UVICS - Univibe In a Crybaby Shell

[R.G.]

I have everything now except the 78L15  (they had a 7812 I picked up)

It will probably work fine with the 7812, although the DC voltages will be proportionately lower, and overload will happen sooner. Not a biggie, really.

If you wanted to make 15V out of the 7812, you could stick a stack of diodes between its ground pin and real ground adding up to 3V, so about 4-5 diodes. PITA. I'd test with the 7812 and get the right thing later.

The dual 100k from RS wont work for the treadle but will for testing.

Yep.

For the discreet darlingtons ,I am a little confused ... Do I use a 2N3904 and 2N5088 together or two 5088's together ? Does it make a difference ?

Some, but tiny. A darlington has the two transistors cascaded for more current gain. The second transistor works at the full current, in this case 1-2ma maybe. It's base current is equal to its collector current divided by HFE, and so is the emitter current of the first transistor. The problem is that HFE falls with decreasing collector current. So if the second transistor is specified at HFE=400 at 1ma, its base current is 2.5uA, and so the first transistor doesn't have that same 400HFE, but something much smaller. That's the theory.

In practice, you can use either 3904 or 5088 everywhere. It seems to work, and I've done it both ways. For best results, I'd use a 5088 for Q1, and for Q4, Q6, Q8, and Q10, because of their higher gain at low currents. Also Q14 if you use Q14. Q14 is NOT stock, but it sets you up for the stereo option if you want to do that. If you don't use it, stick a wire shorting the base and emitter pads of Q14 to leave Q10 like the original.

Q2, 3, 5, 7, 9, can be either 5088 or 3904, or other NPN. The reality is that either way is enough gain to make the slightly-abberational darlington setup in the 'vibe work fine. I'd definitely use a 5088 for the higher gain for Q11 because the LFO darlington needs all the current gain it can get.

[Brymus]

OK neat trick with the diodes and the regulator,I might try it.

I have two 400ma 12VDC warts that are putting out 18.9VDC  and 19.9VDC
A 12VAC wart that puts out 14.5 VAC
and a 18VDC 1A HP laptop supply that puts out 18.2VDC
I think it will be a DC in from one of the 3 above.
The rest of my wall warts are above 24VAC or 9VDC and under.
(now all my wall warts are labeled with actual voltage readings  )

I had to sand down the mounting posts,first to get my front cap to clear the bottom plate.
Then again when I realized I would need insulators (plastic washers) as the traces touch the mounting posts.
No big deal 10 minutes with the dremel and done.

The mounting holes were slightly off,not enough that a little filing didnt fix,may have been my error or my model crybaby.

The shell I am using is pretty old ,it has an 1/8 inch plug for a DC adapter.
I am guessing it to be late 70s early 80s.

I may be able to make the dual 100k pot work off the tredal too with a little ingenuity.

OK now that the power supply and fit/mechanics have been verified time to heat up the iron...

[Beo]

I have to say that redhouse is being a duck. And by u I mean i. I understand open forum and all, but the passive aggressiveness is a big downer. Let's all ignore, and have fun with R.G.'s hard work and contribution.

[RedHouse]

I have to say that redhouse is being a duck. And by u I mean i. I understand open forum and all, but the passive aggressiveness is a big downer. Let's all ignore, and have fun with R.G.'s hard work and contribution.

My applogies, don't mean to be a downer, I'll need to work on my communications skills. 

I'm certain everyone will have fun with RG's hard work, whadda guy he is!.

[R.G.]

I have two 400ma 12VDC warts that are putting out 18.9VDC  and 19.9VDC
A 12VAC wart that puts out 14.5 VAC
and a 18VDC 1A HP laptop supply that puts out 18.2VDC
I think it will be a DC in from one of the 3 above.

I'd use one of the first two. They look just about perfect. If you use a DC adapter, leave out the diode bridge and connect two wires across its space on the PCB. This way you don't lose 1.4V or so to the bridge, and the filter caps still filter for you.

(now all my wall warts are labeled with actual voltage readings  )

Oh, sure, do the smart, useful thing that makes it easy for you in the future. 

Then again when I realized I would need insulators (plastic washers) as the traces touch the mounting posts.

Hmmm. I thought I'd made that big enough. Which places touch. Or do they all? I'll move the traces out to miss the posts.

The mounting holes were slightly off,not enough that a little filing didnt fix,may have been my error or my model crybaby.

Again, if you will post guesses about the direction and amount they were off, I can adjust for the variations in the shells. I just had the one.

I may be able to make the dual 100k pot work off the tredal too with a little ingenuity.

