Modifying Naim K boards for MM cart

[telecine]

My other half killed my Sony XL-88 MC cartridge by cueing up the edge of a record after a few vinos, bending the cantilever and making one channel o/c. She has also done similar things to an Ortofon Red and an Ortofon 510.....It would probably be cheaper to get a divorce at this point, but here is the question...

To get me going for the time being, I dug out an old A&R P77 MM cartridge.

I only have a Naim 42 preamp with (323?) MC K boards.

I changed the input loading to 47K // 100Pf ( Arm leads at 120pF). The sound this gave was pretty muffled.

I then changed the loading to no // c and 68K. This was a bit better, but still not right.

I've also had to reduce the gain a bit further downstream. (r13) The emitter resistor in series with 47 uF has been increased in value to reduce the gain.

Have I missed anything? This doesn't sound right at all and is pretty wooly to my ears. If I can get this working right, I'll replace the stylus on the ortofon Red, as the sound was pretty good (using a different preamp.)

Is what I'm trying to do even feasible?

Thanks in advance.

[sreten]

http://www.neilmcbride.co.uk/mcboard.pdf

Hi,

Due to the voltage divider on the input transistor input impedance can
never exceed around 3K so I don't know where you are going with this.

It will always sound wooly with low input impedance.

rgds, sreten.

[Zeta4]

The first stage of the mc boards gain is far too high for a mm. It will easily overload.
You will have to modify it to that of the mm board. I might have the schematic
somwhere but its a very long time since I played with Naim boards.

By the way the schematic of the mc board is missing a constant current
source on the first stage I think.

[sq225917]

Sreten, you might want to search around a little I seem to remember some comment somewhere that the voltage divider doesn't have the effect one would expect re limiting impedance. I can't remember the exact point but it was well reasoned and supported with measurement.

[sreten]

Originally Posted by sq225917

Sreten, you might want to search around a little I seem to remember
some comment somewhere that the voltage divider doesn't have the
effect one would expect re limiting impedance. I can't remember the
exact point but it was well reasoned and supported with measurement.

Hi,

No. Its you that needs to have any evidence that I'm wrong
before you suggest that is the case. I stand by with what I
said because its right. Upping the loading resistor from 470R
to 47K does not give 47K loading, end of story, its < ~ 3K.

I don't need to search to know my analogue circuit theory.

Due to resistance noise related issues it makes perfect sense.

rgds, sreten.

[Salas]

Here is how to do it without altering the original Riaa shape at all (which is a bit doctored by Naim BTW). The input 5x stage remains on DC bias for the block to function, but you skip the pre pre. The value changes I noted on the schematic are absolutely essential to reshape it back to correct response.

[Salas]

Originally Posted by sq225917

I seem to remember some comment somewhere that the voltage divider doesn't have the effect one would expect re limiting impedance. I can't remember the exact point but it was well reasoned and supported with measurement.

During simulation I used a 60 Ohm source impedance inverse Riaa generator block on the 56dB NA323 original. Had it sine generator value trimmed to deliver 0.5mV on the 470R loaded input. Raised it to 47K and it gave me 0.56mV @ 1kHz across it.

[sq225917]

Sorry Sreten i was only trying to help, I didn't realise it was such a touchy subject.

[telecine]

Hi Salas, I see what you're getting at up until the base of Q2, but why does C4 have to reduce in value? Just trying to learn something!

Thanks also for everyone else's input. Much appreciated.

[Salas]

Because if remaining 47u it makes a longer RC time constant, kicking up the bass when the emitter resistor forming the second stage gain ratio, has to be made higher value dropping gain in that stage too, to cater for MM. C1 has to be 100uF because the input impedance is now lower due to higher bias of Q2. Use a quality 25V polarized cap of low leakage current.