diyAudio

diyAudio (http://www.diyaudio.com/forums/)
-   Analog Line Level (http://www.diyaudio.com/forums/analog-line-level/)
-   -   A better formula for Silonex LDR's (http://www.diyaudio.com/forums/analog-line-level/176706-better-formula-silonex-ldrs.html)

jackinnj 5th November 2010 05:29 PM

A better formula for Silonex LDR's
 
I plugged in about 100 data points for Current (x) and ON-Resistance (y) for the Silonex NSL32SR3 Attenuator -- the errors for the very low currents are close to zero --

http://www.tech-diy.com/Preamplifier..._Predicted.gif

This is much better than the results using the Silonex equations.

Here's the equation - ymmv:

http://www.tech-diy.com/Preamplifier...x_equation.GIF

AndrewT 5th November 2010 05:46 PM

Can I assume that e5 = 10^5?

kristleifur 5th November 2010 05:48 PM

Quote:

Originally Posted by AndrewT (Post 2355435)
Can I assume that e5 = 10^5?

Yep, that is the notational custom - as I know it at least :)

jackinnj 5th November 2010 05:55 PM

I liked it because there is no exponentiation, and only a few terms. The equations with multiple terms offer no improvement in accuracy. A stepwise approximation can probably resolve the low ohms error.

udailey 5th November 2010 06:38 PM

How many different LDRs is this based on John? One with 100 measurements or several with several measurements? Yeah, their datasheet is nothing close to real use.
Uriah

Blues 5th November 2010 08:07 PM

Nice one Jack...btw, x=current in uA. Actual testing/matching is still the way to go but it can serve as cross reference.

Blues 5th November 2010 08:17 PM

On second thought...your formula is pegged down at 47R...it certainly can go lower:)

jackinnj 8th November 2010 01:03 AM

Quote:

Originally Posted by Blues (Post 2355563)
On second thought...your formula is pegged down at 47R...it certainly can go lower:)

the very low resistance points are the insertion loss.

do the math for a 0.5dB 100K+100K attenuator -- you need accuracy in the tens of picoamps in the part of the range where we do most of our listening.

Blues 8th November 2010 05:08 AM

Quote:

Originally Posted by Blues (Post 2355563)
On second thought...your formula is pegged down at 47R...it certainly can go lower:)

And can also go much higher, I might add, and controllable too.

Quote:

Originally Posted by jackinnj (Post 2357836)
the very low resistance points are the insertion loss.

do the math for a 0.5dB 100K+100K attenuator -- you need accuracy in the tens of picoamps in the part of the range where we do most of our listening.

Not quite...with a more precise power delivery "engine" ;). Listening range is mostly in 10s of uA to 100s of uA. With this design working range is from <1uA thru 10s of mA.

I'll probably start a thread...

jackinnj 12th November 2010 03:19 PM

Several forms of equations will work -- but I have to say that the Silonex devices are all over the place!

Here's my setup -- using an LTC or ADI difference amp allows stable current in the picoAmperes:

http://www.tech-diy.com/Components_P...Meas_Setup.gif


All times are GMT. The time now is 10:43 AM.


vBulletin Optimisation provided by vB Optimise (Pro) - vBulletin Mods & Addons Copyright © 2014 DragonByte Technologies Ltd.
Copyright 1999-2014 diyAudio


Content Relevant URLs by vBSEO 3.3.2