Thor Troubleshooting

[mydementia]

Thought I'd start a new topic for this one rather than hijack the ROG announcement thread... ref: http://www.diystompboxes.com/smfforum/index.php?topic=52621.40

Hi guys.
I built the Thor circuit from ROG (http://runoffgroove.com/thor.html) twice now - once on perf following the ROG PCB and the second time using Xavier's DIYLC layout.  Both exhibit the same behavior - when turning the gain pot (1MA) it's clean(ish) from 7-8 - squeels like a pig from 8-4 - huge gain from 4-5 (7=min, 5=max).  I've replaced all the JFETs, the opamp, and the pots - no change... 

Here's a snapshot of the DIYLC I used with a slight correction per Gaussmarkov's comments and a second check of the schematic (Trace between C6 and C9 is cut and e17 connects to e14).  My subs are boxed in red and my voltages are on the right (figured if we're going to troubleshoot, we oughtta do it right!).  I'm including the layout because Q3 and Q4 are swapped compared to the schematic.  (Xavier - if you have a problem with my including your work, PM me and I'll take it down.)


Any help with what I'm screwing up will be greatly appreciated.
Thanks for looking.
Mike

[petemoore]

  I'd get the schematic and the layout printed, take a pen and paper and take notes...
  It looks...like...maybe Q4 doesn't have a G/S cap?...supposed to be a Minibooster bias transistor? Anyway that minibooster has a little twist that may have gotten folded in the translation to layout..I'd check it on paper first.

[mydementia]

Pete - I think the GS cap on Q4 (Q3 in the schematic) is C6 (100nF) - goes thru Q3D then R8 (390R).  Am I reading wrong?
After I read Gaussmarkov's comment that there was a problem with the layout, I went through it with a highlighter tracing the schematic (don't know why I didn't start this project by doing that...lazy, I guess).

Mike

[petemoore]

  I don't think so.
  I started reading in the middle of the layout...still learning the pitfalls of reading layouts.
  Looks to me like Q3/Q4 parts labels are reversed on the schem to layout?
  Ability to be misread is greatly reduced when parts labels from the verified schemata are followed exactly by the layout.
  Schematics are able to be read/comprehended/discussed/debugged/analyzed from any point in the schematic. Layouts require Finding the darn 'thing' your looking for by looking for unique circuitry around the part on the schematic, the looking for that in the layout, often having to ID the input, output, or a PS rail and work from there...twisted...
  Schematic on paper, layout on paper, pen in hand, note the number of connections at each node and note polarized component orientations, continue to next node.
  Count the exact number of connections at that node on the layout, X off each correct node as you go.
  I had Q3/Q4 reversed in my twisted mind...I'll go back and see if I can see straight this time. Reverse mirror is bad enough [layouts is like a house of mirrors], but part mislabeling warps reality.
  If you think a circuit will want to be looked at, a schematic is what is needed, if you think a layout would lend itself to discussion, if it doesn't exactly follow the schematic part label #'s...I don't know what I'm thinking or typing about !..very confusing, excellent opportunity for errors.

[petemoore]

  I'm including the layout because Q3 and Q4 are swapped compared to the schematicHere's Please Excuse my OT, unhelpful, unnecessary, incorrect response mydementia? I became frustrated.
   RG's take on it:
  Schematics are indeed intended to match the electrical design and to make the circuit easy for humans to understand and work with.

Layouts are NOT intended for understanding. Layouts are purely for satisfying mechanical and manufacturing needs. They really are not even slightly intended for humans to understand. It's only because humans are such good pattern recognizers that we can go from layout back to schematic at all. And the step between layout back to schematic is best and least confusingly done without part numbers at all if there is any chance that they can not match. In fact, knowing what is connected to what on the layout and finding that on the schematic is far more reliable than knowing part numbers.

Schematics are for humans; layouts are for machines.
  Audio probe. Transistor voltages ?
  In some cases, the de-powered dogma approach...clip DMM to a rail and start measureing D/G resistors, whatever else can be measured from there, then clip to the bottom rail and measure S/G resistors and continuity at all grounds from there.
  IOW: DMM all Resistors and connections w/schematic there !

[mydementia]

Interesting... I just plugged in my Thor and adjusted the gain pot to maximum squeel.  Then I diddled the trimmers (mostly the Q2 trimmer) until the squeel went away.  Ended up working pretty well with the Q2 drain set around 8.4V and the Q1 drain set around 5V - very strange...  I checked the source resistor coming off Q2 - it is, in fact, the prescribed 1k...  At least the gain pot is useful from zero to max now...
I'm looking forward to some sound clips so I can figure out if what I hear is what was intended...

Mike

[Xavier]

Mydementia,

I will redo the layout when you're done with it. Or I can pm you the .diy file if you want to modify it.

Sorry guys to give you headaches, wasn't my intention.

[petemoore]

  Thanks for providing the layout, it is just a new layout, soon it will be a good verified layout...all is well !

[phil]

Sorry guys to give you headaches, wasn't my intention.

Xavier - Thanks for the layout ... we really shouldn't expect a new and unverified layout to be perfect so don't sweat that it wasn't perfect the first time. Personally, I'm glad you provided the layout so quickly after ROG released the THOR as I didn't want to create a PCB and I haven't quite gotten the art of creating layouts yet. As petemoore said, it may have been new and unverified but as more of us use it, it will get refined and become a good verified layout for members to use.

Thanks again for sharing it!

[liakos]

Sorry guys to give you headaches, wasn't my intention.

xavier you are the man!!!

[mydementia]

Success!
After several Thor builds exhibiting the same squeel problem, Brian at ROG offered to have a look at my circuit.  After much deliberation, it turns out that the proximity of the switch wires was causing my squeel.  This issue was compounded because I thought I was being 'smart' by using a single 'on-off-on' toggle switch.  When I cut the switch out of the circuit and separated the wires, it works great.  Even with the single switch in the circuit, the wires can be positioned such that the squeel goes away...I might try shielding the switch wires on one of my Thor builds... I still don't have any SPST's!!

The circuit sounds great (like all ROG stuff!!).  I built from the ROG PCB and Xavier's perf layout - both work great. 

Thanks so much to Brian at ROG for taking the time to go through my build - definitely above and beyond...
He also walked me through more appropriate grounding scheme (could have been creating ground loop issues - but the switch wire proximity was definitely the main problem here...).

Have you built your Thor yet?
Mike

[grolschie]

Cheers for that. Any chance of some audio clips? I have heard the one at ROG, but would like a better idea of how it sounds. I wonder how it compares to the Eighteen?

[stm]

Cheers for that. Any chance of some audio clips? I have heard the one at ROG, but would like a better idea of how it sounds. I wonder how it compares to the Eighteen?

Now there are three clips available.  The two newer ones show Thor sound alone and shall give you a good idea on how it sounds.

http://www.runoffgroove.com/salvo.html#thor

[grolschie]

Ahhh.... thanks. I am reminded of my Rat2 pedal somewhat. Nice. :-)