Gibson Hawk tremolo problem

[Tom Phillips]

Now that we know that the bulb and LDR are not stock please tell us more about the bulb that you used. Below is detailed information about the measurements I made on the bulb that Gibson used. This is the same information from the thread that Steve A linked above ( Gibson Hawk trem ) and I have restored the photo that went missing from the old thread. I suspect that you may need to use a different bulb. Probably a higher voltage one.

The attached photos show the parts inside a Gibson assembly number "LDR-500" from a ~1965 Gibson GA-77RET Vanguard.

The bulb on the right is original and the wires had corroded off close to the base. I was just able to salvage enough of the stubs to allow attachment of the red lead wires shown in the photo. It’s barely visible in the photo that the bulb had the number “24” printed on it. Actual current draw with 24V applied to the bulb was 18mA and the filament glowed bright orange rather than white hot. Since I salvaged the old bulb I didn't need to source a replacement so my efforts stopped there and some research will be required to find an equivalent bulb.

The LDR on the left was originally inside a glass capsule. The dark resistance is ~10MΩ long term and ~1MΩ short term. Under light the resistance is ~1kΩ.

The use of an incandescent bulb with the proper "pre-heat" adjustment via the 500Ω pot gives a nice smooth tremolo rather than the choppy sound of the Fender circuit which used the neon bulb. You cannot substitute a neon bulb in the trem circuit used in your Gibson Hawk.

Cheers,
Tom

[Chuck H]

Now that we know that the bulb and LDR are not stock please tell us more about the bulb that you used.

From post #1

I made one out of a 6838 bulb (28v/25mA)

[words453]

Olddawg, I will try the flashlight trick, the schematic shows an incandescent bulb so I'm not sure a neon will work. I did try an led but it stayed lit as well. Chuck H, the bulb stays dimly lit when the trem is off and I do have a 12AU7 in the socket. I will check to see if there is DCv on the grid. Tom, the bulb is 28v/24mA, I will try to find a higher voltage bulb and see what happens. Thanks for the advice, I will do some more experimenting and let you know what I find.

[words453]

Update to the ongoing saga; I tried two 28v bulbs in series and parallel and neither worked. I tried a vtl5c4, no go. Out of desperation, I tried a fender style bug, no change. It's the same result every time, the bulb stays lit keeping the resistance low and bleeding signal to ground. Btw, there is no DCv on the grid. I have no choice but to go back to the vibrochamp trem cicuit and try to figure out the parasitic issue. Let this be a cautionary tale to anyone who wants to purchase a vintage Gibson!

[Chuck H]

I don't think you should give up just yet. If the bulb is staying lit such that the trem won't work as intended and you've tried all manor of bulb/bug with the same failure mode then it's likely something about the circuit is, wiring or component values, is wrong causing the floating point of the bulb driver to keep it lit when the oscillator stops. Defaulting to another trem circuit assumes the design that Gibson used doesn't work and I don't think that is the case.

[bob p]

I disagree with the words of caution that suggest we should avoid vintage Gibson amps. The problem is not that the circuit can not be fixed; the problem is that your diagnostic efforts have not yet produced a diagnostic result. This problem is exacerbated by your lack of an oscilloscope. That puts you in a frustrating position.

So far it seems that you've focused on lamp-swapping approach. That approach will only help if and only if only the lamp is the only part of the circuit at fault. I haven't re-read the thread, so pardon me if I ask a stupid question -- Have you already verified that each stage in the tremolo circuit is functioning properly? What is the range of voltage sweep on the oscillator's voltage output?

Back in Post 6 I mentioned that you need to divide the circuit into 3 stages and verify that each stage is functioning properly. To me it seems like you're focusing on the lamp, and you're ignoring the parts of the circuit that are not the lamp. You haven't yet verified the oscillator's function, have you? Or did I miss something?

If you don't have a scope to trace waveforms then an analog voltmeter would be helpful. In cases like this a DMM that has slow response time just won't cut it.

[Chuck H]

Bob is right about the DMM not being able to catch the oscillator function. But you did say the trem in working, so I'll take that to mean that the oscillator is oscillating. So we need to find out why stopping the oscillator doesn't kill voltage to the roach bulb. Can you verify the voltage on V4 pin 8 with the trem switched off please.?.

[words453]

Thanks for the responses. You're right bob p, the circuit is not to blame, rather it's my lack of understanding and skills. Last night out of frustration, I rewired the circuit into a vibrochamp style and it works great, no parasitics! The trem was working fine with the bug, it just lowered the volume too much. When I have some more time and feel froggy enough, I will try to solve the issue again. I don't like admitting defeat, but for now I need to back off. I do see now that the oscillator may not be swinging low enough to kill the bulb, I just couldn't figure why it stayed lit with the switch off. Anyway, thanks for the help and when I try to fix it again, I'll post here as this has been the most helpful forum out of the three that I've visited about this issue.

[Tom Phillips]

Thanks for the status report. When you tackle this again I encourage you to add to this thread so we can reference the background information about this specific amp.

I have never been able to determine the part number and specifications for the original incandescent bulb. As I mentioned in post #14 "Actual current draw with 24V applied to the bulb was 18mA and the filament glowed bright orange rather than white hot." Since the one original bulb only produced the orange glow with 24 volts applied, it seems to me that the voltage rating must have been higher than 24V. I didn't experiment further at the time because I had fixed the problem and I didn't want to add any further stress and take the chance of damaging the only bulb I had. For the same reason, I haven't taken apart the working light bugs in any other Gibson amps. The wire connections are really fragile and the bugs are bonded into metal tubes. It would be really risky to mess with a working example.

I'll keep thinking about this issue and watching for more information as opportunities arise.

Cheers,
Tom