Horrific G7 Problem!

[zebra50]

Good point from Gus.

Pins 4 and 6 swapped? Is the tube wired backwards / as a mirror image? What I mean is have your read the pin diagram as though you are looking from the top, or the bottom of the socket?

Post some photos up.

 

[CurtZHP]

Gus, that's what I'm finding.  It seems as if the plate is shorted to ground.  I've been over the PCB with a fine tooth comb and can find NO solder bridges or other stray connections.

Zebra, I used the Gryaf PCB layouts.  I have the little semi-circular PCB for the tube socket mounted such that the trace side is facing up (toward the capsule) and the component side is facing toward the XLR7.  The tube socket is mounted on the component side.  This puts the tube on the same side of the main PCB as the transformer (which is also on the blank component side of the main PCB).  Does that answer your question?  Pictures on the way.

 

[zebra50]

Just something to look out for really and double check.
When photo-etching one's own PCBs it is very easy to get the acetate flipped over, and when going point-to-point I always have to remind myself which side of the socket the tube pin diagrams represents. Same with XLR pin numbers! I know this from many of my own mistakes!

 

[CurtZHP]

Zebra, I did a bit of checking and found something rather unsettling.  Perhaps Jakob needs to chime in here.  Attached is a photo of my mic's tube PCB.  Compare it with the prototype photo from the Gryaf page:

http://gyraf.dk/gy_pd/g7/gic/gic_construct02.jpg


It looks flipped 180 degrees from what I've got!

When I etched the boards, I followed his instructions thusly (from the Gyraf site):

"The tracks side of the pcb is drawn MIRROR’ED to make closest contact between print and pcb. Think about this when doing the pcb’s. When the text on the board reads the right way around seen from tracks side, then you’re on right track."

(emphasis mine)



Now I just spent a bit of time tracing all the traces on the tube PCB through to the main PCB, and according to the schematic, all my connections are correct, so the discrepancy between the two photos appears to be a non-issue.  What do you think?

 

[CurtZHP]

Another wrinkle:

After pulling the tube out of the socket.  I still measure a dead short to ground from the plate side of the 100K resistor.  I lifted that side of the resistor, so now I'm essentially measuring the continuity of that trace on the PCB.  Sure enough, dead short.  I start following it down the PCB and removing components along the way.  The only other one I come to is the 2u2 cap just before the output transformer.  When I lift the side connected to the same trace as the 100K, I check the continuity again.  This time I show no short.  I put the 100K back and measure the plate connection against ground.  No more short.

Now I'm looking at the capacitor.  Measuring the side of the cap that connects to the transformer, I get a short to ground.........




HOLD EVERYTHING!!!!

I FOUND THE PROBLEM!

While I had the transformer end of the capacitor unattached, I again measured continuity.  No short.  I then barely touched the loose capacitor lead to the PCB and suddenly saw the result I was after.  So what went wrong???

Turns out, the lead was sticking through the hole in the PCB, just as one might expect.  I had snipped it off right at the surface of the other side.  However, there was still a tiny little tip of a lead poking out.  And what, pray tell, was that touching???

The shield can of the output transformer!

After laughing my head off and calling myself everything but a gentleman, I promptly resoldered the cap to the trace side of the PCB, keeping it well away from the shield can.

I powered it up and checked the plate voltage.  SUCCESS!

I put the body back on and connected it to the mixer and put on the headphones.  There was my voice, in glorious in-your-face tubeness!

And did I mention that this mother is QUIET!!

I thank you all immensely for your time.  I raise a frothing pint in your general direction.  Please accept my profuse apologies for bringing you all this way for such a simple thing.  And finally, to Jakob and the crew at Gyraf...

TAK!

 

[zebra50]

Good stuff - sounds like you've had quite a journey, solved all your own problems and you've learned a shitload about trouble shooting. You'll be hooked on the DIY now. Next project coming up.... 

 

[CurtZHP]

Oh, I was hooked years ago. 

Gimme something else to build!!

;D

 

[seavote]

"Please accept my profuse apologies for bringing you all this way for such a simple thing."

seems like most problems are "simple" but finding them's not so simple.
i placed some hard plastic between the mic transformer and the pcb in my build while trouble shooting my G7. sorry i didnt think to mention it. its definitly something to look out for with this project.
congratulations.