Dunlop Univibe UV-1 - No leds No power indication

[saransk]

I just picked up a "dead" Univibe UV-1 with pedal for less than a 10 box of strings - couldn't resist.
Pedal is the "last" revision with the single board - no chip on the switch board, etc.
Board has a 2-12-03 date but I think this is the "1997" version as it says REV A.

No power or LED's - nothing jumps out as a problem - but beyond confirming that power is getting to the regulator, not sure where to go next.
There is 18 vdc on pin 1 of the UA7815 regulator and 15 vdc on pin 3. (Looking at the schematic)  I can't determine how power is turned on - trial and error seem to indicate that a plug had to be in the output jack.) Can't find what controls U4's switching.
One thing I see on my board revision are more diodes in the power section - see attachment, one in parallel with the 100nf dc input cap, and the second is parallel to the 100 ohm dropping resistor for A+
Initial check showed a bad switching contact in the Expression Jack (Middle Contact - Connects U10 (8/9) to On/Off switch (??) but physically closing that contact didn't help.)

Is there a diagram of how the switching works, i.e. this is low when the On/Off is On which turns this Chip on, which turns on he Red LED?
I suspect that U10 was once changed and may still be the issue.
but would like to be able to work through the switching to see if the logic works further down - i.e., is there a way to see if the Chorus/Vibrato logic works or to see if the unit works and it is just the logic of the Effect/Bypass switch that is not.
I'm sure it is a bad solder joint, or U10 didn't get replace well, but there is some information missing from the schematic and this item seems to have dropped off the radar so I'm hoping someon here can save me some frustration ah show me the way
Thanks

[PRR]

> I'm sure it is a bad solder joint, or U10 didn't get replace well

So fix U10. That's gotta be faster and easier on the eyes than squinting the fuzzy schematic.

The logic is simple to a point. U10 pins 8 and 9 are high, unless Exp jack or switch pull them low. U10 is a NOR, so both inputs tied it is an inverter, so if input is high the output is low. U10 pins 5,6 may be a power-up delay, and that cap could be short. U10 1,2 is a basic NOR, truth-table easy to find.

After that.... I think someone was reading the CMOS Cookbook with one hand under the table and got over-excited. There was less logic in my Crown tape-deck than this pedal.

[saransk]

You're right about over think this
Since nothing is quite working one question - is the Effect Bypass (On/Off) closed pulling U10 down when the effect is On?

[Slowpoke101]

Just a couple of things that may help you.

U4 is controlled by the CTL lines.

CTL1 = U4(1)     Controlled by U6 pin 13
CTL2 = U4(2)     Controlled by U6 pin 12
CTL3 = U4(3)     Controlled by U10 pin 12
CTL4 = U4(4)     Controlled by U10 pin 11

Before you go too much further with with tests, etc - In the picture that you posted, I suggest that you find out why a resistor next to one of the diodes that you highlighted is so badly burnt. Finding out why it is so would be highly recommended.

[saransk]

It was #1 on the hit parade  Took a while to find the layout showing A+ and B+ lines

I wonder if anyone has bypassed the over-designed CMOS switching and done a "hardwire" setup
It looks like you could jumper the right U4 switches and rewire the foot switches to match a true bypass
(Use a 3PDT switch to turn the Red LED On/Off) and a DPDT with a dual LED for the Chorus/Vibrato switching.

Figuring out which U4 switches should be closed is the issue - the output circuit is rather complicated vs the original

Mike

[saransk]

Well one thing leads further down the rabbit hole
So I replaced the 100 ohm dropping resistor that "creates" the A+ voltage that looked fried
Just because I used a 1/2 watt instead of the 1/4 watt it replaced.
"cleaned up" the solder connections for U10
did a close inspection of the board.

Fired it up and from how quickly the 100 ohm resistor heated up it is evident that something is drawing way too much current down the A+ line.  Unfortunately nothing visually jumps out so there appears two lines of diagnostic work
1 - disconnect one by one the 470K resistors that tie one leg of the logic gates
2 - Replace the logic IC's

Neither is exactly hard, just tedious. 
Any ideas on how to shortcut this would be welcome

[Slowpoke101]

You appear to have a short circuit (or close to one) on the rail feeding the logic circuits.
If the resistor is not smoking quickly do the finger walk test on the logic chips. At least one will be warm if not actually hot.
The hot one(s) will be the faulty one(s). Replace them. This technique is very rough and ready and not recommended but it does work.
The other technique is to isolate each chip by cutting the power rail to each one and see if the fault clears. You will have to repair the cut tracks after finding the faulty chip(s).
If you have a good low ohms meter (0.1mOhm) you can find which chip reads as shorted.
Finding a short on a multi-distributed power rail can be tedious. There is really no easy way.

A clearer schematic (PDF) can be downloaded from here;

http://milas.spb.ru/~kmg/files/schematics/Dunlop/Dunlop%20-%20Univibe%20UV-1%20-%20Schematic%20(2-2).pdf

[PRR]

> disconnect one by one the 470K resistors

A 470K will NOT fry a 100r. (Do math.)

You have a much harder short. Solder blob. Capacitor. Chip.

If I am reading Slow's plan right, if not stray metal, it has to be the two caps right ON the A+ point. Everything else has 5K or 500K in series, so can't pull enough current to distress a 100r resistor.

[saransk]

Pulling the 470K resistors removes the A+ from the input of the NAND gates
Just looking at the places A+ shows up.
I did the "power up - feel for a hot chip" test and only the 100 ohm resistor got hot - even the 15 VDC regulator wasn't warm.

Looking at the schematic there is one other place that could be a problem
If the "Vintage" switch has gone bad, that is one place that the A+ voltage could be shorted to ground.
It might be worth the desoldering hassle to isolate that potential short.
That A+ connection got lost in looking at the logic switching.
(Remind me again why this is considered a better setup that plain analog switching)

Thanks - Mike