tube preamp to mate with Gainclone

[Joe Rasmussen]

TUBED GAINCLONE

Hi Guys

For those who may be interested here is a...

A TUBED version Gainclone

An externally hosted image should be here but it was not working when we last tested it.

Copyright Joe Rasmussen 2003 - NOT PUBLIC DOMAIN, FOR NON-COMMERCIAL USE ONLY


See thread:

http://www.diyaudio.com/forums/showthread.php?s=&threadid=9557

Actual Circuits, Power Supply etc, lots more. Read posting from:

http://www.diyaudio.com/forums/showthread.php?s=&postid=164935#post164934


Joe Rasmusssen

 

[pedroskova]

I like the dominant second harmonic. I just might have to try this guy out. Any other tubes that work well @ 35V?

 

[Joe Rasmussen]

I like the dominant second harmonic. I just might have to try this guy out. Any other tubes that work well @ 35V?

8 Watts into 8 Ohm (8V RMS). No other easily and so available tube I know off works reliably at such low voltage.

Next installment "PART FOUR" will be posted Monday or Tuesday and will include more measurements/graphs at different power level, did a range from 60mW to 8 Watts and in all of them EVEN order harmonics dominate - a really GOOD sign. Even better, an almost total absence of ODD order harmonics higher than 3rd, and the 3rd never dominate either. So in all, maybe a good pointer to explain, if not totally, why it sounds not so bad!

For the commercial version of this design, have a look at:

http://members.ozemail.com.au/~joeras

Joe

 

[Joel]

This thread does not belong in the "chip amps" forum.

 

[Peter Daniel]

This thread does not belong in the "chip amps" forum.

What makes you think so?

 

[wormturns]

sorta' on topic

In my case I have a tube preamp with cap coupled output and reasonably low output impedance(~1.5k).

I'd like to build an inverted GC without the input pot and cap. Others have pointed out that with the inverted GC the interconnects will be part of the feedback circuit.

Is there any way to modify the input of the GC so the feedback circuit is short and the interconnects are out of the feedback?

Thanks,

Scott

 

[planet10]

Re: sorta' on topic

In my case I have a tube preamp with cap coupled output and reasonably low output impedance(~1.5k).

I'd like to build an inverted GC without the input pot and cap. Others have pointed out that with the inverted GC the interconnects will be part of the feedback circuit.

Is there any way to modify the input of the GC so the feedback circuit is short and the interconnects are out of the feedback?

You might get some info over here on this thread.

The GC is small enuff that maybe you could mount it in the Pre with an external PSU?

dave

 

[ThorstenL]

Re: sorta' on topic

Koinichiwa,

Others have pointed out that with the inverted GC the interconnects will be part of the feedback circuit.

This is a matter of interpretation, IF I may say so. I would take the opposite position, pointing out that the input is outside the feedback loop and in fact simply a straightforward signalpath, which injects a current into a "virtual ground".

The feedback loop is from the output to the inverting input and nothing else, IN CASE of the inverting gainclone. Anyone who has ever designed a summing bus Amplifier for Mixing Desks which take the signal straight of the fader will know the "virtual ground" operation.

The interconnects will allways be in the signalpath, the inverting circuit changes nothing on that..

Is there any way to modify the input of the GC so the feedback circuit is short and the interconnects are out of the feedback?

See above. Some people see a feedback path where is a feed forward ONLY (aka signal) Path, others do not. The revised schematic with the added 22k Resistor from the negative input to ground should address all remaining potential stability issues comprehensively, assuming the speaker cable is low capacitance.

Sayonara

 

[wormturns]

Re: sorta' on topic

Greetings,

Thanks for the replies.
I'm not sure I've got this "quote-/quote" thing figured out yet.

You might get some info over here on this thread.

Yes, it was reading Joe Rasmussen discussing his JTLI that I considered that a pole/node(?) on the input might help isolate the upstream circuit from the feedback.

The GC is small enuff that maybe you could mount it in the Pre with an external PSU?

I've considered the hybrid thing; but want something simple to try first, to see if I will like the sound enough to keep it in the system. Likely do the monoblock thing.

Some people see a feedback path where is a feed forward ONLY (aka signal) Path, others do not. The revised schematic with the added 22k Resistor from the negative input to ground should address all remaining potential stability issues comprehensively, assuming the speaker cable is low capacitance.

I'll give that a try. I breadboarded something similar some time back. Unity gain(?) with 100k, 10k, 100k-NFB and my notes indicate I wasn't impressed with it-sounded thin and over did the bass. I'll try it with the values in the schemo.

Thanks,

Scott

 

[JAZZ2250]

What do you think about this schematic? I have no idea how low the output impedance will be. But, according to the schematic, it provides "low impedance". Sorry for my ignorance. I'm still reading a tube related book. Very slowly.

 

[miguel2]

In this schematic by Joe Rasmusssen there are a couple of changes in the chip circuit relative to Kuei's version. I believe, for quick test purposes, that the tube can be implemented in this inverted version proposed by Kuei. In this case we just have to add the tube. But the gain of the tube may overload the chip so some resistor in between may be necessary. Also, I have +/-28V rails, maybe I can get away with it. The only problem is that I have a problem with catode followers

Miguel

 

[planet10]

But the gain of the tube may overload the chip so some resistor in between may be necessary.

The gain in Joe's tube stage would be just under 1 so no need to pad it down.

dave

 

[miguel2]

Thanks Dave.

I just noticed a 1n3 cap in the chip input. This is something to worry about? Does the tube input can cause some oscillations in the chip?

Miguel

 

[Joe Rasmussen]

What do you think about this schematic? I have no idea how low the output impedance will be. But, according to the schematic, it provides "low impedance". Sorry for my ignorance. I'm still reading a tube related book. Very slowly.

Doesn't seem anyone has responded to this, so maybe I can be of some help.

It's a mu-follower, a variation of SRPP (Single Rail Push-Pull) and as such is a PP circuit (some will disagree with that).

Download this PDF file from TubeCad (allow it a little time if you have a modem connection), it explains both of them in depth, first SRPP, then goes onto the mu-follower variation:

http://www.tubecad.com/articles_2002/SRPP_Deconstructed/SRPP_Deconstructed.pdf

The output Z is not that low, we did a 6922 SRPP and got, from memory, 1600 Ohm.

To be honest, this is probably my least favourite tube circuit!!! It can sound nice but rarely excites, it is prone to vagueness. BUT many love it, maybe you will, after all this is only an opinion. Too me it is almost too tubey, if you know what I mean. I like precision too.

This leaves me with an opportunity to air this hooby-horse of mine. Why is it that I've come across so many 2A3 & 300B SE amps, and yet they use SRPP input/gain stages. Not really Single-Ended IF you use a PP front-end.

Joe

 

[Joe Rasmussen]

Thanks Dave.

I just noticed a 1n3 cap in the chip input. This is something to worry about? Does the tube input can cause some oscillations in the chip?

Miguel

In short... no... and no.

The end product of this is to shape the HF bandwidth and its value is tunable by ear, so once built do experiment.

It is not unusual for SS designers to put a passive bandpass filter on the input stage, to block possible DC and, to me more importantly, reduce potential slew induced distortion. In this case, I believe you will find the end result a warmer sound.

Joe