Testing A Trem Optoisolator

[Daver]

Call it an optoisolator, LDR, trem roach or what have you, how can you tell if it's good? The one in question is in a Gibson Mercury head. It looks like it was replaced at some point. It's a Sylvania LDR packaged in a metal tube, so I can't see inside. The bulb side measures 192 ohms and the resistor side measures infinity on my meter on the 40 Meg setting. Can I substitute a Fender style LDR for this? Thanks.
Dave

[Robert M. Martinelli]

Hi,
since the bulb filament looks ok, you could try to inject a little current with a 9V battery, wired in series with a pot to limit current to a safe value, slowly turning the pot and with an Omhmeter on the LDR side you should be able to see the resistance falling as you increase the current through the bulb. Time ago I had to deal with a VTL5C3 or something on a Fender amp and that was the way I tested it.

It must be said that was an LDR with a LED, not a bulb....

With a 1 K pot you can go from about 7 to 46 mAmps, if you want to play it safe just add a resistor ( say 390-470 Ohms with LED LDRs ) in series and that' ll limit the max current should you go wrong with the pot' s setting.

Hope this helps

Regards

Bob

[Daver]

Bob,
Thanks. I'll try that. The part in the amp now is labeled LDR. It may not be the correct replacement part. If your test shows it's no good, I'll try using a Fender trem roach and see what we get.
Dave

[Daver]

OK, I did the test recommended. I could get the resistance down to only about 86K with the pot down to 0 resistance. The change in resistance is very slow. I can't see how this could keep up with the fast change in voltage the trem circuit presents. Either this thing is no good or it's the wrong spec. I guess I'll order up a Fender trem roach and give it a try. Thanks.
Dave

[Robert M. Martinelli]

Hi Dave,
glad to hear that I have managed to be of some help somehow,
if you orded a Fender trem roach it' ll be almost surely a VTL5C3, ( which, as I was telling you, has an LED instead of a bulb ); searching the internet I found the relevant data sheet at the Perkin Elmer web site. If you wanna test it before installing it keep in mind the maximum forward current is about 20 mAmps, so use a 470 Ohm resistor in series with the pot, this way you' ll be somewhat lower than that at the max current setting.

If the previous LDR was a bulb type and the trem roach you 're putting in is indeed a VTL5C3, be sure to add a series resistor in the trem circuit, right before the LDR LED, measuring the LDR input voltage on the amp and using the LDR datasheet to calculate it. Omitting the resistor would almost surely blow the little thing away. You could also implement the pot ( well, here it is better to use a trimmer ) and resistor network, to better control the effect, shaping the trem depth with the new trimmer as a "depth preset".

Should the new LDR be of the bulb type, it' ll probably work with no modifications.

Good luck

Bob

[52 Bill]

Originally Posted by Daver

I guess I'll order up a Fender trem roach and give it a try.

Do you have the schematic for this amp? Is it solid state or tube? Is there a part number or specs listed for the original LDR?

If you order a standard "Fender trem roach", it will contain a neon bulb and an LDR (Light Dependent Resistor). This will only work in an amp that has a high enough voltage to trigger the neon lamp. This usually means a tube driven circuit.

From the resistance measurement that you listed for the lamp/bulb, it sounds to me to be an incandescent bulb. This usually means a transistor circuit. How much voltage is driving the lamp circuit?

If no specs are available, you can reverse engineer the circuit and build your own replacement. Measure the voltage that goes to the lamp. This will vary depending upon the depth and speed settings. Find an incandescent bulb that works with this voltage range and then heat shrink it together with an LDR.

EDIT: I found a couple of schematics for the tube version of this amp and it shows 6 volts on the lamp. If this is the version that you have, the Fender roach will not work.

[Daver]

Bill,
There is a copy of the schematic at Schematic Heaven. It's the Gibson Mercury Medalist. It opens as a pdf so I can't link to it directly. It's tube driven and has a 150vdc supply. There is no part number on the schematic. The part that is currently in it is a replacement, not the original. I have no idea if it ever worked. The amp came in with no power cord, a missing trem cap and an exploded filter cap. Everything works now but the trem. Thanks.
Dave

[52 Bill]

Originally Posted by Daver

It's tube driven and has a 150vdc supply. There is no part number on the schematic.

If you look at that schematic, the 150 volt supplies the lamp, which supplies the plate of the driver tube. The plate voltage reads 144 volts so there is a 6 volt drop across the bulb. I'd try a 10-12 volt lamp in place of the existing unit to see if the bulb will flash with the oscillator.

[Robert M. Martinelli]

Originally Posted by 52 Bill

If you look at that schematic, the 150 volt supplies the lamp, which supplies the plate of the driver tube. The plate voltage reads 144 volts so there is a 6 volt drop across the bulb. I'd try a 10-12 volt lamp in place of the existing unit to see if the bulb will flash with the oscillator.

Once checked that the trem osc works in the way Bill wisely suggestsed, I would indeed recommend to give a try to a VTL5C3, with a, say, 390 Ohms resistor in series, this way you 're less likely to have problems in the future ' couse LEDs life expectancy is much longer than bulbs'.

