This section isn't quite right for my post, but it looks like the closest match so I'll post this here.
I've got an EMU 1616m with a bit of a strange issue: both of the front mic/line preamps often put out high levels of squealing/noise into the ADC.
Things I have observed so far:
* The squeal is intermittent but frequent.
* It sounds like opamp oscillation. If I touch various spots in the preamp area with my finger it will go away, which is a dead giveaway.
* It can be on either or both the A and B inputs.
* It occurs with various combinations of the A and B inputs open, tied to their pin1 returns, or connected to a mic (I have one Dayton EMM-6 to play with).
* It occurs regardless of whether the 48v phantom power is on or off. It seems to have some sort of DC level dependence, it will go away for about 10 seconds when phantom power is toggled but then it will come back.
* It also seems to be dependent on the input gain level. Either or both channels can squeal, but raising the gain on one channel beyond 3/4 rotation frequently silences one or both channels. Sometimes it only silences the opposite channel, which is really strange. When the squeal is silenced successfully, both the A and B inputs appear to operate as expected.
* It registers on both the A and B level indicator LEDs on the front panel.
* Everything else on the unit works completely fine.
Things I have tried so far:
* Replaced all of the electrolytic caps in the preamp section with new, equivalent caps of the same footprint. No difference.
* Recapped the SMPS on the top board as well. No difference.
* Replaced front input ADC (AK5394) with new part. No difference.
* Figured out that they left two extra headers routed on the boards but unpopulated from the factory. They are connected to the same places as the board-to-board interconnect headers except with mirrored pinouts. I took the liberty of populating them - this allows me to interconnect the two boards with ribbon cables and splay them out on the bench for testing. I can see where the oscillation is introduced, it appears to be in the latter half of the preamp section, which looks like a variable gain stage controlled by the front panel gain pots. Still no idea what the actual root cause is though.
* Replaced all of the opamps in the preamp section. No difference. Tried a couple different potential replacement types, all behaved exactly the same.
* Figured out what each of the SOT-23 components in the preamp section were (they turn out to be MMBT3904, MMBT3906, and BAT54S) and replaced all of them with new parts. No difference.
* I threw in the towel, ebayed a second 1616m, and it had the exact same issue. I wish I was joking.
As you've probably guessed, I do have a usable set of SMD rework equipment as well as a decent assortment of bench gear (100MHz+ CROs, TM500 series FGs/counters/powersupplies, LCR meter, etc.)... but I'm at my wit's end trying to figure this one out. What gives??? Am I doing something really, really stupid and not realizing it? I find it extremely difficult to believe that two of these units would have the exact same issue without other people on the 'net encountering it over the years.
Even better, does anyone know where I might find the schematic for the preamp section in this thing? Trying to fix it has gone way beyond the time and money I wanted to invest, and I'm really not looking forward to trying to reverse engineer the PCB layout (it appears to be 4-layer).
I've got an EMU 1616m with a bit of a strange issue: both of the front mic/line preamps often put out high levels of squealing/noise into the ADC.
Things I have observed so far:
* The squeal is intermittent but frequent.
* It sounds like opamp oscillation. If I touch various spots in the preamp area with my finger it will go away, which is a dead giveaway.
* It can be on either or both the A and B inputs.
* It occurs with various combinations of the A and B inputs open, tied to their pin1 returns, or connected to a mic (I have one Dayton EMM-6 to play with).
* It occurs regardless of whether the 48v phantom power is on or off. It seems to have some sort of DC level dependence, it will go away for about 10 seconds when phantom power is toggled but then it will come back.
* It also seems to be dependent on the input gain level. Either or both channels can squeal, but raising the gain on one channel beyond 3/4 rotation frequently silences one or both channels. Sometimes it only silences the opposite channel, which is really strange. When the squeal is silenced successfully, both the A and B inputs appear to operate as expected.
* It registers on both the A and B level indicator LEDs on the front panel.
