Presonus 22VSL(44vsl 1818vsl etc.) Audio Interface

I just opened this guy out of curiosity. I have also two of 1818vsl for location recording. I have always been satisfied with them and never got why some people regard them as very low end units. Maybe just Presonus' reputation?

Never had an issue with preamps, sound quality, reliability, and they have had near zero latency FX monitoring for a long time now which was the reason i started using them in the first place.

Yes, pres could be a tad quieter, but that never got in the way. Rarely use ribbons, and i use dynamics on very loud sources.

I bought this guy in 2011. Maybe they can be upgraded, it looks like a cool platform for modding since it's not all smt? It uses Jamicon and Elna electrolytics in the signal path, Cirrus cs4272 converters Apogee uses...

Maybe someone got a schematic?
It's their xmax pres. I did find Digimax schematics, but they don't have the same compononts.

Vsl1818 with 8 pres can often be bought used for peanuts.

Anyway, just wanted to repport....

Angle

Opened 1818. It's all Jamicons. WL(low impedance), TK(wide temperature range), NK(just regular ones).

There are some smd electrolytics beside converter chips.

So i was wondering what could i use to upgrade them. Especially TK and NK (blues anf blacks) in the preamp area? I guess WL are ok?

There is room for regular caps instead of smd beside converters.

Any suggestions?

Converter area

Cap goes to pin called Filt+ (Positive reference voltage for internal sampling Circuits).
On CS4272 datasheet in typical Connection diagram 47uF is bypassed with 0.1uF, which is not the case here, as far as I can see.

http://alsa-project.org/~james/datasheets/www.cirrus.com/en/pubs/proDatasheet/cs4272-1.pdf

Kind of nuts to keep talking to myself here, but i guess someone might find usefull things i post.

These XMax preamps seem to be all over the place design and component wise. The same pres should be in Digimax, but they are not the same. Voltages in VSL44 and VSL1818 don't come near 30V. They are more like 3V around the transistors. Also older Digimax used an IC before transistors, these newer use two pairs of transistorss. 2C smd NPNs and 2N4403 PNP.

Today i tried a kind of Blacklion mod to add bypass caps to all electrolytics, and that didn't do a thing in regards to noise or distortion. I even removed phantom blocking caps (bypassed them, the Elna ones on vsl44) and that didn't do a thing either.

Might try to replace all the electrolytics with something fancier as there aren't many in vsl44, i'll repoort if anything happens.

Does anyone know what "propriatery converter decoupling" might be. It's an advanced mod Black Lion does. It is supposed to minimize digital noise in audio path, if that makes any sense.

All i see on a really bad screenshot of their digi 002 mod is some joined pins on converter chip and a bunch of electrolytics around it. The ones marked with blue arent there on stock one.

You could take some REW spectra snapshots before any changes for reference later.

I mean, if the perf measures out really good already, there might be no need to change, except of course, if you want to.

I would expect it to measure up pretty good; if it is working properly, it should be clean and clear. I put up some basic characteristic snaps of a few units here somewhere if you are interested.

I found some of the older models I have (Motu from around 2005 or so) develop some mains hum issues -  probably filtering electros in the psu gone old. They develop some 100Hz spike not present when new.

Apart from that sort issue, the main difference between models are reasonably minor ...  perhaps a little more or less THD at different freqs, perhaps slight differences in the 'hiss' profiles.

Nothing much to measure different in freq or phase, generally speaking. I suppose some can hear differences in the sound of different converters and their associated circuitry - except in some few cases, I can't say that I can discern.

Anyway - have fun; see what you see and listen to what's what!  People often like reading this stuff but don't say much.

Thanks man. Did take snapshots with REW. I guess 002 had some issues and could be resolved. But now i see they upgrade latest Presonus interfaces, have no idea...

kingkorg said:

On CS4272 datasheet in typical Connection diagram 47uF is bypassed with 0.1uF, which is not the case here, as far as I can see.

Click to expand...

Hard to tell....doesn't look like it but, sometimes it's hard to see????At least with multi layer boards... Easy enough to test with continuity on some of those caps to find out..... ? I think  sometimes things aren't done "by the book" for various reasons too...

MC33079 op amps??  I like those..... Kinda sloppy bass (dual 078) in some old converter circuit I've tried them in with my set up but I like the sound over all....

interesting topic....Curious too about the converter decoupling....Kinda vague info....read somewhere they just upgrade the caps on the converter rails but who knows.....

As far as i can tell by checking continuity this thing is pimped up version of Behringer ADA8000 circuit. Although i am sure this design originates from some earlier device.

Presonus version has quieter components and way less THD IC. It has PNP/NPN pairs instead of NPN/NPN. Direct in doesn't go through the preamp.  So some of major ADA8000 issues are addressed,

If i am not wrong, coupling capacitors between discrete stage and IC are two 3.3uF non polarized electrolytics. So i guess something better could be used here. I might try upgrading these and see what happens.

Otherwise i don't think there can be done much more without some serious pre redesign. Maybe use a transformer and go directly to IC 33179.