or Connect
Head-Fi.org › Forums › Misc.-Category Forums › DIY (Do-It-Yourself) Discussions › MOSFET-MAX tube biasing issues (Solved, now about GrubDAC issues)
New Posts  All Forums:Forum Nav:

MOSFET-MAX tube biasing issues (Solved, now about GrubDAC issues) - Page 3

post #31 of 55
Thread Starter 

Yeah, I was incredibly dumb to attempt SMD soldering with that equipment I had. I've succeeded in soldering all the SMD parts on my gamma1/gamma2 boards now with my new soldering station though, so the grubDAC won't be a problem once I get the new PCB.

post #32 of 55
Thread Starter 

OK, so after getting the first one to work (the grubDAC worked on the 2nd try, had to reflow some joints the first time around, but after fixing that it was all smooth sailing :D) I've now finished the second Mosfet-max. Left channel works perfectly, however the right one is giving me some issues. I think I burned out one of the Mosfets (QB9R) due to my multimeter leads not being connected right when biasing, so I replaced that, and now I'm getting some static noise on the right channel (there is sound, too, though, it's just interrupted by the static sometimes and the static's always there in the background), and the DB bias voltage on the right channel fluctuates within a margin of +-50mV. I've already replaced both MOSFETs (got that issue after replacing QB9R so I figured why not replace QB8R as well, since I had some spares) and I know for a fact that all the transistors are oriented the right way etc... could it be one of the electrolytes? (I hope not, I'd have to reorder them from beezar) Or is it one of the MOSFET solder joints (tried to redo those like 10 times, but I can't really reach them from the top on QB9R, and the grounded one on QB9R is the only one that looks a bit objectionable)?

 

Tube bias has been working fine since the first start-up by the way. Also, I can't take any usable pictures right now since I've lost the charger for my camera and my cellphone can't be convinced to take some sharp pics.


Edited by hitman47 - 11/11/11 at 8:55am
post #33 of 55
Quote:
Originally Posted by hitman47 View Post

OK, so after getting the first one to work (the grubDAC worked on the 2nd try, had to reflow some joints the first time around, but after fixing that it was all smooth sailing :D) I've now finished the second Mosfet-max. Left channel works perfectly, however the right one is giving me some issues. I think I burned out one of the Mosfets (QB9R) due to my multimeter leads not being connected right when biasing, so I replaced that, and now I'm getting some static noise on the right channel (there is sound, too, though, it's just interrupted by the static sometimes and the static's always there in the background), and the DB bias voltage on the right channel fluctuates within a margin of +-50mV. I've already replaced both MOSFETs (got that issue after replacing QB9R so I figured why not replace QB8R as well, since I had some spares) and I know for a fact that all the transistors are oriented the right way etc... could it be one of the electrolytes? (I hope not, I'd have to reorder them from beezar) Or is it one of the MOSFET solder joints (tried to redo those like 10 times, but I can't really reach them from the top on QB9R, and the grounded one on QB9R is the only one that looks a bit objectionable)?

 

Tube bias has been working fine since the first start-up by the way. Also, I can't take any usable pictures right now since I've lost the charger for my camera and my cellphone can't be convinced to take some sharp pics.

"know for a fact" has led to many a downfall ... wink.gifwink.gif  I'm not trying to be mean, though - on the contrary - I know for a fact wink.gif that I've gotten a part misplaced and/or oriented it incorrectly and stared at it for a couple of dozen times before figuring out I got it wrong.  The process with our PCB's and parts is similar to proof-reading a written document.  It often takes someone else reading it before the mistakes are noticed.  There are techniques for this sort of thing: make a table, list and confirm every part, turn the PCB so that it's reversed and check them again, etc., etc.

 

I say all of this because I doubt seriously you burned one of the MOSFETs up.  Most likely it's a resistor or small transistor in the wrong place.  The small transistors will fry almost without fail if they're soldered in wrong.  The most likely thing would be either swapping a 2N5087 with a 2N5088 or getting a couple of resistors out of place.  However, the static is most likely a fried transistor.  That's my guess.  I don't know anything else that would actually cause the DB bias to fluctuate as you describe and the static.
 

