Spud-Assist: Totem-Pole Current-Mirror PP Hybrid - Page 3 - diyAudio
Go Back   Home > Forums > Amplifiers > Tubes / Valves

Tubes / Valves All about our sweet vacuum tubes :) Threads about Musical Instrument Amps of all kinds should be in the Instruments & Amps forum

Please consider donating to help us continue to serve you.

Ads on/off / Custom Title / More PMs / More album space / Advanced printing & mass image saving
Reply
 
Thread Tools Search this Thread
Old 19th September 2008, 03:23 AM   #21
diyAudio Member
 
Join Date: Nov 2007
Location: Dallas
Back to MJK's question about maintaining a long term current
balance in the face of temperature and other drifts.

I probably threw a few too many conservative tweak parts at it
than needed. If both FETs are on the same heatsink, probably
doesn't need the obsessive extra diode and 2R drop equalizers..

Basic concept is biasing up the Triode with a bypassed CCS.
That just happens to have the same drift as the Anti-Triode.

Its a little different than the "brute" version, but not much.
Still needs a big cap to bridge two low impedance nodes.

---------------------------------------------------------------------------

I'd probably choose to drive IN-, just because a gate won't
forward conduct. Cap on the CCS may "breathe" the same
no matter which input is overdriven, or not, I'm not sure.
Attached Images
File Type: gif tempcomp.gif (30.3 KB, 630 views)
  Reply With Quote
Old 19th September 2008, 04:03 AM   #22
diyAudio Member
 
Wavebourn's Avatar
 
Join Date: Sep 2006
Location: Pleasant Hill, CA
Send a message via Skype™ to Wavebourn
Looks nice, except an odd diode.
__________________
The Devil is not so terrible as his math model is!
  Reply With Quote
Old 19th September 2008, 04:34 AM   #23
diyAudio Member
 
Join Date: Nov 2007
Location: Dallas
I just went overboard a little trying to equalize the power
dissipation of the two FETs, so they would drift equally.

Odd diode merely to drop the same as the BJT cascode's
emitter. That degree of voltage nitpicking went clearly
beyond the absurd. "Do Nothing" 2R on the Anti-Side is
in the same category.

Currents were already equal by design. If the voltages
idle the same, then we got equal power, equal heat,
equal drift... Blah Blah Blah etc etc....
  Reply With Quote
Old 19th September 2008, 04:37 AM   #24
diyAudio Member
 
Wavebourn's Avatar
 
Join Date: Sep 2006
Location: Pleasant Hill, CA
Send a message via Skype™ to Wavebourn
Yes, but if you put them on the same equal heatsink...

Did you breadboard it?

Looks more elegant than my Alligator.
__________________
The Devil is not so terrible as his math model is!
  Reply With Quote
Old 19th September 2008, 04:57 AM   #25
diyAudio Member
 
Join Date: Nov 2007
Location: Dallas
I'm still breadboarding with an actual wooden breadboard.
A new HV power supply, that is...
  Reply With Quote
Old 19th September 2008, 06:28 AM   #26
diyAudio Member
 
Join Date: May 2007
Michael,
As we speak my friend Hoktuna is actually building the SE-version:

Click the image to open in full size.

Then we can easily move the cap to try out the other possibility. Actually it was kenpeter who, in another thread, adviced me to move the cap. Moving up the currentmirror with an unbypassed R below as kenpeter suggests could also be an option.
Attached Images
File Type: png kp.png (65.0 KB, 577 views)
  Reply With Quote
Old 19th September 2008, 06:53 AM   #27
diyAudio Member
 
Michael Koster's Avatar
 
Join Date: Jan 2008
Location: Eureka, CA
Quote:
Originally posted by kenpeter
I just went overboard a little trying to equalize the power
dissipation of the two FETs, so they would drift equally.

Odd diode merely to drop the same as the BJT cascode's
emitter. That degree of voltage nitpicking went clearly
beyond the absurd. "Do Nothing" 2R on the Anti-Side is
in the same category.

Currents were already equal by design. If the voltages
idle the same, then we got equal power, equal heat,
equal drift... Blah Blah Blah etc etc....
This is a nice variation on the circuit. Other possibilities are
bypassed resistor and shunt regulator, but the CCS keeps
good balance.

I think the circuit would operate much the same with the BJT
removed, but the cathode is a convenient upper control terminal
reference voltage for the cascode (very clever), which will also
improve the dynamic performance. Edit: of course, the cascode
itself equalizes the dissipation and keeps the drift between
MOSFETS within reason.

The bypassed CCS will have a time constant and will allow the
cathode voltage to increase under large-signal conditions as
needed to keep the average current at Cset. The tone burst
response will be such that the average current through the
tube will be greater than the average current through the
mirror early in the burst, the extra current going through the
capacitor into R until the cap charges and the tube average
current drops to 1/2 the total. at this point, the midpoint of
the tube's grid signal will have moved downward on the load
line. When the tone burst ends, the cathode will return to
the quiescent value at the time constant of the RC formed
by cathode impedance and C.

