What are the chances of getting this kind of support...?

[slowpogo]

I have an Echo Mia sound card, which I believe is (at least kind of) pro audio level. Anyway, I wanted to do some experimenting, so I determined which electrolytics were the output caps, and then replaced them with some Nichicon Muses. It sounded better in a lot of ways but was way too bassy; I had to use EQ to tone it down.

I then decided to try some Elna Silmics. I removed the Muses, and soldered in the Silmics. So far so good. However, suddenly I noticed a little speck on the soldering iron -- it turns out some errant movement of the iron had removed a tiny surface mount capacitor. I attempted to gently scrape it off the iron, which it was stuck on, and it bounced CRAZY off the table and onto the carpeted floor.

I spent half an hour looking for the sucker, but it's gone, lost in the abyss of carpet threads. Yes, a VERY dumb screw up on my part.

If I knew what value that capacitor was, I could easily get a replacement and solder it in. But there's no way of knowing without a schematic. It had nothing printed on it, unlike all the SMD resistors on the card. It's from the output section, but the L/R paths aren't perfect mirror images of each other, so I can't just find the analogous part on the other side and measure it.

I emailed Echo, told them I'd made a mistake and knocked off C17. I told them I understand they probably don't support modding of their products, etc. etc. (basically a lengthy, "I know how this sounds, I know what I'm asking, BUT....") I asked if they had a schematic they could send me, or if someone could just tell me the stock value of C17.

Do y'all think there's any chance they'll help me out? If not, I'm pretty pissed; one little slip of the wrist, and my entire sound card is bunk, when all I need is a 50-cent part to fix it. This was long, thanks for reading and any insight..

 

[Rescue Toaster]

I have an Echo Mia but I don't see how I could determine the value of that capacitor, really. The extra circuitry would probably throw off the value. I can give it a shot though I guess.

 

[slowpogo]

Rescue Toaster, you would be a lifesaver were you willing to do that.

If measuring the cap end to end doesn't yield a sensible value, wouldn't measuring from either end to ground *have* to work?

Anyway, a few minutes of your time would really save me a lot of $ and worry..not to overdue it but THANK YOU.

 

[Rescue Toaster]

Didn't work. At 400nf scale I get over range, and at 4uf it sits at zero. No clue.

 

[slowpogo]

Well, thanks for trying. I may just have to try and educate myself in hopes of being able to determine which is the analogous resistor on the other channel, and measure that. As it stands I don't know where to start to do that. Any lurkers with good ideas, I'd appreciate it...

 

[bhjazz]

You might get a reply, but it will only be explaining the value of C17. I highly doubt they would give out the schematic. What if you were the competition searching for an answer? Unfortunately, they honestly don't know you. No worries. At the worst, you'll need to buy another card, maybe used, and grab the cap off of there if they don't tell you what it is.

But hey: 100 bonus points for taking a solder iron to it! Guts, man!

 

[slowpogo]

I don't know what Echo's turnaround time for support emails is, but it's been almost two days now. I wasn't sure about the "competition" angle, because hell, Fender posts schematics of most of their guitar amps on their website. Anyone can download them. Maybe that's different somehow, I'm not sure.

Still, by studying the pinout of a typical SOIC dual opamp, and examining the Mia, I'm 95% sure that C16 is the same as C17. There are caps going from the V+ and V- of the opamp to ground, and they throw off some of the symmetry of the circuit. It was a lot easier to figure out than I thought.

So now I just have to measure C16, and if necessary remove it from the board to do so (carefully). Trouble is, I no longer own a DMM that measures capacitance.

I know there are really cheap basic multimeters out there for like $5, and I know ones that measure uf are usually more expensive. Is there, I don't know, a $10-15 capacitance meter out there somewhere? *or some easy clever way to measure uf with just a regular DMM?

 

[flecom]

if they are going from V+ to gnd to V- those sound like filter caps, you could just replace them with whatever value the datasheet of the ic calls for...

