Tube DIY Asylum

Do It Yourself (DIY) paradise for tube and SET project builders.

Return to Tube DIY Asylum


Message Sort: Post Order or Asylum Reverse Threaded

differential input, single ended output

98.250.33.227

Posted on December 8, 2015 at 17:17:15
rage
Audiophile

Posts: 494
Joined: December 17, 2010
is there a way to do a differential input and drive a single ended output, without using a transformer?

I'd like to use a source that is balanced into a differential pair then drive my SE 2a3.. preferably without coupling caps.

tube would most likely be 12at7.

any suggestions?

 

Hide full thread outline!
    ...
RE: differential input, single ended output, posted on December 8, 2015 at 17:26:53
JKT
Audiophile

Posts: 454
Location: Midwest
Joined: November 26, 2002
Why not use an input transformer at the input to your SE 2A3. I did that with my Sun SV-2A3.


"It is better to remain silent and thought a fool, then speak and remove all doubt." A. Lincoln

 

RE: differential input, single ended output, posted on December 8, 2015 at 17:33:12
Eli Duttman
Audiophile

Posts: 9355
Location: Monroe Township, NJ
Joined: March 31, 2000



Piece of cake! Use differential gain block, AKA LTP. You even get the option to experiment with "absolute polarity". ;>)

Each channel gets a twin triode. Don't forget the tail CCS.

The line stage schematic I've uploaded shows the kind of circuitry needed. You would drive, instead of grounding, the non-inverting triode's grid.

Eli D.

 

RE: differential input, single ended output, posted on December 8, 2015 at 17:37:00
rage
Audiophile

Posts: 494
Joined: December 17, 2010
jkt - what transformer are you using?

eli - that looks really good. what happens if I remove the coupling caps and grid resistor? Right now I'm using 12ax7 direct coupled to 2a3..so B+ is 500 and I've got around 250v on the cathode of the 2a3.

 

RE: differential input, single ended output, posted on December 8, 2015 at 18:34:45
Eli Duttman
Audiophile

Posts: 9355
Location: Monroe Township, NJ
Joined: March 31, 2000
You certainly can DC couple the 2A3's grid to a LTP plate. Notice in the preamp that the MOSFET source follower is DC coupled to the non-inverting plate.

That 500 V. of B+ makes for healthy sized LTP plate load resistors, which is good for keeping gain up and distortion down. 200 to 210 V. on the 'T7 plates and 3 mA. of plate current per plate sounds pretty darned good. A 6 mA. tail CCS is what you need.

Eli D.

 

That's not a differential input, posted on December 8, 2015 at 19:11:06
Triode_Kingdom
Audiophile

Posts: 5768
Location: Texas
Joined: September 24, 2006
If the purpose is to amplify a differential input that isn't referenced to ground, he needs a true differential amplifier. Why not use a version of the usual circuit (below) and take single ended output from only one of the two tubes?






--------------------------
Buy Chinese. Bury freedom.

 

RE: That's not a differential input, posted on December 8, 2015 at 21:06:29
Eli Duttman
Audiophile

Posts: 9355
Location: Monroe Township, NJ
Joined: March 31, 2000
Read carefully. I said he had to drive, not ground, the non-inverting grid.

BTW, regardless of the amplifying device used, a tail CCS is in order.

Eli D.

 

RE: That's not a differential input, posted on December 8, 2015 at 22:45:55
Triode_Kingdom
Audiophile

Posts: 5768
Location: Texas
Joined: September 24, 2006
Yep, sorry, I missed that. I will go on record though as being particularly uncomfortable (I'm being diplomatic here) with all those SS devices you toss into vacuum tube circuits. You probably already know that. :)


--------------------------
Buy Chinese. Bury freedom.

 

RE: That's not a differential input, posted on December 9, 2015 at 08:56:53
Stuben
Audiophile

Posts: 513
Location: Guber Ohio
Joined: December 30, 2005
Come on TK, Try it...your ears will like it ;>)

 

RE: That's not a differential input, posted on December 9, 2015 at 12:13:03
Eli Duttman
Audiophile

Posts: 9355
Location: Monroe Township, NJ
Joined: March 31, 2000
You've never seen me use a SS voltage amplifier. As of now, the only place I'd be comfortable with SS voltage amplification is in Allen Wright's hybrid cascode LOMC phono setup.