Get that camera fixed and some pics posted, if you can.  In the meantime, check the transistor locations again and check the resistance of all the resistors on that side of the DB (power off).  Luckily, almost none of them are in parallel, so you should get fairly consistent readings that are close to their intended value.  If that doesn't work, we'll look at your pics and see if we can come up with some other ideas.

 

BTW - a bad electrolytic capacitor is painfully and messily obvious.  I doubt that you have anything wrong with that. 

 

 


Edited by tomb - 11/11/11 at 11:16am
post #34 of 55
Thread Starter 

Well, the reason I thought it was a MOSFET was that 1. before it happened, everything looked fine, 2. when it happened, I was cranking up the DB bias while still getting a 0 reading from my multimeter (I'd bought new leads that don't really fit the connections), 3. when it happened, I heard a crack and there was some smoke coming from approximately where QB9R is and everything kinda powered down and 4. after it'd happened, when powered, the amp would just turn on, PS voltage would go to around 18V and then everything would go very dim again with PS voltage dropping to like 2-4V, indicating a short somewhere (I was assuming in the MOSFET). Now that I've replaced the MOSFET, that stuff doesn't happen anymore. I guess I did fry it but also got something else burned up, any idea which one it could be (maybe the CCS transistors?)? Or should I just go ahead and replace all the small transistors?

 

Anyway, thanks for your help! I'll try to get some pictures ASAP.

 

Edit: Well, this is really weird, but I just powered the amp up and the problem seems to have disappeared... No static, no fluctuating bias voltage, nothing. It all works as it should. I'll let you know if I run into any issues again though


Edited by hitman47 - 11/12/11 at 4:39am
post #35 of 55
Quote:
Originally Posted by hitman47 View Post

Well, the reason I thought it was a MOSFET was that 1. before it happened, everything looked fine, 2. when it happened, I was cranking up the DB bias while still getting a 0 reading from my multimeter (I'd bought new leads that don't really fit the connections), 3. when it happened, I heard a crack and there was some smoke coming from approximately where QB9R is and everything kinda powered down and 4. after it'd happened, when powered, the amp would just turn on, PS voltage would go to around 18V and then everything would go very dim again with PS voltage dropping to like 2-4V, indicating a short somewhere (I was assuming in the MOSFET). Now that I've replaced the MOSFET, that stuff doesn't happen anymore. I guess I did fry it but also got something else burned up, any idea which one it could be (maybe the CCS transistors?)? Or should I just go ahead and replace all the small transistors?

 

Anyway, thanks for your help! I'll try to get some pictures ASAP.

 

Edit: Well, this is really weird, but I just powered the amp up and the problem seems to have disappeared... No static, no fluctuating bias voltage, nothing. It all works as it should. I'll let you know if I run into any issues again though


OK - that sounds pretty definite.  You must had that bias cranked up very, very high.

 

The static still concerns me.  I'd double-check every connection and make sure the PCB is as clean as you can get it before powering it up again.  It could a loose connection or residual flux, etc.
 

 

post #36 of 55
Thread Starter 

Well, now it seems the grubDAC I made took a dump while I wasn't looking... It wouldn't play any sound when I plugged it in today, so I reflowed all of the connections and now it won't even get recognized by the PC. All the connections look perfect. The LED glows, it flickers a bit sometimes though. I'm guessing it's some flux hiding under the chips, I've cleaned the surface very thoroughly but I'm gonna try giving it an iso bath tonight. Pics will follow when I get my hands on a camera.

post #37 of 55
Quote:
Originally Posted by hitman47 View Post

Well, now it seems the grubDAC I made took a dump while I wasn't looking... It wouldn't play any sound when I plugged it in today, so I reflowed all of the connections and now it won't even get recognized by the PC. All the connections look perfect. The LED glows, it flickers a bit sometimes though. I'm guessing it's some flux hiding under the chips, I've cleaned the surface very thoroughly but I'm gonna try giving it an iso bath tonight. Pics will follow when I get my hands on a camera.


Maybe that's where the static was coming from?
 

 

post #38 of 55
Thread Starter 

No, I never had the GrubDAC attached to that amp. It's always been connected to the first Mosfet-Max I made.

