• WARNING: Tube/Valve amplifiers use potentially LETHAL HIGH VOLTAGES.
    Building, troubleshooting and testing of these amplifiers should only be
    performed by someone who is thoroughly familiar with
    the safety precautions around high voltages.

Tube LTP CCS?

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
This is a circuit I’m working on and wanted to implement a tube CCS as I have (2) 9 pin sockets
to where I was going to run a couple dual triodes for this setup of something like 6CG7’s, 12BH7’s
and or 12AU7’s depending on how things shake out, which for this driver and LTP leaves me one
triode open so I wanted to see how a tube CCS would work out and kinda like what I see here. :)

The resistors R,17 and R,14 are because I wanted a 190ohm resistor so I had to parallel the 510
and 300 to get basically 189. Don’t worry about the input as I have that toned down since there
is no feedback implemented here yet my gain is quite high so it looks better on the scope. :)

So I’m curious if my tube CCS is implemented correctly and if there is any glaringly obvious errors
in my concoction here. Or if it’s not the traditional way to do it but looks like it would work I’d like
to know that also, as I have no experience with a tube CCS but would like to try one.

I’m curious what the gurus or anyone else that would like to chime in have to say.
(If the circuit is to tough to follow with all my meters I can take them out and reprint) :)
This is done in the free 45day version of Multisim with my modded tube models.

Thanks.
 

Attachments

  • vta_6j5g_pi_6j5g_drv_410v_tube_ccs.gif
    vta_6j5g_pi_6j5g_drv_410v_tube_ccs.gif
    42 KB · Views: 631
Ex-Moderator
Joined 2003
kegger said:
(If the circuit is to tough to follow with all my meters I can take them out and reprint) :)

That's quite an achievement, you've managed to draw a valve circuit that I can't follow! Your second stage is a LTP, but for the life of me I can't work out how the other two valves are configured. A scanned back of an envelope sketch would make life much easier.

You seem to have a 410V positive HT and are using 6J5. 6J5 need plenty of HT, so although you have enough, you can't really afford to waste any by raising the cathodes of the LTP to 100V or so to allow a valve CCS in their tail. If you can provide a negative HT, then an EF184 pentode would make a nice CCS for the LTP. Otherwise, just make a semiconductor cascode CCS (search for DIYaudio CCS board).
 
Allright I'll work on getting the meters out of there and making it easier to follow, it can be a little daunting I'm sure.

But it's basically a 6J5 as a driver with a pair of 6J5's as the LTP with another 6J5 as the CCS for the LTP. (2 6SN7's really)

These are all single 6J5's to spread them out to work on it but it will be 2 twin triodes, a 12bh7 driver for one
of it's halves and the other halve as the CCS for a 6CG7 LTP splitter, at least that's what I'm thinking as of now. :)

I have (2) 9 pin sockets to work with so I wanted to stay to a pair of dual triodes as I have them and they seem
to fit the application, I wanted to try and do it all tubed with no solid state parts and I wanted to try a triode CCS.

I've never seen a setup like this so it made me even more intrigued to try it. (There probably is I just haven't seen it)

Usually you see as you say, a pentode or solid state CCS but I wanted it to be different and fit it in my 2 sockets.

I'm just not sure if I've got it all worked properly or if it will work right, but it sure seems to look right in the sim, I think. :)

Thanks for your reply and I'll try to get an easier to follow schem up shortly.
 
Ex-Moderator
Joined 2003
Although you can make a CCS using a triode like a 6J5, it won't be a terribly good one. See if you can't find another 9 pin socket. Beware that it's only worth going to all this trouble if you can be sure that the load on the LTP will never stray out of Class A.
 
Ex-Moderator
Joined 2004
Although you can make a CCS using a triode like a 6J5, it won't be a terribly good one.
I agree, I wouldn't call it a CCS at all. You need a sharp cutoff pentode with high gm. A pentode makes an excellent CCS and won't blow up without warning, like some SS devices can do.

I'm sure you don't intend to go out of class A, so you're well advised to use a CCS in the tail. It should give almost perfect balance and will make a big difference.
 
Here it is cleaned up.

I took out the driver and just left the LTP and Tube CCS. With the tubes I'm thinking of using, or simular.

If the Sim is any indication I get perfect balance with equal plate resistors and nicely in the plate curves.
 

Attachments

  • ltp_12au7_tubeccs_12bh7.gif
    ltp_12au7_tubeccs_12bh7.gif
    14.8 KB · Views: 538
Well I have see it done both ways now that yu mention it.

With the resistor you get a negative voltage and without the resistor you get
a positive voltage. Which way is better or propper at this point I am not sure.

My thinking was, if you use a "Tail" to ground you get a negative voltage and if
you use a larger resistor creating more negative voltage your "Tails" work better
giving you even voltages, more drive, even plate resistors and whatnot but you can't
go to large as it starts to come apart, so if we apply a negative voltage our tail can be
larger and more effective.

Here is a schem I just found on the net and a link to a very popular 6V6 amp "POINZ" where
in both cases the resistor is used and the negative voltage technique is employed.

http://ekingsbu.users4.50megs.com/audio.html

Again I don't know which way is a better solution. Maybe someone can break them down for us. :D
 

Attachments

  • neg_ccs.gif
    neg_ccs.gif
    12.7 KB · Views: 280
kegger said:
Again I don't know which way is a better solution. Maybe someone can break them down for us. :D

Here is how I did this:

807 Amp Schemo

Here, you can see a 6SL7 being used as a differential phase splitter. Since this is DC coupled to the previous stage, the 6SL7 grids operate at some 55Vdc above ground. That's more than sufficient to allow the use of a SS CCS with no need for a negative DC rail under the tail.

By cascoding the BJTs, we gain the advantages of extra voltage margin, although the MPSA42's used here have a V(CEO) rating that's more than adequate: 300Vdc. Cascoding also improves the CCS impedance by avoiding Early Effect, the parallel capacitance is reduced so that high frequency audio currents aren't as likely to flow around the CCS, thus defeating the purpose.

Triodes aren't constant current devices, and so won't work as a CCS. Pentodes require extra tail voltages, and a negative DC rail to operate them, they don't have nearly the gain as a BJT, and don't offer as high an impedance as a BJT based CCS.

This is one of those cases where silicon gets the job done better than glass.
 
"Triodes aren't constant current devices, and so won't work as a CCS. Pentodes require extra tail voltages, and a negative DC rail to operate them"

I'm curious why you would say this, when a pentode used in this situation is strapped into triode mode isn't it?

Also how are the few I have now seen doing it? I recently found a Dehaan-Heuvelman amp using a 5687 dual triode in this manner.
 
kegger said:
I'm curious why you would say this, when a pentode used in this situation is strapped into triode mode isn't it?

No. The screen is always given a positive voltage (or more accurately less negative voltage than the cathode) of its own so that the pentode CCS is operating as a true pentode.

Also how are the few I have now seen doing it? I recently found a Dehaan-Heuvelman amp using a 5687 dual triode in this manner.

Where was that? Are you sure that this dual triode wasn't wired up as a cascode? Two triodes stacked in a cascode will operate in much the same manner as a pentode, and, therefore, will make a CCS.
 
OK Good at least I got the correct thing going here. :)

I want to try this just to see it work if not anything else, not worried to rip it back apart and do again.

Thanks for all the pointers and cautions guy's.
I'll let yu know if I blow anything up, if it works or how well it may work.
 
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.