GM70 PP

What people are working on at the moment
User avatar
IslandPink
Amstrad Tower of Power
Posts: 10041
Joined: Tue May 29, 2007 7:01 pm
Location: Denbigh, N.Wales

#46 Re: GM70 PP

Post by IslandPink »

Nice work ! Glad you like them .
"Once you find out ... the Circumstances ; then you can go out"
User avatar
izzy wizzy
Old Hand
Posts: 1496
Joined: Fri Nov 02, 2007 7:02 pm
Location: Auckland NZ
Contact:

#47 Re: GM70 PP

Post by izzy wizzy »

IslandPink wrote: Sun Aug 04, 2019 8:11 pm Nice work ! Glad you like them .
Thanks. Would have been a bit annoying if I didn't having got 8 of them 😜

I got stereo GM70 and it's sounding very nice.

The fixed bias circuit is very basic so wondering if there's much to improve there and what to do about it. Any ideas welcome.

Cheers,
Stephen
User avatar
izzy wizzy
Old Hand
Posts: 1496
Joined: Fri Nov 02, 2007 7:02 pm
Location: Auckland NZ
Contact:

#48 Re: GM70 PP

Post by izzy wizzy »

Time to consolidate and see where I've been and where to go and maybe answer some questions I asked. I could ramble a bit :roll:

There's not many PP GM70s about on the web so it's hard to get ideas. I've only seen one transformer coupled example but the previous stages to the finals stage were single ended. Most amps using transmitter valves are SETs and that goes for the DIY community as well. It would seem PP GM70s is not a popular choice but then neither is PP845, 211, 813 etc and I can see why. SETs are where a significant amount of people feel that architecture floats their boat for music. I'm not anti SET but it's not an architecture for me. I was/am swayed significantly by Lynn Olson's writings on the subject and also the heritage of these transformer coupled circuits.

I like what the circuit looks like too. Another reason is I don't like capacitors and I resent having to spend significant money on good ones. I'd rather spend that money on some iron and ditch the caps. Plus caps are such a lottery as to what is going to sound good and in what position. When the output valve cost serious money, adding another is a big deal and caps are not. If the valve is a GM70 (or was 813 a while ago) then caps cost more than the extra for PP architecture so not as big a deal and makes other things simpler i.e. the power supply.

I've come at this amplifier in a kind of backwards way. It's here and now I have to figure out from a technical point of view how and where to now. I've had to learn about load lines, biasing, gain, power and so on which at best have been suck it and see to date.

In my search for more info on PP load lines and biasing, one thing kept coming up that I hadn't considered completely; grid current and A2. Grid current I thought I had considered as in, keep the grid resistance under a certain value so that grid current doesn't cause the valve to run away. Grid current being a feature of most valves at some point especially if they are being pushed near the extremes. It lowers the bias, causing more anode current to pass, causing more grid current and so on. I have in the man cave an example of a PL519 where the glass has been sucked in by the vacuum it got so hot when this happened.

Then there's A2. Some transmitter valves only run in A2 which is the operating area where grid current flows as part of normal operation. Under these conditions the driver has to provide or sink the current. In my case, the driver is isolated by the transformer from this other than the impedance change that occurs during A2 but the grid circuit of the GM70 and therefore the bias supply isn't. The bias supply has to deal with this grid current situation and there seem to be a number of solutions.
User avatar
Nick
Site Admin
Posts: 15707
Joined: Sun May 06, 2007 10:20 am
Location: West Yorkshire

#49 Re: GM70 PP

Post by Nick »

The bias supply has to deal with this grid current situation and there seem to be a number of solutions.
A shunt regulator seems to be the simplest IMHO. Couple of VR tubes at its simplest, active shunt to get more complex.
Whenever an honest man discovers that he's mistaken, he will either cease to be mistaken or he will cease to be honest.
User avatar
izzy wizzy
Old Hand
Posts: 1496
Joined: Fri Nov 02, 2007 7:02 pm
Location: Auckland NZ
Contact:

#50 Re: GM70 PP

Post by izzy wizzy »

Nick wrote: Thu Aug 08, 2019 10:21 am A shunt regulator seems to be the simplest IMHO. Couple of VR tubes at its simplest, active shunt to get more complex.
This is one of the options but I'm not that happy about having to introduce a regulator especially into the grid. One of my prejudices. Another way to address it, in a very simplistic way is to stick a big cap (47 to 100u maybe) from the bias supply end of the grid winding to ground; from what I've read. There goes another one of my prejudices; I've avoided caps for the most part but this "issue" will have me facing a number of prejudices. I suppose at least a VR tube is in keeping.

