JFET strain gauge amplifier with RIAA correction
I now implemented a circuit with a passive correction network. The first stage provides all the gain while a JFET follower buffers the network.
The attached picture shows the test bed, and a three position switch can be seen.
This switch selects between "flat", ideal displacement transducer, and Panasonic SG transducer.
"Flat" means no correction. (But the 50.05 Hz point is implemented for all three positions.)
The correct strain gauge correction according to calculations is implemented , see e.g. "RIAA Equalization for Displacement-Sensitive Phono Cartridges" by Gary Galo in Linear Audio Volume 11 April 2016. I implemented this with a shelving filter as done by Mr. Galo. The shelf filter starts to rise by 6 dB/octave at 500 Hz ending at 2122 Hz. The difference in gain between the flat portions is then 12.5 dB. This is the correction for an ideal displacement transducer.
The site "Audio Investigations" has looked into the equalization of the Panasonic strain gauge and found that a cut of 9.5 dB below 2122 Hz is optimal.
I used Siegfried Linkwitz' formulas for shelving highpass which is more a filter lifting above 500 than cutting below 2122, but for now this is good enough (for rock'n'roll). And also easier to implement.
The late Lou Dorren replied this about the strain gauge:
"The absolute realization of the ideal curve in the cutter pre-emphasis and playback de-emphasis networks is impractical. The compromise is to insure that the networks are at the 0dB point when crossing 1KHz.
To insure that the semiconductor pickup follows the curve a slight tweak is added to the pickup elements to provide the 0dB point crossing. Euphonics and Technics used this method to simplify the pre-amplifier. Soundsmith puts their tweak in the pre-amp. Either method will yield proper results."
The complete conversation can be found here at the Quadraphonicquad forum.
I did some listening to my solution and compared to a Red Book FLAC CD-rip of Steely Dan's "Gaucho". The so-called "flat" position with no correction sounded very different from the CD-version, and very unnatural. The two corrected positions were more difficult. Maybe the 9.5 dB version was more similar to the CD. More listening is necessary.
I must also admit I did not do any kind of HF roll-off. But comparing to the CD, it did not seem very urgent.
I used my old JVC SRP-473E with the Technics EPC-450C strain gauge cartridge for the listening.
I urge everyone with some interest to read Mr. Galo's article in Linear Audio, as indicated above.
See also this thread: Playing With Panasonic Strain Gauge Cartridges (And A Dedicated Phono Stage)
I now implemented a circuit with a passive correction network. The first stage provides all the gain while a JFET follower buffers the network.
The attached picture shows the test bed, and a three position switch can be seen.
This switch selects between "flat", ideal displacement transducer, and Panasonic SG transducer.
"Flat" means no correction. (But the 50.05 Hz point is implemented for all three positions.)
The correct strain gauge correction according to calculations is implemented , see e.g. "RIAA Equalization for Displacement-Sensitive Phono Cartridges" by Gary Galo in Linear Audio Volume 11 April 2016. I implemented this with a shelving filter as done by Mr. Galo. The shelf filter starts to rise by 6 dB/octave at 500 Hz ending at 2122 Hz. The difference in gain between the flat portions is then 12.5 dB. This is the correction for an ideal displacement transducer.
The site "Audio Investigations" has looked into the equalization of the Panasonic strain gauge and found that a cut of 9.5 dB below 2122 Hz is optimal.
I used Siegfried Linkwitz' formulas for shelving highpass which is more a filter lifting above 500 than cutting below 2122, but for now this is good enough (for rock'n'roll). And also easier to implement.
The late Lou Dorren replied this about the strain gauge:
"The absolute realization of the ideal curve in the cutter pre-emphasis and playback de-emphasis networks is impractical. The compromise is to insure that the networks are at the 0dB point when crossing 1KHz.
To insure that the semiconductor pickup follows the curve a slight tweak is added to the pickup elements to provide the 0dB point crossing. Euphonics and Technics used this method to simplify the pre-amplifier. Soundsmith puts their tweak in the pre-amp. Either method will yield proper results."
The complete conversation can be found here at the Quadraphonicquad forum.
I did some listening to my solution and compared to a Red Book FLAC CD-rip of Steely Dan's "Gaucho". The so-called "flat" position with no correction sounded very different from the CD-version, and very unnatural. The two corrected positions were more difficult. Maybe the 9.5 dB version was more similar to the CD. More listening is necessary.
I must also admit I did not do any kind of HF roll-off. But comparing to the CD, it did not seem very urgent.
I used my old JVC SRP-473E with the Technics EPC-450C strain gauge cartridge for the listening.
I urge everyone with some interest to read Mr. Galo's article in Linear Audio, as indicated above.
See also this thread: Playing With Panasonic Strain Gauge Cartridges (And A Dedicated Phono Stage)
Attachments
Last edited:
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.