diyAudio

diyAudio (http://www.diyaudio.com/forums/)
-   Tubes / Valves (http://www.diyaudio.com/forums/tubes-valves/)
-   -   A2 positive bias push pull possible? (http://www.diyaudio.com/forums/tubes-valves/207528-a2-positive-bias-push-pull-possible.html)

wicked1 25th February 2012 04:08 PM

A2 positive bias push pull possible?
 
I'm thinking about using some small transmitting tubes for an amp. Local electronics surplus had 3c24's for very inexpensive, so I grabbed a bunch.

I thought I wanted to run them PP, and started to draw up some loadlines last night.. then I decided to look around the web for ANY info having to do w/ positive bias and PP, and couldn't find much.. Found an old post here or on another forum that said it wasn't worth trying due to something w/ the grid current, and that's why you usually see single ended transmitting tube amps. Then I went to bed......

SO.. Before I go any further.. Is it possible? (I'd be using mosfet followers before each 3c24) It's looking to me like it's really not much different from plotting the loadlines for a class A amp.. and actually I was thinking about AB2, which in this case w/ an always positive biased amp, I think seems a lot like plotting an AB loadline.

I know the path I'm going down w/ these particular tubes is illogical due to filament requirements and low efficiency and all that... But is it also close to impossible for another reason I'm unaware of?

Miles Prower 25th February 2012 07:31 PM

Quote:

Originally Posted by wicked1 (Post 2922123)
I'm thinking about using some small transmitting tubes for an amp. Local electronics surplus had 3c24's for very inexpensive, so I grabbed a bunch.

I thought I wanted to run them PP, and started to draw up some loadlines last night.. then I decided to look around the web for ANY info having to do w/ positive bias and PP, and couldn't find much.. Found an old post here or on another forum that said it wasn't worth trying due to something w/ the grid current, and that's why you usually see single ended transmitting tube amps. Then I went to bed......

SO.. Before I go any further.. Is it possible? (I'd be using mosfet followers before each 3c24) It's looking to me like it's really not much different from plotting the loadlines for a class A amp.. and actually I was thinking about AB2, which in this case w/ an always positive biased amp, I think seems a lot like plotting an AB loadline.

I know the path I'm going down w/ these particular tubes is illogical due to filament requirements and low efficiency and all that... But is it also close to impossible for another reason I'm unaware of?

It's quite possible, and done a lot.

"Found an old post here or on another forum that said it wasn't worth trying due to something w/ the grid current..."

This applied back in "the day" as it was quite difficult to go into Class *2 operation with good sonic performance, so it was mostly regelated to applications such as AM plate modulators or PAs where fidelity wasn't necessary or impossible, and where the main concern was lotsawatts and efficiency.

These days, MOSFET source followers makes Class *2 operation with sonic excellence very possible. A power MOSFET has a much lower rd(on) than any hollow state device, so grid current sourcing isn't a problem, and the source follower has a much lower Zo as well, and that also helps improve sonic performance.

There's really no big mystery here. Draw up the PP loadline, determine your Q-Point, and you're done. NBD.

tubelab.com 25th February 2012 10:16 PM

Quote:

SO.. Before I go any further.. Is it possible?
YES. I have done 45's, 300B's, 307A's, 6L6GC's, KT88's, and a dozen or so others in P-P A2 or AB2. Soon there will be some 4D32's kicking it in AB2. I won't be happy until I see 200 watts from a pair.

The old breadboard seen in post# 565 in the 6L6GC in AB2 thread went into the trash can Thursday night. The new one is BIGGER and better. It holds green driver boards, and red driver boards, and........

This thread details the development of an AB2 amp using 6L6GC's. I have run all the tubes listed above in it. It's worth a read if you have lots of time.

http://www.diyaudio.com/forums/tubes...ight=6L6GC+AB2

wicked1 26th February 2012 04:07 PM

Thanks guys! I have been reading that 6l6 thread. Still more to go....
I've seen plenty of amps biased into ab2, but still starting below 0. Just wanted to be sure there wasn't anything over my head about starting above 0... Guess not!

I might post the loadline I come up w/ later, so you can tell me if I'm on the right track.
Thanks for answering all my simple questions.. It's my first time I'm trying this w/out a preexisting schematic to start with, or 100 other designs for the same tube to steal ideas from.

wicked1 26th February 2012 07:04 PM

2 Attachment(s)
Ok, so I do have some questions about my loadlines.. I was using edcor tx choices as a starting point (simply to have a starting point). One graph uses 400v B+ @ 60ma, 4k trafo. The other uses 350b+ @75ma, 2.5k trafo.
I'm planning on paralleling tubes, so I double the TX value for my calcs. (same result as doubling/re-labeling the plate current on the chart)
So, my class B loadline for chart 1 is 400/2000, and I draw the line from 400v to 200ma. Class A would be 400/4000. I draw a line from 400v to 100ma, and shift it up to my Q point.

Is all of that correct?

Then, I use some calcs in a post by by Miles P about loadlines and figuring out class A power vs class B power.
The intersection of the Class A and Class B loadlines is the transition point. Read the volatge there as: 180V. The Class A power becomes:

Vp= 400 – 180 = 220Vp
Vrms= Vp / sqrt(2)= 220 / sqrt(2)= 155.5Vrms

Po= V^2 / R= (155.5^2) / 2000= 12.1W per VT (pair). Since two VT(pairs) contribute to this, the total Class A power is: 24.2W.

I get about that for the 350v 2.5k tx.

350V -200v = 150v =106.8Vrms
106.8^2 / 1750 = 6.4w per parallel pair, so 12.8 class A watts.

If all of that is correct, one more thing confuses me.. Does this tube cut off at 0 grid v, or at the bottom of the chart? I've read a few posts that say these tubes cut off “well before 0v on the grid” (they were running 300vB+, but still, the chart shows some negative grid voltage there). If that's the case, it looks like I switch in to class B a lot earlier than I've been accounting for above.

Well... Am I on the right track?

--forgive the fat lines and rough placement on the charts.. I'm just trying to see if I've got the correct idea, at this point.
red=25W
Blue= class A at my Q
Purple= class B


All times are GMT. The time now is 04:02 AM.


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