Maybe the time constant would need to be tuned, but this
looks like a viable design.

Lars:
The voltage drop across your 750R will moderate the current
shift with signal level and will be between a fixed bias scheme
and the above.

Another place to put the self-bias is under the constant current
common node. This scheme would require an OPT with a small
air gap.

Michael
Attached Images
File Type: png deadsimple-ccm-selfbias.png (5.4 KB, 558 views)
  Reply With Quote
Old 19th September 2008, 07:02 AM   #28
diyAudio Member
 
Michael Koster's Avatar
 
Join Date: Jan 2008
Location: Eureka, CA
Quote:
Originally posted by revintage
Michael,
As we speak my friend Hoktuna is actually building the SE-version:

Then we can easily move the cap to try out the other possibility. Actually it was kenpeter who, in another thread, adviced me to move the cap.

Oops, our posts crossed. This is very interesting as you can
compare to see if the sound is improved or degraded by activating
the drive on the current mirror. One thing, though, is that turning
on the mirror drive raises the load impedance seen by the triode
by a factor of 2. I think you would need to change transformer
ratios or secondary taps to get a direct comparison. Both versions
should have the SE harmonic structure.

Michael
  Reply With Quote
Old 24th September 2008, 10:24 PM   #29
diyAudio Member
 
Michael Koster's Avatar
 
Join Date: Jan 2008
Location: Eureka, CA
Default Synthesis of parallel + totem pole

I think the totem pole loads may be a good solution to DC balance
the dead-simple signal mirror. Any slight unbalance in this circuit will
be carried by the OPT, but it should be very small if the CCS all drift
together. Cascodes would probably be a worthwhile refinement.

The tube sees load equivalent to the impedance across the entire
primary, 5K in this example, since the circuit operates in class A and
the load is shared 50/50 across the ends of the totem pole.

I think it's statically and dynamically stable, preserving the large-
signal asymmetric current swing of the triode throughout.

Time to clear off the workbench again...

Michael

PS it's a lot like Don's #3 but should be DC-stable without adjustments

http://www.diyaudio.com/forums/showt...009#post638009
Attached Images
File Type: png quadse.png (7.9 KB, 512 views)
  Reply With Quote
Old 25th September 2008, 02:42 AM   #30
diyAudio Member
 
Join Date: Nov 2007
Location: Dallas
I would think twice before building that latch.

Suppose the differential Antitriode feels it's side isn't quite
pulling its fair share of DC total current? It draws DC across
the OPT, and lowers V-Plate. Now the Triode's plate will draw
less current, and the differential anti-triode feels even more
strongly that it isn't pulling its share. etc. etc. meltdown.

Or just as easily latch the other way. Maybe your triode gets
the full 80mA? The differential Antitriode would go into total
shut down. What brings it out of this?

Let me be clear which anti-triode I speak of as the "differential".
The anti-triode on the lower right, sitting just below the totem
ANTI-anti-triode. And not the totem anti-triode on the top left.
Clear as mud? I thought so...

I think I would want to halve the totem resistors and DC ref
the totem gates to the lower plate and drain. Even this may
not yet be fully DC stable? I havn't quite wrapped my head
completely around this problem yet.

Too bad I don't actually know how to use LTSpice's ability to
sim this circuit. Much less pretend there is a drift? For me, its
just a nifty drawing utility. You might ask Rev if he can sim for
latchup with tail slightly too big / tail slightly too small.
  Reply With Quote

Reply


Hide this!Advertise here!
Thread Tools Search this Thread
Search this Thread:

Advanced Search

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off
Trackbacks are Off
Pingbacks are Off
Refbacks are Off


Similar Threads
Thread Thread Starter Forum Replies Last Post
Slones Totem Pole Design goodlife Solid State 17 3rd July 2014 07:57 AM
URL collection of Totem Pole and CSPP (Single Ended related) Solid State Output Stage tiefbassuebertr Solid State 7 10th May 2012 05:02 PM
Current Mirror: The V-mirror rtarbell Solid State 1 19th November 2008 02:27 PM
no current mirror, current mirror, cascoded current mirror, bjt or jfet VAS? maxpou Solid State 31 17th June 2008 04:31 PM
Values for OTL totem pole (cathode followers) 00940 Tubes / Valves 2 10th November 2007 10:02 AM


New To Site? Need Help?

All times are GMT. The time now is 09:06 PM.


vBulletin Optimisation provided by vB Optimise (Pro) - vBulletin Mods & Addons Copyright © 2014 DragonByte Technologies Ltd.
Copyright 1999-2014 diyAudio

Content Relevant URLs by vBSEO 3.3.2