Yes, I know it John. The amp is one of the audio legends, so I can imagine, that the problem is in my calculation, but I don't know where. That's why I wrote I'm not fully sure in myself.
Best Regards
Attila
First, it is a transconductance amp. The signal level varies with frequency with cap. The cap ONLY changes the voltage out with frequency. It does NOT effect the current drive of the transconductance amp. That is determined by the Id capability of the input devices.
On the other hand, the error, which I calculated can be easily corrected with some selection. For example if the output impedance is 10 kohm, the 770 ohm, and 105 nF looks work fine. Or maybe there is an unwanted variable left in the program? I will check it.
Not really John, I only want to fully understand the design, and want to get know, what is wrong in my calculation. And if I see good ideas, I'd like to use it in my design, If I will ever have.
Best Regards
Attila
You are over-precise in your calculation and we most probably have greater than 10,000 ohms drive impedance because it is a cascode.
Ok John, I think. The other thing, what I did not consider, is the input capacitance of the next stage. I think, that you did know what you did.
But only one question John. Is it possible, that my calculation is correct, even if it is over-precise? I would be glad, if I could use it.
Thanks
Attila
SCP-1A frequency roll-off ( -3dB at 2120Hz ) simulation results
Hi John and aszoke,
I hope these attached files helps something about.
eD
Hi John and aszoke,
I hope these attached files helps something about.
eD
eD:
I'm in no position to verify the accuracy of the circuits but there are a few interesting items in your model around the moving coil cartridge and the loading circuit. I measured a few mc cartridges some time back and I think I posted them. I would recheck my numbers but I'm on the other side of the world from my test equipment for the next two weeks. I believe the inductance is much lower as is the resistance, only getting near the 20 Ohms on the high output mc cartridges. I don't think it will affect your results however.
I may be way off base with this but to explain another mystery, at low frequencies the output impedance will decrease from the feedback of the "servo", which seems to be how the 50 Hz corner is introduced (very clever, John) ending at a gain of -1.
I'm in no position to verify the accuracy of the circuits but there are a few interesting items in your model around the moving coil cartridge and the loading circuit. I measured a few mc cartridges some time back and I think I posted them. I would recheck my numbers but I'm on the other side of the world from my test equipment for the next two weeks. I believe the inductance is much lower as is the resistance, only getting near the 20 Ohms on the high output mc cartridges. I don't think it will affect your results however.
I may be way off base with this but to explain another mystery, at low frequencies the output impedance will decrease from the feedback of the "servo", which seems to be how the 50 Hz corner is introduced (very clever, John) ending at a gain of -1.
OK, (I still think its an interesting idea. . .) I was going from the pdf, not having seen the actual schematic. I have been using a passive 2120 Hz. low pass filter on the output for years in a similar idea (+ up to a 42 dB reduction of the 7th Harmonic above the corner).
Hi Demian,
the 50uH series 40 ohms are some specifications of Denon DL-304 moving coil cartridge. They recommends a 100ohms load for it. I never see the capacitor C3 ( 100pF ) at any schematic in this forum. I don´t know if John uses someone there. I put it in for simulations analysis.
In the sweep frequency simulation the V6 stimulus source does not effect and the 50Hz corner is on the second part of the circuit. See attached file.
regards
eD
Last edited:
Thanks eD,
Is it possible to test the amlification factor of the first stage in the original setup, and with 10 megs closing impedance instead of the final R||C tag?
Hi aszoke,
here is the result of just the 10Mohm resistor at output of first satge, without 750ohm // 100nF.
eD
here is the result of just the 10Mohm resistor at output of first satge, without 750ohm // 100nF.
eD
Last edited:
Thanks eD,
If I could understand the plots, the difference is about 50 dB, which means more than 100 times of amplification. Similar to that, what I supposed previously.
Interesting. First I was sure I can't use such type of equalization, now, I begin to think, that I will.
Thanks again,
Attila
Please, everyone, it is trivial, especially for those who don't believe in 1% matching. I cannot say that it is not measurable, it may well be.