Please check out http://www.tubefreak.com/mk1.gif
This is a schematic of the Mesa Boogie Mark I. The phase inverter seems pretty much Fender derived. However, the 330k grid load resistor on the grounded grid side (1meg in Fenders) seems odd.
Why would one want to do that? Do you think it was done to better balance the PI or to deliberately unbalance it? Has anyone tried it instead of the regular 1meg/1meg?
Thanks, Stephan
Phase inverters once more
Moderators: pompeiisneaks, Colossal
-
Darkbluemurder
- Posts: 584
- Joined: Tue Jan 03, 2006 7:28 pm
Phase inverters once more
Last edited by Darkbluemurder on Thu Aug 23, 2007 8:28 pm, edited 1 time in total.
Re: Phase inverters once more
Hmm, the link is broken.
On the Mk2 and many other 80s designs he used 100k on one side and 150k on the other. No good reason I can see or think of. The newer ones all use 1M/1M (the gainier the better
)
--mark
On the Mk2 and many other 80s designs he used 100k on one side and 150k on the other. No good reason I can see or think of. The newer ones all use 1M/1M (the gainier the better
)--mark
Re: Phase inverters once more
Lose the period at the end of the link's address......
HTH,
D
HTH,
D
-
Darkbluemurder
- Posts: 584
- Joined: Tue Jan 03, 2006 7:28 pm
Re: Phase inverters once more
Removed the period in the link - it works now as intended - sorry for the inconvenience.
Re: Phase inverters once more
> the 330k grid load resistor on the grounded grid side (1meg in Fenders) seems odd.
It's insignificant. Maybe they had too many 330K resistors hanging around.
This resistor should be 1Meg or less (tube rating), and much more than 10K so it doesn't load the 10K in the NFB network. Actually it could be near 10K and the NFB adjusted; but another constraint is the coupling cap must be affordable. Also, the grid resistor should not be real comparable to the 15K longtail resistor. The exact relations are complex because one side bootstraps the other, but a 1Meg resistor can use less than 0.01uFd and couple well to 17Hz (the 0.001uFd+1Meg on the other side is probably -3dB at 60Hz), so a 330K could use 0.03uFd, but I guess they also had a crate of 0.1uFd caps or discovered some benefit to sub-sonic NFB (which does have some subtle effects).
These older M-Bs were not designed with engineering rigor. They were much-much-modified extensively played breadboards which, when deemed "good", were turned over to Production. In hindsight, some of it doesn't have to be exactly like that. OTOH, none of their "odd" detailing adds more than a buck to the cost, and the cost of the hand-build M-Bs was lots of very fastidious labor and inspection/playing rather than parts.
I suspect you could throw in 1Meg and 0.03uFd (or 0.05 or 0.02uFd) and Mr Boogie himself could not tell.
But I could be wrong.
It's insignificant. Maybe they had too many 330K resistors hanging around.
This resistor should be 1Meg or less (tube rating), and much more than 10K so it doesn't load the 10K in the NFB network. Actually it could be near 10K and the NFB adjusted; but another constraint is the coupling cap must be affordable. Also, the grid resistor should not be real comparable to the 15K longtail resistor. The exact relations are complex because one side bootstraps the other, but a 1Meg resistor can use less than 0.01uFd and couple well to 17Hz (the 0.001uFd+1Meg on the other side is probably -3dB at 60Hz), so a 330K could use 0.03uFd, but I guess they also had a crate of 0.1uFd caps or discovered some benefit to sub-sonic NFB (which does have some subtle effects).
These older M-Bs were not designed with engineering rigor. They were much-much-modified extensively played breadboards which, when deemed "good", were turned over to Production. In hindsight, some of it doesn't have to be exactly like that. OTOH, none of their "odd" detailing adds more than a buck to the cost, and the cost of the hand-build M-Bs was lots of very fastidious labor and inspection/playing rather than parts.
I suspect you could throw in 1Meg and 0.03uFd (or 0.05 or 0.02uFd) and Mr Boogie himself could not tell.
But I could be wrong.
Re: Phase inverters once more
+1, the 330k is close enough to 1M in that place in the circuit so as to not make any audible or other significant difference.
--mark
--mark
-
Darkbluemurder
- Posts: 584
- Joined: Tue Jan 03, 2006 7:28 pm
Re: Phase inverters once more
Thanks, guys, very interesting.
I was curious because most "accepted good" amps use 1meg for both grid loads. Many SF Fenders and also the 80s Concert use 330k for both. You can read almost anywhere that they are best changed to 1meg (Torres, Weber etc.). In fact when I changed my Concert from 330k to 1meg I really didn't notice a difference.
I should have guessed that this was a moot point, otherwise M-B would have used 1meg/330k also in the Mark I Reissue and not 330k/1meg.
As for the rest of the circuit I believe they wanted a bit more gain with the 15k tail resistor than the 22k BF Fender value but more balance compared to the 6k8 brown Fender value. Marshall as well as many others use a 10k there. HAD used a 24k but I am not sure I would ever hear the difference between 22k and 24k. M-B went back to 22k in the Mk II and later models.
So the consensus appears to be that the M-B Mk I phase inverter is nothing special really. Hence, the amp's ascribed legendary tone must come from the first gain stage ahead of the Fender style preamp (ignoring output stage and the Altec speakers for a moment).
I was curious because most "accepted good" amps use 1meg for both grid loads. Many SF Fenders and also the 80s Concert use 330k for both. You can read almost anywhere that they are best changed to 1meg (Torres, Weber etc.). In fact when I changed my Concert from 330k to 1meg I really didn't notice a difference.
I should have guessed that this was a moot point, otherwise M-B would have used 1meg/330k also in the Mark I Reissue and not 330k/1meg.
As for the rest of the circuit I believe they wanted a bit more gain with the 15k tail resistor than the 22k BF Fender value but more balance compared to the 6k8 brown Fender value. Marshall as well as many others use a 10k there. HAD used a 24k but I am not sure I would ever hear the difference between 22k and 24k. M-B went back to 22k in the Mk II and later models.
So the consensus appears to be that the M-B Mk I phase inverter is nothing special really. Hence, the amp's ascribed legendary tone must come from the first gain stage ahead of the Fender style preamp (ignoring output stage and the Altec speakers for a moment).