I also did a layout for a 1" by 2" board that holds the single pot circuit and the 9V upverter. It fits behind a single wah pot velcro'd to the side wall. It's designed to be cut in half so you can get one or the other circuit, or both. I'll get that posted.

OK now that the power supply and fit/mechanics have been verified time to heat up the iron...

Kewl.

[Brymus]

Got it about half populated last night,hopefully the rest tonight.

The traces that concern me are:
* from the bulb where it goes under R48,(that looks like a pretty simple fix)
* and the trace from C13 that goes under C15 ,(that looks like a tough one to sort out)
I think they may just clear with careful postioning,but I dont want worry about a jolt in transit causing a short there later.
So a plastic washer or maybe just a little hot glue or both is easy assurance.

I should have layed it in my shell then marked the holes for the mounting posts,or double checked them again.
I drilled those a little off to begin with anyway.
But like I said no big deal ,just a little file work and it lines up fine.
(The holes for the treadel tension lined up just fine)

One thing slowing me down is the schematic has pads marked W1 - W14
And the board is labeled A ,B,J,H,X,ect

I would like to add a rate indicator LED under the treadel for visual effect,so I am thinking about that too.
Well I should have the board completely populated tonight,and maybe some of the off board wiring.

[Brymus]

I forgot to ask
For the high impedence input.
Do I just change R3 to 1M or higher ?

And leave out R2 ?
I only need one input.
Thanks,Bryan

[Brymus]

R51 is mis labeled as R49 on the layout (there are 2 R49's)
R51 is the one belowe R30 and the A1 ,A pads.

[Brymus]

For R35,36  is 86K OK or 68K even ?
Thats the phase mix at TR3 IIRC

And for R4 and R7 would 1.5M or 1M be OK ?
If not I can make 1M2 with a 1M and 220K if thats better.

I was thinking If A1 and A are to get two inputs and most people wont want that.(I think)
A and its trace could be omited and that would free up a bit of room.

[R.G.]

The traces that concern me are:
* from the bulb where it goes under R48,(that looks like a pretty simple fix)
* and the trace from C13 that goes under C15 ,(that looks like a tough one to sort out)
I think they may just clear with careful postioning,but I dont want worry about a jolt in transit causing a short there later.
So a plastic washer or maybe just a little hot glue or both is easy assurance.

I'll check and see if I can give it more space.

One thing slowing me down is the schematic has pads marked W1 - W14
And the board is labeled A ,B,J,H,X,ect

I'll go label. Actually, they are labeled on the original PCB layout file, but this was lost in the conversion to PDF. OK, I can do manual retouch. 

For the high impedence input. Do I just change R3 to 1M or higher ?

Yes.

And leave out R2 ? I only need one input.

Yes.

For R35,36  is 86K OK or 68K even ? Thats the phase mix at TR3 IIRC

Either way works fine. That is not a fussy part of the circuit.

And for R4 and R7 would 1.5M or 1M be OK ? If not I can make 1M2 with a 1M and 220K if thats better.

It messes with the biasing a bit. It will probably still work, but in the spirit of testing, try a 1M and 220K if you have them. It will confuse issues less when you go to debug.
Just put one end of each into a hole in the PCB, solder, then twist the free ends together and solder in the middle. Leave yourself room to clip them off when you want to put in a single resistor.

I was thinking If A1 and A are to get two inputs and most people wont want that.(I think)
A and its trace could be omited and that would free up a bit of room.

Yeah. It's that old what to put in, what to leave out thing. I was actually thinking that I may retire the existing Neovibe layout, no new updates. It's over a decade old now. I was thinking that I could put enough of the circuitry into this layout that all new Neovibes could just go on this PCB layout.

Even then, the second input is an holdover from the dark ages. No one uses the second input these days. If they did, they could tie a resistor onto a jack and hack-wire it into the PCB. Just thinking out loud.

Any change in the circuit that can't be changed back to the original circuit, no matter what flaws it corrects, will be roundly d@mned by the "vintage is not the most important thing, it's the only thing" crowd. Happens every time I tinker. I wonder if there will be all-out effects wars between the "vintage is god" crowd and the BUMS sufferers some day. 

[R.G.]

Here's the translator cheat sheet from "W" numbers to lettered pads. This is in physical order, from the center of the board running along the edge toward the filter capacitors end.

