Apogee Rosetta DC voltage at input

Help Support GroupDIY:

scott2000

Well-known member
Joined
Feb 21, 2015
Messages
2,550
Location
Sunny...Sometimes Florida- USA
I have an Apogee Rosetta200 that has -12.5 ish Vdc at it's input xlr pins 3 only. All the supply rails are where they should be everywhere as far as I can tell (-15/+15 etc). Pretty sure all various rail caps to ground are fine but it's hard to tell without starting to remove stuff.

Apogee can't provide schematics and,I was told in many cases, they can't provide service for this because of parts unavailability . They have been great and corresponded with me with some checks and I'm waiting to hear back from them on my findings .But they are really busy and it's taking a little while longer than I can justify keeping in limbo if there's another way. I'm hoping to hear something back from them this week.

Apogee hasn't said if this DC voltage is normal or not. They just asked me to check my supply rails because they said the 12v should've been 15 but, it is 15 where it should be. It's not clear if they meant, that the pin 3 should've been -15Vdc...but I wouldn't think so.

Depending on the load at the inputs, the negative dc voltage will vary. With a 5k pot for example it'll vary from almost zero to -2.5Vdc....

There is an issue that when nothing is connected to the inputs or outputs, and selecting analog as a source for digital outs, or analog as a source for analog outs, the meters will jump and stick at -4 on the front panel..not sure if it's a clue

I can use the Aptomizer function which will "learn" an input and adjust the voltage during recording. When doing this with nothing connected, the negative dc voltage at pin 3 will drop to about -7Vdc...not sure if it's a clue.

I started trying to trace the input circuit in hopes that something will stand out. I'm including a sketch as best I could do with my limited abilities and not knowing what these circuits do.


The main CS5361 adc chip is further down the circuit buffered by a couple OP275 chips but I don't think it's relevant at this point since the issue is this first stage it appears?

The unit works fine as far as I can tell and, as long as something is connected to the inputs that brings this dc voltage down, it doesn't seem to be an issue. But there are some things that may not do this depending, and there may be a concern with magnetizing any output transformers with dc if the chance arises?

Any ideas what may be going on? Bad switch? Bad Ad8513? Both left and right channels are behaving the same.
 

Attachments

  • 8513 HOOKUP3.2b.jpg
    8513 HOOKUP3.2b.jpg
    61 KB · Views: 25
Last edited:

pucho812

Well-known member
Joined
Oct 4, 2004
Messages
12,464
Location
third stone from the sun
I can't see how -12VDC on pin 3 at the input would be normal. I would suspect a cap is shorted. not blown up with the mushroom top or anything drastic, just failure from heat over time. those are harder to find as they usually do not have any visible signs of fail.
 

scott2000

Well-known member
Joined
Feb 21, 2015
Messages
2,550
Location
Sunny...Sometimes Florida- USA
Thanks. I've checked a number of them but can't find anything obviously wrong and it's difficult with some being in circuit with other capacitances..The fact that both channels act the same doesn't help when looking for differences. . The tiny tiny caps are unreadable but seem to be doing what they should at least from this part of the circuit... Not looking forward to start removing things. Just weird how everything works well enough.

Lots of tant caps in this if I'm understanding what they are ...interesting...
 

radardoug

Well-known member
Joined
Oct 10, 2009
Messages
1,029
Location
North New Zealand
Its likely to be a faulty multilayer board. I had that problem with an Apogee once. People, dont buy Apogee! Useless firm, no service backup, no schematics, just useless.
 

scott2000

Well-known member
Joined
Feb 21, 2015
Messages
2,550
Location
Sunny...Sometimes Florida- USA
Its likely to be a faulty multilayer board. I had that problem with an Apogee once. People, dont buy Apogee! Useless firm, no service backup, no schematics, just useless.
Yeah I thought about that but hope that's not the case. Definitely weird stuff going on. Maybe if I can decipher enough I can jump some things as a workaround if that happens to be the case.. Schematic would save time for sure. At least Apogee is keeping in contact with me more or less for now.

