gk71 Matrix amp

[Paul Barker]

Ive just read something from a prominent name in diy audio that was a view ofanother miniscule player relative to the author of the comment, which aligns with how I found things 20+ years back, first time I’ve read an opinion that matches mine on that.

“ thorsten loesch is an asshole...”

[Paul Barker]

Now dust has settled: my goodness aren’t GK71’s a bargain? Found a US supplier 35 usd. I don’t need any these pair new ones I bought years ago have never been in a valve base. The bases I bought at time are unused packed together with valves.

How close is Pentode GK71 to Beam 813 ? Well g2 limited to 400v in pentode mode: but it compares favourably to beam 813 at the circuit g2 of 750v, so that’s one parameter to set 400v not 750v, the B+ is fine at 800 and the bias for 100mA won’t need much tweaking, maybe slightly less than -60v. As the curves at 800v are horisontal there’s nothing to be gained pushing dissipation from 80 watts to 125 watts using voltage and holding voltage down to 800 gives power supply benefits. If I raise the current I may also compromise in power supply. So only change is g2 = 400v Im thinking this is because gk71 is a pentode and 813 is a beam valve. The beams probably handle more voltage and more remote.

Another point of interets is the G3 connection, Stephie puts the 813 at filament centre, GK71 curves all show it at +50v, but this can be manipulated for effect. more on thta later.

[IslandPink]

Now dust has settled: my goodness aren’t GK71’s a bargain? Found a US supplier 35 usd.

Unbelievable - I think they were £12 each when I bought a pair.

[Paul Barker]

Probs same for me, I only got two of these and ten gm70’s cant recall their price, perhaps 50. Now graphite cheaper. Copper 100 usd:bargain.

Any who, back to gk71 matrix!

Stephie does so much above my paygrade I have to read and re-read everything, then out pops a litle snipet I never noticed until its almost too l8!

First mistake swerved: Going off schematic ive ordered two 6em8 now I read 6gx7 better at lower levels and overall cleaner also overall distortion against power better for 6gx7. Now ill have to get those and make pinout changes. parameters same. If u build off schem get 6gx7 and relable pin connections.

Nugget of new learning! _ Second mistake swerved: If you look at the negative 100v master supply which everything else references to; you see an unexpected link between white led and vr tube which defines the master voltage for evry voltage level. Sort of indicated on schem as optocoupled. ( Paul thinks ‘erm Stephie have you gone mad?’) so Paul dismisses the led as he doesn’t yet understand it. Later, reading the discussion; “sometimes VR tube firing voltage increases with no or low ambient light. I used a white LED optically coupled to the VR tube to prevent this.” Iam also wondering if theres something else not mentioned in the discussion backing up this led. Theres a flippin light bulb in there. Unless thats used as a resistor and I’ve missed it’s position in the circuit so far. If anyone spots its purpose please enlightne me. There is so much different about this deisgn my dyslexia means its very hard to keep an eye on all the balls some drop.

Ive only just started, theres plenty more hidden in there I haven’t yet found.

[Nick]

A link to the full circuit would help.

[Paul Barker]

sure, theyre seperate links

Signal section

Negative and filament supplies

B+ & g2 supplies

Discussion

[Paul Barker]

To continue findings about adapting to gk71. On final page of gk71 charts is a very handy chart showing anode current and g2 current at varying g3 voltages. All the other charts hold g3 at +50v.

Swerved mistake three while trying to comprehend this chart, confused because the numbers on the chart to add up with the title of the chart which states g3 is at -50v This is a misprint, on the Russian text this is shown as g1 -50v , which now makes the chart right!


Fortunately the 750v b+ and the -50v g1 land us pretty square where we want to be. Stephie shows combined current g2 and anode 100mA for -60v. If we go with b+ 750 g3+50 we can see anode current 90mA g2 current 13mA we’re there. So last page of charts says it all. As on other charts the pentode curves are horizontal at this point, operating at 800v won’t make any difference to current, OTOH operating at 125 watts anode dissipation we could run it at 1388v if we so desired, but the opt’s are the main distortion element and Stephie alreday says 60 watts is their limit. So we’ll target 750 to 850v wherever within thta range is fine, unaffecting naything else, so for me thats great as I can make a different Stephie b+ supply hwich I’ve used before first at very first Eggborough a resistorless capacitorless px25 power supply. I’ll rebuild that for this broad target. Listen folks if youve never built a resistorless capacitorless power supply, truts me its. alot easier with a broad target. Yes its regulated (shunt) so when youve made it fit in the target window it holds it’s place. Trust me this is light years easier to build than one which has to hit 800v on the dot. I just need a break, this voltage isn’t critical so lets not beat ourselves up.

Note to arrive at the bias I have two options for adjustment now 1/ bias supply 2/ g3 level, plenty of knobs to twiddle for effect!

[Nick]

This may help, its the how it works page

https://jacmusic.com/techcorner/SBENCH- ... 60amp.html

[Paul Barker]

Have you noticed how versatile this gk71 is?

Others here have already proved its a solid contender triode connected, may actually sound as good as or better than a gm70, some here will think that and others not. I don’t know, but gk71 as triode is a solid contender.

Now we have realsied we have an exceptionally versatile valve which is a true pentode like Mullard originally made the EL34, and were reasons for this, but probbaly main reason most so called pentodes are beam tubes is the difficulties involved with winding g3. In this valve the reason g2 is held down to 400v is obvious it’s harder to not get arcing between g2 and g3 than to not get arcing g2 to sheets of metal as I imagine they are physically more distant from g2.

But look g2 is 25 watts similar is 801a, so you can run an experiment triode connecting g2 used as anode grounding plate and g3 quiescent point 400v for up to 62mA for 25 watts dissipation. No idea where the chart would be, but someone could just plop that into a 801a circuit and see what it sounds like. Cheper than 801a and you don’t yet know, but it may sound as good (but much cheaper) or better! OK so I’ll do a spin off one day just for curiosity.

[Paul Barker]

This may help, its the how it works page

https://jacmusic.com/techcorner/SBENCH- ... 60amp.html

Thanks.

[Nick]

but probbaly main reason most so called pentodes are beam tubes

I thought it was because Phillips had the patent

[Paul Barker]

I read somewhere the Mullard EL34 special because of the quality of the g3 winding.

[pre65]

but probably main reason most so called pentodes are beam tubes

I thought it was because Phillips had the patent

That is as I understood things.

Also that EL34 and EL84 were designed specifically as audio valves.

[Paul Barker]

You done good

[Paul Barker]

But look g2 is 25 watts similar is 801a, so you can run an experiment triode connecting g2 used as anode grounding plate and g3 quiescent point 400v for up to 62mA for 25 watts dissipation. No idea where the chart would be, but someone could just plop that into a 801a circuit and see what it sounds like. Cheper than 801a and you don’t yet know, but it may sound as good (but much cheaper) or better! OK so I’ll do a spin off one day just for curiosity.

Which is a true triode