Mackie Onyx Satellite - phantom power whine fix!

[Scodiddly]

Noticed a reference to the "Mackie Onyx phantom power whine" thing over in the classified section, and since I recently dealt with this I thought I'd (start to) post how to fix that.

I bought a second-hand Satellite recently (no hate to the seller - it sounds good without the phantom and it was a fun challenge to figure out how to fix the whine), and noticed a really annoying high frequency whining noise when the phantom was engaged. A few emails with tech support actually got me all the service info, including schematics. They've kind of abandoned the satellite at this point. Anyway, the problem (which is apparently shared by the 400f rackmount thing) is that they designed the DC-DC converter to get 48vdc (phantom power voltage) from the lower Firewire voltage by running at audio frequency. A DC-DC converter works by changing DC to AC, pushing the voltage up either by a transformer or a doubler-type circuit, and then converting back to DC. Best if you run that at a very high frequency, so it's kind of stupid for Mackie to have done this at 6KHz.

So, how do you fix this? The service docs told all, though not in a very easy form. The DC-DC circuitry is pretty crude, a 555 timer being used as an oscillator to push AC into a sort of voltage doubler/tripler/something stack - I think this may be what's referred to as a "charge pump". The 555 frequency is set by a capacitor and resistor, and the last revision (according to the docs) changed the critical resistor from 10K (might have been some 1K versions) all the way down to some 301 ohms, changing the frequency from 6K to 160K. That's a big change, but a lucky one for us solder-savvy end users. Because all you have to do is replace one resistor, and the problem is pretty much gone. The downside is that you have to deal with surface-mount, but that's the worst of it.

I'll post photos pretty soon here, but you're looking for R5 on the smaller circuit board in the Satellite. Change it to something around 300 ohms (mine is a hack, two 150 ohm leaded resistors in series because that's what I had lying around) and the frequency should go well up above the audible range. I also covered a lot of the circuit board with plastic tape and then copper foil tape to keep the noise from leaking anywhere else.

[Scodiddly]

Noticed a reference to the "Mackie Onyx phantom power whine" thing over in the classified section, and since I recently dealt with this I thought I'd (start to) post how to fix that.

I bought a second-hand Satellite recently (no hate to the seller - it sounds good without the phantom and it was a fun challenge to figure out how to fix the whine), and noticed a really annoying high frequency whining noise when the phantom was engaged. A few emails with tech support actually got me all the service info, including schematics. They've kind of abandoned the satellite at this point. Anyway, the problem (which is apparently shared by the 400f rackmount thing) is that they designed the DC-DC converter to get 48vdc (phantom power voltage) from the lower Firewire voltage by running at audio frequency. A DC-DC converter works by changing DC to AC, pushing the voltage up either by a transformer or a doubler-type circuit, and then converting back to DC. Best if you run that at a very high frequency, so it's kind of stupid for Mackie to have done this at 6KHz.

So, how do you fix this? The service docs told all, though not in a very easy form. The DC-DC circuitry is pretty crude, a 555 timer being used as an oscillator to push AC into a sort of voltage doubler/tripler/something stack - I think this may be what's referred to as a "charge pump". The 555 frequency is set by a capacitor and resistor, and the last revision (according to the docs) changed the critical resistor from 10K (might have been some 1K versions) all the way down to some 301 ohms, changing the frequency from 6K to 160K. That's a big change, but a lucky one for us solder-savvy end users. Because all you have to do is replace one resistor, and the problem is pretty much gone. The downside is that you have to deal with surface-mount, but that's the worst of it.

I'll post photos pretty soon here, but you're looking for R5 on the smaller circuit board in the Satellite. Change it to something around 300 ohms (mine is a hack, two 150 ohm leaded resistors in series because that's what I had lying around) and the frequency should go well up above the audible range. I also covered a lot of the circuit board with plastic tape and then copper foil tape to keep the noise from leaking anywhere else.

[Scodiddly]

Getting the Satellite open is a pretty good challenge to begin with. Here's how you do it: On the rear panel, remove the XLR screws and the nuts on the 1/4" jacks. On the front panel, remove all the allen head screws. You should now be able to pull the front panel free along with the innards.

Finding R5:


There should be a marking on the circuit board for "R5". The two big ugly resistors are my addition; you can order 301-ohm surface mount resistors or just do it the dumb way like I did. It's not especially easy to attach surface mount stuff, but it's a tiny bit easier to attach regular resistor(s) to the same points once you've removed the original R5. DON'T TRY THIS IF YOU DON'T HAVE A GOOD (regulated) SOLDERING IRON.

In the photo you can also see the copper foil tape. I don't know if this made a big difference or not, but the phantom power part of the circuit board is rather close to the multipin connector to the base, which includes the mic inputs. That would help explain why some versions don't have the whine when removed from the base, though.

