Tremolo problem

[bluesguitar]

I fired up my '66 Twin Reverb for the first time in months. The tremolo roach is working because I can hear a ticking, but the tremolo effect is barely noticeable. However, the interesting thing is that when I turn my intensity knob down there is a corresponding volume increase in the amp. When the tremolo knob is turned all the way off there is quite an increase in amp volume. I've never experienced a relationship like this with the intensity pot and the amps volume. It acts like the intensity pot is really draining some serious volume from the amp, but doing very little in terms of it's designed tremolo effect. Any ideas about what the problem would be?

[Phil_S]

I take your post to mean it is a real 1966 TR and not a reissue. I wonder if the board has become conductive. This is one of the known problems with the material that was used. It will absorb water over nearly 50 years.

It is not supposed to tick. That may be a lead dress issue.

Finally, look at this, 2nd picture...really...worth a look and more: http://www.timeelect.com/jbl-twin.htm

[pdf64]

I suspect that the opto resistor part of the roach is dead, permanently low resistance.
The ticking indicates that the neon part of the roach is ok.

[sliberty]

Its always a good idea to try a different tube before doing much else when it comes to trem issues. But if that solves nothing, I would also suspect the roach, regardless of the ticking.

[Leo_Gnardo]

I suspect that the opto resistor part of the roach is dead, permanently low resistance.
The ticking indicates that the neon part of the roach is ok.

Right, the photocell is kaput or disconnected. Lots of old ones going bad these days. Lots of amps ticking like Boris Badenov's time bomb too. Instead of sending away for a pre made roach/bug, I make my own. And so can you. Instead of a neon lamp, I use a hi brightness LED, yellow or orange. (just bought some red, we'll see how they work out.) Lots of photocell choices, you can drive yourself crazy. I just get the ones intended for Morley pedals, from Antique (tubesandmore.com). Slide into a piece of heat shrink, shrink it, attach LED long lead to the 100K, short lead to 10M, photocell where it was, you're back in biz, and NO CLICK! Woo hoo!

So far some of the ones I made have lasted 4-5 years, no sign of trouble.

If anything the trem may be deeper, more cutoff, than the old neon bug. So, you can dial up less than 10, or put a resistor in series with photocell to limit depth of signal attenuation.

Some have asked "doesn't that high voltage hurt the LED? or tube?" There's a 10 megohm resistor in series with LED. Hard to hurt anything with that in the way. Feel free to increase its value if you're worried.

Try it you'll like it.

[Stevem]

X2 with Leo, the roach is toast!

[pdf64]

Leo, that's inventive!
Just to note that the neon / LED is in series with the 100k plate resistor; the 10M resistor also connects the plate to HT, ie it's in parallel with the 100k+neon/LED http://bmamps.com/Schematics/fender/twi ... _schem.pdf
So the (almost) full plate current passes through the neon / LED.
The trem may seem deeper partly because the LED may be brighter, but also because it may be on for longer per cycle, as the neon would need~90V across it before it strikes, whereas the LED just needs ~2V.
Perhaps the neon would be better simulated by also adding a suitable zener in series with the LED?

Whatever, if it works reliably as it is and the customer is happy, then all is good.

[Leo_Gnardo]

Leo, that's inventive!

Naah, just stolen, aherrrmm, borrowed from an amp I met one day.

Just to note that the neon / LED is in series with the 100k plate resistor; the 10M resistor also connects the plate to HT, ie it's in parallel with the 100k+neon/LED http://bmamps.com/Schematics/fender/twi ... _schem.pdf
So the (almost) full plate current passes through the neon / LED.

True 'nuff, just had another peek at a typical Fender vibro. Plate & lamp all floating at near 400V. Better idea to dim the LED: increase the 100K instead of the 10M (drat I thought I could use up some hi R resistors... got way too many.)

The trem may seem deeper partly because the LED may be brighter, but also because it may be on for longer per cycle, as the neon would need~90V across it before it strikes, whereas the LED just needs ~2V. Perhaps the neon would be better simulated by also adding a suitable zener in series with the LED?

Def. longer light flash per cycle, neon rides a sawtooth, voltage rises until the neon "breaks down", gas becomes conductive, burst of current & light drops the saw almost vertically & the cycle continues. LED a bit more linear, yes longer flash time. Those interested may fiddle with zener in series, or very low value zener, or a couple of cheap diodes in series across LED (voltage clamp), or increase the 100K value to idealize vibrato characteristics. But so far nobody's complained, most enjoy the no-click first & foremost. Lots of Fender owners have no pedal, don't want one but do want vibrato, then I put a cheater plug in the vibrato control jack.

Whatever, if it works reliably as it is and the customer is happy, then all is good.

Right on!

[xtian]

Want to add my experience--very good!

I bought both traditional neon "bug" and LDR only, to replace the dead roach in my Bandmaster Reverb.

The traditional neon bug works, but ticks loudly.

I made my own bug with the LDR and a bright white LED. IMPORTANT! I burned out several LEDs experimenting with orientation. The anode must go to the 100K resistor, and cathode to the 10M resistor.

The LED bug works really well. No tick, and nice deep tremolo.

[tubeswell]

...Instead of sending away for a pre made roach/bug, I make my own. And so can you. Instead of a neon lamp, I use a hi brightness LED, yellow or orange. (just bought some red, we'll see how they work out.) Lots of photocell choices, you can drive yourself crazy. I just get the ones intended for Morley pedals, from Antique (tubesandmore.com). Slide into a piece of heat shrink, shrink it, attach LED long lead to the 100K, short lead to 10M, photocell where it was, you're back in biz, and NO CLICK!

That is cool

[martin manning]

For experiments, you can still get an assortment of CdS LDR's at Radio Shack (if you still have a Radio Shack), and LED's of various colors and MCD output. The seemingly high voltage isn't a problem for the LED since the current through it will be limited to a couple of mA by the driver triode and its 100k anode resistor. The LED won't see more than its nominal forward voltage drop, unless you reverse it. In that case, it will see the voltage drop across the 10M resistor, hundreds of volts, where LED PIV is usually only around 5V.

[xtian]

See? Martin knows stuff I find out by experimental *ahem* method.