John Curl's Blowtorch preamplifier part II

[BV]

gerhard
http://www.diyaudio.com/forums/showp...83&postcount=9

[syn08]

Originally Posted by gerhard

The test board

... and now back to paid development work

Off topic, I'm really surprised, since when is Maxim doing LT1028? There's no trace of that in their catalogue, though your picture seems to show a genuine Maxim part...

[john curl]

Gerard, I don't know how you did it, but you completely missed what we were arguing about. At first I thought you were clarifying the situation, but you didn't put in the IM byproducts in actual contention. You appear to have ignored them.

Please everyone, the two most obvious 'pips' on the graph that really depart from being an IM byproduct are alleged to be: 2F2-8F1, and 2F2-6F1.

They compute to be: 4.56 KHz for 2F2-8F1, and 10.92 KHz for 2F2-6F1.

Do they appear on the actual test (fig 3) to be these frequencies?
Did I do my math wrong?

This is the actual problem.

[ikoflexer]

Originally Posted by gerhard

The questionable frequency pair in high resolution:

Squarewave = 3180 Hz ==> 1380 Hz
Squarewave = 3210 Hz ==> 1110 Hz

IMD product wanders 270 Hz down for input step of 30 Hz

=====> This is IMD with 9 * square wave frequency.

Methinks this argument is settled.

The not so funny thing is that everybody with a tone generator and a soundcard
could have done this in one evening, but nobody did it.

What a breath of fresh air! Nicely done!

[Steve Eddy]

Originally Posted by john curl

Please everyone, the two most obvious 'pips' on the graph that really depart from being an IM byproduct are alleged to be: 2F2-8F1, and 2F2-6F1.

They compute to be: 4.56 KHz for 2F2-8F1, and 10.92 KHz for 2F2-6F1.

Do they appear on the actual test (fig 3) to be these frequencies?

They don't appear to, no.

Even though the peaks in fig. 3 are skewed such that they don't match up with the actual numbers, when 2F2-8F1 and 2F2-6F1 are added to the graph I made previously, their positions are inconsistent with the skewing of the peaks in fig. 3.

When the graph is scaled to match fig. 3 from 0Hz to 15.9kHz, the lower frequency peaks in fig. 3 gradually skew further and further to the left of the peaks of the graph.

However when I add 2F2-8F1 and 2F2-6F1, instead of being to the right of the peaks in fig. 3, they are to the left.

So in spite of the skewing of fig. 3, it's rather evident that the two peaks of interest here are NOT at 2F2-8F1 and 2F2-6F1, but rather are higher in frequency.

Unless of course your math is wrong.



se

[Steve Eddy]

By the way, when I made the graph, I did so assuming that the leftmost point on the X axis in fig. 3 was 0. I got to thinking that may have not been a good assumption. So I've since scaled the graph between 0.9kHz and 15.9kHz and that gives a near perfect fit and it appears that the leftmost point in fig. 3 is 20Hz.

Now that the graph is properly scaled, I can get a much better estimation of the frequencies of the two peaks in question, and they're at approximately 5kHz and 11.3kHz.



se

[john curl]

Thanks Steve, and if the frequencies don't fit, we have to look elsewhere for the cause. How about FM modulation?

[Steve Eddy]

Originally Posted by john curl

Thanks Steve, and if the frequencies don't fit, we have to look elsewhere for the cause.

Or as the late Johnnie Cochran might have said, "if the frequencies don't fit, you must acquit."

How about FM modulation?

I'll let y'all fight over that one. I just wanted to try and get to the bottom of what exactly was represented in fig. 3.

se

[john curl]

That's fair, and I couldn't spell 'acquit' so I left it out.
I did note that the unknown frequencies (at least 4) appear to be symmetrical about the fundamental and the 3'rd harmonic, respectively.

[janneman]

Originally Posted by Steve Eddy

Or as the late Johnnie Cochran might have said, "if the frequencies don't fit, you must acquit."



I'll let y'all fight over that one. I just wanted to try and get to the bottom of what exactly was represented in fig. 3.

se

Steve, I had an appointment with Jan Lohstroh today and took the opportunity to go through his files from way back. Came up with this:

http://www.linearaudio.nl/Miscellane...poo%201976.pdf

I haven't gone through it in detail, just scanned and posted it, but it does offer some other perspectives on the DIM method and results.

jd