[Scodiddly]

Getting the Satellite open is a pretty good challenge to begin with. Here's how you do it: On the rear panel, remove the XLR screws and the nuts on the 1/4" jacks. On the front panel, remove all the allen head screws. You should now be able to pull the front panel free along with the innards.

Finding R5:


There should be a marking on the circuit board for "R5". The two big ugly resistors are my addition; you can order 301-ohm surface mount resistors or just do it the dumb way like I did. It's not especially easy to attach surface mount stuff, but it's a tiny bit easier to attach regular resistor(s) to the same points once you've removed the original R5. DON'T TRY THIS IF YOU DON'T HAVE A GOOD (regulated) SOLDERING IRON.

In the photo you can also see the copper foil tape. I don't know if this made a big difference or not, but the phantom power part of the circuit board is rather close to the multipin connector to the base, which includes the mic inputs. That would help explain why some versions don't have the whine when removed from the base, though.

[Electricide]

ha, I solved this problem by buying an external phantom box. I had whine with the 4041, but not the 4040.

Scodiddly, I'd like to repost this over on the Mackie Onyx forum if you don't mind.
_________________
"Looks like I picked the wrong week to quit amphetamines"
formerly cman548

[Scodiddly]

Sure, go ahead.

[Scodiddly]

Sure, go ahead.

[Andy Peters]

[mixedupsteve]

I know it isn't the case here, but I'm just curious. Will a step up transformer even work well at 100khz+ ??
_________________
Steve

www.myspace.com/rollingrecorders

[Scodiddly]

[Scodiddly]

[sad iron]

Hey, that used to be my Satellite. I never even used it enough to know here was a problem. Good work!.

Of course, right after I sold it I had a remote session, which was why I bought it and then never used it for a year. I had to make due with a Lexicon Omega that was hissy as hell. What a terrible piece of gear. I don't know how I'm going to save those tracks. Maybe I'll buy the Satellite back...
_________________
New music: www.sadironmusic.com

Studio site: www.sadironstudio.com

[Odie]

I just saw this post about the Satellite - I have an Onyx 400F with the exact same problem and am trying to figure out a fix myself.

I see you have a great fix for the Satellite, but for some reason I can't find which is the corresponding resistor in the charge pump for the 400F. I'm looking at the schematics now, to little avail - in any case I don't see a 10k resistor here. How can I recognize the 555 timer on the schematic and the R-C that set the frequency?

Anyone who's able to help would be much appreciated!

[/code]

[Scodiddly]

Here's what the 555 timer circuit looks like in the Satellite schematics:



The frequency-setting resister is R5, which in this version is shown as 1K ohms. You might check for revision notes around the edges of the schematics.

[Scodiddly]

Here's what the 555 timer circuit looks like in the Satellite schematics:



The frequency-setting resister is R5, which in this version is shown as 1K ohms. You might check for revision notes around the edges of the schematics.

[Odie]

Thanks, man!!!
I think I am beginning to see what's going on.

From the schematics I have, it appears like the 400F is made in a slightly different way. I don't see the LMC555 thing in the Power section, but instead there's something called LM2651-3.3 which looks similar and has a resistor at 15 kOhms. But I think I'll have to contact the Mackoids first to be sure.

Alternatively, it may be that the 400F is not using a charge pump at all but is converting to +48V DC in some other way. I'll have to check.

Although, I doubt I could screw this unit up more than they have already

Many thanks for the help, again.

[Andy Peters]

[Scodiddly]

Actually I think that aside from the whine problem the Satellite is a pretty decent interface. The satellite part, anyway. The base station has pretty much been sitting in the box.

On the other hand there's this similar Presonus Inspire 1394 that I got cheap. Now there's a piece of crap - the headphone and line outs can barely drive anything without losing a bunch of bottom end, the 1/4" inputs somehow have backwards polarity, etc.

[Scodiddly]

Actually I think that aside from the whine problem the Satellite is a pretty decent interface. The satellite part, anyway. The base station has pretty much been sitting in the box.

On the other hand there's this similar Presonus Inspire 1394 that I got cheap. Now there's a piece of crap - the headphone and line outs can barely drive anything without losing a bunch of bottom end, the 1/4" inputs somehow have backwards polarity, etc.

[Odie]

Andy, if the 48V comes from a tap in the transformer, that would say that the reasons for the whine in the 400F and that in the Satellite are entirely different. From the original post I was led to believe that the problem in these 2 units is the same, but I may be wrong.

I keep getting the impression that the FireWire has something to do with it too because when it's on, the unit becomes even noisier. I will see if there's some way to post a schematic here to illustrate what I'm takling about.

[Scodiddly]

Actually I haven't looked at the 400F schematics, I had just heard that they had the same circuit and the same problem.

When I was researching this I did run across references to "Mac Firewire whine" as a different problem.

[Scodiddly]

Actually I haven't looked at the 400F schematics, I had just heard that they had the same circuit and the same problem.

When I was researching this I did run across references to "Mac Firewire whine" as a different problem.