W8 = A1- one input; ditch one of these
W7 = A - second input
W9 = B - ground for input jack
W14 = N - ground for output jack
W12 = L  - vibrato output to "chorus/vibrato" switch
W13 = M - "chorus" output to "chorus/vibrato" switch
I ( no "W" marking) "cancel" switch connection from original, probably not used
X - (no "W" marking) ground for "cancel" from original, probably not used
W6 = H - depth pot cold connection
W5 = G - depth pot wiper connection
W1 = F - depth pot "hot" connection
W2 = E - common to the speed pot
W4 = D - one of the speed pot leads
W3 = C - one of the speed pot leads
W10 = K - ac input, or (+) if you connect a DC supply and don't use the bridge
W11 = J - ac input, or (-) if you connect a DC supply and don't use the bridge

[Brymus]

Thanks RG that clears alot up !

I will make 1m2 resistors to keep the bias right.

The last thing I have discovered is I dont have a 500R trimmer 
I thought I had 1K and I was going to parallel a 1K resistor to make it 500R.
I do have 2K .
Would adding a 500R resistor parallel to a 2K trimmer still have a proper taper for adjusting the bulb ?

[R.G.]

Hmm. Could work. It's not so much taper as total current through the parallel combination. Try a 680 ohm fixed resistor. 680 parallel with 2K is close to 500.

[head_spaz]

R.G.
Shouldn't D7 (protection diode) be reversed in your schematic?
Looks backwards to me.

Thanks for all the work you've dedicated to this project over the decades,
and special thanks(!!!) for supporting this community with your wisdom and
expertise.

You've probably put more man hours into this project than the all of the
original Univibe developers - combined!

I nominate R.G.Keen for the DIYstompbox community's ...
"Heart and Soul Award"

[R.G.]

Shouldn't D7 (protection diode) be reversed in your schematic?
Looks backwards to me.

Not in this case. It's there to protect the 15V regulator from sudden shorts in the incoming bulk DC line. Without it, the regulator can die when capacitance on its output is left charged if the input voltage suddenly goes to zero. But you raise a point that needs addressed. If you do the thing I was suggesting, putting wires across the BR1 holes with a DC input, there then is *no* polarity protection on this thing. Bad juju indeed.

What ought to happen is a diode across the + side of the bridge printing, a wire on the other. This gives series diode protection for reverse voltage, and leaves the (-) side at ground.

Thanks! I wouldn't have caught that without your note.

Thanks for all the work you've dedicated to this project over the decades,

You're welcome, and thanks for the kind words.

[RedHouse]

...
I will make 1m2 resistors to keep the bias right.

The last thing I have discovered is I dont have a 500R trimmer  
I thought I had 1K and I was going to parallel a 1K resistor to make it 500R.
I do have 2K .

Would adding a 500R resistor parallel to a 2K trimmer still have a proper taper for adjusting the bulb ?

The 1M resistors will work just fine, if you are worried about it upsetting the natural balance of the vibe circuit you can also adjust the 6k8 (UVCIS part# R6) to 5k6 as Bob Sweet did with his MojoVibe, that mod actually works quite well when using 2N5088/89 in the preamp section. I've tried it and can confirm it works just fine.

Speaking of mod'ding the preamp, if you use the vibe for clean stuff (ala, Gilmour) you may experience a kind of "blatty" blocking kind of distortion when hitting barr chords (on a Strat) or most often when using humbucker equipped guitars into a stock vibe front end.
(and almost always whith 2N5210 as Q1)

If you find you have this distortion swap a higher value for the vibe's 22k input resistor (UVICS part# R1,R2). I have found 27k on up to 76k to be very effective when matching a vibe to a customers favorite axe.

I can confirm the 500 Ohm bulb bias trimmer TR1 doesn't need to be exactly 500 ohms as RG said previously, to expand on this notion I can say that in reality it all depends on what bulb you choose to use at the voltage you run at. For instance if you are running a 7815 v-reg and you're using the Radio Shack 272-1139A which is a 1.5v-25mA bulb you will need the upper portion of the range (300-500 Ohms) but on the other hand if you're using say a 14v-40mA bulb then the needed trimmer range drops way down to the 10-300 Ohm range, in fact when using higher voltage bulbs you can drop the accompanying emitter resistor (UVCIS part# R48) altogether, though it's always good to keep a minimum resistance to help you not blow the bulb when adjusting it too far.  



Unrelated to the current topic(s) there is something you should be running into very soon so I thought I'd post a heads-up and maybe you can plan around it, regarding the Dunlop cry baby rack-n-pinion system and it's geometry "issues". It wasn't really designed with pot rotation in mind and has been a "flaw" since day-1.

