Hi everyone.
Sorry if this question is far too basic, but I haven't found a clear answer, and I cannot measure it right now.
An output of 5 mV rms from a MM cartridge seems a common value. But this is for a 1 kHz signal. Would, due to the RIAA equalization, a 20 kHz signal come out at 50 mV rms? (and 20 Hz signal, at 0.5 mV)?
In preamp design "articles/web sites/forum threads", all gain and noise considerations are based on nominal 1 kHz pickup output level, but, after reading a couple of decades of references, I didn't find a clear answer.
Thank you.
Sorry if this question is far too basic, but I haven't found a clear answer, and I cannot measure it right now.
An output of 5 mV rms from a MM cartridge seems a common value. But this is for a 1 kHz signal. Would, due to the RIAA equalization, a 20 kHz signal come out at 50 mV rms? (and 20 Hz signal, at 0.5 mV)?
In preamp design "articles/web sites/forum threads", all gain and noise considerations are based on nominal 1 kHz pickup output level, but, after reading a couple of decades of references, I didn't find a clear answer.
Thank you.
I'll start the ball rolling with these non-expert thoughts - let's see where they take us!
The output is stated at a particular frequency and recorded velocity e.g. 5.0mV @ 1kHz, 5 cm/s.
The output of an MM cartridge is proportional to the recorded velocity.
An LP recorded with an RIAA characteristic limits the maximum recorded velocity and hence the maximum output at high frequencies.
The output is stated at a particular frequency and recorded velocity e.g. 5.0mV @ 1kHz, 5 cm/s.
The output of an MM cartridge is proportional to the recorded velocity.
An LP recorded with an RIAA characteristic limits the maximum recorded velocity and hence the maximum output at high frequencies.
That means working with 0.5 mV rms (in the bass region), which, in my humble opinion, is something to take into account (and many seem to ignore it, and be happy with the 5 mV rms assumption).
A circuit might deliver good enough performance for a particular input level, but not that good for a -20 dB input.
Generally speaking a cartridges output on a clean disc will be in the about 20mV RMS maximum depending on how loud the disc in question is cut.
However, if there is some surface damage to the disc or even a little dust in the grooves then the velocity sensitive output will be much higher and you can expect transients to reach 100mV maximum. Think of how quickly the stylus accelerates as it's jarred by these bits of debris and undesirable anomalies. These transients mainly exist in the 5-20kHz region. Thankfully they are attenuated by a good 10dB or so by the RIAA curve upon playback.
The catch here is, that your phonostage has to be able to take those 100mV kicks without overload otherwise they will generate lower frequency intermodulation products that will be more audible and subjectively worse... A little click becomes an obtrusive and distracting pop quite easily. This is why overload margin is very important.
Are you designing a phonostage?
However, if there is some surface damage to the disc or even a little dust in the grooves then the velocity sensitive output will be much higher and you can expect transients to reach 100mV maximum. Think of how quickly the stylus accelerates as it's jarred by these bits of debris and undesirable anomalies. These transients mainly exist in the 5-20kHz region. Thankfully they are attenuated by a good 10dB or so by the RIAA curve upon playback.
The catch here is, that your phonostage has to be able to take those 100mV kicks without overload otherwise they will generate lower frequency intermodulation products that will be more audible and subjectively worse... A little click becomes an obtrusive and distracting pop quite easily. This is why overload margin is very important.
Are you designing a phonostage?
Thank you. Even though I think I know what the article is about, I just printed, and will read it thoroughly before going on.
Yes, I want to experiment and have some practical knowledge.
I already aimed at having enough headroom to handle 200 mV peak. The problem right now is the lower end. What I don't know is if the lower nominal level is 5 mV (as stated for 1 kHz signal) or 0.5 mV (for the bass region, due to -20 dB respect to 1 kHz, RIAA equalization).
The cartridge can only reproduce what the groves in the disc dictate and by that I mean what outputs from the cartridge before it reaches the pre-amp and that's why you should read the PDF link I provided .
Once it passes through the pre-amp RIAA stage its de- emphasized , do I take it you mean the voltage output from the cartridge as it plays an LP ?
If so whats stopping you buying a hi-fi test record and measuring the LF output directly from the cartridge on a test band of LF ?
Once it passes through the pre-amp RIAA stage its de- emphasized , do I take it you mean the voltage output from the cartridge as it plays an LP ?
If so whats stopping you buying a hi-fi test record and measuring the LF output directly from the cartridge on a test band of LF ?
Yes, in some ideal world.
Real speech/music does NOT have full level highs. It is hard to make big highs. It would burn our ears. Real recorded programs fall-off above ~~~3khz. Record/Play equipment is built for real work.
And yes, bass is tiny until RIAA boosts it up. That's not a huge problem becuse bass-boost does not hiss. (We always have some or more boost-limit so hum and 1/f noise does not intrude.)
