The Tube Swapping Thread

I must admit this is turning into a very interesting topic for me. As an engineer I have been interested in how H&K have set up the GM since I bought it, but of course they are very close mouthed about it and opening my own GM up to explore isn't high on my list at the moment. I think your post has raised a very useful issue Namklak. So much so I'm going to open a new thread with this question as its point. This one can remain for its original valve swapping thoughts and I'll post my own answer and open it up to schematic thoughts elsewhere.

We shouldn't talk about that here, it will hijack the thread from valve changes.

How about posting this info on the thread I've just started to do with the design of the amp in general?

Same here Raf as I've said many times.  I'll be VERY interested to hear your impressions of the 5751 swap.  You've chosen the right way to go with that alternative.

There was a suggestion a while back that the GM may have an integral solid state Tube Screamer type of overdrive setup before the valves which is used as an integral part of the voicings.  This has not been confirmed or refuted in any way as far as I am aware.  If that were to be the case then reducing the gain of the valve stages could create a significant shift in the balance of the distortion creating mechanisms.  There would be less valve overdrive available before the SS clipping kicked in which may make for a very different sound through the onset of distortion.  It may not be an unpleasant thing, (or it might be!)  It may also be that that is effectively the "Boost" option, an additional solid state circuit which only engages with the Boost button, effectively giving us a boost/overdrive/distortion pedal setup within the amp. That's a very clever approach if it is true. It's effective, exactly like we do already, and hence it doesn't break the "amp must be all valve or it sounds crap" rule! It is all valve but it has a standard addition before it.

Anyone else have any knowledge of this area, or any ideas as to whether it makes a difference?

That would be great Jeff.  Tread lightly so as not to worry them! Bear in mind that the H&K team are a bit mystified by anyone's desire to modify their baby in this way.  When I originally spoke to them about valve rolling and its implications in the GM36 circuit they were very kind in confirming that the usual 12AT7 and 5751 swaps would not cause any undue distress to the amp but, as I said, they were puzzled as to why we would want to.  They had voiced it to perfection!

I find I am increasingly turning to higher power soak/lower gain sounds to get near to where I want to be and this will be another variable to help find my own sweet spot.  (Or maybe just cloud the issue even more.)  I think it may have been yourself who suggested this as another approach to try out.  It is certainly opening up for me.  And if it turns out to have a SS preamp boost/distortion stage, lower input signal and gain should take away some of the effect of that.

Well my 5751s appeared at the door today.  I opened up the GM and had a look and, what do you know, I had my NOS Mullard 12AT7 still in place as V1, and a Fender 12AX7 in V2.  I don't remember doing that....

Anyway, I swapped out V1 for a 5751 and turned on.  It's instant.  I loved it.  So I went the whole hog and swapped out both V1 and V2.  I loved it even more.  I will be using this setup for a while now in order to burn them in and see where it goes but I can't seem me going back.

Everyone here will be bored to Hell with my bleating on about the 12AT7 being a s**te choice as a substitute for a 12AX7.  They are so far apart in many things that really matter it's untrue.  The average Joe can't see that, while it seems to be a "good thing" to drop your "gain" (mu) from 100 to 60, the problems caused by other parameters changing is more profound.  If you haven't already seen it, have a look at page 4 of this thread.  At the top I have posted a simple comparison of the 3 valves.  The parameter Plate or Anode Resistance, (Ra or Ri depending on where you live), has a tremendous effect on the sound.  To change this by a factor of 5-6 times without knowing the consequences is stupid.

The output of an amplifying valve stage has an output capacitor and this sets the low frequency breakpoint for that stage along with the overall anode resistance.  If it has a 100k resistor in the circuit and a Ra of 80k (12AX7), then the valve is really seeing those two in parallel (100k//80k) as its load, that's 44k.  The capacitor will be chosen to give a cutoff not too far below the bottom E string 82Hz, say 40Hz for easy calculation, (it isn't a bass amp and too much low end extension sounds bad for guitar).  Changing the 80k for 15k (12AT7) means it now sees 12.6k.  That means the low frequency cutoff for that stage has just been raised to almost 4 times its original frequency, about 139Hz!  That's potentially much more of an effect than lowering the gain a tad.  With a 5751 the difference is 34k and 51Hz.  Much closer to the original so the tonal characteristic will remain much as it was while the gain changes.

So to the differences.  Well the gain has obviously changed, 100x100=10000, 70x70=4900, not that simple in real life but we can guess it will be pretty much halved.  The Gain control now spreads its effectiveness over a much greater travel.  I love it!  It means I can now more easily control those voicings where I want to use say the Lead channel with a cleaner sound.  The Crunch channel has become more sensitive and easier to control too and the sweet spot is easier to find.

The bottom end has changed and filled out.  That damned 12AT7 must have been thinning things out way more than I realised.  My latest speaker mod of my Mitchell foam rings, (reported here on this forum), is now almost too much over the top.  If I can't learn to compensate to live with it I will have to go back and remove some of that.  I use a 1/2" and a 1/4" thick pair for each speaker so there is plenty to play with.

The tone controls at both ends, bass and treble, have kicked in a little more, always a good thing with the GM.  I wasn't expecting the top end to change but there seems a little more sparkle to it without any added harshness hence the treble perhaps has more to work with.

Another unexpected one, even though the gain may have gone down, pinch type harmonics have now appeared on higher but not extreme gain settings.  They are not my finest feat I'm afraid.  I'm no metal player so have never really looked for them in my playing but now I can get say ZZTop type sounds more easily.

It's definitely a success in my opinion and blows away the 12AT7 substitution, NOS Mullard or not.  That valve was recommended by an independent valve expert as the one which would offer the least thinning out of sound, so God knows what you guys using any old 12AT7 are hearing!  It's much more a case of getting the benefits I wanted from reducing the gain while keeping the best characteristics of the amp with its design 12AX7 bottles.  And don't be frightened to swap out both if you have the chance.  I can't speak for the hardcore metal freaks who want so much rasp they can't identify the pitch of the notes   , (see what I did there - rasp?), but for all else you do not end up with too little gain to develop a full range of overdrive/distortion tones and feel.

Yes both V1 and V2 are now 5751. My repertoire is mainly classic rock from when rock was really classic! Very little from new bands after about '80. So the amount of overdrive I feel I need is not the same as a lot of more modern stuff. Even with the reduced gain two 5751s gives me, I can't imagine anyone except the most flesh shredding extreme black death doom speed thrash metal exponent would find something missing.