Why mention the seemingly obvious? well, just as when adjusting the pot in a wah circuit and paying particular attention to the end of sweep of the pot itself when securing the pot's locknut one naturally sets the pot in the bracket so to allow for that extra movement (1/8" to 1/4") for the bypass switch ...right?... sure, of course, but you'll quickly notice (in a vibe) it robs you of some noticable amount top end LFO speed.

The first solution I applied in my Vibe-Baby pedals is to use smaller upper limit resistors than the 4k7 (UVICS part#s R42, R43) normally used. I've found that 3k3, or even 2k2 work very well for this but having tried 1k5 resistors in those positions found that value was too low and the LFO can (and did) stall when you click into bypass (foot switch) which usually requires a power cycle to get it going again.  

While this solution actually offers faster than normal operation speed, before we get too exited recall we then subtract that bit of treadle travel the bypass switch rob's us of and this mod actually only pulls you back to where we would have been had we been building a normal vibe, but does solve the problem.  In my Vibe-Baby (retail pedals) I've moved away from using the 'ol stomp switches for bypass and use a momentary switch that toggles a relay. Since my Vibe's run on power supplies the few extra mA for relays don't matter, we still have absolute bypass (no FET or CMOS hiss) and minimizes the amount of treadle travel lost when using the push-push foot switch.

Hope any of this info is helpfull with your build.

PS: I'm confident Beo will be so kind to let me know if this (informational) post has turned into a "duck-downer" bumming out his/her forum enjoyment experience. I've been practicing my posting skills ...is it working yet?  

I nominate R.G.Keen for the DIYstompbox community's ...
"Heart and Soul Award"

I'll second that, as we raise our beer mug's ...hip-hip-hooray!  

[Brymus]

OK then thanks for the heads up.

Everything is populated except the V regulator.
Which way should the diodes face ,and 4 X 4148 or 1n4001 ?

I added a 100R across were the diode bridge would go for AC in.(+ side  - side = jumper)
Then added a 1N4001 for polarity protection and a .1 ceramic in parallel to the 2 X 1000uf 35V caps (axial)
I used for the power DC in filtering.
Plenty of room for different filter cap options and adding the .1 and 1N4001.
I am very impressed with the flow of this layout.
And really thank you RG ,this is a very neat project !!

I used all 1uf polar electros checking with the schematic for proper polarity.
Even subbing and such everything still fits fine.  
All the Qs are socketed so is the LDRs and bulb.
The LDRs I used go to 200-300R in bright light and 1M or higher in darkness.
The others I have are the same 200-330R in bright light but have lower ceilings in darkness like 100K,50K ,10K

I want to try all 5089s later ,right now it has all 5088s save for the MPSA13
I could even verify if just throwing NPNs at it will work after its tested.
I wanted to socket the 4 phase caps and not the Qs but will try that next time.

So all thats left is off board wiring and the V reg.
And the light shield,
will get some pics later tonight.

EDIt: I decided to skip the stereo mod for now.

[Brymus]

Light sheild = done
1" diameter PVC pipe ,a piece 5/8" long lined with chrome ducting tape.
If I hadnt socketed the LDRs and bulb a 1/2" long piece would have been plenty.

[R.G.]

Which way should the diodes face ,and 4 X 4148 or 1n4001 ?

Either diode should work. The band on the cathode end matches the band on the parts sketch. D1 and D2 are OK as long as there is one each in both directions, back to back. D7 has its band toward the filter-caps end of the board.

I added a 100R across were the diode bridge would go for AC in.(+ side  - side = jumper)
Then added a 1N4001 for polarity protection and a .1 ceramic in parallel to the 2 X 1000uf 35V caps (axial)
I used for the power DC in filtering. Plenty of room for different filter cap options and adding the .1 and 1N4001.
I am very impressed with the flow of this layout. And really thank you RG ,this is a very neat project !!

Thanks. That's how you always hope it will go - and it's nice if it does sometimes. 

The LDRs I used go to 200-300R in bright light and 1M or higher in darkness.
The others I have are the same 200-330R in bright light but have lower ceilings in darkness like 100K,50K ,10K

Yep, use the wide ones.

[Brymus]

Sorry RG
What I meant was, how do I face the diodes (what type, how many) for making the 12V reg a 15 V reg ?

Good news ,I was able to make the RS dual pot work off the treadel with a little grinding with my dremel.
Sanded the mounting casting were the pot mounts in the shell to make it thinner.
And filed a flat spot in the pot shaft to take the gear for the rack.

Now it just needs the Vreg and some off board wiring and its ready for testing 
Some pics will explain alot,when I get them uploaded.