"5mV @ 1kHz" is kinda mid-meter. There will be peaks 10dB-20dB higher (but not too many, they reduce time-per-side). If you are trying to pass the needle signal "flat", you certainly want to pass 70mV to be 99.9% clip-free.
I severely doubt that
. My own experiments transferring 78RPM records using high output DJ cartridges (for high tracking force) have not generated more than 100mV even when tracing severe damage (1930s discs with steel needles dropped on them 60 years before I was born) where the angular velocity is almost twice as high as one would expect with modern pressings.Overload margin has more to do with circuit topology than amplifying devices themselves...
The peaks were short but they was at 500mV. And i do not doubt aI severely doubt that
Overload margin has more to do with circuit topology than amplifying devices themselves...
tektronix memory scope, it shows in green on black what has happened.
Using "normal" or "cheap" scope might miss these transients.
Imagen what happens with a decive that is designed to be linear with 10mV
input when 500mV comes by. It will saturate and in addition charge
the PN layers with additional charge that will take a while to discharge.
A tube won't store any charges thus it will work fine the microsecond after
the peak has passed.
Passive equalization, no feedback. My aim is not to build a best performer, but to experiment and get to an usable preamp.
I read the full PDF. It really didn't anwer my question.
Yes, I mean the voltage output from the cartridge. To design the first stage, I find useful to know what the lowest nominal level is at low frequencies (from a noise and S/N point of view).
Even though bass does not hiss, isn't S/N ratio significantly degraded? Both because of low voltage from cartridge and flicker noise?
I suppose headroom limits the highs, but preamp's noise performance limits the lows. Am I right?
What you aren't taking into account is the human ear which is not sensitive to very low frequencies .
Taken from a doctorate on the human ear and hearing-
"There is a steep decline of sensitivity ( up to 18db/octave ) in human hearing for frequencies between 100 and 20Hz . This steep cut-off means that to hear a stimulus at 5Hz it must be presented at 105db higher level than one at 500Hz "
If you want to wade through this scientific Paper then-
Responses of the ear to low frequency sounds, infrasound and wind turbines
What I am getting at is how much does it matter if noise etc is worse at LF compared to higher auditory frequencies ?
Even your sense of stereo effect is lost (progressively ) as one low frequency woofer is all that is required in many domestic situations.
Taken from a doctorate on the human ear and hearing-
"There is a steep decline of sensitivity ( up to 18db/octave ) in human hearing for frequencies between 100 and 20Hz . This steep cut-off means that to hear a stimulus at 5Hz it must be presented at 105db higher level than one at 500Hz "
If you want to wade through this scientific Paper then-
Responses of the ear to low frequency sounds, infrasound and wind turbines
What I am getting at is how much does it matter if noise etc is worse at LF compared to higher auditory frequencies ?
Even your sense of stereo effect is lost (progressively ) as one low frequency woofer is all that is required in many domestic situations.
So are you saying that you need a special oscilloscope to measure transients well under 100kHz? The cartridge inductance along with the load capacitance and resistor form a second order low pass filter with a cut-off around 20-30kHz.
Phono preamplifiers should be designed to be linear with 100mV+, not 10mV regardless of amplifying devices used.
Yes, less sensitive to low frequency noise, nut less sensitive to low frequency signal (music) too.
Isn't S/N ratio what matters?
The arm /cartridge resonance is around 10Hz so that too interferes with a scientific deduction of very low frequency noise as does the "rush " of the needle as it traces out the grove in the disc.
Also VTA/tracing angle has to be taken into account .
Very few records are cut with a signal lower than (approx ) 20Hz and LP,s will have their bass summed to mono below about 100Hz to prevent exciting the resonance .
Take into account too micro-warps ( above 10Hz ) that too is affected by the resonant frequency .
Do you have the Perfect Record Deck with perfect LP and a perfect cartridge /needle ? if not how do you expect to achieve a figure that can be used as an Industry Standard ?
Also VTA/tracing angle has to be taken into account .
Very few records are cut with a signal lower than (approx ) 20Hz and LP,s will have their bass summed to mono below about 100Hz to prevent exciting the resonance .
Take into account too micro-warps ( above 10Hz ) that too is affected by the resonant frequency .
Do you have the Perfect Record Deck with perfect LP and a perfect cartridge /needle ? if not how do you expect to achieve a figure that can be used as an Industry Standard ?
You need to load the cartridge with 50k and a few 100pF before any voltages measured are meaningful - with only a x10 probe as load a cartridge may be strongly resonant at the top end.
mHz is thousands of a hertz! Not sure Tektronix would appreciate their scope being downgraded 9 orders of magnitude there
Have you heard flicker noise?
I have, and have had trouble with it, it is a real thing. But in most sane situations it is mild and benign.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.