* Everything else on the unit works completely fine.
Things I have tried so far:
* Replaced all of the electrolytic caps in the preamp section with new, equivalent caps of the same footprint. No difference.
* Recapped the SMPS on the top board as well. No difference.
* Replaced front input ADC (AK5394) with new part. No difference.
* Figured out that they left two extra headers routed on the boards but unpopulated from the factory. They are connected to the same places as the board-to-board interconnect headers except with mirrored pinouts. I took the liberty of populating them - this allows me to interconnect the two boards with ribbon cables and splay them out on the bench for testing. I can see where the oscillation is introduced, it appears to be in the latter half of the preamp section, which looks like a variable gain stage controlled by the front panel gain pots. Still no idea what the actual root cause is though.
* Replaced all of the opamps in the preamp section. No difference. Tried a couple different potential replacement types, all behaved exactly the same.
* Figured out what each of the SOT-23 components in the preamp section were (they turn out to be MMBT3904, MMBT3906, and BAT54S) and replaced all of them with new parts. No difference.
* I threw in the towel, ebayed a second 1616m, and it had the exact same issue. I wish I was joking.
As you've probably guessed, I do have a usable set of SMD rework equipment as well as a decent assortment of bench gear (100MHz+ CROs, TM500 series FGs/counters/powersupplies, LCR meter, etc.)... but I'm at my wit's end trying to figure this one out. What gives??? Am I doing something really, really stupid and not realizing it? I find it extremely difficult to believe that two of these units would have the exact same issue without other people on the 'net encountering it over the years.
Even better, does anyone know where I might find the schematic for the preamp section in this thing? Trying to fix it has gone way beyond the time and money I wanted to invest, and I'm really not looking forward to trying to reverse engineer the PCB layout (it appears to be 4-layer).
Sounds like a production, design, or layout problem, perhaps with the NFB loop. I've designed variable gain mic preamps
from 0 to 50 dB without such problems. Couldn't find a schematic for this, though.
Last edited:
At this point I'd turn down headphone level, try the other 1616m's power supply, and try another computer if possible (which should give you an opportunity to configure the card from scratch). It is odd that noone else should have mentioned such a problem.
A squeal should be too high in frequency to pass through both ADC and DAC, but still: Does sample rate setting have any effect?
If the problem should persist, it might be something nasty like a bad ground connection. 4-layer PCB sounds like the connections to the pots ought to run in there somewhere, otherwise I might have suspected bad shielding.
A squeal should be too high in frequency to pass through both ADC and DAC, but still: Does sample rate setting have any effect?
If the problem should persist, it might be something nasty like a bad ground connection. 4-layer PCB sounds like the connections to the pots ought to run in there somewhere, otherwise I might have suspected bad shielding.
Last edited:
The 2nd unit had to be returned while I had the chance to get a full refund, however I had similar thoughts and tried these things before I returned it. Results:
* Headphone level does not matter. The squeal is full-scale on the inputs, so if they are monitored by the headphone output the squeal is immediately apparent at all headphone volume levels. Anything much above the minimum headphone level is deafening with the headphones on one's head. If the A/B inputs are not monitored by the headphone output, the output is dead silent, but the level meters show the squeal is still there.
* Both 1616m units did the same thing using both of the power bricks (all 4 combinations).
* The 1616m kit I kept has a PCI card; the one I returned had a Cardbus laptop card. I also had 2 EDI cables. On the laptop I installed the software from the original CDs on a fresh Win2k install; on the PC I used the most recent downloads available from Creative on a fresh WinXP install. All combinations of PC/Laptop, EDI cable, and 1616m had the same issue. I tried electrically floating the laptop (disconnected all other cables including the power) and that had no effect either.
I'm pretty sure I tried this too, same result at all rates. The oscillation is not a pure tone and apparently has AF-band components that propagate through the signal path.
