For the most part this phono stage is within tight tolerance, but I was troubled that the response rises so much after 20kHz -- not that there is that much information on vinyl save for the fact that a "snap, crackel or pop" will overload the next stage. It takes way too much additional Cgk to tame the ultrasonic response -- or is it OK, or should I even worry as the nominal response is down ~ 20dB at 20kHz vs 1kHz?
An externally hosted image should be here but it was not working when we last tested it.
The question is actually which RIAA curve you want to follow. There is a recent fashion for stopping the rolloff at 50kHz, as the commonest cutting head stops the boost at 50kHz. My personal take is to say that a preamp should follow the standard, but the design that you have built has an extra component to limit rolloff in the ultrasonics.
R5 is the culprit. If you remove it, you will get the right curve, without using a compensation cap. Leave it in if you want ultrasonic boost relative to RIAA, to compensate for cutting head bandwidth limitations.
R5 is the culprit. If you remove it, you will get the right curve, without using a compensation cap. Leave it in if you want ultrasonic boost relative to RIAA, to compensate for cutting head bandwidth limitations.
Too many rice crispies
The input capacitance seen by the cartridge is going to be humungous! Surely your rising 20kHz response is simply an indication that 1k is too big for the resistor below the 3200pF capacitor? As for worrying, yes, you're right to worry, excess ultrasonic response could exacerbate snap crackle and pop.
The input capacitance seen by the cartridge is going to be humungous! Surely your rising 20kHz response is simply an indication that 1k is too big for the resistor below the 3200pF capacitor? As for worrying, yes, you're right to worry, excess ultrasonic response could exacerbate snap crackle and pop.
Wavebourn said:
The data in the graph are the actual results, not simulated -- I just used the Multisim schematic to illustrate the layout.
With R5 = 625R -- (RED) -- and various other values
An externally hosted image should be here but it was not working when we last tested it.
I've been following your posts on this circuit for some time, and I'm rather impressed with the results you've been able to achieve! So much so, that I'm seriously considering building it for one of my PAS units.
What values did you use for the violet trace in your last post? Looks good to me! It's essentially flat to 20K, and 1db down at 50K. I tend to think, from a practical standpoint, that this would be preferable to a rising HF response. I have just two questions..... Is R10 calc'd for a certain following load (volume control), and... how does it sound to you?
Thanks!
What values did you use for the violet trace in your last post? Looks good to me! It's essentially flat to 20K, and 1db down at 50K. I tend to think, from a practical standpoint, that this would be preferable to a rising HF response. I have just two questions..... Is R10 calc'd for a certain following load (volume control), and... how does it sound to you?
Thanks!
The magenta trace is 1k//1k = 500R.ArtG said:
I have no idea how the circuit will perform over time -- the anode follower and its components might be very sensitive to tube aging, and all I have done so far is assemble the RIAA amplifier -- it hasn't gone live with vinyl yet.
Adjusting the value of R6 tilts the curve a bit clockwise -- which makes the deviation from RIAA even lower.
I have some cases for the preamp and its power supply -- have been busily rewinding a transformer to fit in the case -- this will use the LR8N3 as localized regulators.
The top trace is the RIAA preamp with LED's in both cathode positions, R5=625R, R6=201K
The two lower traces use 200R bypassed with 1,000u -- the magenta is the stock kit, the blue trace places 625R at R5 and 198K for R6. In the original R5 is 1K and R6 is 220K
The two lower traces use 200R bypassed with 1,000u -- the magenta is the stock kit, the blue trace places 625R at R5 and 198K for R6. In the original R5 is 1K and R6 is 220K
An externally hosted image should be here but it was not working when we last tested it.
- Status
- Not open for further replies.