 

Well, after giving it a good iso bath, the grubDAC is now recognized by the PC, but still doesn't output any sound... Any ideas what connections I should be looking at? I've reflowed every single joint on the board, but I'm not getting anywhere.

 

Edit: well, now after doing nothing but switching between 2 cables a few times, I'm getting sound from it using either cable, but it's extremely distorted... Anything I should check out? Where's the bad connection (I'm assuming that's what it is) most likely to be at? Also, does it matter if some of the pins on the two ICs are connected to those little holes/test points around the pads? Because I've got a lot of that and it's the only thing I can think of now that could be wrong with the DAC


Edited by hitman47 - 11/12/11 at 1:48pm
post #39 of 55
Quote:
Originally Posted by hitman47 View Post

No, I never had the GrubDAC attached to that amp. It's always been connected to the first Mosfet-Max I made.

 

Well, after giving it a good iso bath, the grubDAC is now recognized by the PC, but still doesn't output any sound... Any ideas what connections I should be looking at? I've reflowed every single joint on the board, but I'm not getting anywhere.

 

Edit: well, now after doing nothing but switching between 2 cables a few times, I'm getting sound from it using either cable, but it's extremely distorted... Anything I should check out? Where's the bad connection (I'm assuming that's what it is) most likely to be at? Also, does it matter if some of the pins on the two ICs are connected to those little holes/test points around the pads? Because I've got a lot of that and it's the only thing I can think of now that could be wrong with the DAC

 

Those little holes are "vias" that are used to connect the circuit trace to the other side of the PCB.  It shouldn't hurt anything - in principle - if you get solder into them.  It's whether the solder touches other traces that are not normally connected to those vias.

 

Sorry for guessing wrong about the use of your GrubDAC, but it's sometimes tough to suggest solutions with descriptions like, "[it] still doesn't output any sound ..."  I would check the signal ground connection now on the GrubDAC PCB.  That can be a cause of the symptom you describe.  It can sometimes be difficult to check, too.  I've had a DAC PCB where the ground connections came loose, but for all appreances seemed to be physically connected.  Try checking for zero resistance using a DMM. 
 

 

post #40 of 55
Thread Starter 

Well, I'm not quite sure I understand... What do you mean by checking the signal ground connection? I checked all the ground pins I could find on the board and they all had a very low resistance to the output ground... Also, I've reflowed all the pins of U1 and U3 twice more (each time giving it a good iso wash afterwards) and now the symptoms have changed to no sound at all and varying mileage when it comes to getting it recognized by the PC. The results always vary, are different between the 2 mini USB cables I own and even when using the same one I get different results from time to time; sometimes it just doesn't get recognized, sometimes a "device not recognized" message pops up, sometimes it's correctly recognized as a "USB audio DAC" and sometimes it's recognized as a "USB audio device". When it does get recognized correctly, the output from it - if there is any (that varies as well) - is a very fast staccato of clicks/pops. I'm assuming that means a capacitor in there is somehow doing something it shouldn't be doing, but that's all I can guess... This is really getting frustrating, especially considering that it was working great just a week ago and that I didn't actually do anything that could have made it stop working... Thanks for your patience with a noob like myself!

post #41 of 55
Quote:
Originally Posted by hitman47 View Post

Well, I'm not quite sure I understand... What do you mean by checking the signal ground connection? I checked all the ground pins I could find on the board and they all had a very low resistance to the output ground... Also, I've reflowed all the pins of U1 and U3 twice more (each time giving it a good iso wash afterwards) and now the symptoms have changed to no sound at all and varying mileage when it comes to getting it recognized by the PC. The results always vary, are different between the 2 mini USB cables I own and even when using the same one I get different results from time to time; sometimes it just doesn't get recognized, sometimes a "device not recognized" message pops up, sometimes it's correctly recognized as a "USB audio DAC" and sometimes it's recognized as a "USB audio device". When it does get recognized correctly, the output from it - if there is any (that varies as well) - is a very fast staccato of clicks/pops. I'm assuming that means a capacitor in there is somehow doing something it shouldn't be doing, but that's all I can guess... This is really getting frustrating, especially considering that it was working great just a week ago and that I didn't actually do anything that could have made it stop working... Thanks for your patience with a noob like myself!


