1969 Univibe Issue

[GuitarFreak302]

I am currently working on a 1969 Univibe for someone and it has me stumped. When I opened it up I noticed that one of the LFO transistors was shorted from base to collector. I replaced it with a 2N3904 and made sure the pinout was the same as the 2SC828 that I pulled out. When I turned the Univibe on and used the footswitch the bulb was oscillating like it should but then it would stop after messing around with the footswitch position. I noticed that the transistor I just put in became shorted from base to collector just like the one that was in there previously. So once again I replaced that same transistor and the oscillation came back. Then when the speed of the bulb oscillation was changed, the bulb would get brighter and the oscillation would speed up but then like before it would stop. I checked that transistor and it was shorted from base to collector again. That is the only transistor in the LFO that is shorting out and dying. The bulb driver still works and the other transistor in the LFO Darlington pair is also still good. I checked the resistors around it and all of them are within spec. The diodes are also still good. The bulb driver was changed to a 2N3904 also and it does get hot but I have heard that other Univibes have that issue. I will post voltages when I get a chance to take the readings. I am hoping it is something simple that I have overlooked but your help would be appreciated. Thanks.

Ryan

[Danich_ivanov]

There might be something helpful.
http://www.geofex.com/Article_Folders/univibe/uvfrindx.htm

[GuitarFreak302]

Here is the schematic I am using. Q12 is the transistor that keeps shorting out. Some of the resistor values in this schematic are different than what is actually in the unit but the ones in the unit are still original from the factory. I did not create this schematic I am only using it as a reference.

[Kevin Mitchell]

I'd recheck the pinout again and again and again... The trouble you're having would drive me nuts lol

I've made a clone of a Shin-ei Univibe and used 2SC539 for the LFO transistors.  I wouldn't be surprised if your 69 model takes after the Honey Univibes that came before the Shin-ei. There are some minor cap and resistor differences between the two. Like using .33uF caps-(if I recall correctly) in the LFO instead of the 1uF. This gave way better speed range in my opinion.

Anyways... Could we perhaps get some photos? 

[GuitarFreak302]

I'd recheck the pinout again and again and again... The trouble you're having would drive me nuts lol

I've made a clone of a Shin-ei Univibe and used 2SC539 for the LFO transistors.  I wouldn't be surprised if your 69 model takes after the Honey Univibes that came before the Shin-ei. There are some minor cap and resistor differences between the two. Like using .33uF caps-(if I recall correctly) in the LFO instead of the 1uF. This gave way better speed range in my opinion.

Anyways... Could we perhaps get some photos? 

Here are some pictures I took last night.

Sent from my SM-G950U using Tapatalk

[Danich_ivanov]

Looks awesome!

I have a weird question though. Have you considered making a different LFO, not the one that original univibe had, but something more approachable, like a phase 90 relaxation lfo type, or in that ballpark? Because it seems to me that lfo that univibe has is...sort of...ridiculous. I know that it's a uni-vibe, and it says it all, but just think about it. I haven't personally build a univibe as it is, but I did a very similar thing based around jfets with trigger integrator lfo, and it does univibe thing really well.

Just saying, I know that it's a touchy subject. 

[Kevin Mitchell]

Looks awesome!

I have a weird question though. Have you considered making a different LFO, not the one that original univibe had, but something more approachable, like a phase 90 relaxation lfo type, or in that ballpark? Because it seems to me that lfo that univibe has is...sort of...ridiculous. I know that it's a uni-vibe, and it says it all, but just think about it. I haven't personally build a univibe as it is, but I did a very similar thing based around jfets with trigger integrator lfo, and it does univibe thing really well.

Just saying, I know that it's a touchy subject. 

Modifying an original univibe is a thought of nightmares. I'd be after blood if someone tampered with an original Shin-ei like that. Besides, the design of the LFO plays a massive part in what we identify as a univibe sound - apposed to your garden-variety phaser. Well... that and a few other attributes.

