hey yes the schematic in post #17
the first 2 stages are the phono stage, the final srpp stage deals with other inputs like cd etc
I have 2 stages so i could split the network into 2 stages- 1st being c3g then output of d3a. I have read many things about the value of the first resitor in the riaa stage- what is its approximate value? Should it be 10x the rp of the c3g ? Ejp suggested i increase my resistors(in the riaa network and the 13.6k resistor is the first in the network) by a factor of 1000 and decrease caps by a factor of 1000 which means my 13.6kOhm resistor from the anode of the c3g should be 13Mohms. I am a tad confused now so not quite sure where to go from here
the first 2 stages are the phono stage, the final srpp stage deals with other inputs like cd etc
I have 2 stages so i could split the network into 2 stages- 1st being c3g then output of d3a. I have read many things about the value of the first resitor in the riaa stage- what is its approximate value? Should it be 10x the rp of the c3g ? Ejp suggested i increase my resistors(in the riaa network and the 13.6k resistor is the first in the network) by a factor of 1000 and decrease caps by a factor of 1000 which means my 13.6kOhm resistor from the anode of the c3g should be 13Mohms. I am a tad confused now so not quite sure where to go from here
Do you studies schematics in #18 and #19?
"the little higher impedance correction chain helps to smoothing out frequency resolution"
The second one is from my built C3g, C3g, D3a CF phono.
I think the R-C values are good as they are. 10x R and 0.1x C would be still OK, but 1000x seems extreme. Try simulate the frequency response.
Hey Euro could you , if possible, model 10 x r values and 0.1 x c values. If (and I don't know how time consuming it is) you could model 1000 x r values and .001 x c values i would be thankful. My next question is, and I ask this respectfully, how accurate are the sims with respect to reality?
In my opinion the best way is to drive the riaa network with low impedance as:
The first stage is possible to set with proper gain and it sees always an high Z; the the CF can drive a low Z riaa network wth a lot of benefit also noise included.
In this case with 81 and 88 I got around 42 dB with low THD, low noise, high Vin max
With ecc83 the gain will be aroun 46-47 dB, little bit noiser.
This is the freq. response, +/ 0,4 dB
The first stage is possible to set with proper gain and it sees always an high Z; the the CF can drive a low Z riaa network wth a lot of benefit also noise included.
In this case with 81 and 88 I got around 42 dB with low THD, low noise, high Vin max
With ecc83 the gain will be aroun 46-47 dB, little bit noiser.
This is the freq. response, +/ 0,4 dB
My understanding is the c3g Z out is around 3k? How does this compare to the cathode follower arrangement (i.e. Z out) in your schematic?
In my case there is a High Zin (the CF) as load so the value of the previous stage is not rilevant
Walter
C3g as triode has Ri=2.3k.
The output impedance of C3g first stage in your sketch is 2k3//6k i.e. 1.66k.... really near 1k9...2k, due to the non-ideal AC blocking in cathode.
The theoretical output impedance of the CF is 1/gm.
ECC88 as CF has about theoretically 80R output impedance, really about 100-110R.
If you design the RIAA compensation, these values must be taken into account as source impedance.
Since rp changes as the tubes age (if they are used, not sitting in the Siemen's box for 40 years), that's a good idea.
so assuming i wish to use the c3g/d3a what do you suggest or put another way are the operating points anywhere near acceptable
The operating points are about right, the phono stages don't require larger current than these -about 10mA-.
The anode swings (voltage and current too) are low, the Ri relatively stable.
If you take into account the first stage's output impedance, IMO the larger impedance RIAA module would be OK.
You can't escape not measuring it.
Minimum one signal generator (with appropriate attenuator/inverse RIAA module, which provides 7mV RMS output signal at 1kHz) and AC RMS voltmeter requiring.
Sometimes I use PC/laptop soundcard (and inverse RIAA module for example or better) for tuning/measuring.
One of my favorite -fast- program is the Audiotester.
This is one -not well compensated- phono measuring.
Tuning for example the #19 schematic R17, R18, C3, possibly R9 elements, the curve visibly changing, so balanced FR available.
The anode swings (voltage and current too) are low, the Ri relatively stable.
If you take into account the first stage's output impedance, IMO the larger impedance RIAA module would be OK.
You can't escape not measuring it.
Minimum one signal generator (with appropriate attenuator/inverse RIAA module, which provides 7mV RMS output signal at 1kHz) and AC RMS voltmeter requiring.
Sometimes I use PC/laptop soundcard (and inverse RIAA module for example or better) for tuning/measuring.
One of my favorite -fast- program is the Audiotester.
This is one -not well compensated- phono measuring.
Tuning for example the #19 schematic R17, R18, C3, possibly R9 elements, the curve visibly changing, so balanced FR available.
Last edited:
I think that in the riaa chain the impedance of the network must be low to reach the best s/n
Of course the selection of the tube helps
Walter
It must be as low as possible given what the previous stage can deive. This one is far too low for a valve.
62dB at 20Hz. In theory. This is a lot of gain for a valve stage, and depending on the loading you may not get it all.
In my case the gain stages have each around 30 dB ( the selection of tubes move +/- 1 or 2dB )
So the final gain is easy to find