Head-Fi.org › Forums › Misc.-Category Forums › DIY (Do-It-Yourself) Discussions › Pimeta V1 power problems (fried buffers?)
New Posts  All Forums:Forum Nav:

Pimeta V1 power problems (fried buffers?)

post #1 of 5
4/17/11 at 4:30am
Thread Starter 

Having lurked here for a good two years - and as a result having been inspired to revisit an old boyhood hobby of building audio, I decided on a Pimeta - simple (I thought......)

 

However - and I hope someone can help here, I have major problems!

 

I built the circuit (V1) on a official board and tested the rail voltages etc w/o the opamps and buffers (LMH) (or the big rail capacitors as they'll be mounted off board for height reasons). All well at this stage showing  around half of the 18 v supply (two NiMH batteries).

 

After this all went pear-shaped :(

 

My first attempt at SMD (a big leap from messing around with valves all those years ago) seemed to go ok - LMHs orientated correctly - left pin 1 soldered on R11 not connected anyway - but, I neglected (stupidly) to remove the pin 5s.

 

1/ Then I tested the rail voltages with opamps and buffers installed (no elec rail caps yet) and found that after a second or so the voltage dropped from approx 8v to just below 3v.

 

2/ Trying to trace the problem I found the connected pin 5s and removed them - took the opamps out as well.

 

3/ tested rail voltages again - same result - 8v and after a second or so down to around 2.7v.

 

Looking at the circuit board I've found the following:

 

BUFR - pin 5 is connected to -ve

BUFG - connected to OG

BUFL - not connected to anything

 

As pin 5 on the LMHs is GND (unlike the BUF634 where it is NC) I've effectively connected my BUFR GND  to -8v. I assume that BUFG is ok?

 

I assume that I've fried BUFR - so thats it's shorted out causing a huge voltage drop? At present I'm reluctant to remove all the LMHs as I suspect my removal technique will render them useless - if they're not already!

 

Can anybody please offer any advice to rescue my Pimeta?

 

 

 

 

Reply
post #2 of 5
4/17/11 at 6:10am
Quote:
Originally Posted by londontrot View Post

I neglected (stupidly) to remove the pin 5s.

 

I don't believe you can kill an LMH6321 in a PIMETA v1 by soldering down pins 1 and 5. It will just cause the amp to not function until you cut those connections.

 

Since your problem may lie elsewhere, I suggest posting high-res pics of both sides of the board.

 

 

 

Quote:
the voltage dropped from approx 8v to just below 3v

 

That usually indicates a short of some kind. It could be either a solder bridge or something inside one of your chips.

 

 

Quote:

Looking at the circuit board I've found the following:

 

BUFR - pin 5 is connected to -ve

BUFG - connected to OG

BUFL - not connected to anything

 

Yes, you can see that in the layout image. The reason for the difference is simply that the layout uses pin 5 on the buffers as a short-cut to make the route nicer since the pin isn't used on the BUF634. That's why you have to cut this pin when substituting the NatSemi buffer: it does use pin 5 for something, so these short-cuts cause problems for BUFR and BUFG. Again, not something that's likely to be fatal, just an impediment to success as long as the connections remain.

Reply
post #3 of 5
4/18/11 at 5:13am
Thread Starter 

Well, here are the photos - I used meths for cleaning in the absence of IPA - not sure that it's as good. A tad untidy pcb - I suspect my eyesight is not what it once was!

 

Link to photos on Dropbox

 

 

Reply
post #4 of 5
4/19/11 at 3:13am

Pure methanol should be a fine rosin flux solvent. I suspect yours isn't pure, however, which is where all that white schmutz comes from. It's probably electrically fine, just unaesthetic.

 

I don't see anything that jumps out and yells "newbie mistake!" at me. Here are just some things to look at:

 

  • BUFL pin 5 looks like it may still be connected
  • The resistor sockets are a risk. Make sure the resistors are all firmly seated.
  • Check the rail voltages with the power switch off.
  • If those batteries aren't new-bought, try something fresh. You can use alkalines, temporarily, or a bench supply, or AA holders, or... It could simply be that your 9V NiMHs have been sitting on a shelf too long and have killed themselves with cell reversal. It happens.
Reply
post #5 of 5
4/19/11 at 3:49am
Thread Starter 

British meths is purple - apparently to prevent people from drinking it - and it does leave a residue when burnt.

 

1/ Sorted - did a little more work on pin 5 BUFL and it's, for sure, not connected.

2/ Reseated the resistors

3/ checked rail V - )v with power switch off

4/ Swapped for two fresh akalines.

 

Rail V now reads around 9v and steady - no drop at all. The NiMHs are on charge and one is indicating an error on the charger - so I suspect that. 

 

I'll give it another test later - but I'm sure it's ok.

 

Many many thanks for your advice! :) I do greatly appreciate the help.

Reply
New Posts  All Forums:Forum Nav:
  Return Home
Head-Fi.org › Forums › Misc.-Category Forums › DIY (Do-It-Yourself) Discussions › Pimeta V1 power problems (fried buffers?)