Noise from Ultra A16 - can someone please help by testing

orbita · Post by **orbita** » Fri Sep 20, 2002 9:07 am

Hi
Ive had noise issues with my Ultra A16s for ages now and I would appreciate it if someone could test their setup to see if I am simply expecting too much or whether there is actually a problem with my units or card. I've even been suggested that my power supply might be the cause of the problem.

So if you have a few minutes to do this simple test and report your results I would be extremely grateful..

Pick a channel on your A16 with nothing connected to its audio input. Connect ONLY that channel to the 2448 mixer using ADAT Source or Zlink (Please do the test for both if you can). You should find that the bottom green led on the channel lights to show there is some signal.

Now, insert a gate on the channels first insert. Set the Attack to 0, the Release to 0.2, the Hyst to 0, the Floor to -96, the Threshold to -96 and the Gain to 0.

Now adjust the threshold gradually until the gate kicks in - the blue bar will appear and if you are listening, your should hear the noise drop.
Adjust the threshold until the blue bar is flickering on/off.

If you could do this using -10 and +4 settings and using ADAT and Zlink.

Just so you know your in the right ball park, I'm mostly getting values of between -75db and -85db. This seems rather noisy for a unit that is supposed to have a 105db signal:noise ratio and I'm you can get close to -100db of noise.

thanks in advance

[ This Message was edited by: orbita on 2002-09-20 10:10 ]

Post by **garyb** » Fri Sep 20, 2002 12:00 pm

on the 2448,you SHOULD see the bottom green led lit when something is connected to that channel.

kimgr · Post by **kimgr** » Fri Sep 20, 2002 12:36 pm

You should also connect something to the input... Without connections, your a/d inputs are nothing more than strange looking radio antennas !!!

Kim.

Steve-o · Post by **Steve-o** » Fri Sep 20, 2002 1:51 pm

n1 Kim!

orbita · Post by **orbita** » Fri Sep 20, 2002 2:04 pm

hmm strange looking radio antennas are cool, in fact I collect them!

but seriously I was just making the point that the noise wasnt due to my synths or cables as thats what people normally suggest but im sure you realised that

kimgr · Post by **kimgr** » Fri Sep 20, 2002 5:36 pm

I'm (hopefully) installing my new A16 Ultra next week, so I can test it then.
In the meantime I could test my Fostex VC-8's !!! Anybody interrested ???

Kim.

_________________
Design & Media
http://www.d-media.dk

[ This Message was edited by: kimgr on 2002-09-20 18:37 ]

orbita · Post by **orbita** » Fri Sep 20, 2002 7:20 pm

hehe

I really quite suspicious of these Ultra A16s now. Sometimes you have to wiggle the cables to get a good connection too. I might see if I can borrow an RME convertor and see if they have better SNR. They are a bit pricey though.

Post by **garyb** » Fri Sep 20, 2002 8:51 pm

chasing dragons!my frontier converters behave in the same way.also,those are balanced connectors.are you SURE the cables are wired correctly?(if you're using unbalanced cables,you must rtm as they must be wired properly)

kimgr · Post by **kimgr** » Sat Sep 21, 2002 4:35 am

I went ahead and tested the Fostex's anyway:
First of all the gate isn't exactly a precise measurement instrument, but for comparison in this case we'll just ignore that...

A) With nothing connected to the Fostex, the gate won't even open. The input peakhold reading is about -85.5dB.

B) With Ensoniq TS-10 connected, but switched off, the gate opens at -93.4dB, and the input peakhold reading is about -83dB.

(Guess the threshold is calculated using RMS, not peak).

Kim.

orbita · Post by **orbita** » Sat Sep 21, 2002 5:05 am

Kimgr,
Interesting to see it was gating at -93db. That shows that you have a lower noise floor with the fostex that I get with my Ultra A16. Your right, the gate method is pretty inaccurate but I couldnt think of an alternative that didnt involve recording and and analysing a waveform - got any ideas?

Garyb,
I get the same results whether I have cables plugged in or not. The cable are hosa balanced 8 way snakes which should be fine.

Retro · Post by **Retro** » Sat Sep 21, 2002 5:24 am

Interesting exercise Orbita. I agree with Kim that the gate is not exactly a precision instrument, but here are the results from my A16 Ultra:

With nothing connected (antenna mode!) the gate was kicking in at -83db (-10V input) and -81db (+4V).

With a Juno 60 connected (but not switched on) the threshold reached -82db (-10V) and -78db (+4V).

As you can see they are only slightly better results than yours, but I'm sure an oscilloscope would tell a different story. From experience I can definitely say that the SFP devices are far from accurate. Note also that I could only test with ADAT links as my cards don't have Z-link.