I'm not comfortable with smd stuff and make nightmares out of my previous attempts but may try shot gunning at some point. Have some of that chip quick like stuff on the way .
 

Khron

Well-known member
Joined
Apr 8, 2010
Messages
1,931
Location
Finland
Tantalums are notorious for failing, and I've read of more than a few Ensembles that died due to those. A couple years back i grabbed one with some noise on the mic inputs - turned out to be a shorted tantalum cap that had also toasted the 10 ohm series resistor upstream of it.

Makes you wonder what exactly you're paying a premium for, really...
 

scott2000

Well-known member
Joined
Feb 21, 2015
Messages
2,550
Location
Sunny...Sometimes Florida- USA
So Apogee said:

check the voltage on U3 "IC 5361", if pin 9 is low and pin 24 is +3.5vd, should be +5VD (U3 is bad)

I'm getting about 1 to 1.2vdc on pin 9 and 5vdc on pin 24.

Anyone know how this might translate to what Apogee is saying?

Not sure how long before they get back to my response.

Attaching pinout of the 5361
 

Attachments

  • 5361 pinout.jpg
    5361 pinout.jpg
    277.3 KB · Views: 27

Khron

Well-known member
Joined
Apr 8, 2010
Messages
1,931
Location
Finland
Pin 9 should be best viewed on an oscilloscope, it's the digitized audio output of the ADC.

With absolute silence on both inputs, you would expect that to be permanently low. With the inevitable noise coming into the analog inputs (because real world), you should see at least SOME activity there.

For the nominal value of FILT+ check the specs in the datasheet, but the 5V you're seeing sounds plausible.

Either way, if you're measuring the DC voltages that you mentioned, AT THE INPUTS, it's a pretty safe bet the ADC chip is not the issue, because it's nearly infinitely unlikely that there's NO other circuitry between the physical input connectors and the ADC.
 

ccaudle

Well-known member
Joined
Mar 18, 2010
Messages
71
Location
Houston
If the power supplies are correct at the input 8513 op-amps, my first guess would be shorted ESD protection cap inside the 8513 which was connecting negative supply to the input pin. Replace the input op-amps.
 

scott2000

Well-known member
Joined
Feb 21, 2015
Messages
2,550
Location
Sunny...Sometimes Florida- USA
If the power supplies are correct at the input 8513 op-amps, my first guess would be shorted ESD protection cap inside the 8513 which was connecting negative supply to the input pin. Replace the input op-amps.
Thanks.

Yes I've always been suspect of the 8513. I've seen similar things happen in the past but the fact that both channels and their respective 8513 behave the same, gave me pause. But it is used in stereo always so maybe not out of the ordinary .

Waiting for some quick chip stuff to get here before I try to tackle those chips.

Pin 9 should be best viewed on an oscilloscope, it's the digitized audio output of the ADC.

With absolute silence on both inputs, you would expect that to be permanently low. With the inevitable noise coming into the analog inputs (because real world), you should see at least SOME activity there.

For the nominal value of FILT+ check the specs in the datasheet, but the 5V you're seeing sounds plausible.

Either way, if you're measuring the DC voltages that you mentioned, AT THE INPUTS, it's a pretty safe bet the ADC chip is not the issue, because it's nearly infinitely unlikely that there's NO other circuitry between the physical input connectors and the ADC.
Thanks. What would I be looking for on a scope? any settings to use?
Apogee said the ADC chip test was for the meter issue I'm having. The meter issue doesn't affect the negative voltages on pin 3 when they peg with nothing plugged in.....but actually it does in the way that when the dc level is taken down on the inputs, the meters reflect this......


Just realized that the soic-14pin AD8513 is a bit hard to find atm. Any sources for the soic? Replacement possibly?
 
Last edited:

Khron

Well-known member
Joined
Apr 8, 2010
Messages
1,931
Location
Finland
The data rate will be something in the MHz region, a squarewave signal, of either 3.3v or 5v magnitude (check the ADC datasheet for what digital supply voltage it uses).

