Improvements in this riaa tube

[widi]

Well, Ca goes from Ra to ground, but not in your schematic.
Mona

Then the modification would be like this, as attached.
Or still it is not correct?

 

[Ketje]

From where did you get the 100nF for Cb ?
Mona

 

[widi]

The mod's i sugested where intended to get more headroom and output drive to prevent signal top cutting with strong signals, not to change the frequency response.
The correction network was ok.
What you do now is another way of making the filter, if well dimensioned, same result.
The Rout from the first stage is somewhat less than the 6k8 anode resistor, take 6k.
The Rin is the grid resistor of the next stage, 1M.
Resulting , with the same capacitors Ca=10n and Cb=100n, in Ra=67k and Rb=10k5.
Mona

From here, I thought this is what you wanted to tell me

 

[widi]

And is it valid?

 

[popilin]

As I have no a decent signal generator, always do an LTSpice simulation with a Laplace Transform on the signal source

This saves me a lot of time.

 

[popilin]

And is it valid?

As the time constants are wrong, ms instead of μs, this question is difficult to answer.

Your calculator seems to not take into account stray capacitances, so use it as an approximation, and then run an LTSpice simulation, depending on mathematical models accuracy, it should be more precise.

 

[Ketje]

From here, I thought this is what you wanted to tell me

You are right, my mistake
Only excuse is the time, was allmost midnight overhere
Cb has to be 30nF.
Mona

 

[widi]

You are right, my mistake
Only excuse is the time, was allmost midnight overhere
Cb has to be 30nF.
Mona

I will install the new component and inform you of the results.

 

[widi]

I will install the new component and inform you of the results.

 

[Ketje]

Why did you change Zout from 6k to 54k ? It is 6k not 54k !
Mona

 

[popilin]

Why did you change Zout from 6k to 54k ? It is 6k not 54k !
Mona

Maybe to compensate Zin=Rload=1M ?

Needless to say that input capacitance of next stage will ruin all those predictions...

Old values: Ra=75K, Rb=10k5, Ca=10nF, Cb=30nF, Cc=22n, Zout=6K Zin=∞, Rload=1M, give a decent result if taken into account Cin of next stage.

 

[kevinkr]

The 12AX7A also will have a very high output impedance configured as an SRPP, not significantly lower than that of the 5751 in fact.. (Still north of 12K) What is the load impedance you are driving with this thing?

 

[popilin]

Hi Kevin, good point!

I missed that because always do the phono and line stage into the same enclosure, anyway 12K is a lot.

 

[popilin]

As Kevin pointed out, the 12AX7 in SRPP has an output impedance of about 12K bypassed, unbypassed it is about 30K (!), so the OP should think in another valve or add another stage, e.g. a cathode follower.

 

[widi]

Maybe to compensate Zin=Rload=1M ?

Needless to say that input capacitance of next stage will ruin all those predictions...

Old values: Ra=75K, Rb=10k5, Ca=10nF, Cb=30nF, Cc=22n, Zout=6K Zin=∞, Rload=1M, give a decent result if taken into account Cin of next stage.

Thanks Popilin for your help, those values are not the old ones but the new ones calculated with the help of our friend Ketje.
The old ones are: Ra 91K, Rb 7K87, Ca 7.5nf and Cb 7.5nf. The values of the capacitors, being parallel / serial the value of Ca and Cb are the same, as you can observe the original circuit differs to the spreadsheet by the arrangement of such components including the resistor Rb.

 

[widi]

As Kevin pointed out, the 12AX7 in SRPP has an output impedance of about 12K bypassed, unbypassed it is about 30K (!), so the OP should think in another valve or add another stage, e.g. a cathode follower.

Thanks again Popilin for your help, and also KevinKr.
Excuse my ignorance, but I do not know what CPSR means, I would be very grateful if you explain it to me.
Right now keeping the original configuration at the output of the 12AX7 I have a resistance of 59.40K at rest. Again I do not know what OP means.
At this point the phono preamplifier is poorly designed from the beginning?

 

[widi]

Why did you change Zout from 6k to 54k ? It is 6k not 54k !
Mona

Sorry Ketje are 6K, 54K that are testing different curves of the RIAA

 

[popilin]

Thanks Popilin for your help, those values are not the old ones but the new ones calculated with the help of our friend Ketje.
The old ones are: Ra 91K, Rb 7K87, Ca 7.5nf and Cb 7.5nf. The values of the capacitors, being parallel / serial the value of Ca and Cb are the same, as you can observe the original circuit differs to the spreadsheet by the arrangement of such components including the resistor Rb.

OK, just keep the results of post#31, they will work.

Thanks again Popilin for your help, and also KevinKr.
Excuse my ignorance, but I do not know what CPSR means, I would be very grateful if you explain it to me.
Right now keeping the original configuration at the output of the 12AX7 I have a resistance of 59.40K at rest. Again I do not know what OP means.
At this point the phono preamplifier is poorly designed from the beginning?

Your circuit is very good, I like it, a cascode as input stage and an SRPP (Shunt Reguated Push Pull) as output stage, just choose another valve for the latter, the 6922 should do the work quite nicely, if gain is not enough just changing Zout and Ra such that Zout+Ra=constant the problem should be solved.

I have no LTSpice model for the 6C45P (adequate here due to its high transconductance), so please obtain the model (maybe opening another thread) and I will run an LTSpice simulation for you to optimize the circuit.

I hope Ketje do not care. Anyway corrections and critiques will be welcome.

SRPP is almost foolproof, so it will work with most resistor values, on condition that not exceed valve limits, of course there are some tricks to make it sing better.

Salas shunt regulator is amazing, so you can build a very good phono preamp.

Edit: Your new circuit is wrong, see capacitors Ca, Cb conection.

 

[widi]

Thanks, I hope your results impatient with LTSpice,

 

[popilin]

As said before, I do not have LTSpice models for the valves which you want to use, so I used the E88CC/6922

THD is about 0.18%, mostly 2nd harmonic

Frequency response is reasonably flat, about ±0.2dB from 20Hz to 20KHz

Note that I did use the Laplace transform without the missing time constant 3.18µs as you did so far because the circuit does not allow for it.