GM70 PP

What people are working on at the moment
Post Reply
User avatar
izzy wizzy
Old Hand
Posts: 1496
Joined: Fri Nov 02, 2007 7:02 pm
Location: Auckland NZ
Contact:

#61 Re: GM70 PP

Post by izzy wizzy »

Nick wrote: Fri Aug 09, 2019 10:48 pm Once the grid goes positive, it becomes an anode.
Thanks Nick. Obvious really isn't it.

So that makes me wonder why I see some circuits have caps in that position (bias point on the IT to ground) and some don't. Maybe I should go and play in the sim and try and answer that.

In my head, I've kinda been thinking should I put some in to ensure the AC impedance of both halves are exactly the same independently of where the bias pots are set i.e. bypassing that resistance so the signal is developed only across the IT not the IT and the bias resistance. But then I'm adding caps in series with the signal and that doesn't feel right.

Cheers,
Stephen
User avatar
Nick
Site Admin
Posts: 15704
Joined: Sun May 06, 2007 10:20 am
Location: West Yorkshire

#62 Re: GM70 PP

Post by Nick »

Sometimes a little more complexity allows for simpler and therefore better operation.
Or, sometimes more complex is better full stop. Without needing the mental gymnastics making black into while so it fits in with a dogmatic position.

"It doesn't matter how beautiful your theory is, it doesn't matter how smart you are. If it doesn't agree with experiment, it's wrong. Richard P. Feynman"
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:

#63 Re: GM70 PP

Post by izzy wizzy »

Nick wrote: Sat Aug 10, 2019 6:11 pm Or, sometimes more complex is better full stop. Without needing the mental gymnastics making black into while so it fits in with a dogmatic position.
That's why I like it round here. So many other forums decend into a pissing contest of what's "right" ... whatever that is.
User avatar
IslandPink
Amstrad Tower of Power
Posts: 10041
Joined: Tue May 29, 2007 7:01 pm
Location: Denbigh, N.Wales

#64 Re: GM70 PP

Post by IslandPink »

Nick wrote: Sat Aug 10, 2019 6:11 pm .... If it doesn't agree with experiment, it's wrong. Richard P. Feynman"
Now then Nick - remember what we said over email, about citing Richard Feynman against people in an argument ?
"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:

#65 Re: GM70 PP

Post by izzy wizzy »

IslandPink wrote: Sat Aug 10, 2019 6:44 pm
Nick wrote: Sat Aug 10, 2019 6:11 pm .... If it doesn't agree with experiment, it's wrong. Richard P. Feynman"
Now then Nick - remember what we said over email, about citing Richard Feynman against people in an argument ?
I wasn't game to tackle that one ;)
User avatar
Nick
Site Admin
Posts: 15704
Joined: Sun May 06, 2007 10:20 am
Location: West Yorkshire

#66 Re: GM70 PP

Post by Nick »

IslandPink wrote: Sat Aug 10, 2019 6:44 pm
Nick wrote: Sat Aug 10, 2019 6:11 pm .... If it doesn't agree with experiment, it's wrong. Richard P. Feynman"
Now then Nick - remember what we said over email, about citing Richard Feynman against people in an argument ?
Not sure I was citing it "against" anyone (other than myself), and I am not sure I was even arguing :-).
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

#67 Re: GM70 PP

Post by RhythMick »

Ok I overstated. Fair cop. I'm well chastised.

"Everything should be made as simple as possible, but not simpler" isn't right either - it references what's possible but not what is optimal. And of course everyone will have their own view of what is optimal.

I was simply trying to say that if it works and sounds fantastic, why introduce additional complexity? But then I don't live by that guidance myself, as my regular experiments to see where the envelope can be stretched demonstrate. That's how I learn and I've no doubt others in the group do similar.

In other words ignore me. I'll get me coat.
User avatar
izzy wizzy
Old Hand
Posts: 1496
Joined: Fri Nov 02, 2007 7:02 pm
Location: Auckland NZ
Contact:

#68 Re: GM70 PP

Post by izzy wizzy »

Hooking up the SMPS direct to the filaments was never going to be the final solution but it got me started. Put in 8mH common mode chokes between to great effect. Then you can hear the signature of the SMPS on the sound once a fair portion of the ripple has gone. Spec for the PSUs are 120mV of ripple and I'd guess that will be mostly HF stuff.

Looking at the spec sheet for the choke cites 100R impedance at 1k. With the filament at about 7R, that should equate to roughly 6mV of ripple, less at higher frequencies. That's a decent reduction and it sounds it. Plan is this will feed Lehane VCCS filament supplies. With a cap following the CM choke, I'm hoping the VCCS will get a relatively smooth ride. These chokes have good response out to 300kHz.

