Compound amplifier with constant current sources and DC coupling between stages

[JR Cho]

I found 'Compound amplifier with constant current sources and DC coupling between stages' in TUBECAD Journal. With single IXYS 10M45S or 10M90S I think I may be able to implement simple CCS's. This looks similar issue with CCS's for CCD line amp?

Any comment and advice would be appreciated.

 

[Alllensoncanon]

I don't see the two 5K resistors do much other than reducing the constant current output voltage swing by 50V (aka: Increase the drop off voltage by 50V. Other than that, it has the same function without these two resistors. Why put it in?

 

[Sodacose]

Those resistors are there to limit the voltage across the (solid state) CCSs. It's a stepping stone to tube CCSs in the article. Typical TubeCAD compound arrangement madness.

This is from SRPP deconstructed.

The Tube CAD Journal: Mixed Mode Amplifiers

 

[Hearinspace]

What J.B.'s motivation was I don't know but there are some ideas about added series resistance here.

 

[Alllensoncanon]

Those resistors are there to limit the voltage across the (solid state) CCSs.

IXYS 10M45S has a 450V limits. The 50V reduction from the 5K is not needed here.

 

[JR Cho]

Hi all,
With or without 5k resistors. Is that okay to connetct directly b/w V1 and V2 without coupling capacitors?
I'm considering to build the circuit 12AU7 as V1 with 10M90S CCS and 6922 as v2 with resistor load (w/o CCS) and also considering capacitor coupling b/w V! and V2 for stable working (I'm not sure interaction b/w V1 and V2 in case of Direct Coupled..).

 

[Alllensoncanon]

Is that okay to connect directly b/w V1 and V2 without coupling capacitors?

Yes. All you need to do is adjust V1 plate voltage via the CSS active load and/or V1 cathode resistor such that it equal to the negative bias voltage that you want for V2. In your example schematic that will be 100V for a -2V of V2 grid bias. If you want a -5V bias, you will need to set it to 97V.

 

[Alllensoncanon]

All you need to do is adjust V1 plate voltage via the CSS active load and/or V1 cathode resistor such that it equal to the negative bias voltage that you want for V2.

More specifically,

1. Replace V2 cathode follower CSS+5K with a 10K resistor. This should yield the ~102V cathode at idle for the 6922.

2. Replace the 200R V1 cathode resistor with a 330R. This will drop the 12AU7 Ip to 6mA from the 10mA Ip of the 6DJ8. This is to retain the -2V Vg bias and the Vp of 100V.

3. Adjust 10M90S CCS R to yield near 6mA "and" the require -2V V2 bias. Note, The -2V bias is more important than the 6mA in this case. 10M90S also needs in circuit adjustment. The data sheet number is only an approximate.

 

[JR Cho]

Compound amplifier with constant current sources and DC coupling between stages 2

Hi,

I have follow-on/similar questions of the forum I posted -
Compound amplifier with constant current sources and DC coupling between stages

Recently I built 'Morgan Jones OTL headphone amplifier' with attached schematic.



It works well but I would change the tube of first stage with smooth and old style one such as 6cg7,6fq7,12au7, etc..

And I would also apply plate loaded CCS instead of R2, 27K for those first stage tubes with keeping direct coupled b/w stages.

I would apply Hybrid Mu Stage of Alan Kimmel as CCS of the first stage. I focus on 'fixed bias of the Mu stage' which can manage plate voltage of first stage tube and it may maintain the operation point of direct coupled.

My expectation is if I'd connect fixed bias resistors with pot, I could manage bias voltage of first stage-plate voltage of first stage and grid voltage of second stage. And I would easily manage 2~3 V difference b/w grid and cathode of second stage for direct coupling b/w stages.

PLS refer to the schematic I would try to apply..

Any recommendations / comments would be appreciated.

 

[savu]

Why not try Ale's Gyrator PCB

Schematic: http://www.bartola.co.uk/valves/tag/mosfet-gyrator/

PCB for sale: http://www.bartola.co.uk/valves/for-sale/gyrator-pcb/

From what you can find online it seems to be the most liked and used Hybrid Mu-Follower.

Regards,
Silviu

 

[JR Cho]

Hi Silviu,

Thank you for your comments.

What I'd try is not Mu Follower but fixed bias CCS using Alan Kimmel hybrid mu follower. Morgan Jones headphone amp has white cathode follower and additional follower function is not required.
And my focus is operation of DC b/w stages using fixed bias CCS, not the quality of CCS.

The Gyrator looks very attractive and it has fixed bias, so I can use it as CCS. But my 'Morgan Jones' has no space to accommodate big CCS/Mu follower circuits.

Regards,

 

[jhstewart9]

The 2nd stage of the Morgan Jones Headphone amplifier is a White Cathode Follower, well known since the beginning of time.


It is useful where both +ve & -ve going signals must have fast response. The ordinary CF does not do that well.