Like most things in this world - they work beautifully when they're working right.

I'd check out the capacitors next. Other than that - I'd be mindful about any damage to the phenolic (paper based) PCB board since many old univibe boards have been compromised due to moisture over the years.

The issue also could be super obvious. I'm just giving advise as if I were in your shoes.
Totally jealous that you get to work on one of these.

[GuitarFreak302]

I was thinking about changing out the caps in the LFO next, like you suggested because that is usually the first thing I do on vintage pedal repairs. My customer was actually selling this Univibe on Reverb in non-functioning condition and I messaged him to see if he would want it repaired. I knew it was a transistor issue right when he told me it wasn't oscillating. I do see myself lucky to be working on such a sought after pedal. I have been doing pedal repairs for over 5 years now so I can for the most part tell what is wrong before going through them. I have never repaired or even seen one of these Univibes in person before so I thought I would ask for help if I needed it.

Sent from my SM-G950U using Tapatalk

[Kevin Mitchell]

Just a thought... that's not an original bulb. Perhaps other parts were replaced and should be checked - assuming you're not the one who replaced it.

Shoot me a PM if you need some parts. I think I have spare transistors from my clone build kicking around.

Just remembered I still need a enclosure for the damn thing 

[anotherjim]

Possible excessive volts? Is it a 110/120v model being used on 220/240v? Should be about 24vDC Q12 collector to ground

Possible short somewhere?
Without switching it on, meter resistance from Q12 emitter to ground. Play with controls that you think made it fail before. Do you see any suspiciously low resistance?
For example, there could be something in a pot shorting to its case and then to ground.

[R.G.]

[...] I noticed that one of the LFO transistors was shorted from base to collector. I replaced it with a 2N3904 and made sure the pinout was the same as the 2SC828 that I pulled out.[...]  the transistor I just put in became shorted from base to collector just like the one that was in there previously. [...] That is the only transistor in the LFO that is shorting out and dying. The bulb driver still works and the other transistor in the LFO Darlington pair is also still good. I checked the resistors around it and all of them are within spec. The diodes are also still good. The bulb driver was changed to a 2N3904 also and it does get hot but I have heard that other Univibes have that issue. I will post voltages when I get a chance to take the readings.

Most bipolar deaths are from only a few causes. The biggies are overvoltage, over heating, and overcurrent, which is really a version of overheating. You might possibly be having issues with reverse breaking of the base-emitter of Q12. Maybe.

I would look at the power supply in great detail. These things have a crude power supply, and if it's intermittent, it might be punching through the collector on Q12. I would also put a small signal diode like a 1N914 or 1N4148 with anode to emitter, cathode to base on both Q11 and Q12. This immunizes the base emitter from reverse transients.

Another thing to look at is that string of capacitors from Q12 emitter to ground. If these go intermittent-shorted or low resistance, it will over-current Q12 and kill it. It would be good practice to replace those three caps just on speculation.

And let us know those voltages.

Because it seems to me that lfo that univibe has is...sort of...ridiculous. I know that it's a uni-vibe, and it says it all, but just think about it. I haven't personally build a univibe as it is, but I did a very similar thing based around jfets with trigger integrator lfo, and it does univibe thing really well.

It is very odd. It's a current-mode phase shift oscillator. It looks very strange to people used to looking only at voltages, but it does a good job in this situation while taking up very few parts to adapt its output to the needs of the lamp driver. The earlier variant of what would become the Univibe was the RT-18 (IIRC) "Resly-Tone". This one did have a square wave oscillator in it, an early variation of the integrator/trigger. oscillator.

[GuitarFreak302]

Thanks for the replies. I will swap out those 3 caps, check voltages, and add those diodes for the transient protection. I will report back my findings. Why would the power supply only take out Q12 and not Q11 as well? Is it because of the Darlington setup they are in?