However having read this, there is a very real possibilty I have the theory wrong. I only went for the theory after listening to the reduction of the SMPS effect on the sound. Maybe more work to do on the theory.

Modelling the bias supply in PSUD shows 150mV of ripple. That seems a lot. I've often wondered why fixed bias supplies are so crude in general and then all this gets injected into the grid(s) of the final(s). A LC of 10H and 20uF drops this to about 1mV so that could be my next move.

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

#69 Re: GM70 PP

Post by izzy wizzy »

Decided to revisit the bias supply as it is very simple and crude. As are just about any bias supply for a PP amp I've ever seen. Sure there is the Millet regulated supply but most are nothing more than I've done or in many cases, just half wave rectified and then presets, sometimes if you're lucky, to set the finals grid bias.

In my EARs this was the case and the later models had full wave rectified bias and I thought that better but it's hardly great. For the GM70, I added a 10H choke and 20u met film cap after and have been very surprised by the result. PSUD predicted about 160mV of ripple for the first instance and 1mV of ripple for the second. Sticking 160mV of ripple up the finals grid doesn't sound a great idea but I would wager many designs, commercial included are far worse. Relying on PP cancellation doesn't sound a great idea coz it isn't perfect but I think many rely on it for this case.

I wish I could measure things a bit better at high voltages or had a singal gen to measure frequency response as the change was quite large in the bass. Sure big bass can be fun but the flab around 50 to 100Hz (a guess) has largely gone. As it was there with the EAR509s and this amp, I thought it was the speakers but it would seem maybe not. Is this flab part of having crude fixed bias circuits? Is this maybe why some people prefer self bias? Have I made a mistake?

Overall I'm impressed with the lower noise, clearer musical lines throughout, hearing interactions between instruments and just plain hearing things I've not heard before. All typical of lowering noise in a circuit.

Has anyone had experience of the hammond chokes from Bluebell, say something like the 60H @ 8mA or some of the other cheapish ones? This one could get the ripple down to about 180uV but would there be some other issues I haven't thought about? The 10H choke in there at present is a 10H 100mA Danbury jobbie but I'd rather use that for other things like the new front end supply. I have some 100H Lundahl chokes lying around to see the effect of a huge inductance but don't want to use them as surely they have a better use one day.

Cheers,
Stephen
User avatar
Nick
Site Admin
Posts: 15704
Joined: Sun May 06, 2007 10:20 am
Location: West Yorkshire

#70 Re: GM70 PP

Post by Nick »

The little Hammond chokes are fine. Buying from Bluebell less so.
Whenever an honest man discovers that he's mistaken, he will either cease to be mistaken or he will cease to be honest.
Cressy Snr
Amstrad Tower of Power
Posts: 10552
Joined: Wed May 30, 2007 12:25 am
Location: South Yorks.

#71 Re: GM70 PP

Post by Cressy Snr »

Mouser usually have some in. I bought a couple of Hammonds last year from them with no probs.
Sgt. Baker started talkin’ with a Bullhorn in his hand.
User avatar
Nick
Site Admin
Posts: 15704
Joined: Sun May 06, 2007 10:20 am
Location: West Yorkshire

#72 Re: GM70 PP

Post by Nick »

The problem with Bluebell is between me and him from years ago now, normal sales should be fine.
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:

#73 Re: GM70 PP

Post by izzy wizzy »

Thanks Steve and Nick. Would rather support a local one band band than a corporation but if he can't do, then Mouser have 7 in stock.
User avatar
Nick
Site Admin
Posts: 15704
Joined: Sun May 06, 2007 10:20 am
Location: West Yorkshire

#74 Re: GM70 PP

Post by Nick »

I am sure Bluebell can do it, its just the owner got into the habit some years ago of calling me up to borrow money off me, then eventually did it one more time and never paid it back. Only £250 but enough to make me never use him again.
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:

#75 Re: GM70 PP

Post by izzy wizzy »

Nick wrote: Thu Oct 10, 2019 2:01 pm I am sure Bluebell can do it, its just the owner got into the habit some years ago of calling me up to borrow money off me, then eventually did it one more time and never paid it back. Only £250 but enough to make me never use him again.
There seems to be a bunch of stories in a similar vein over the years; money paid, no goods or very slow goods. Seems he has some issues that get in the way now and then. I've had a few transformers in the past no problem.
Post Reply