Massive pops and clicks can occur if your signal ground connection is not good.  I'm talking about the wire leads to your output connectors - the ones that would be soldered into the pads labeled "OG."  Check to see that there's zero resistance between the ground on your connectors and some ground point on the PCB.  Don't measure to the OG pads themselves, because depending on where you touch the DMM probe, you could just be touching part of the lead.

 

As for not recognizing the GrubDAC on the PC, it's possible that a loose ground connection on the output pads can also cause this.  While we have no reports at all of the GrubDAC getting fried connecting and disconnecting from certain amps (like the Starving Student), the GrubDAC can get thrown offline on a fairly regular basis - with connecting/disconnecting the output leads.  In my experience, it usually comes back immediately, but your media player may need to be re-started to work.  However, if your ground connection is intermittent (again, I'm talking about through the output leads), then it might cause the symptoms you describe.

 

Again, just some guesses ... but I don't seem to be batting very well in your case.confused.gif

 


Edited by tomb - 11/13/11 at 5:19am
post #42 of 55
Thread Starter 

I'm getting around 20 ohms of resistance from the output ground wire to the 0V point right behind the USB connector. Is that too much or is that point not a real ground reference? I always got 1.3 or so when measuring to the ground pins on the PCM for example.


Edited by hitman47 - 11/13/11 at 7:34am
post #43 of 55
Quote:
Originally Posted by hitman47 View Post

I'm getting around 20 ohms of resistance from the output ground wire to the 0V point right behind the USB connector. Is that too much or is that point not a real ground reference? I always got 1.3 or so when measuring to the ground pins on the PCM for example.


No, the 0V point is not Ground.  Pick any place where the solder pad is connected to the ground plane.  The ground place is the thin metal sheet you can see that covers almost the entire surface of the top of the PCB.  Almost every resistor and capacitor has one end connected to the ground plane (as in, "Ground") and the other end connected to a trace (one of those thick shiny lines that connects the parts under the surface of the PCB).

 

If you get 1.3 ohms measuring at true ground, then maybe the issue is not in the signal ground leads.  As I said, I'm not guessing too well with your problems. 
 

 

post #44 of 55
Thread Starter 

Is it possible that what caused the DAC to "break" is liquid flux corroding a contact or messing with things in some other way? I didn't clean it as well when I first assembled it as I did when I reflowed the joints the last 2 times, and it just kinda seems weird to me that it would break while not being used..

post #45 of 55
Quote:
Originally Posted by hitman47 View Post

Is it possible that what caused the DAC to "break" is liquid flux corroding a contact or messing with things in some other way? I didn't clean it as well when I first assembled it as I did when I reflowed the joints the last 2 times, and it just kinda seems weird to me that it would break while not being used..


I'm sorry, but none of this is really getting us anywhere - we've turned this into a guessing game.  Is it possible?  Yes.  Is it very likely?  I don't think so, not in the time you've built and been using the DAC.  It's just as likely that your re-flowing "lifted" some of the solder from the PCM or WM chip pins.  I've done that a number of times myself when re-flowing pins on SMD chips and had to go back and add more solder.

 

Try to post some pics.  If you can't do that, look at the troubleshooting guide on the GrubDAC website and try to go through a systematic approach that you document and report back to us:

 

  • What are the voltages at the test points and output?
  • Can you ohm out all of the pins on each of the DAC chips?  You can do this by placing a probe (power off, please!) somewhere a distance away on a trace that's supposed to be connected to one of the pins.  Then carefully place the other probe on the pin itself.  If you can read zero resistance, then that pin is connected to the trace.  You can go through each pin and troubleshoot this way.\
  • Did you get solder from the four side pads on top of the crystal chip?  If so, and the solder is contiguous, it will short out the crystal chip.
  • Is the LED on?


There are other things you can check, too, this is a small suggestion.  Pics are important, though, and you will have the benefit of dozens of eyes looking at it.

 

New Posts  All Forums:Forum Nav:
  Return Home
Head-Fi.org › Forums › Misc.-Category Forums › DIY (Do-It-Yourself) Discussions › MOSFET-MAX tube biasing issues (Solved, now about GrubDAC issues)