Excact RIAA numbers?
Hi!
I was hoping someone could shead som light over the RIAA response curve.
The points are as usual the 50Hz, 500Hz, 2120Hz and maybe the 50kHz.
My question is:
My guessings are:
20Hz = +20dB
50Hz = +17dB (20-3dB)
500Hz = +3dB
1kHz = 0dB
2120 = -3dB
20kHz = -20dB
50kHz = -25dB (??)
but Im not sure. At least not about that 50kHz...
I recently found this Excel RIAA table, showing numbers that obiously where calculated from one heck of a formula, and the numbers from that table compared to my points up there where not identical.
So, what are the REAL numbers to use? Are this Excel Table accurate? I think so, but I would appreciate some comments, especially on that 50kHz point.
http://www.bonavolta.ch/hobby/en/audio/riaa.htm
Cheers;
Hi!
I was hoping someone could shead som light over the RIAA response curve.
The points are as usual the 50Hz, 500Hz, 2120Hz and maybe the 50kHz.
My question is:
My guessings are:
20Hz = +20dB
50Hz = +17dB (20-3dB)
500Hz = +3dB
1kHz = 0dB
2120 = -3dB
20kHz = -20dB
50kHz = -25dB (??)
but Im not sure. At least not about that 50kHz...
I recently found this Excel RIAA table, showing numbers that obiously where calculated from one heck of a formula, and the numbers from that table compared to my points up there where not identical.
So, what are the REAL numbers to use? Are this Excel Table accurate? I think so, but I would appreciate some comments, especially on that 50kHz point.
http://www.bonavolta.ch/hobby/en/audio/riaa.htm
Cheers;
After the rock throwing contest ends......
Somewhere I have the RIAA document, from the Beast itself, with the exact numbers. Maybe I'll look for it.
Meanwhile.......this might be fun. [joke]
Looks like Frank is becoming a Texan. Amazing the influence I have on people who actually pay attention.
Jocko
Somewhere I have the RIAA document, from the Beast itself, with the exact numbers. Maybe I'll look for it.
Meanwhile.......this might be fun. [joke]
Looks like Frank is becoming a Texan. Amazing the influence I have on people who actually pay attention.
Jocko
actually ...
Texas Here I Come.
Hi,
Mon petit pois chiche?
Just hope it's a two way street so one day I can show you what tubes and vynil can do to cure you of digititis?
How on earth did you guys catch this booby trap anyway?
So the travelling plan is California, Texas, must check some New York ears too ( for onions) and than...Canada???
Cheers guys,
Hi,
Mon petit pois chiche?
Just hope it's a two way street so one day I can show you what tubes and vynil can do to cure you of digititis?
How on earth did you guys catch this booby trap anyway?
So the travelling plan is California, Texas, must check some New York ears too ( for onions) and than...Canada???
Cheers guys,
If you include the 3.18us, the replay numbers are...
DC 19.909
20 19.272
50 16.944
100 13.087
200 8.218
500 2.646
1k 0
2k -2.583
5k -8.168
10k -13.566
20k -18.979
50k -24.536
infinity -27.543
This does not include the IEC 7950us time constant, which is a naff replay-only recommendation aimed at reducing turntable rumble.
DC 19.909
20 19.272
50 16.944
100 13.087
200 8.218
500 2.646
1k 0
2k -2.583
5k -8.168
10k -13.566
20k -18.979
50k -24.536
infinity -27.543
This does not include the IEC 7950us time constant, which is a naff replay-only recommendation aimed at reducing turntable rumble.
Thanks guys!! This is very appreciated!
So, I've been comparing the different suggestions from the different pepole, and there seem to be some disparity between them,
For example for the 50Hz there's -16,96 from Jocko Homo, -16,944 from EC8010, and -17,03 from the excel at Bonavolt's site...
