Granted that ultimately precision and tweaking of the RIAA is done on the circuit with proper set up and equipment and not only driven by the simulation's tool.
What I find is that, usually, you can compensate a little bit with resistors and having custom values for caps and resistors is the optimal solution.
Once you determined on the circuit your ideal values that gives you the best RIAA recontruction you order the custom or you try to get the closest one and see if you can adjust with resistors.
What I find is that, usually, you can compensate a little bit with resistors and having custom values for caps and resistors is the optimal solution.
Once you determined on the circuit your ideal values that gives you the best RIAA recontruction you order the custom or you try to get the closest one and see if you can adjust with resistors.
I do two stage RIIA´s and one stage too. I general i prefer the simplest signal path as possible provided that distortion is under the audible limit.. Usually it put the 75usec in the first stage because that gives a better overload margin where it counts, suppressing bad effects from scratches and dirt.
By the way i do not have audible problems with electrolytics and coupling caps when they are used in such a way that they do not produce distortion. Servos have their own issues.
Best would be DC coupled without servo but that is hard to do with high gain circuits. I have done that in my line stage and in my power amps that have very good thermal stability because i use lateral mosfets. This kind of circuits can be hand trimmed without much drift.
Best would be DC coupled without servo but that is hard to do with high gain circuits. I have done that in my line stage and in my power amps that have very good thermal stability because i use lateral mosfets. This kind of circuits can be hand trimmed without much drift.
I agree
The network that gives the 2122Hz corner (or 75uS) should be placed first because the signal coming off disc has its highest amplitudes at high frequencies, so we wish to avoid high frequency overload in our gain stages so we filter it before it becomes to large. The second reason is that we wish to defer filtering the low frequencies until we have amplified them up to a reasonably healthy level.
Now the big question...stefanoo removed the EL caps and got lower amplification but Joachim told me these are responsible for the bass response.. What am I missing ?
Joachim, thanks for your contribuite.
I haven't experimented with this topology yet in term of sound, and I am going to have a paradise's prototype ready for listening by tomorrow and then start swapping parts.
If from one side I agree with you, on the other hand I think that capacitor most of the time leave a sonical trace and most of the time is not measurable at all not in terms of frequency response nor distortion wise.
For example, different PSU filter cap.
You can explain certain things with ESR, capacitance value Ripple current, but, given pretty much the same specs, common practice suggests that they sound some time pretty different one from another.
This is my take so far and it can be wrong or impartially incorrect but it is what I know so far 🙂
Said that I can't imagine the boosting capacitor not leaving any significant trace on the sound and if it doesn't I will be extremely pleasantly surprised.
A straight test I am going to run is to have same values but different type, perhaps Panasonic FC and some cheapo nasty cap.
If there is no significant impact then I will change my approach on caps.
The network that gives the 2122Hz corner (or 75uS) should be placed first because the signal coming off disc has its highest amplitudes at high frequencies, so we wish to avoid high frequency overload in our gain stages so we filter it before it becomes to large. The second reason is that we wish to defer filtering the low frequencies until we have amplified them up to a reasonably healthy level.[/QUOTE]
That is a good thought behind it! There is no end to learn; I would have done the other way around to filter better at the output high frequency noise.
However it just makes a ton of more sense to do it the way Joachim suggested.
That is a good thought behind it! There is no end to learn; I would have done the other way around to filter better at the output high frequency noise.
However it just makes a ton of more sense to do it the way Joachim suggested.
Maybe there was a sort of misunderstanding between you and Joachim.
In short, I am unable at this point to tell you exactly what the rule of these capacitors on the transfer function is unless I calculate it (which I might due when I get some spare time).
What I can tell you is that these capacitors are responsible for the bass response but at the same time, if you take them off your gain drops down, and you need a way to set your AC gain which you do it by inserting the resistor from the midpoint to GND.
You can see the effect of it if you model the circuit on the simulator, which is what I have been describing above.
I have heard it argued that the stages should be the other way around because putting the HF filtering second will help reduce noise - this is possible but just ignores the reasoning above.
Last edited:
yes I fully agree with you and after all all the high frequencies are attenuated anyway by a 20dB factor.
Would you post an LT Spice file so I can study it ?
It would save me a lot of trouble... I am a novice with simulators 🙂
PS: How did you build the paradise... p2p ???
I designed the layout and had it done professionally and mount it.
I will post some pictures as well.
To make work easier for everybody who is going to join this 3d and hope it is appreciated a little bit since it quite of a work to create all the models 🙂 ...here it is...
Copy library.txt on the same folder as the simulation file and then run it.
I will post some pictures as well.
To make work easier for everybody who is going to join this 3d and hope it is appreciated a little bit since it quite of a work to create all the models 🙂 ...here it is...
Copy library.txt on the same folder as the simulation file and then run it.
you are very welcome.
you should also have the inverting filter set inside of the function generator so you can check RIAA accuracy against the current RIAA standards
you should also have the inverting filter set inside of the function generator so you can check RIAA accuracy against the current RIAA standards
The Pardise needs the big elcaps to make an AC path at the low frequencies. The high path is formed by the emitter impedance of the input stage in series with the electrolytics.. The formula is simply R x C.
So when R is low you need a big electrolytic or the gain drops in the bass.
So when R is low you need a big electrolytic or the gain drops in the bass.
How is the masterpiece coming
Joachim, you are so kind and considerate to help Stefanoo design a RIAA that supercedes your own simpler version. I guess in this case, the student will surpass the teacher
Joachim, you are so kind and considerate to help Stefanoo design a RIAA that supercedes your own simpler version. I guess in this case, the student will surpass the teacher
I use a DCB1 for preamp so I can apreciate what a DC coupled, no servo thing can do.... über soundstage... super speed and detail... I am never going back
As for EL caps, they all leave it´s signature .... BG's are different from Panas etc...
In the simplistic first stage, I experimented degenerating with a big EL // Rs and guess what... I got more gain but lost all the magic....
How is the masterpiece coming
Joachim, you are so kind and considerate to help Stefanoo design a RIAA that supercedes your own simpler version. I guess in this case, the student will surpass the teacher
you are not welcome here either!! Thanks for stopping by though but no need for arguments this is a serious thread, go somewhere else!
Thanks!
- Status
- Not open for further replies.