An alternative to the 2520 for 312 projects - Test Results

[Flatpicker]

[quote author="Category 5"]...I only hope my tests are accurate enough to draw meaningful conclusions...[/quote]What you say you heard speaks volumes to me!

 

[Bear]

Can anybody give me a read for where the Yamaha discrete opamps fall? I've got five of them I can put to use. Fun with solder suckers.

Bear

 

[Category 5]

Not familiar with the Yamahas. Can you give more information? Picture?

you can get the SCA 2520s at

http://www.seventhcirclestudios.com/SCA/SCA.htm

 

[Bauman]

Very nice work Shane!
I'd love to have a original 2.5.2.0. to compare to my 3 different boards and circuits, but I don't...

All I can tell you, soon, is how different they sound from each other...

Hey Shane, I can send you some boards, are you in to stuff them and compare?

cheers!
Fabio

 

[Category 5]

[quote author="Bauman"]Very nice work Shane!
I'd love to have a original 2.5.2.0. to compare to my 3 different boards and circuits, but I don't...

All I can tell you, soon, is how different they sound from each other...

Hey Shane, I can send you some boards, are you in to stuff them and compare?

cheers!
Fabio[/quote]

Bring it on! I'd love to get ahold of some of your boards. I inquired a couple months ago, but you had said they weren't ready for prime time yet.

I'd like to get as many otions as possible and run some more formal tests. For one, I am looking for the best way of getting my soundcards output attenuated, and to appear as 150ohm source to the preamps. This should give me slighlty more meaningful frequency curves. Also, the way i am doing it now the noise and dynamic range measurements are useless. I hope to change that soon.

Shane

P.S. FWIW (OT) I tested my modified Presonus MP20 (Jensen transformers and AD797AN chips) and it is literally FLAT from 25-18k with a .5dB peak at 20Hz and a .2dB drop at 20k. I am really convinced my mods improved the heck out of that sucker.

 

[Category 5]

Well...very different things are happening when the preamps are loaded wih a 150ohm input instead of just patching the high impedance sound card output into the mic pre.

the 2520 and SC25 are like twins in frequency response. They only differ in THD, which is quite similar to say the least. Have a look!

Full Spectrum

http://www.pbase.com/sbushman/image/35932471/original.jpg

Highs

http://www.pbase.com/sbushman/image/35932472/original.jpg

Lows

http://www.pbase.com/sbushman/image/35932473/original.jpg

And THD

http://www.pbase.com/sbushman/image/35932474/original.jpg

These have been corrected for my soundcard. The second channel was the soundcard patched back into itself, and after correction I get a straight line. Also, keep in mind that the scale is WAY zoomed in to show the differences. Frequency response wise I think there is no appreciable difference anymore. Of course, both 2520s and both SC25 are identical in pairs, so the small difference IS real.

If anyone want I can send the RightMark files and you can zoom in and out of all the graphs on your own.

Shane

 

[tk@halmi]

Thanks Shane, excellent information!

Tamas

 

[DrFrankencopter]

What's the deal with the THD spectrum around 2-3 Khz.....there are non integer harmonics in there. Was the amp being run into clipping for this test? Maybe that is more a function of the A/D performance of the RME card. Perhaps try running a signal intentionally hot (i.e. make it clip), and pad it down into the RME, then see if the distortion characteristics are similar.

Cheers,

Kris

 

[Category 5]

The amp was not clipping during any of the tests. If the amp was clipping the only way I could get it into the RME WOULD BE to pad it down. It has WAY more headroom than any converters I have ever used.

When i get some more time I will try padding on the way in and see what happens. Those harmonics between 2 and 3k are not likely introduced by the RME since they fall on different frequencies, and the same channel and cabling was used for both tests. They are unique to each op-amp. THD is the area of greatest difference between these 2 parts. Tim has done a good job of recreating the 2520 IMO.

Perhaps they have to do with my carbon resistor pad. I need to remake the device using metal film, and matched resistors.

I need to conduct more listening tests with sources in my booth too. I am considering patching a recording played through monitors back through a single mic, and compare the 2 gain blocks after gain matching. Using the EXACT same mic and source material can yield much more acurate listening comparisons. Even then 3d Pre Cd has a high margin of error due to performance differences.

Shane