It's all relative (really!). If you take the 1000Hz point as reference, 0dB, the curve gives you the attenuation or gain at other frequencies relative to the gain at 1000Hz. (See Salas' table).
If you have a preamp with a gain of say 40dB (100x), that means that the gain at 1000Hz is 100x. So there is no '0dB' crossing' as such. From the table you can see that in the situation of a 40db gain preamp, the attenuation at 20kHz, which is the largest in the audio band, is about 20dB. That means there is still 40-20=20dB gain left at 20kHz.
If your preamp had a design gain of 20dB at 1kHz, than at 20kHz the gain would be down to 20-20=0dB. That *could* be considered '0 crossing' but it is a meaningless thing, because, you see, it's all relative
jan didden
The 0 db point of the RIAA cruve is arbitrary. It is only the shape which matters. The concept is that it should be an exact compliment to the curve used in the recording process. The RIAA curve standard is only one FR curve in the chain that must be taken into consideration. Optimal complex impedence of the electrical load the cartridge terminates in including tonearm wiring is another. Unless all of them are correctly "equalized" the FR between the input to the recording cutting head amplifier and the output of the playback preamplifier will not be flat.
Hi,
the original RIAA curve is defined by the 3 time constants 3180µs (pole), 318µs (zero) and 75µs (pole). These time constants can be represented by the frequencies 50Hz, 500Hz and 2122Hz. Later a second zero at 7950µs (zero) was added. This is the so called RIAA/IEC variant.
Since this preequalization curve raises till infinite with rising frequencies and since this is non practical with reallife cutterheads the company Neumann introduced a variant with an additional zero at ~3.18µs. The amplitude response of the preequalization curve levels out above 50kHz. This is called the RIAA-Neumann variant, which afaik has never become a official standard.
In playback RIAA and RIAA/IEC differ in that the IEC-variant adds a subsonic-lowpass filter at 20Hz. The Neumann variant differs to the RIAA-standards in that its bandwidth limit raises from app. 20kHz to higher values.
For a sufficient description of the curve a reference point in frequency and level is needed. Usually 1kHz and anormalized-to-0dB-level is taken. But it could as well be taken from 20Hz or 20kHz or whereever You want and at any reference level you want.
jauu
Calvin
ps. sreten, could You explain what you mean? To me it doesn´t make sense
the original RIAA curve is defined by the 3 time constants 3180µs (pole), 318µs (zero) and 75µs (pole). These time constants can be represented by the frequencies 50Hz, 500Hz and 2122Hz. Later a second zero at 7950µs (zero) was added. This is the so called RIAA/IEC variant.
Since this preequalization curve raises till infinite with rising frequencies and since this is non practical with reallife cutterheads the company Neumann introduced a variant with an additional zero at ~3.18µs. The amplitude response of the preequalization curve levels out above 50kHz. This is called the RIAA-Neumann variant, which afaik has never become a official standard.
In playback RIAA and RIAA/IEC differ in that the IEC-variant adds a subsonic-lowpass filter at 20Hz. The Neumann variant differs to the RIAA-standards in that its bandwidth limit raises from app. 20kHz to higher values.
For a sufficient description of the curve a reference point in frequency and level is needed. Usually 1kHz and anormalized-to-0dB-level is taken. But it could as well be taken from 20Hz or 20kHz or whereever You want and at any reference level you want.
jauu
Calvin
ps. sreten, could You explain what you mean? To me it doesn´t make sense
Last edited:
Hi,
The gain of a practical RIAA stage is 20dB+ at 1kHz, in
fact more typically 30dB+, for around 150mV+ output.
My bad, I did not say what I meant, and got the number wrong.
rgds, sreten.
Hi,
well the needed gain depends on the cartidge´s output voltage and the required circuit output voltage. For MM-Pickups one can assume as a rule of thumb around 3mV-5mV (@5cm/s, 1kHz) of signal voltage, for MC-Pickups one can assume 0.2mV-0.5mV, hence ~20dB less.
If the phono stage is internal to an preamp one can assume output voltage levels of 100mV-500mV. This is app. +40dB above MM-input-level and app. +60dB above MC-input-level. If the phono stage stage is an external device the output voltage could be 2V or more, which means roughly another +20dB of gain. So everything is possible between >+40dB and <+80dB. This means gains from less than 100 up to 10.000 times.
jauu
Calvin
well the needed gain depends on the cartidge´s output voltage and the required circuit output voltage. For MM-Pickups one can assume as a rule of thumb around 3mV-5mV (@5cm/s, 1kHz) of signal voltage, for MC-Pickups one can assume 0.2mV-0.5mV, hence ~20dB less.
If the phono stage is internal to an preamp one can assume output voltage levels of 100mV-500mV. This is app. +40dB above MM-input-level and app. +60dB above MC-input-level. If the phono stage stage is an external device the output voltage could be 2V or more, which means roughly another +20dB of gain. So everything is possible between >+40dB and <+80dB. This means gains from less than 100 up to 10.000 times.
jauu
Calvin
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.