So is it implied that this ADC is only used for the metering, and there's another separate ADC for the actual audio path?
 

JohnRoberts

Well-known member
Staff member
Moderator
Joined
Nov 30, 2006
Messages
20,971
Location
Hickory, MS
if pin 9 is serial digital data it will look like a square wave but not stable fixed up/down periods.

Start looking in 1 uSec timebase range (guess).

JR
 

scott2000

Well-known member
Joined
Feb 21, 2015
Messages
2,550
Location
Sunny...Sometimes Florida- USA
The data rate will be something in the MHz region, a squarewave signal, of either 3.3v or 5v magnitude (check the ADC datasheet for what digital supply voltage it uses).

So is it implied that this ADC is only used for the metering, and there's another separate ADC for the actual audio path?
Thanks. I don't think they are implying that ...Maybe just that the meter uses it. Here's the quote in context

"Seeing the Front Panel "meters on -4 thru -50 lite up"...check the voltage on U3 "IC 5361", if pin 9 is low and pin 24 is +3.5vd, should be +5VD (U3 is bad)."

if pin 9 is serial digital data it will look like a square wave but not stable fixed up/down periods.

Start looking in 1 uSec timebase range (guess).

JR
Thanks. Should everything be connected or disconnected from any digital/analog inputs or outputs while looking?
 

MisterCMRR

Well-known member
Joined
Jul 1, 2019
Messages
119
Location
Ventura, CA
One failure mechanism that computes is that the input transistor at the + input of the A section (pins 1, 2, and 3) of the op-amp has a base-emitter short - that would pull the operating current of the input pair from that input (something your voltages with a 5 kΩ pot connected confirms - about 500 µA flowing out of the pin instead of the normal nA range current). And this would likely drive the output to the negative rail - and take the - input with it through the feedback resistor. These symptoms might be due to a missing + rail, but this A section is powered by via the same IC pin as section B, which appears to be working normally. Bringing these op-amp inputs to a connector with nothing to clamp even static charges just seems to be irresponsible design - just begging for something like this to happen! Oh, unless there's something unusual about the DAC chip, I'll bet it's a bit unhappy about having it ref input pin at a high negative voltage. I almost looks like the designer of this circuit didn't bother to read the "absolute maximum" voltage limits for the ICs!
 

scott2000

Well-known member
Joined
Feb 21, 2015
Messages
2,550
Location
Sunny...Sometimes Florida- USA
if pin 9 is serial digital data it will look like a square wave but not stable fixed up/down periods.

Start looking in 1 uSec timebase range (guess).

JR
Thanks .I'm still unclear what I'm supposed to be looking for. Sorry . I'm surprised I could get the scope to work after sitting for so long waiting to be learned.

Here's a shot of what I have with these settings on a Tek 2225 probe x10
Bringing these op-amp inputs to a connector with nothing to clamp even static charges just seems to be irresponsible design
There are some protection diodes at the inputs I didn't put in but I'm not sure if this is what you are talking about. I will update the sketch to show these . edit... You said DAC chip...nevermind....attaching data sheet for it
 

Attachments

  • 5361 pin 9 a.jpg
    5361 pin 9 a.jpg
    107.1 KB · Views: 22
  • 1590f-1271520.pdf
    250.9 KB · Views: 3
Last edited:

scott2000

Well-known member
Joined
Feb 21, 2015
Messages
2,550
Location
Sunny...Sometimes Florida- USA
Apogee responded and said they feel pretty good they can possibly easily repair this if I send it to them, and didn't feel it would be easy enough to explain to me through emails what the process may be. I did ask if they could submit the directions they might take and I could submit them here for others to understand if I couldn't but, will have to see. I gave them the link to the thread here. They gave me a pretty good offer to send it in but with the caveat that I'd get a call if it were going to cost more so I need to weigh this out a little longer regardless.
Will update .
Thanks for all the help so far.
 
Top