Ive got a -2,50 at 2000Hz, and -2,61 (Jocko Homo) and -2,583 (EC8010)
But than again, theres probably a roughly 0,1dB difference max between the three, Where Jocko Homo's seems to be the "brighter" one, Bona's the "darkest" one and EC8010 in the middle. (but closer to Bona's)
I guess it would be quite meaningless to start making calculations for a filter and then worrying about a 0,1dB difference, but I really want to be shure that I'm using THE one riaa cure as a reference. (despite the fact that in real life my actual curve will NOT be so close to the calculations)
Anyway it is interesting to observe that Jocko Homo's and Bona's equations are almost identical. Why are they not?
Come to think of it, it really IS meaningless to compare the different curves, it really is more important that the final amp plays music that pleases the ear. If I get there by intense calculation of values or lenghty tweaking it makes no differense.
So thanks for the numbers guys! (the 50kHz in particular)
So, I've been comparing the different suggestions from the different pepole, and there seem to be some disparity between them,
For example for the 50Hz there's -16,96 from Jocko Homo, -16,944 from EC8010, and -17,03 from the excel at Bonavolt's site...
Ive got a -2,50 at 2000Hz, and -2,61 (Jocko Homo) and -2,583 (EC8010)
But than again, theres probably a roughly 0,1dB difference max between the three, Where Jocko Homo's seems to be the "brighter" one, Bona's the "darkest" one and EC8010 in the middle. (but closer to Bona's)
I guess it would be quite meaningless to start making calculations for a filter and then worrying about a 0,1dB difference, but I really want to be shure that I'm using THE one riaa cure as a reference. (despite the fact that in real life my actual curve will NOT be so close to the calculations)
Anyway it is interesting to observe that Jocko Homo's and Bona's equations are almost identical. Why are they not?
Come to think of it, it really IS meaningless to compare the different curves, it really is more important that the final amp plays music that pleases the ear. If I get there by intense calculation of values or lenghty tweaking it makes no differense.
So thanks for the numbers guys! (the 50kHz in particular)
Mine came from the RIAA.....
So, I assume that it is accurate.
I have trimmed RIAA networks to +/- 0.05 dB, relative to the curve using an FFT with fancy interface. A bit of a pain, even with a passive network. (Feedback one would be harder.)
Most RIAA networks have a hump of about 0.25 - 0.5 dB in the 250 - 500 Hz range, which is hard to take out. Most of our customers prefered the one with the hump.
Jocko
So, I assume that it is accurate.
I have trimmed RIAA networks to +/- 0.05 dB, relative to the curve using an FFT with fancy interface. A bit of a pain, even with a passive network. (Feedback one would be harder.)
Most RIAA networks have a hump of about 0.25 - 0.5 dB in the 250 - 500 Hz range, which is hard to take out. Most of our customers prefered the one with the hump.
Jocko
It made the bass sound fuller. One dealer was especially upset, until I pointed out that the CD of the LP he loved so much didin't have that flabby bass sound.
After I cured him of that bad habit, I built 2 identical RIAA preamps: except that one had his favourite mystery wire du jour, and the other with wire that was neutral.
"You can keep that one, it sounds all nasty and rolled off in the top end. What is the difference between them?"
Opened the lid, and showed him. It was weeks before he talked to me again.
Little weasel that he is.............
Jocko
After I cured him of that bad habit, I built 2 identical RIAA preamps: except that one had his favourite mystery wire du jour, and the other with wire that was neutral.
"You can keep that one, it sounds all nasty and rolled off in the top end. What is the difference between them?"
Opened the lid, and showed him. It was weeks before he talked to me again.
Little weasel that he is.............
Jocko
Hmmm, I'm not really sure I know what you mean? for the first example I used 50Hz for ALL the three numbers, and 2000Hz for all the next numbers...
I DID notice that at 50Hz, I wrote -xdB, it should of course be +.
Very interesting about the "hump" in the 250-500Hz area, though I can't say that I've spotted it...
The phonoamp that I have buildt is quite sharp in the high, so much that my ears get tired. It needs serious trimming, so I've been doing some hefty simulating (yeah yeah) to see what components I should be trimming. (aint got any fancy measuring stuff
) Mine does not invorporate the 3.18us time constant for the moment.
Simulations showed that it had a -18,48dB at 20kHz an -1,97dB at 2000Hz so that'll explain it. (if one would trust the software
)- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.