where to test pre amp voltage

[frankiebones 10+ year member]

i have a pioneer deh-p7800mp that advertises 5v preamp outs. i was wondering if anyone knew at what volume (i.e., ##/60) the HU got its cleanest output.

could i test the voltage by the amp by putting the test pins of my DMM to the RCA cables?

Click to read more...

 

[thch 10+ year member]

there is limited testing that can be done.

basically, play a 0dB test tone and a -3dB test tone.

at each volume setting record the result. (or every other volume setting)

for each result, find a logarithm. (eg, log(x) or ln(x)). for this case, it won't matter what type of logarithm you choose.

so now you will have two sets of real values, and two sets of logrithms of these real values.

take the log(values) from the first run (0db) and subtract the log(values) from the second run (-3dB). for each test setting, the difference should be equal, up until there is clipping on the output. at that point, the difference should start to decrease.

eg, if there was a difference of 0.5 for settings 1 to 20, then at setting 21, log(test@0db) - log(test@-3dB) was less then 0.5, clipping has probably started at setting 21. (0.5 is really arbitrary. the numbers you are getting will seem arbitrary, but increasing)

also, on a peice of graphing paper, plot the logrithmic plots vs setting. eg, the X-Axis on the bottom is setting #, and the Y-Axis on the side is Logrithm(value@this setting). plot both traces and you will get mostly straight lines.

this should trace out a straight line for both cases, up until clipping sets in, where the traces will get closer together.

 

[frankiebones 10+ year member]

there is limited testing that can be done.
basically, play a 0dB test tone and a -3dB test tone.

at each volume setting record the result. (or every other volume setting)

for each result, find a logarithm. (eg, log(x) or ln(x)). for this case, it won't matter what type of logarithm you choose.

so now you will have two sets of real values, and two sets of logrithms of these real values.

take the log(values) from the first run (0db) and subtract the log(values) from the second run (-3dB). for each test setting, the difference should be equal, up until there is clipping on the output. at that point, the difference should start to decrease.

eg, if there was a difference of 0.5 for settings 1 to 20, then at setting 21, log(test@0db) - log(test@-3dB) was less then 0.5, clipping has probably started at setting 21. (0.5 is really arbitrary. the numbers you are getting will seem arbitrary, but increasing)

also, on a peice of graphing paper, plot the logrithmic plots vs setting. eg, the X-Axis on the bottom is setting #, and the Y-Axis on the side is Logrithm(value@this setting). plot both traces and you will get mostly straight lines.

this should trace out a straight line for both cases, up until clipping sets in, where the traces will get closer together.

WOW! any links anybody has for anything detailing this process?

 

[AJ²06 5,000+ posts]

there is limited testing that can be done.
basically, play a 0dB test tone and a -3dB test tone.

at each volume setting record the result. (or every other volume setting)

for each result, find a logarithm. (eg, log(x) or ln(x)). for this case, it won't matter what type of logarithm you choose.

so now you will have two sets of real values, and two sets of logrithms of these real values.

take the log(values) from the first run (0db) and subtract the log(values) from the second run (-3dB). for each test setting, the difference should be equal, up until there is clipping on the output. at that point, the difference should start to decrease.

eg, if there was a difference of 0.5 for settings 1 to 20, then at setting 21, log(test@0db) - log(test@-3dB) was less then 0.5, clipping has probably started at setting 21. (0.5 is really arbitrary. the numbers you are getting will seem arbitrary, but increasing)

also, on a peice of graphing paper, plot the logrithmic plots vs setting. eg, the X-Axis on the bottom is setting #, and the Y-Axis on the side is Logrithm(value@this setting). plot both traces and you will get mostly straight lines.

this should trace out a straight line for both cases, up until clipping sets in, where the traces will get closer together.

THIS IS sweet //content.invisioncic.com/y282845/emoticons/tongue.gif.6130eb82179565f6db8d26d6001dcd24.gif

 

[frankiebones 10+ year member]

actually, before i read your reply, i tried to see if i was getting any sort of reading from the RCAs and couldnt using a pretty high end DMM on a very low range setting.

also, what frequency should my test tone be?

 

[frankiebones 10+ year member]

can preamp V only be measured at the back of the HU?

 

[frankiebones 10+ year member]

found this:

http://www.carstereo.com/forum/showthread.cfm?tid=10401

 

[thch 10+ year member]

i don't think anyone uses the method i posted. either that or it is too time consuming or hard to understand or hard to convey. it also assume that the HU manufacturer doesn't have signal processing to defeat this, and that they use an exponentially increaseing signal output (which should be common).

It should be trivial to make a speadsheet so one would put in test1, test2, and have it calculate log(test1), log(test2), log(test1)-log(test2)

 

[dkguitarist 5,000+ posts]

i never did get logarithms in hs. i failed horribly at them. //content.invisioncic.com/y282845/emoticons/crap.gif.7f4dd41e3e9b23fbd170a1ee6f65cecc.gif