InverseRIAA Generators

LT spice, Curve Captor, PSUDII and whatever other sims you can think of.

Post Reply
sbench
Posts: 296
Joined: Fri Jan 28, 2005 3:45 pm

InverseRIAA Generators

Post by sbench »

I noticed we have several essentially identical Inverse RIAA generators being used, so I thought I'd do a sort of universal one. Place the attached INVRIAA.ASY file in the spice library symbol directory (under "special functions" is a good choice), and the attached INVRIAA.TXT file in the "sub" directory.

I have implemented both with and without the 50kHz pole generators. One set has a built in AC and sine generator, so all you have to do is hook up the output you want. Another set has an input so you can use your own source to drive the inverse RIAA generators.

You can use either or both outputs or either or both types of generator. If you don't want to use something, simply don't hook anything to the pins. (Yes, she thought to internally terminate the generators so you can leave open pins).

The inverse RIAA generators are scaled so that there is "unity gain" at 1kHz. So if you put in a 1mV signal, you will get 1mV at 1kHz.

The ones with the internal generator produces 1kHz, 1mV if you do transient analysis, and 1m "gain" for AC analysis. (So that at 1kHz, with no GAIN, you will get -60dB at 1kHz. In other words, the output is scaled to 0dB = 60 dB gain.

Enjoy,
Stephie <3
Attachments
invriaa.asy
NEW VERSION. Slightly cleaner than the one I uploaded before.
(1.47 KiB) Downloaded 578 times
INVRIAA.TXT
(583 Bytes) Downloaded 430 times
Last edited by sbench on Mon Dec 12, 2011 10:18 pm, edited 2 times in total.
HiFiTubeGuy
Posts: 1
Joined: Mon Dec 12, 2011 8:33 pm

Post by HiFiTubeGuy »

Very nice!

Thanks Stephie; for this and all of your excellent contributions; very cool! :D

Kerry
docali
Posts: 58
Joined: Thu Oct 16, 2008 7:16 am

Post by docali »

Hi Stephie,

cool thing!

For those of us who needs more digits accuracy I found in german book about Spice simulation that the normalization factor is 0.101 :wink:

LOL

Best regards!
sbench
Posts: 296
Joined: Fri Jan 28, 2005 3:45 pm

Post by sbench »

The 0.1 estimator is ever so slightly low, as you pointed out. (by .089 dB). If you want it EXACT, then the normalization should actually be 0.10105. You can set it up in LTSpice easily enough... just do a frequency response (AC) analysis from 999 to 1001 Hz, using 1000 or 10000 steps. The 0.1 factor (embedded in the text file as a divide by 10 in the behavioral sources) can be changed to *.1 or *.101 or *.10105. You'll find the .10105 to go thru -60.0000dB at 1.000kHz.

Of course, theres always the 3180 vs 3183 debate :D

Stephie <3
Post Reply