BTW, I do own a Klyne model 1 LOMC "head" amp, which is 100% SS, but that unit has a well proven track record. Some folks can make SS "dance", Stan Klyne and Nelson Pass to name 2.

FETs, being high impedance, voltage controlled, devices, mate well with tubes. FETs make excellent current amplifiers (voltage followers). While it is an over simplification, think of a FET as a heaterless pentode. :>D

I'm a pragmatist. If a lump of galena yielded good sound (it definitely does not), I'd use it.

Eli D.

 

There is a simpler way to do this., posted on December 9, 2015 at 13:20:29
Ralph
Manufacturer

Posts: 3164
Location: Minnesota
Joined: April 24, 2002
Contributor
  Since:
April 1, 2002
You can can simply use a single-ended circuit as the input voltage amplifier.

Since you are are not going to be running loop feedback, you could drive the cathode of the tube in addition to driving its grid.

Its not balanced (the cathode input being fairly low impedance), but it is differential in every aspect of the word. So it will accept and process a balanced input.

In order to drive the cathode, you would use an electrolytic capacitor of exactly the value one would use for cathode bypass. You would tie it to pin 3 of the XLR, and the grid to pin 2 of the XLR. If you run single-ended, you would put a jumper between pin 3 and pin 1 of the XLR. Or just let it hang.

Don't laugh till you try it- this was how the original opamps designed by George Philbrick worked way back in the late 1940s. I've been using the technique for years.

 

RE: That's not a differential input, posted on December 10, 2015 at 09:39:10
JKT
Audiophile

Posts: 454
Location: Midwest
Joined: November 26, 2002
I have a lot of experience using CCSs with LTPs. Cascode MOSFETs are way better than single MOSFET based CCS. IXYS 8N50D2/1N100D better sounding than DN2540s. But for currents less than 10mA a cascode BJT (BC549C for example) out performs any of the MOSFETs.

I guess if you HAD to use tubes a pentode based CCS would work well. That would be quite the pain in the ass though to implement. However, looking around this forum I do see many members of the "Hair Shirt" DIY audio bent. ;~)


"It is better to remain silent and thought a fool, then speak and remove all doubt." A. Lincoln

 

RE: That's not a differential input, posted on December 10, 2015 at 18:44:56
rage
Audiophile

Posts: 494
Joined: December 17, 2010
TK,

If I do that will still have the benefits of being balanced? (common mode rejection?

 

RE: That's not a differential input, posted on December 13, 2015 at 14:24:30
dtut
Audiophile

Posts: 111
Location: central coast california
Joined: March 27, 2006
I read that as "heartless pentode" and thought, damn, that's harsh.


Doug

 

RE: That's not a differential input, posted on December 13, 2015 at 15:22:51
Eli Duttman
Audiophile

Posts: 9355
Location: Monroe Township, NJ
Joined: March 31, 2000
Somebody's "sand allergy" got the better of them. FETs make excellent "spear carriers" for tubes. Pick your spots carefully.

Eli D.

 

RE: That's not a differential input, posted on December 20, 2015 at 17:24:15
rage
Audiophile

Posts: 494
Joined: December 17, 2010
I'm over my sand allergy!

I am loving Cree schottky diodes and constant current sink cascodes with mosfets.....

 

RE: That's not a differential input, posted on December 21, 2015 at 00:46:38
beautox
Manufacturer

Posts: 297
Location: Auckland
Joined: July 9, 2013
Try the unbalancer circuit from John Broskie

http://www.tubecad.com/2011/03/blog0203.htm

 

RE: differential input, single ended output, posted on January 10, 2017 at 17:08:47
rage
Audiophile

Posts: 494
Joined: December 17, 2010
I am still kicking this idea around doing it how TK had mentioned... a differential pair from a 12a7 and just taking the signal from one of the triodes for the output tube.

my questions are:

- will the second harmonic generated in the LTP be canceled out like in a push pull amp?

- if I use a hammond 156c as choke in the tail (instead of a CCS) - should that go to a negative supply or to ground?

 

Page processed in 0.034 seconds.