Monday 17 June 2013

Tuesday 4 June 2013

Tremolo Sound Clips

I really must buy a better mic.

Tremolo Troubles

I'd read that I might have some problems with the tremolo channel and so was prepared for it not to work immediately.  There are various discussions about the right kind of ECC83 to use in V3, with the consensus being that Sovtek is the way to go.

Initially I had no tremolo effect on the channel, no matter what the speed and intensity knobs were set to.  This was using the standard TungSol valve, the same as I've got in V2 (I paid the extra seven quid for a "premium tested" TungSol for V1).  I don't have a whole lot of valves lying around* but I do have a JJ ECC83S from my 5w amp.  I duly tried this and, once again, no tremolo effect.

I knew there was an ECC83 in my little blackstar, and I'd heard they used Sovtek valves as standard so I took the screwdriver to it.

Nice little amp but I'm not sure I like seeing valves next to ICs.

Anyway, once I'd forced the Sovtek from there into V3 (it seemed "fatter" and I really struggled to get it into the shielded base on the 1973) still no tremolo.

It was then I began to doubt my own sanity, and in particular my understanding of the "Cliff" type jack socket - and which lugs are connected at which time.  The picture below is a bit blurry but it clearly shows the incorrectly wired tremolo switch socket - the ground is wired across so with no jack in it it's going to be off.

Without powering down the amp, I very carefully snipped the diagonal wire with my insulated snippers (don't try this at home) and suddenly I was awash with tremolo. 

I tried it with all the valves and it worked strongly with all of them and perhaps too strongly with the JJ ECC83S.  At one point I started getting a bit of interference so I gave the wires to V3 a bit of a prod with my insulated prodding tool and it all settled down.

*Thinking about it, I probably have a total of 12 valves in the house which is probably more than average.

Saturday 1 June 2013

Calculating the plate dissipation

How hot is your amp running?

The EL84 was designed to be running at about 12w - that's to say that at this output it will last pretty much forever.  Obviously part of the appeal of valve amps is the overdriven sound at various points - clipping preamp stages, power tubes running hot and (to a lesser extent) speaker distortion.  So running the EL84s at beyond their rated dissipation adds to the tone, at the expense of their operational lifespan.  This can be controlled by changing the value of the cathode resistor, but it's probably a good idea to know how to calculate it and make a judgement based on that, rather than just eyeballing the tubes to see if they're glowing red hot.  These are my calculations which I think are correct, based partly on the Amp Maker testing notes.

                                                   Cathode voltage (D)
Cathode current  =                 --------------------------
                                               Value of cathode resistor (120 ohm)

                                          Voltage over Cathode resistor (D-E)
Screen current   =            -----------------------------------------
                                              Value of can cap resistor (1500 ohm)

                                            Cathode current - Screen current         
Plate Current =                   ------------------------------------

Cathode-to-plate voltage  =  Plate voltage - Cathode Voltage

Plate dissipation   =   Plate Current  x  Cathode-to-plate voltage


It's alive!

Just a quick check for output from the amp.  The mic on this phone is crap but I don't have a better one.


Point B has 0v when it should have 83v.

Tracing the circuit back, the voltage at B is going to be coming from V2 pins 2 & 7 (grid) and pins 8 & 3 (cathode).

Tracing it back further, the anodes on V2 (pins 1 & 6) go back to the 100k resistors (18 & 19).  Which are connected to...  Hang on a minute...

Completely missed that connection on the blue dashed wire (running under the board).

Time for an ugly jumper.

My voltage readings

I think I've found the problem.