[DrewWinters]

I was wondering if it was possible to burn the RMAA test tones to a CD and then use the input on your computer sound card to run the tests.

I am curious if it is possible to measure the differences between some popular "tweaks" and measure the differences in tubes (I have the PPX3 tube rolling pack right now). But I wanted to ask around before I try to mess with this. So, if the answer to my first question is "yes", is there a reason that this is a bad idea? I know that it isn't a good idea to run the PPX3 without headphones hooked up (something about impedence I am sure) - but could it be run with its output hooked up to the input of a soundcard without problem? (This may be something I need to ask Mikhail) I know the test tones are very loud - would this setup put the tubes in any danger? I certainly wouldn't mess with someone else's tubes unless I was sure that it wouldn't...

Lastly, I am aware that the inputs on the sound card would affect the results; but it would affect all results in the same way - meaning that comparisons should be possible. Right?

 

[breez]

Yes it is possible. The signals are loud, but there's no more possibility for damage than listening to a loud CD (fullscale signal is only fullscale signal, doesn't go any further than the maximum voltage from the cd-players which is probably 2Vrms or something).

You could hook headphones as a load with a splitter, although I'm not sure if this is a valid testing method (have seen these kind of benchmarks done with RMAA over at the DIY forums). Just try not to feed extra sensitive headphones too much voltage.

 

[Clown12]

I've measured lots of stuff this way, tho there is a knack to clicking 'record' at the right percentage

The input side does have an effect, try recording the same signal with your onboard sound's line-in (or a 2nd pc) to see the differences.

 

[DrewWinters]

I am working on it now. I picked up a 1/4" - 1/4" cable today (I'll use the 1/4"-> 1/8" adapter that came with my 650's) and do some testing. I have a baseline test stored from my soundcard's output->input. However, the sound card is affected that signal twice, while it will only be affecting the incoming signal once. I will try to also make a baseline from my DVD player, bypassing the amp, then start testing....

Mikhail has informed me that running the amp without headphones attached is only a bad idea if done for long periods of time: my interpretation would be that the few seconds it takes for this test - and at the VERY low volumes from the amp (otherwise I toast my sound card probably) - that it would not be an issue.

 

[DrewWinters]

Well, I've run a test of several different tubes... the measured differences are, not surprisingly, very minor. In fact, I'd speculate that they are WELL within the margin of error on a test like this. For instance, here is the frequency response graph from the first round of tests...

You'll notice that the differences in frequency response are TINY. The dramatic slopes at the upper and lower ends of the spectrum are not only drastically exagerrated by the scale of the graph - but were also present in my baseline test with my soundcard - in fact, they were worse, which, not surprisingly, indicates that my source and amp are more linear than my soundcard.

So in summary - absolutly nothing interesting going on at all. Any theories as to why the tubes *sound* like they have a different response curve without any measurable difference? Sure, one could sound better than another with the same curve - but it sure sounds like the Cleartops have more treble, etc.

EDIT: The reason for the "warm" designation on the GE tube in the graph is a leftover from an earlier test where I tried the same tube a minute or two after turning it on and after about half an hour (don't worry - I plugged my headphones in to warm them up... didn't leave it "naked")... The only substantial differences in measurement was in intermodulation - the warm tubes had about half the intermodulation.

 

[gaboo]

Can you post an IMD graph? TIA.

 

[DrewWinters]

Here you go... this is the four tubes I tested - NOTE: I have only tested these ONCE.

 

[DrewWinters]

Oh, I forgot to specify, the difference in cold and warm tubes was in intermodulation - not intermodulation distortion.

 

[DrewWinters]

Here's a bigger comparison between Ei, GE, and Tung Sol tubes + my baseline test with just the soundcard. This is just frenquency response.