Sent from my SM-G950U using Tapatalk

[GuitarFreak302]

I replaced the 3 capacitors in the LFO. I added the diodes from the emitter to the base. I then replaced the blown 2N3904 with a MPS8099 because I ran out of 2N3904 transistors. When I turned on the Univibe I had the intensity knob fully counter clockwise. The bulb seemed really dim. I measured the voltage after the half wave rectifier and I got 17.8 volts DC. After this I noticed the bulb was brighter. I then measured the voltage after the half wave rectifier again and it was around 22 volts. The voltage to the circuit input was around 17 volts. I checked the MPS8099 that I put in and it was open from base to emitter.

Sent from my SM-G950U using Tapatalk

[GuitarFreak302]

Here are some more voltages. The order from left to right is ECB.

Q1: 1.081, 2.311, 1.507
Q2: 1.677, 5.27, 2.307
Q3: 4.62, 11.81, 5.27
Q4: 4.78, 16.87, 5.24
Q5: 4.13, 12.69, 4.79
Q6: 4.76, 16.91, 5.22
Q7: 4.08, 12.89, 4.75
Q8: 4.85, 16.92, 5.31,
Q9: 4.20, 12.79, 4.85
Q10: 5.91, 16.88, 6.43
Q11: 9.41, 21.35, 8.73
Q12: (Open B to E) 60mV, 21.18, 9.70
Q13: 3.37, 17.08, 3.93

Power supply: 21.35



Sent from my SM-G950U using Tapatalk

[Kevin Mitchell]

Your measurements don't look quite right.

The voltage on the cathode of the diode at the power supply should be the same voltage on the collector of Q11, Q12 and one lead of the bulb.

Did you replace the bulb or was this done prior to you obtaining the unit? Just curious.

[GuitarFreak302]

Your measurements don't look quite right.

The voltage on the cathode of the diode at the power supply should be the same voltage on the collector of Q11, Q12 and one lead of the bulb.

Did you replace the bulb or was this done prior to you obtaining the unit? Just curious.

It was replaced prior but I do have the original bulb. I was planning on putting the original bulb back in and then going from there. The owner replaced the bulb driver transistor with a 2N3904 but he did not use the correct pinout. He just put it in the way the original was oriented. I replaced this with a new 2N3904 but with the correct pinout. I noticed the orientation and fixed it before the other issues but I wasn't sure if it might have played a role in how it is currently acting.

Sent from my SM-G950U using Tapatalk

[GuitarFreak302]

Last night I put the original bulb in, replaced the LFO transistor again and then I removed the cap that goes from that transistor to the intensity pot. With this removed, the power supply stayed at the correct voltage and the LFO transistor that I was having issues with was sustaining. I figured that the bulb driver was causing the LFO to be dragged down. I then put the capacitor that I removed back in and I put a MPSA13 Darlington transistor in for the bulb driver. Now the bulb driver is not getting as hot and the LFO is working so I now have oscillation. There is an issue with the footpedal though. It only causes the bulb to oscillate when the footpedal is moved like a wah pedal. Also when I wiggle the cable, it causes the bulb to get brighter. I am thinking that the pot may be dirty or bad and the cable may have a bad connection.

Sent from my SM-G950U using Tapatalk

[RedHouse]

Here are some pictures I took last night.

Who is your friend?, I worked on that unit about 4 years ago. 

[GuitarFreak302]

Here are some pictures I took last night.

Who is your friend?, I worked on that unit about 4 years ago. 

That must have been one of the previous owners. I messaged him on reverb because he was selling it as non-functioning and I offered to fix it. It is working perfectly now.

Sent from my SM-G950U using Tapatalk

[pinkjimiphoton]

sometimes, radically mismatched gain in a balanced circuit can mess it up.

828's are not the same pinout or gain as the average 3904. i've seen opposite pinouts on 828's before, so i check everything with a meter to be sure of pinout.

glad ya got it working. if ya need a couple 828's just for authenticity sake, pm me your snail addie.