 

[ericj]

Yeah, those are probably .1uf or .01uf.

 

[slowpogo]

Sorry, C16/17 are not the filter caps, I was just saying that the position of the filter caps on the board throws off the symmetry and makes it hard to see that C16 and C17 are performing the same function.

They are in some kind of loop, along with a few resistors, between Output and -In of each channel. This loop "exits" through the 10uf output caps.

*speaking of datasheets, I can't find one anywhere for the opamp, JRC 2068. I believe it's discontinued. Anyone know where I might find an archived datasheet? It could be helpful. I can't even find a website for "JRC" whoever they are.

 

[~n00beR]

Quote:

Originally Posted by slowpogo /img/forum/go_quote.gif Sorry, C16/17 are not the filter caps, I was just saying that the position of the filter caps on the board throws off the symmetry and makes it hard to see that C16 and C17 are performing the same function. They are in some kind of loop, along with a few resistors, between Output and -In of each channel. This loop "exits" through the 10uf output caps. *speaking of datasheets, I can't find one anywhere for the opamp, JRC 2068. I believe it's discontinued. Anyone know where I might find an archived datasheet? It could be helpful. I can't even find a website for "JRC" whoever they are.

If it is a capacitor going from output to -In, it sounds like a bypass capacitor., and the other resistors would be part of the feedback loop.

IIRC JRC is the "Japan Radio Company" and I think the have changed their name to NJR "New Japan Radio".

Since they have changed names i presume the op amp prefix will also have changed so from the NJR site we have the "NJM 2068"

http://semicon.njr.co.jp/njr/hp/prod...TAIL_MOVE_SPEC

On the above page you will find the application data for the device under the "related data heading"

Looks like it could well be a 0.01uf capacitor

Good luck and have fun

 

[slowpogo]

Thanks noober!

I'm not finding an application schematic that matches exactly what I'm seeing. The resistors in the loop are 4.2k and 2.3k. Can I ask what makes you suspect it would be .01uf?

 

[threepointone]

The two resistors you see likely are part of the feedback for the op-amp on the board. The exact value is not important; they can be changed for different circuits. The capacitor is likely a compensation capacitor meant to ensure the stability of the op-amp; typically, in audio applications, .01uf is typically used. I don't know if you've been doing any work with headphone amps, but the same compensation capacitor (IIRC) is in tangent's PIMETA amp.

Of course, unless you could take a pic of the board and the traces just so happen to be on the top or bottom of the board, it might be something entirely different. If you're lucky, though, this capacitor might not even be necessary (not recommended, though)

 

[slowpogo]

The pic below is the best I have right now, taken a while back. The upper left is the area in question.

C17 (the one I'm missing) is in a loop with R22 and R23. C16 is in a loop with R20 and R21. C15 and C29 go between V+/V- and the ground plane.

I've built two PIMETAs, and besides rail capacitors (and optional input caps) the only other capacitor is on the ground channel opamp, a small ceramic meant "for stability." Maybe that's what you're referring to?

I'm getting ahead of myself a little but assuming the caps are .01uf, would using a film cap instead of SMD ceramic provide any benefit to the sound? Changing the output electrolytics made a very obvious difference.

 

[threepointone]

I just rechecked the tangent website, and what I was talking about actually isn't in the PIMETA--it's in the cranky opamps document (http://tangentsoft.net/audio/hs-opamp.html, C4).

It seems like the capacitor I'm thinking about in the picture, though I'd need a closer picture to be more sure. Unfortunately, a bunch of the traces look like they're under the board or even worse within the board, so I still can't really be sure. I would think .01uF is okay, but if you want to be more sure, I'd take C16 and measure its value, as it looks like it's the same as C17.

I did a bit more reverse engineering with the DAC chip and everything, and I have a question: does this sound card really have two channel outputs, or do they mean that they have two outputs for one channel? It doesn't look like the DAC is actually capable of two stereo channel outputs.