Cheers,
Retro

orbita · Post by **orbita** » Sat Sep 21, 2002 11:31 am

Yep, having tried a couple of times its really not very accurate since there is a range of several db where the gate kicks in by different amounts. But still it gives a general idea and your results suggest that you are getting similar noise levels to myself. Its interesting that if you look at the stats on the Ultra A16 in the manual and on the website they never quote Signal to noise ration. Only dynamic range. Other manufacturers like RME quote SNR values of 110+ dbA. It would certainly be interesting to compare.

Post by **garyb** » Sat Sep 21, 2002 12:29 pm

this method does NOT give you s/n RATIO.it only suggests the level of the noise FLOOR.all electronic devices have SOME noise.(listen to jimi hendrix kick in distortion boxes on axis:bold as love.)this minor noise should not cause problems.(do what you like tho!)

Retro · Post by **Retro** » Sat Sep 21, 2002 9:54 pm

Chill out gary

No one said this experiment had anything to do with S/N ratio. I think Orbita brought it up as a seperate but vaguely related issue.

[ This Message was edited by: Retro on 2002-09-22 02:23 ]

mano · Post by **mano** » Sat Dec 07, 2002 2:26 pm

On 2002-09-20 21:51, garyb wrote:
chasing dragons!my frontier converters behave in the same way.also,those are balanced connectors.are you SURE the cables are wired correctly?(if you're using unbalanced cables,you must rtm as they must be wired properly)

Actually the A16 manual states that the "Ultra" version accepts both BAL (TRS) or UNBAL connections.

This said, I also have a ton of noise on my A16 Ultra. I know I must be doing something wrong, but frankly I tried everything and there is always too much noise... Even if I use a gate in front of each channel... It does help, but when the gate opens to let the notes through, there's quite a noticeable amount of SCHHHHH.

Some sounds give better results if they are played loud (and I mean pound on the keys) but I feel that it shouldn't be that way. And if the sounds have a long release, forget about it!

I get -72db in silent parts, which some would say is not bad, but when the notes come in they barely cover the noise which could be heard at -72db.

sucks

Maybe my sources are supposed to be louder, and thus I am not supposed to "crank" it up enough to hear the -72db noise floor?

If you are a sound engineer feel free to post

[ This Message was edited by: mano1.com on 2002-12-07 19:50 ]

mano · Post by **mano** » Mon Dec 09, 2002 3:32 am

Wow forget my previous post!

All I needed was 3 good days of non-stop tweaking and learning! I've solved all my problems, the sound is excellent.

My drums are now loud and snappy, my choirs are crisp and clear, etc. Thanks to you all for your great posts!

... and Vinco + Psy Q rock! So does Optimaster. Excellent. The A16U works great

Of course I still need to figure out a lot, but at least I see what direction to take.

[ This Message was edited by: mano1.com on 2002-12-09 03:37 ]

orbita · Post by **orbita** » Mon Dec 09, 2002 3:50 am

feel free to leave some hints tips on what you learnt while tweaking so we can gain from your experimentation. thanks

mano · Post by **mano** » Mon Dec 09, 2002 4:13 am

On 2002-12-09 03:50, orbita wrote:
feel free to leave some hints tips on what you learnt while tweaking so we can gain from your experimentation. thanks

Sorry didn't mean to sound like an ass

I am in the middle of getting it right, I thought I'd share when I am sure of the why and what

The main thing that helped with drums was the "VST Dynamics" device. Drop this puppy on top of each audio track, and enable the AUTO GATE. Play with the settings there. Enable the AUTO LEVEL and SOFT CLIP, etc.

On my master bus, I've chained a PSY Q with some subtle settings (careful this one is addictive

) and a VINCO S.

Of course this is "obvious" stuff (gate, compressor...) but for some reason they never appeared as efficient than during the last few days (as I said I took a few days off to focus on music)...

I still can't get a clean piano out of my XV, "SCHHHH" each time a note is let thru. Oh well, for now I'm loading one in KONTAKT and it sounds great... Until I get better at capturing my XV's.

I'll get back to it now

piet piel · Post by **piet piel** » Tue Dec 10, 2002 8:23 am

Helluw y'all,

I don't have the Ultra A16 but the previous version, the 'normal' Creamware A16 converter. I have it mounted above my CD-player. When I have the CD-player turned on, the noise from the converter gets much louder. I have balanced cables. So my converter itself is not interference-free. If the Ultra isn't, too, make sure it's mounted free of any interference field.

Dunno if it's mentioned but off course you should make the input for the converter as loud as possible. So don't make your mix BEFORE the converter but AFTER the converter.

Hope diz helps.

Regards, Manoah