mediatechnology wrote:Have you ever viewed the Schoeps circuit as a "current loop" interface? I say this because like a conventional industrial current loop, the load - in this case the emitter load - is distant from the microphone and resides in the preamp at the far end of the loop as a 6K81 resistor. The microphone electronics are not a voltage source - like a transformer output would be - but instead a current sink.
The impedance at the emitters is very low. Though it depends to some extent on the impedances at the bases as JR points out, the example with 5mA total has about 26R/2.5mA = 10R4 + a bit on each leg.

Because the bases do not see the same impedances, the Schoeps is rarely perfectly balanced. The pic shows the impedance on pins 2 & 3 of a Schoeps circuit with various configurations including transistor hfe. Vertical scale is dB with the bottom of the graph 1R so all examples are between 20dB & 30dB wrt 1R ie 10 - 30R. The rise at LF is due to various coupling capacitors from the FET to the bases. In each case, the follower connected to the source has lower Zo compared to the one connected to the drain.

But because the impedances are small, the absolute imbalances are also small.

Whitlock (and I think Muncy too) in several of his papers points out that it is the absolute imbalance that determines CMR. ie 10R 10% imbalance gives the same CMR as 100R 1% imbalance. His balanced bridge viewpoint is a good way to look at this.

Though not quite as good in real life as a properly implemented transformer output (the transformer adds a big passive filter to stop RF getting to the active devices), the Schoeps is probably the best of the simple electronic 'balanced' mike outputs for EMI. Dip. Ing. Wuttke put his RFI bits in the same places as I did in 1980 though the Calrec is a very different circuit. I'd like to see where the RFI bits are in a modern Schoeps.
Attachments
Output Z on pins 2 &amp; 3 of Schoeps circuit for various configurations.
Output Z on pins 2 & 3 of Schoeps circuit for various configurations.
SchoepsZo.gif (7.88 KiB) Viewed 11969 times
User avatar
mediatechnology
Posts: 5444
Joined: Sat Aug 11, 2007 2:34 pm
Location: Oak Cliff, Texas
Contact:

Re: JFET Follower Amplifier Cancels Distortion

Post by mediatechnology »

The figures for this article have largely disappeared from the web and the original archive making only the text useless. I've reconstructed the original article and am bumping this thread for a re-crawl. Thank you Dimitri for all of your work. Wayne, November 2018.
Post Reply