[beerguy0]

Quote:

Okay guys I was hoping my next reply would be a good one but here goes...
 
I replaced the LM317 (lot easier than I thought it would be), moved the leads back over to the "Fuse" pads, cleaned and dried the flux up, waited a bit, put a new 1amp slo-blo in, attached power and POP! I blew the fuse twice in a row. I'll heed the earlier warning and not bypass the fuse but once again, I'm open to suggestion so fire away guys...
 
-Chris

PM sent.
 

 

[Nomnomnom]

Switched ground and voltage pads (GND now in the middle, V+ next to fuse), blew the fuse yet again... I've gone over this thing with a magnifying glass and I just can't see anything physically wrong with it. I know my work isn't perfect but it's far from an atrocity so this is rather perplexing to me...

 

[tomb]

Quote:

Switched ground and voltage pads (GND now in the middle, V+ next to fuse), blew the fuse yet again... I've gone over this thing with a magnifying glass and I just can't see anything physically wrong with it. I know my work isn't perfect but it's far from an atrocity so this is rather perplexing to me...

All your resistors look OK in the pics.  Check your MOSFETs, though - do you have the Z24's on the outside heat sinks and the 9Z34's on the inside?  Also, go over the TO-92 transistors and make certain that you haven't mixed them up.  Look at the PCB pics on the MOSFET-MAX website and remember that even-number transistors (2N5088) goes on odd-number silkscreen (QB5, QB7) and odd-number transistors (2N5087) goes on even-number silkscreen (QB4, QB6).  Also check to see that QB1 is a 2N5486.  You can bend them slightly sideways to read the imprint on the transistor, then bend them back.
 
The offer I gave you still stands ...

 
 
 

 

[jdkJake]

TomB,

Perhaps he should pull the poly fuse to separate the amp from the power supply and give it a go. I know without a load of some sort (100k resistor, resistor/LED, etc) the 317 will not be able to regulate very well, but, at least it will take the PS out of the equation.

Thoughts?

 

[jdkJake]

Chris,

One more thing I cannot tell from the pictures. When you put a fuse in the fuse holder, can you verify the fuse holder itself does not make contact with the body of CR1A?

The parallax on the pics makes it hard to see for certain.

 

[Nomnomnom]

The fuse and the fuse holder are both a comfortable distance away from CR1A...
 
Just throwing this one out there. Could one or more of the rectifiers (DR1A-DR1D) be toasted from the previously backwards install of the first two? And if they are toast, could they be tripping the fuse?
 
Still reading approximately 26.5V at the power supply
 
-Chris
 
EDIT: Tom - All TO-92s, MOSFETS and QB1 are installed according to layout

 

[tomb]

Quote:

The fuse and the fuse holder are both a comfortable distance away from CR1A...
 
Just throwing this one out there. Could one or more of the rectifiers (DR1A-DR1D) be toasted from the previously backwards install of the first two? And if they are toast, could they be tripping the fuse?
 
Still reading approximately 26.5V at the power supply
 
-Chris
 
EDIT: Tom - All TO-92s, MOSFETS and QB1 are installed according to layout

No, I do not believe toasted rectifiers could cause a short.  Rectifiers are very simple devices: either the current goes through them, or it doesn't.  If it doesn't, then the heat will build until something blows.  In your case, it was the fuse.  So, honestly - I would not suspect the rectifiers.  Besides, you measured 34.5VDC coming from them, so they're working just fine.

 
26.5VAC from the walwart is also good - just about right without much of a load.  So, I'd say we can maybe dismiss any issues with the power supply.  The tube circuit is very simple - only a few resistors, the tube sockets, and the CCS transistors and resistors - plus the heaters and resistor.  IOW, it's very unlikely anything in the tube circuit could cause this kind of short.
 
That's why my last post was focusing on the buffer.  If you have a short in the buffer, it can do all of what you've experienced - high current, blown fuse, etc., etc.  In my experience, something like this is usually found to be one of the power output transistors or MOSFETs (depending on which type of MAX it is) that are shorting out - maybe one of them is swapped with the other kind, or someone put them in backwards, etc., etc.
 