Hope this helps

Regards

Bob

[Daver]

Bill,
OK I'll give that a try. I'd rather get something in one package than put something together myself. Maybe the VTL5C3 that Bob suggested is the thing to try. It would be nice to have an original part number, but that's life. Thanks much.
Dave

[52 Bill]

Originally Posted by Daver

I'd rather get something in one package than put something together myself. Maybe the VTL5C3 that Bob suggested is the thing to try. It would be nice to have an original part number, but that's life.

Dave:
I looked at all of the Gibson schematics that I have, and they used this circuit design on a lot of different models. The only part number that is listed on the schematics is something meaningless like LDR500. Something I did notice was that on some of the schematics, the voltage drop across the bulb is as high as 14 volts, so if you try to build one, use a higher rated bulb.

[Daver]

OK, all I have around here are #47 bulbs. I wired two of them in series and hooked it up. I got nothing. No light at all. These must not be the correct thing to use. I got 15 volts on either side of them. Something is weird. With the original/replacement opto I had 150V supply and and 144V on the plate of the tube just like the schematic. The tube tests fine and all surrounding circuitry is correct and in spec. The schematic shows a bulb not an LED. I obviously don't have the right type of bulb. I'll check things over again. Any other suggestions for a correct opto would be great. I can find the VTL5C3, I just want to be sure it will work. Thanks!
Dave

[Robert M. Martinelli]

Hi Dave,
I would definitely give the VTL5C3 a try, just remember to keep the voltage across the LED at about 2VDC, and to limit the current at no more than 20 milliamps. I don' t have the specs for the #47 bulbs you tried, but since the voltage across the original LDR was 12-14 V, and your bulbs just don' t seem to fit, I would also test the trem osc using an LED and a resistor, or as already stated, a resistor and a trimmer in series...say 470 Ohm for the resistor and 1 KOhm for the trimmer - ( mind the LED's polarity and be sure to keep fwd voltage and current within the LED specs ), with the tremolo osc in good working order, you should be able to see the voltage swing reflecting in changes in the LED' s emission. Once verified the thing works, just keep the resistor and trimmer that way and replace the LED with the emitter pins of a VTL5C3, ( which BTW is a pretty cheap component, should cost less than 5 USD ).
Good luck
Bob

[Robert M. Martinelli]

Hi,
I checked the schematic diagram and the tube curves, hope the results are helpful...and correct
According to the schematic, you have 144 VDC on the plate ( quiescent ) 141 V drop across the tube ( the cathode bias voltage being 3 VDC, this gives the grid a bias of -3VDC ) you have a quiescent current of about 11,5 mAmps, the gm is about 2,150 mA/V ( indicating you will have a plate current change of 2,150 mA for a change of 1 Volt at the control grid ). The grid voltage cannot exceed 0 V, thus the maximum voltage swing caused by input signal at the grid will be 6 V ( 0 to -6 V ), and the plate current will swing from 'bout zero to 24 mAmps ( with the quiescent point in the middle ). In case you didn' t do this before, adjust the 500 Ohm cathode trimmer to get the proper bias ( voltage between the cathode and gnd should be 3VDC ).
Hope this helps
Regards
Bob

[Daver]

Bob,
Yes I have adjusted the cathode trim pot. That hasn't helped. I can see the oscillator section working on a scope and the 0A2 regulator tube flashes with the trem. I'll set the cathode voltage to 3V just to be sure. I'll order up a VTL5C3 and see what happens. Thanks for all the help.
Dave

[slorinczi]

Thanks for all the excellent info gents! I'm struggling to fix the vibrato on a Sunn Sceptre. The bulb fires just fine, but the vibrato effect is very weak. I measured the LDR resistance, but under both light and dark it was in the high Megohms, making me think it's gone bad.

Seems like the "standard" VTL5C3 should work, but I do want to make sure I'm in the ballpark as far as voltages go. Excuse my inexperience, but if I'm reading this schemo correctly, the bulb expects to see 4.5V:



According to Vactec, the unit specs are:

LED Reverse Breakdown Voltage: 3.0V
LED Forward Voltage Drop @ 20mA: 2.0V (1.65V typ)

So does this mean I have to "kill" 2.5V or so to avoid burning out the bulb prematurely?

Thanks,
Seth

[loudthud]

The problem with the Sunn circuit is that when the bulb goes bad, it kills the signal because the LDR is in a high impedance state. I have attached a schematic of the later model 190L. The changes that were made didn't change the basic operation of the circuit but the LDR was moved to the next stage between the tube and the tone stack. In this position it is after the Reverb pick-off so the modulated guitar signal isn't going thru the reverb. (That must be kind of strange.)

The driver transistor is probably close to being saturated so the current is set primarily by R114 in the 190L schematic. Adjust it as needed to make the new part work.

[slorinczi]

Hmm, got it, and thanks for the info! Now I'm thinking that I may try to insert a generic (i.e. Radio Shack) CdS part into the photocell circuit and see if that does the trick, given that the bulb fires fine.... Not sure of the specs I need, I'm guessing high MegOhms when dark and under 100K when light?