For once though, I'm going to try and understand the problem before deciding what to do. Also, is there a problem albiet even a small one? The amp will run class A1 most of the time, maybe even all of the time on my speakers. Given the system gain structure and where the volume is set, I doubt I swing more then 10 to 20V on the GM70 grid at sound levels where the bass on dance music and the like makes you realise your breathing is being modulated and it's a bit of a physical assault. Bloody loud is another way of putting it. Am I even getting anywhere near A2? This is the area I need to figure out and from what I've seen, there are graphs for the GM70 that show where the onset of grid current occurs.

Sure, if I want an amplifier that will run flat out into a dummy load, then there might be an "issue" to fix or maybe if I drag the Acoustat ESLs out, that too might require some A2 but that isn't my day to day at present. I don't want to try and solve something that doesn't really exist.

Nick, if you have any examples of this type of thing you could point me at, would appreciate it.

Cheers,
Stephen
User avatar
Nick
Site Admin
Posts: 15707
Joined: Sun May 06, 2007 10:20 am
Location: West Yorkshire

#51 Re: GM70 PP

Post by Nick »

Well, you raised a question about a problem. I suggested a solution, if its not a problem you need to solve, then you don't need a solution :-).

I can't see how a cap would help, it would just get charged up by the grid current. You just need a supply that can sink as well as source current, and that's what a shunt reg is good at. When I have done A2 in the past I have used a cathode follower (or variant) driver so that is happy handling the grid current.
Whenever an honest man discovers that he's mistaken, he will either cease to be mistaken or he will cease to be honest.
User avatar
izzy wizzy
Old Hand
Posts: 1496
Joined: Fri Nov 02, 2007 7:02 pm
Location: Auckland NZ
Contact:

#52 Re: GM70 PP

Post by izzy wizzy »

Nick wrote: Thu Aug 08, 2019 12:42 pm Well, you raised a question about a problem. I suggested a solution, if its not a problem you need to solve, then you don't need a solution :-).

I can't see how a cap would help, it would just get charged up by the grid current. You just need a supply that can sink as well as source current, and that's what a shunt reg is good at. When I have done A2 in the past I have used a cathode follower (or variant) driver so that is happy handling the grid current.
Well it was always the question of is there a problem and if so, is it relevant to my situation. Many designers propose all kinds of issues they have to get over, get on top of or generally be the master of to make sure their amplifier does whatever they want it to do. If that means taking a valve to its limit, then they have to solve that issue. I've done a lot more reading and for now have come back to my original place in that A2 operation is so far at one end of what is needed, that I don't think I need to bother. A1 will do and I'll have oodles of headroom. I don't need the claim that this amp can put out some silly number of watts by pushing it into A2.

As for the cap helping or not? I dunno. I read it on DIY Audio a number of times without much thought. The claim was it offered a low impedance when grid current flowed. I thought the grid current was an AC situation so the cap would be a low impedance. But I also read the grid acts like a diode under these conditions and so I can see the AC might be only in one direction and so as you say, it charges up. Again, if I haven't a problem with grid current, I don't need a cap/solution.

That still leaves the question of how clever does the bias supply have to be? It could hardly be simpler at present so many things to try. I see Pete Millet has a shunt reg design http://www.pmillett.com/bias_supply.html and Rod Coleman does a CCS based bias supply from what I can gather http://lyrima.co.uk/Bias/Bias-Regulator-V1.pdf; two quite different approaches. Better do some more reading and then put some ideas into practice.

