Modify SRPP Phono Stage

[vinylnvalves]

I have a phonostage which i built sometime back (20 years+). I am now using a Nadja DSP to actively feed my speakers, the AD input stage has a 1V limit before clipping. There is too much gain from the phonstage and the 10:1 SUT i use with my MC cartridge.
I see i a number of options

1) add a attenator (~ 6db)
2) use the phonstage without the SUT and load the MC correctly, as bright on the 47Kohms.
3) reduce the gain of the phonstage, and use the SUT to load the MC.

The later is the one my favorite, with the second option next. Can someone advise if it is an easy task to reduce the gain (see attachment) BTW- don't use the line stage shown on the schematic. Concerned that the RIAA equilasation may get screwed.

Advise equally for reducing the input resistance of the input down to around 150 ohms. I don't think this can be done but am willing to be proved wrong.

[Paul Barker]

the question carries very easy adjust on test answer and very complicated must look right when measured answer.

Three elements to be considered in your question:

1/Impedance loading of cartridge
2/capacitive loading of cartridge
3/ Affect of output resistance of first valve stage and or input resistance of second valve stage on RIAA curve.

At the moment your SUT is loaded by 47k on the secondary. This is reflected to the primary as 470 ohms. (primary turns over secondary turns = square route of (primary impedance over secondary impedance). You are fortunate with a 1:10 step up because it is a simple function of 100. 47k devided by 100 = 470 ohms.

470 ohms is unlikely to be what your cartridge is specified to be loaded by.

A simple first solution on loading is to remove the 47pf capacitor which the majority of diyers don't use for MC.

Find out the loading the cartridge is specified for, calculate the secondary resistor.

A few examples using the 100 factor 20 ohm load = 2k 30ohm 3k etc. Get the point?

A big however, and Neil may help here as he did a lot of work on discovering what worked best for him in this situation.

Some people find that to leave the secondary not paralleled with any resistance sounds better. For that method, you snip out the 47k resistor.

Some people put the actual load resistance the cart. is specified for on the primary of the transformer.

So what you have to do is try different resistances on the primary and or the secondary until you have the sound you like best. that is tha AOT method.

People who like machines to show pretty pictures of a least distortion approach will probably do something similar but stop when the lines on their displays look best to them. Normally there is a discrepancy between the last two methods, seldom would the lines agree with the ear. Others may challenge that value judgement.

There are folk who say the capacitor is required for MC.

So there is a lot to try out.

then comes the gain withing the phono stage with least alteration to the RIAA.

Why not as a first go at it swop out the second stage for the third stage which you didn't build. So you put an ecc82 in place of an ecc81 and you change 1k cathode resistors for 1k5. the beauty of this is that the RIAA is mostly affected by the output impedance of the stage driving it. You haven't messed with that.

The value of C7 may require a little trimming as it is parallel with the miller capacitance of B2V2 bottom half. but it is swamped by the value of C7 18nF which at the moment is parallel with 98pf (0.098nf) and will become 27pf (0.027 nf) so no big issue. The tolerance of your 18nf is probably far wider than this alteration.

I think, the input impedance of the second stage only affects the gain, in your case less gain, which you are aiming for anyway.

Now it gets tricky. This gives quite a difference to the gain calculation. But it is such an easy valve swap that I would try this and see if there is sufficient gain.

If not you need to find a valve middle for diddle. I am thinking E88CC. But just try ECC82 first.

Good luck playing with cartridge loading!

[Paul Barker]

Just needs mentioning that what I held in mind while coming up with this solution was

1/ you do need the transformer or the valve stages alone will be much too noisy.

2/ you are best achieving maximum gain right off the bat and the last thing you want to mess with is the RIAA. Out of 2/ we learn, leave the ECC83 well alone.

Otherwise we are in need of an entirely new phono stage. That is not the solution you asked about.

[vinylnvalves]

I should have mentioned the cartridges i am currently using are ortofon kontrapunkt b's, which require 50 to 200ohm loading. I remembered that i hadn't changed the loading resistor when i changed from Dynavector.

To the other questions - the phono stage is quiet with or without the SUT.

I do have the line stage in the phono amp, its just not connected currently, so if i am reading your suggestion correctly connect the first stage to the line stage for reduced gain.

The third option appeals also (not financially through) as the 19" rack unit with diel mono PSU's takes up alot of room. From what i read ditch valves for JFets for phono's seems to be the trend.

[Paul Barker]

I generally find the lower values of cartridge loading sound better but give less gain. In your case less gain is good, so try the 50 ohm setting.

Regarding up to date opinions on jfet input phono stages people like Nick and Island Pink need to advise you there. Not my expertise. Always fancied it but didn't do it. Those two did there may be others here that have also.

[vinylnvalves]

Paul

When replacing V2 with the ECC82 - would it be better to replace R7 and R8 with 1k5 resistors. I also have the 350V supply to hand, so could use that as well.

I have tried another phonstage recently with completely variable cartridge loading, and cannot hear any differences below 200ohms, also changing the capacitance didn't seen to affect anything either. Got a EAR 834P coming to visit to see if thats any better, and how loading affects it.

[IslandPink]

If your phono is quiet enough for your needs without the SUT then I'd go with that . You'll get better low bass, because you lose the inductance-limited bass rolloff from the SUT , and there should be some more low-level tone and microdynamics in the mix too . I do remember the EAR 834P was almost usable without SUT when I had one , years ago, although my standards for noise might be a bit more stringent these days .

[vinylnvalves]

The stage is quiet enough without the SUT. The SUT does allows me to impedance match (closer) for my cartridge. Unless i am missing something, i did not think i would be able to go putting a 50 - 400 ohm resistor in place of R1?
Ironically this SS phono stage i am playing with has less bass than i like, not made my mind up whether its more accurate, and i am used to flabbier bass.

[Paul Barker]

Paul

When replacing V2 with the ECC82 - would it be better to replace R7 and R8 with 1k5 resistors.

and you change 1k cathode resistors for 1k5.

[IslandPink]

The stage is quiet enough without the SUT. The SUT does allows me to impedance match (closer) for my cartridge. Unless i am missing something, i did not think i would be able to go putting a 50 - 400 ohm resistor in place of R1?

I don't think there's a problem with this .
You're just tying the grid to ground a little more 'strongly' .
The cartridge has to work harder to lift it up and down
Paul or Nick can correct me if there's any other issue. It's sure to be safe, so you can try it anyway .

[Paul Barker]

The stage is quiet enough without the SUT. The SUT does allows me to impedance match (closer) for my cartridge. Unless i am missing something, i did not think i would be able to go putting a 50 - 400 ohm resistor in place of R1?

Read my first post on the thread again please so i don't have to keep quoting it?

At the moment your SUT is loaded by 47k on the secondary. This is reflected to the primary as 470 ohms. (primary turns over secondary turns = square route of (primary impedance over secondary impedance). You are fortunate with a 1:10 step up because it is a simple function of 100. 47k devided by 100 = 470 ohms.

470 ohms is unlikely to be what your cartridge is specified to be loaded by.

A simple first solution on loading is to remove the 47pf capacitor which the majority of diyers don't use for MC.

Find out the loading the cartridge is specified for, calculate the secondary resistor.

A few examples using the 100 factor 20 ohm load = 2k 30ohm 3k etc. Get the point?

[vinylnvalves]

[quote="vinylnvalves"]Paul -
When replacing V2 with the ECC82 - would it be better to replace R7 and R8 with 1k5 resistors. I also have the 350V supply to hand, so could use that as well.quote]

Sorry this was a stupid question - i had forgotten that both halves of the valve were used. Both are cathode loading resistors -doh.

With the cartridge loading - i am aware and have added a secondary resistor to get the right reflected load. But only experienced with adding larger resistors to match MM cartridges. Logically as long as the resistor is at least a decade large there should not be any issues with the source driving the stage.
This brings me onto something i have just throught about and the ECC82 probably brings benefit. The output impedance of the ecc81 is probably too high to drive the input stage of the Nadja, where as the ~ 2k output impedance of the ecc82 will be better. In my previous setup where the WOT preamp had a 500k input it was not an issue.

Just for curosity - what phono stage are you currently using Mark?

[Paul Barker]

Sorry this was a stupid question -

Not stupid. don't worry. Just prompting you to read my post in more depth for a second time as you must have scated over the features you missed.

Rather than refer to resistor numbers I gave a commonly used descriptor which meant you had to speed up and learn some terms but after learning those terms all your future discussions of valve circuits will be more grown up.

With the cartridge loading - i am aware and have added a secondary resistor to get the right reflected load. But only experienced with adding larger resistors to match MM cartridges. Logically as long as the resistor is at least a decade large there should not be any issues with the source driving the stage.

I see what you are worried about but you have that angle covered when you offer the very load which the cartridge is suited to. do you get that? The MC cartridges only want a low load because they are only low ohms output resistance.

This brings me onto something i have just throught about and the ECC82 probably brings benefit. The output impedance of the ecc81 is probably too high to drive the input stage of the Nadja, where as the ~ 2k output impedance of the ecc82 will be better. In my previous setup where the WOT preamp had a 500k input it was not an issue.

True

But when you are talking output impedance of a standard configuration Resistor loaded common cathode ECC82 stage and the output resistance of an SRPP with output taked from the cathode of the top valve you are looking at much lower output impedance, and I don't think the ECC81 as a SRPP with that output option would have dificulty driving 100k ohm input impedance. Though prorata ECC82 will be better for that reason alone but not necessarily better altogether because of the misterious world of valves in which impedance matching is not a stand alone contribution to sound. Different valves sound different for reasons of their own. Many people like the sound of the ECC81. Can't say I am a fan. But I am a lesser fan of the ECC82. I would be using the 6SN7 (or alternative based valves of similar properties like 7n7 etc etc) instead.

Mark will speak for himself but he has had lots of really good phono stages and has a project on hold I think as he went quiet on it, to build his best yet hybrid FET valve phono stage. There is an old thread of his on that subject which brought Jo Roberts out of hiding.

[vinylnvalves]

Just for information - the impedance of the input for the analogue stage on the Nadja is 8K. So lower impedance will be the order of the day. I think this sort of impedance is outside of a valves without a transformer?

[IslandPink]

This could be tricky, as you say. 8K is quite a low input impedance.
Typical phono valves with high current can get you down to 1.5K or so, but that's marginal for an 8K load.

Paul - my next build , after I can get an 'acceptable' speakers solution, will be an LCR phono based on the footprint of the current one, with a beefed-up power supply , using Dave Slagle's LCR chokes. I have a circuit sketched out and planned, but Nick has given me serious pause for thought wiht his new Mk.3 unit wiht the direct input into D3a-pentode ( Gm-stage ) .
I won't be using the jFet/triode cascode as the experiements, although it was pretty close to being right - it just didn't quite get there .
http://www.audio-talk.co.uk/phpBB2/view ... sc&start=0