The main problem in Rods RIAA amp is - in my opinion - that the second correction network (the high frequency lowpass, a passive R/C network) is situated at the output of the amp and the filter characteristic is therefore always influenced by the following circuit. Everything below 100k is a clear no-no then. A ZEN V4 is definitely not suitable, although it inhibits a higher input impedance than earlier Zen versions. There is also not enough amplification level to drive a Zen amp.
Your RIAA amp, miguel, is probably also oscillating. It must sound immediatly "right", otherwise anything has failed.
My suggestions (for beginners) was:
(therefore this thread)
1. Use a well designed simple, but not too simple circuit (like the c't circuit or Rods riia amp)
2. Use a well designed PCB (to overcome all those construction problems, like the c't or Rod's PCB)
That make is so more simple and will have more time for other projects...
That does not refer to all real high end projects, because here we will have to did much deeper in all aspects.
Werner, I can confirm your suggestions. I have good experience with the "classical" set of a 10yF good quality elcap close to the LM317's output, and then a 27R-47R series resistor and another 47yF elcap to ground for each supplied stage. This may not be the best ever ( concerning superregs and you name it ), but that is basically the setup companies like Naim used successfully for all so many years...
Klaus
Your RIAA amp, miguel, is probably also oscillating. It must sound immediatly "right", otherwise anything has failed.
My suggestions (for beginners) was:
(therefore this thread)
1. Use a well designed simple, but not too simple circuit (like the c't circuit or Rods riia amp)
2. Use a well designed PCB (to overcome all those construction problems, like the c't or Rod's PCB)
That make is so more simple and will have more time for other projects...
That does not refer to all real high end projects, because here we will have to did much deeper in all aspects.
Werner, I can confirm your suggestions. I have good experience with the "classical" set of a 10yF good quality elcap close to the LM317's output, and then a 27R-47R series resistor and another 47yF elcap to ground for each supplied stage. This may not be the best ever ( concerning superregs and you name it ), but that is basically the setup companies like Naim used successfully for all so many years...
Klaus
Great little RIAA!
Thanks Werner!
I was just some hours ago trying to find some values
in RIAA feedback filter.
Thinking of using 5534 like you! (it has offset adjust possibilities)
I was thinking of using two 5534
with little less amplifications in first stage.
And, if needed a cap between stages to correct the DC-offset.
I used 10nF 33nF and was thinking of 100k and 8.2k
And have 470ohm something to the ground.
Here you use 10/33nF 110k and 6.8k
Is that 18ohm//120ohm to ground?
/halo
Thanks Werner!
I was just some hours ago trying to find some values
in RIAA feedback filter.
Thinking of using 5534 like you! (it has offset adjust possibilities)
I was thinking of using two 5534
with little less amplifications in first stage.
And, if needed a cap between stages to correct the DC-offset.
I used 10nF 33nF and was thinking of 100k and 8.2k
And have 470ohm something to the ground.
Here you use 10/33nF 110k and 6.8k
Is that 18ohm//120ohm to ground?
/halo
It is not my design. RIAA is not very accurate. But it sounds good for a cheap preamp.
Switching in that 18 Ohms resistor brings gain up to 54dB or so, enough for some MCs.
Something interesting to try then:
use one NE5532 for each 5534. Load both opamps in the 5532 with a current sink to pull them into modest class A, 1-2mA should suffice.
The first opamp gets the normal input, output and RIAA. Its output goes to the outside world and into the second opamp. The second opamp is configured as an inverting unity gain amplifier. It is loaded with 2k7, like the first one.
See what happens?
Switching in that 18 Ohms resistor brings gain up to 54dB or so, enough for some MCs.
Something interesting to try then:
use one NE5532 for each 5534. Load both opamps in the 5532 with a current sink to pull them into modest class A, 1-2mA should suffice.
The first opamp gets the normal input, output and RIAA. Its output goes to the outside world and into the second opamp. The second opamp is configured as an inverting unity gain amplifier. It is loaded with 2k7, like the first one.
See what happens?
RIAA Values
The values of the resistors & caps are not accurate.
if you want to use the .047 mf cap the resistor value should be 67.6K by deviding 3180 by the value of the cap.
The second part of the EQ. is the value of the resistor which in the schematic is 6.8K which means that your cap in parallel with it is going to be .011 mf by deviding 75 by 6800 Ohm.
The mid range curve is the combinations of both Eq's the high & the low put together. in this case your mid rage friquency is way high at (352) . But the correct mid constant is supposed to be at(318).
So to correct your RIAA curve & make it as accurate as poissible
you can use these values to put it at the closest you will ever get
The 1st resistor value is 67.6 K in parallel with .o47mf cap
that will give you the perfect (3180).
The second resistor value is 5.76K in parallel with .013 mf cap
you can get this value by putting .012 & .001 in parallel.
Your two values will come to a perfect (75)
Now when you put both of the Eq's together you will come to
a closest perfection of (315) for mid range curve. The
the formula ask for (318)
Good luck
Ray Samuels
The values of the resistors & caps are not accurate.
if you want to use the .047 mf cap the resistor value should be 67.6K by deviding 3180 by the value of the cap.
The second part of the EQ. is the value of the resistor which in the schematic is 6.8K which means that your cap in parallel with it is going to be .011 mf by deviding 75 by 6800 Ohm.
The mid range curve is the combinations of both Eq's the high & the low put together. in this case your mid rage friquency is way high at (352) . But the correct mid constant is supposed to be at(318).
So to correct your RIAA curve & make it as accurate as poissible
you can use these values to put it at the closest you will ever get
The 1st resistor value is 67.6 K in parallel with .o47mf cap
that will give you the perfect (3180).
The second resistor value is 5.76K in parallel with .013 mf cap
you can get this value by putting .012 & .001 in parallel.
Your two values will come to a perfect (75)
Now when you put both of the Eq's together you will come to
a closest perfection of (315) for mid range curve. The
the formula ask for (318)
Good luck
Ray Samuels
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.