Welcome! Need support, you got it. Or share your ideas and experiences.
166.250.0.62
In Reply to: RE: Note to folks performing my PAS RIAA mod posted by Joe Curcio on March 23, 2011 at 08:43:26
Joe,
My comment was a bit of extrapolation. The R component of the 3180us
pole is off by a factor of 2.5X, and that's the pole that's at issue.
I saw the graph in your white paper. The other (parallel) components added
were 3% and 5% of the original component values. So ALL the components
on the phono board (nearly, anyway) better be 5% (or 1%) components and
in spec, otherwise the changes are indistinguishable from the values of
the original components. If the components are original, a large number
can be expected to be out of spec; a practical mod would have to re-stuff
the whole board first. Then we get to impedance match the triodes...
and I don't see that mentioned in the mod.
If someone is solder challenged, the mod isn't trivial.
(That's why you had to include in one of the postings of the mod a warning not to remove the 47K feedback resistors - why would anyone
do that?... maybe they can't read the schematic, don't know the resistor
codes, or don't know how to trace the board layout :-( ). But I think
most modders can locate a pair of 4.7Mohm resistors reliably.
Stock response on the graph, with the exception of the rise below 100hz, looks like it fits within a 1db range. If the RIAA is within +/- .5%,
that's darn good (and I think it will be, if the 4.7M ohm -> 2.0Mohm),
because I doubt the rest of the circuit is that good. The other issue is can you HEAR differences as little as .5db at typical listening
volumes of 70-90db? The speakers with the flattest frequency response
I've auditioned are something like +/-2.5db, about as good as it gets.
They have non-linearities of 2-3db, and sound quite flat.
I can't take the debate any further without measurements, which I don't
have at this point. I'd like to mod-the-mod to preserve Decca/78RPM
equalization, so I'm looking for an 80/20 solution that'll minimize the
ripple effects of this mod (like requiring 100K ohm volume pots, which
means forgoing the loudness function, losing the other equalizations,
re-stuffing the board...) but substantially reduce or eliminate the
non-linearity around the 50hz pole. How that 4.7Mohm resistor got there
I think had something to do with offbeat 78RPM equalizations. You don't
think David Hafler made a "mistake" like that!
Stuart
Follow Ups:
You are right - all of the components (but for the 2M resistor) are within 5% - however, in total the response departs from spec significantly especially below 100Hz. Plus, those values already account for 5% of error so when added to the tolerance of the new part, you may be off by 10% or more. So it is misleading to suggest that only the 4M7 resistor be changed.
I have not addressed the tape (NAB) eq since most folks will want to run one or more turntables and my mod provides this capability. If you plan on making a collection of recordings from a tape head I would suggest that you make all of your recordings using the stock PAS in the tape position and then perform the mod for your subsequent Phono recordings. If OTOH you will need the use of a direct tape head input frequently, you may want to not perform the mod.
> > the response departs from spec significantly especially below 100H
Other than the 2M resistor, and a 3% change to the 2750uuf cap, the
rest of the mod is to the RC time constants of the other poles/zeros.
These don't affect the response below 100hz. I'd like to see the
response curve with just a 2.0Mohm 1% resistor in place.
The mod is a quite large change if more phono inputs is the goal. The kit
instructions include alternate wiring for "special" as a second phono
input with just a change of a couple of wires. A few more wires change
can convert the line level Spare input to a third phono input, if need be.
Many "other" turntables are MC anyway, so the additional phono input(s)
won't help.
Stuart
Once the wires are cut (as described) from the center section of the selector switch, all of the low level inputs will exhibit the same modified RIAA response (see the bold print on the first page of the white paper).. Therefore up to 3 turntables can be used.
Post a Followup:
FAQ |
Post a Message! |
Forgot Password? |
|
||||||||||||||
|
This post is made possible by the generous support of people like you and our sponsors: