Re-designing my phonostage

[rho]

Hi,

Im working on a re-design (upgrade) of the phonostage I currently use. (Rod Elliot's Project 06 )
I want to add an extra gainstage for MC's. For this I picked the SSM2019 out of my component stock. Specs look very nice and this thing seems very easy to intregrate in the design.

I was thinking about the following:
I put the SSM2019 as a first gainstage and give it a good amount of gain. Lets say 20dB. This 20dB of gain will be subtracted form the orriginal 31dB of gain in the second stage of the Project06, leaving it with the 11dB I still need. This will result in the same amplification for MM's but with better noise figures (I hope).
Now I want to add a restsor to the one that sets the gain on the SSM2019 to give an extra 20dB of gain to be able to use low output MC's. This will be done by use of a relais (Hamlin HE721R0510 => Farnell order code: 433-2155).

Loading of the cell will be done by selecting one out of 6 possible sets of R and C. I'm thinking about the C&K A20615RNZQ (Farnell Order Code : 443-6234).

Are these good ideas? Will the shifting from the gain from the second (third in my situation) to the first stage work like I intend it to be? Is there any reason not to do it like this?

Thanks for the advice!

[Havoc]

AFAIK, it is always best to put the most gain at the front-end.

[peranders]

It's not true at all I'm afraid, particulary in RIAA amps. If you have more than one amp stage you must match the gain in each stage in order to get highest possible input clipping voltage. This is very important if the RIAA network is all passive.

[rho]

If an MC puts out about 0.5mV and the gain in the first stage is 40dB I will never get to the ~12V the first stage can give.

Hmmmm, wait a minute....the 0.5mV that's at 1kHz, right (looking at the specs of some cartridges) ? So the output will be higher at higher frequencies. 20dB extra at 20kHz, right?
That would result in a maximum output after the first stage of 0.5mV + 60dB. That would be 0.5V. Not realy a problem, I think.

Can anyone confirm that?

[Havoc]

It's not true at all I'm afraid, particulary in RIAA amps. If you have more than one amp stage you must match the gain in each stage in order to get highest possible input clipping voltage. This is very important if the RIAA network is all passive.

This is very true of passive stages I'm afraid. Since you have attenuation after the first stage, any noise from that stage will be amplified by following stages. So the less they have to amplify, the better.

If you don't want your first stage to clip, then design it that way.

Rho: I think your reasoning is correct.

[s2kov]

Using rod elliots project06 phonostage circuit http://sound.westhost.com/project06.htm, what sonic difference will i get if i use 47uF instead of 22uF?

[janusz]

Hi,

First s2kov: R and C determine the low frequency –3dB corner of the first stage. It should be set well below 20Hz. You may calculate that frequency (f) for any combination of R and C from this formula:

F = 1,000,000 /(2*Pi*R*C) where R is in ohms and C in micro Farads

Elliot set that frequency to 1.5Hz. Increasing C to 47uF will reduce that corner frequency to about 0.7Hz. I doubt one could hear the difference but you may try.

It is important, however, to use best quality BP cap (Black Gate, ELNA, Nichicon ES) here and a low noise resistor. You may also choose to parallel the BP with a high quality metalized polypropylene cap (100-220nF/100V) but again with high quality BP in the circuit I don’t think one would hear any difference. Replacing TL072s with better IC such as OPA2132PA would be more audible than changing the cap value.

Rho, I understand that you are adding a “pre-stage” to the existing phono amp, which already has two stages, to cater for a MC input. When choosing the right gain for this stage you have pay attention to PSR, CMR, distortion and the overall noise level. Increasing the gain you improve PSR and CMR but also increase distortion and probably total noise level. In most two stage passive RIAA designs using this kind of ICs (THAT1512 is somewhat better than SSM2019) most people set the gain to about 100. One must also pay attention to stability of these ICs.

If you choose to add this stage you may try to redesign the existing two stage phono amp. I would redesign some sections of RIAA, namely I would move the 75us stage to just past the added MC stage. This has the advantage of cutting down frequencies above 2.1kHz before the next two amplification stages. Next, you could add the fourth filtering stage at the output (using the existing space left by moving the 75us stage forward) setting its low frequency –3dB corner to 12Hz but then calculating the associated resistor you must take into account what comes after the phono preamp. The third stage gain has to be substantially cut down and then probably you could remove the 22uF cap.

You might also choose to redesign the active RIAA and reduce the 4k7 resistor to a lower value generating less noise. Alternatively with three amplification stages you might opt for a passive RIAA and replace TL072 ICs with better ones.

Cheers,

[rho]

That's alot of changes you suggest. I'll have to read some more about how to design good RIAA-filters before I go ahead with all the changes. I like to realy know what I'm doing.