Cheers,
Stephen
User avatar
IslandPink
Amstrad Tower of Power
Posts: 10041
Joined: Tue May 29, 2007 7:01 pm
Location: Denbigh, N.Wales

#53 Re: GM70 PP

Post by IslandPink »

A while back, I'd concluded from a bit of searching that the GM70 was very well behaved on grid current and could go at least to 0v bias without problems.
"Once you find out ... the Circumstances ; then you can go out"
User avatar
izzy wizzy
Old Hand
Posts: 1496
Joined: Fri Nov 02, 2007 7:02 pm
Location: Auckland NZ
Contact:

#54 Re: GM70 PP

Post by izzy wizzy »

IslandPink wrote: Fri Aug 09, 2019 8:31 pm A while back, I'd concluded from a bit of searching that the GM70 was very well behaved on grid current and could go at least to 0v bias without problems.
I've come to the same conclusion from what I've seen/read in all my rummaging around the interweb which is a relief so thanks for confirming that.
User avatar
Nick
Site Admin
Posts: 15707
Joined: Sun May 06, 2007 10:20 am
Location: West Yorkshire

#55 Re: GM70 PP

Post by Nick »

But I also read the grid acts like a diode under these conditions and so I can see the AC might be only in one direction and so as you say, it charges up.
Once the grid goes positive, it becomes an anode.
Whenever an honest man discovers that he's mistaken, he will either cease to be mistaken or he will cease to be honest.
RhythMick
Old Hand
Posts: 747
Joined: Fri Jun 23, 2017 12:24 pm

#56 Re: GM70 PP

Post by RhythMick »

I'm a ways behind you in terms of really developing my understanding of the behaviour of my amp in all conditions. However my ears have told me similar things. I prefer the sound of interstage coupled, all DHT, PP architecture. I prefer to minimise caps in favour of iron and I instinctively avoid silicon in the signal path, though I absolutely love Andrews DC filament boards.

I should figure out how to observe and measure grid current in my amps. None of the valves have grid resistors, they are grounded directly through the transformer secondaries which drive the grids.

The bottom line is, do you like the sound of your amps on your system? Are you chasing a problem that doesn't actually exist? Perfectly valid to seek a better understanding of its behaviour, but that doesn't necessarily mean you want to change anything.
User avatar
izzy wizzy
Old Hand
Posts: 1496
Joined: Fri Nov 02, 2007 7:02 pm
Location: Auckland NZ
Contact:

#57 Re: GM70 PP

Post by izzy wizzy »

It might sound as though I'm agonising over things that don't exist but it's more like trying to understand what is going on.

I'm over the moon with how this has turned out. So so happy with the sound. It's not only having what sounds like a new music collection but also having it expressed to me in a more meaningful way. The 813 I thought good but the GM70 is on another level.

And given this thing is so basic, there is the excitement of what it can be for the future.

Cheers,
Stephen
RhythMick
Old Hand
Posts: 747
Joined: Fri Jun 23, 2017 12:24 pm

#58 Re: GM70 PP

Post by RhythMick »

Simplicity rules
User avatar
IslandPink
Amstrad Tower of Power
Posts: 10041
Joined: Tue May 29, 2007 7:01 pm
Location: Denbigh, N.Wales

#59 Re: GM70 PP

Post by IslandPink »

NO.... the DHT filament supplies prove that !
Remember what Einstein said ?
"Once you find out ... the Circumstances ; then you can go out"
User avatar
izzy wizzy
Old Hand
Posts: 1496
Joined: Fri Nov 02, 2007 7:02 pm
Location: Auckland NZ
Contact:

#60 Re: GM70 PP

Post by izzy wizzy »

I remember a discussion involving jc once where there is an illusion of "simple" being better coz of lower component count where in fact the path for the electrons in real life is anything but simple. Therefore simple can be an illusion when seen on paper but very complex in reality.

The Lehane regulators being a case in point. They might seem complex but their function allows the filament to behave in a more simple way so the complexity leads to a simpler situation. It was a major prejudice I had to overcome almost insisting that AC would work but in fact operationally, it was more complex for the valve.

Sometimes a little more complexity allows for simpler and therefore better operation. I wish I understood things to the level jc does but I haven't the time and probably not the brain for that.

Cheers,
Stephen
Post Reply