GK-71 - economical build.

pre65 · Post by **pre65** » Thu Jul 04, 2013 7:16 pm

On my GM70 the filament supply was 18V 120va toroid, cheap bridge, smoothing caps and dropper resistors (1R or 2R from memory). Mine were monoblocks.

I did try a 10mH 5amp choke but did not seem to make any audible difference.

JamesD · Post by **JamesD** » Thu Jul 04, 2013 7:39 pm

Paul,

I take your point but part of this was cost driven too - or at least andrews was. If you just use a laptop supply, then a 20V 6A supply on ebay is about £13 upwards... I tried to make it more sophisticated and ended up with the design I showed - and now I think its probably too elaborate... I guess I start with the cheap laptop supply and then add the bits I have in the bin and see if they do improve it or not...

Phil,

It does seem that the GM70 and GK71 are pretty immune to the quality of the filament supply and that a simple one will work well! Thank you for your account its helping me decide to go back to simple implementation a la Andrew I...

Steve must be laughing his socks off as his KT120 is doing everything this promises to do and for similiar outlay and simpler implementation... except for old british mesh plates!

James

Paul Barker · Post by **Paul Barker** » Thu Jul 04, 2013 7:47 pm

Yes Andrew I's sounded fine and Phil and my 833a's on bare smps filament supplies sound great.

Not to say that they can't be improved. Just to say they needn't be improved to make great music. Sure; everything can sound better, but not everything has to sound better.

JamesD · Post by **JamesD** » Thu Jul 04, 2013 8:24 pm

I hear you brother!

and i think you are right...

So back to a bare cheap SMPS for now and play with the naughty bits later

IslandPink · Post by **IslandPink** » Thu Jul 04, 2013 8:37 pm

I'm interested in the noise-cancellation methods, I've not read up on much of that, but I remember some things by Thorsten and Broskie , no doubt pulled out of the early archives when caps were small etc.
I guess some of them assume you can use ripple from the B+, which is hard if you have a shunt-reg power supply

What other methods were there ?

Worth trialing anyway - I guess you can use a 6A CMC between the toroid and the filament to knock of some HF noise too .

This could be a cheap option too, done properly .

JamesD · Post by **JamesD** » Thu Jul 04, 2013 9:11 pm

I had a crazy idea...

Filament bias!!!

72V bias at 3A filament current only 24 ohms!!!

and 216Watts - Ooooooopppppppsssssssssssssssssssssssss!

Oh well I said it was a crazy idea...

JamesD · Post by **JamesD** » Fri Jul 05, 2013 4:00 pm

Revised circuit diagram with simplified filament supply and a couple of values added.

J

Paul Barker · Post by **Paul Barker** » Tue Jul 09, 2013 7:32 am

We have been comparing anode structure to gm70 in another place, which inspired me to add the 813.

this reminded me of the gritty sound of the 813 triode connected compared to the 211. There is a vast difference.

I am wandering whether the GK71 will also lack finesse against the GM70. Just conjecture AFAIK non if us has tried the GK71 triode connected yet.

But if it does have the same issue as the 813 have you set plans in your mind to pentode connect it with either the anode resistor feedback or output transformer feedback?

Paul Barker · Post by **Paul Barker** » Tue Jul 09, 2013 7:45 am

Just found this impression I had in 2002 of the 813 connected low power with the screen as the anode. Most of the rest I wrote at that time was a little over enthusiastic. With the wisdom of hindsite the 813 doesn;t really make a great triode compared to what I have experienced since those days.

"Using screen as plate it has more of a soft edge, but not over soft, like drums and things are still tight but introduced and finished with less firmness."

That was at a time when the 813 was my first large valve experience. The 211 completely displaced it.

JamesD · Post by **JamesD** » Tue Jul 09, 2013 7:54 am

I know that the 813 is compared to the GK-71 and vice versa but the triode curves are different. The 813 slots in between the GM-70 and the GK-71 - and pretty much exactly half way between them. The pentode curves are similarly different so although they would both work in each others circuits they would bias to different operating points and hence different performance from each...

They still might sound similar but the chances are that they will sound different...

We will have to wait and see

pre65 · Post by **pre65** » Sun Aug 18, 2013 10:49 am

Do GK-71 have getters ?

I ask because there were some advertised as "have no getters".

Paul Barker · Post by **Paul Barker** » Sun Aug 18, 2013 1:31 pm

I can't easily look at my pair today but will look when I can.

IslandPink · Post by **IslandPink** » Sun Aug 18, 2013 5:51 pm

They appear to have getter flash around the lower end of the bulb, just where it's waisting-down into the base.

andrew Ivimey · Post by **andrew Ivimey** » Sun Aug 18, 2013 9:19 pm

I tried GK against GM a year or so ago here if anyone cares to look backwards; I re-posted photos recently. I concluded that GM sounded better. GK was not 'gritty' as some have found 813; it just didn't supply as much of a good sound as GM. I was driving with a 12GN7 at, if I remember, 20ma ... may have been more-> 200H choke.

The 813SE amp I sold Ali did not sound at all gritty - Marco fell in love with it instantly and had to have his fingers surgically removed from the amp (this is not necessarily an endorsement you understand) that saturday night in Owston after we returned from the bar totally pissed. Did we then engage on the whole of Live Dead before bed?

When I get round to it I'll rig up the doddingtonGK new circuit and I will expect improvements - just haven't got the time right now - this is a pity.

IslandPink · Post by **IslandPink** » Mon Aug 19, 2013 6:25 pm

Is this the kind of operating point you were using, Andrew ?

I like the little speech-bubbles !

audio-talk

GK-71 - economical build.

#31

#32

#33

#34

#35

#36

#37

#38

#39

#40

#41

#42

#43

#44

#45