10M45 Mu-Stage Idea?

[sgerus]

This is an Idea for a Mu-Stage based on the 10M45 chip in place of a Mosfet
(Ref: MU stage Philosophy by Alan Kimmel)

I have not tested this… need to order some parts.

Just wanted to get some feedback from any one that had made a SS Mu-Stage

In general… think this will work?
Also, if I understand it right, the Mu-Stage sets the current 3 or 4 times higher than what V1 needs, then just drains off the extra current via RS1 (the 42K in my schematic)

What is the benefit of setting the current 3 or 4 times higher and just draining off the excess?

[darkfenriz]

Benfit is that anode loading is resistor to ground parallel with cathode resistance. More linear but less gain. Also without this resistor one would be forced to use DC feedback to set anode DC current of the same to CCS.

[rdf]

I recall tubelab tried it and found it worked suprisingly well.

[sgerus]

I got the idea of the 10M45 as a ccs from tubelab, and use it now as a ccs for my 6sl7
Grounded cathode amp... Yes, it works great.

I did not find anything on tubelab where the audio signal actually comes off the ccs device…. That is what I’m trying to accomplish here.

[tubelab.com]

I routinely use the IXYS chip for plate loads, but I have not tried this Mu Stage circuit. I tried several similar circuits that used pentodes wired as CCS's but none of them could drive a DHT well into A2, which was the reason for my experiments at the time. After I added a mosfet follower (PowerDrive) I never went back and experimented with other CCS loaded circuits.

I have seen a circuit similar to this on the web somewhere, except it did not have the current bleed resistor.

When attaching the CCS directly to the plate, a cathode bypass capacitor is needed. You may, or may not need it in this circuit. It affects the gain and the third harmonic distortion.

[rdf]

I use them them too and also learned of them from tubelab, but I'm sure he mentioned it in a one-liner somewhere. I'm guessing it wasn't pursued in depth because he already has his Powerdrive circuit.

3 minute late edit: Ooops, now there's bad timing. I know someone tried this.

[sgerus]

Thanks Guys...

Just ordered the parts, I will let you know in a week or two
if this works or not!

[mach1]

You would do a lot better using two 10M45S in a casocde arrangement (lowers the shunt capacitance) and taking the output from the top of the current set resistor.

The attachment shows DN2540 depletion mode FETs used in this manner. However the circuit is equally applicable to the 10M45S(an enhancement mode FET with an error amplifier which mimics a depletion mode FET). This arrangement works beautifuly and has a very low output impedance.

[dsavitsk]

Might be wirth reading (http://www.audioasylum.com/forums/bo...es/108492.html) the sub thread "What does this mean"

[mach1]

I think Doc B's bin' overdosing on the triple chocolate fudge sundays again ! The 10M45s has an impedance of over 10Meg over much of the audio range, and measures around 7 Meg @10Khz.

See this link for details
http://www.pacifier.com/~gpimm/ccs_performance.htm

This figure is substantially increased in the cascode arrangement.

As for them blowing up all the time, all I can say is that I have had zero failures in over 60 chips using plain old 510R carbon comps as 'gatestoppers'.