Jake's idea is a good one and worth trying just to completely eliminate the power supply as an issue.  It's fairly simple - just melt the pad to one of the leads of the polyfuse - PF1 - and remove that lead.  That will completely isolate the power supply from the rest of the PCB.  If we can measure and adjust to something around 27VDC, then we can completely eliminate the power supply as any cause of the problem.  To check this: if you remove one of the PF1 leads, simply touch your V+ DMM probe to the inside pad of PF1 (toward the VREG heat sink) and the other probe in GND.  If you can measure something in the 20-30VDC range and then adjust it to ~27VDC with the trimmer, then the power supply section can be completely dismissed and any source of the problem. 
 
 

 

[Nomnomnom]

Well, I was a little worried for a second. I put the last fuse I have available until my next Rat Shack Raid into the holder after removing a leg of PF1, powered it up, shielded my eyes from the expected spark and.... Nothing... Which in this case was good. Using the trimpot it was easy to adjust up to 27VDC.
 
- I'm assuming that's all that was supposed to happen? No LEDs lit up...
 
-Chris

 

[tomb]

Quote:

Well, I was a little worried for a second. I put the last fuse I have available until my next Rat Shack Raid into the holder after removing a leg of PF1, powered it up, shielded my eyes from the expected spark and.... Nothing... Which in this case was good. Using the trimpot it was easy to adjust up to 27VDC.
 
- I'm assuming that's all that was supposed to happen? No LEDs lit up...
 
-Chris

Yes.  You have confirmed that there's nothing whatsoever wrong with the power supply section.
 
Just to be prudent, go over the PCB again.  Check every resistor with the silkscreen numbering and check every transistor.  Something is out of whack or you wouldn't be getting a short.
 
Just for safety's sake, adjust the buffer trimmers (RB12-L/R) by turning them all the way clockwise - 25 turns clockwise, at least.  Go over those transistors once again.  Try to make sure that everything is installed correctly.  Then see if you can apply power to the amp - watch that fuse!!  Let us know what happens.  If the fuse blows again, please accept my apology (I'll send you some more to make up for the loss.).  There are other things we can check (but it takes a lot of effort), but I'm counting on the hunch that perhaps the MOSFET bias is turned up way too high.  Otherwise, it's something we've all missed and we've had four people looking at it for two days, now.
 

 

[jdkJake]

EDIT: Tom - All TO-92s, MOSFETS and QB1 are installed according to layout

I still think it would not be a bad idea to ensure the tabs of one (or more) of the MOSFET's has not broken through the thermal tape and is making contact with a heat sink, thereby inducing a short.

Should be very easy to test. Using your DMM and with the board powered OFF, measure resistance from the tab of each MOSFET to the corresponding nut used to hold the part onto the heat sink. They should all measure as open (infinite resistance).

 

[tomb]

Quote:

I still think it would not be a bad idea to ensure the tabs of one (or more) of the MOSFET's has not broken through the thermal tape and is making contact with a heat sink, thereby inducing a short.
Should be very easy to test. Using your DMM and with the board powered OFF, measure resistance from the tab of each MOSFET to the corresponding nut used to hold the part onto the heat sink. They should all measure as open (infinite resistance).

This is a good thought - and it costs nothing to check it before you do what I suggested.
 
 

 

[Nomnomnom]

Just one quick question before I power this beast up. I installed those trimpots backwards because of a post I read elsewhere referring to the proximity of the adjustment screw to the tube and the casework involved. Will this effect the way I turn it to increase or decrease the bias?
 
Jake - All of the MOSFETS read 0 resistance

 

[tomb]

Quote:

Just one quick question before I power this beast up. I installed those trimpots backwards because of a post I read elsewhere referring to the proximity of the adjustment screw to the tube and the casework involved. Will this effect the way I turn it to increase or decrease the bias?
 
Jake - All of the MOSFETS read 0 resistance

Unscrew them - 25 turns counter-clockwise.
 
 

 

[jdkJake]

Jake - All of the MOSFETS read 0 resistance

"0" or does it display dashes or "INF" or something. Typically, a zero indicates a direct short.

What type of meter are you using?

 

[tomb]

Quote:

"0" or does it display dashes or "INF" or something. Typically, a zero indicates a direct short.
What type of meter are you using?

If he's measuring resistance, you're absolutely right!