I have considered blaming/replacing the pots, but am not really sure how it would consistently cause the same issue on both channels of both units. Might be worth trying if I can get exactly the right replacement part.
Last edited:
.
Thought from random passerby, it seems maybe the EMU 1616m isn't at fault after all. Might the noise be coming in from some previous stage or device.
Or it's kinda-sorta possible to conceive of some filtering arrangement on the output that might cause feedback.
Or there's always Rule One: plug everything into the same wall receptacle. Not a power strip or surge suppressor, but the same wall receptacle, or a plain ol' extension cord.
Thought from random passerby, it seems maybe the EMU 1616m isn't at fault after all. Might the noise be coming in from some previous stage or device.
Or it's kinda-sorta possible to conceive of some filtering arrangement on the output that might cause feedback.
Or there's always Rule One: plug everything into the same wall receptacle. Not a power strip or surge suppressor, but the same wall receptacle, or a plain ol' extension cord.
It happens with all the inputs and outputs on the EMU 1616m completely disconnected.
It happens with the output disconnected, the headphone volume pot at the minimum, and nothing patched through to the DACs (i.e. they are driving silence). If I connect stuff to the outputs when nothing is patched through, they really are silent.
Tried that too (EMU 1616m 48v brick plugged into wall, laptop running from battery, NOTHING else connected). No dice.
This one is a real headscratcher, which is why I decided to ask you guys
This one is a real headscratcher, which is why I decided to ask you guys
Could even be bad pcbs with shorts, opens, or bad smt reflow.
What are the Patchmix settings? Have you tried a default session and muting all unused inputs? Also try removing all mixer strips except for one ASIO and one Wave host.
Have you tried using a shorting cable on the inputs? That way at least eliminate or focus on the first stage.
Usually using a small homebrew 100 ohm resistor and a short length of cable will help you work out where the issue is originating. There are usually no coupling caps except right at the input (if any), so you'll have to go banzai stage by stage. The 100 ohm stays soldered to ground, and the short pin can be used to short the respective opamp output to ground.
If it's on both channels though, it would be a more involved issue like a failed via (the board is a multilayer board with separate power planes, at least the 1010 card is - not sure about the breakout box). You also might try a replacement cable. The squeal sounds like a feedback path, and since replacing the discrete bits did not help I am assuming it could be a controller issue.
Sorry for the bunch of random ideas. Good luck with your troubles.
Have you tried using a shorting cable on the inputs? That way at least eliminate or focus on the first stage.
Usually using a small homebrew 100 ohm resistor and a short length of cable will help you work out where the issue is originating. There are usually no coupling caps except right at the input (if any), so you'll have to go banzai stage by stage. The 100 ohm stays soldered to ground, and the short pin can be used to short the respective opamp output to ground.
If it's on both channels though, it would be a more involved issue like a failed via (the board is a multilayer board with separate power planes, at least the 1010 card is - not sure about the breakout box). You also might try a replacement cable. The squeal sounds like a feedback path, and since replacing the discrete bits did not help I am assuming it could be a controller issue.
Sorry for the bunch of random ideas. Good luck with your troubles.
Yeah that's basically what I was trying, all unnecessary strips have been removed. No dice. The A/B level meters show the squealing coming from those inputs only.
I wired up an XLR-M connector to do that. Plugging/unplugging it disturbs the operating point of the input stage (especially if phantom power is on) and makes the squeal go away for a few seconds, then it comes back.
Hm, that might be worth trying.
The exact same failure on both channels of two units though? That would really surprise me.
Tried that, no difference.
I think the remaining parts of potential interest that I haven't replaced yet are:
* Linear regulators feeding preamp power rails (I actually did identify the parts and have exact replacements on hand)
* All the rest of the electrolytic caps on both boards (I made up a complete recap kit to cover these too, just haven't installed yet)
* All of the passives in the preamp. Been avoiding this one since I think a failure here is unlikely and I really don't want to go through the trouble of figuring out exactly what every part is.