Original Schematic Uni-Vibe (old one)

Discussion regarding early stompbox technology: 1960-1975 Please keep discussion focused and contribute what info you have...
User avatar
amp_surgeon
Information
Posts: 18
Joined: 05 Jul 2011, 18:59
my favorite amplifier: 1955 Gibson GA30
Been thanked: 32 times
Contact:

Post by amp_surgeon »

R.G. wrote:Getting a calibrated light source is *hard*. These are rarer than voltage reference sources. Back when I was doing this, I picked up some green LEDs, and decided that *this one, right here* was my reference source, whatever it did. This let me test LDRs on a relative basis, as they'd be different with different LEDs. I also picked a reference LDR, so I could back-test LEDs whenever I might need to pick and calibrate a new reference.
The Perkin Elmer app notes say that the standard light source is a tungsten filament lamp with a color temperature of 2850K with a known candlepower output at a specific voltage and current. While these can be obtained from NIST at considerable cost, they suggest a suitable substitute is a conventional frosted 100W incandescent bulb. They say a bulb of this type should produce about 90cp at 120VAC, with a color temperature between 2700K and 2800K. I'm putting a light meter sensor next to the LDR socket just to confirm the illumination. I don't plan to vary the voltage to the lamp, other than to switch it on and off for the rise/fall time tests, because running it at a lower voltage will change the light spectrum. I'm making the lamp moveable instead. PE recommends a distance of at least 5X the bulb diameter, so I'm making it adjustable out to 10X the diameter.

To test the calibration of the box I plan on producing curves from some Perkin Elmer LDR's and comparing my charts with the ones Perkin Elmer published. If my charts are pretty close then I'll go ahead and produce charts for the four LDR's in my Uni-Vibe and average them out. This should give me a starting point for tracking down a suitable substitute. I'm not expecting to find an LDR with an exact match over the entire curve, but that's not really necessary. It just needs to match fairly closely in the portion of the curve where the LDR's will be operating in the Uni-Vibe circuit, or I'll look at making adjustments to the light driver circuit to see if I can get an off-the-shelf LDR working within the right portion of it's operating curve. Mike Fuller and Bob Sweet purportedly went to the extreme of having custom LDR's fabricated. I'm hoping this isn't necessary. Frankly, I wouldn't know where to go to have that done. Perkin Elmer isn't making LDR's anymore, apparently.

User avatar
Mr. Bill
Information
Posts: 19
Joined: 11 Mar 2010, 21:16
Has thanked: 1 time
Been thanked: 12 times

Post by Mr. Bill »

Back in the late '80s I started building the big 3 Hendrix boxes, as there were a lot of guys that kept asking for them. The Univibes were obviously the most difficult ones due to the ldrs. What I found in general was that minimum resistance circuit value could be tweaked by removing or replacing the 4k7 resistors that are in series with the photocells.

I also found that the cell's response time and memory were probably more important than the min/max resistance values. Even if the cell had great minimum and maximum specs, if it couldn't change fast enough it sounded weak and blurry.

The cells that sounded best for me were Vactec cells removed from one of their Vactrol modules. We were an Acoustic Control dealer at the time, and they used these as channel switches in the G100 and G60 amps. They had 2 cells and a 9 volt lamp in a small thimble sized spun aluminum case. The bulbs worked great in the Univibe circuit as well.

User avatar
The Rotagilla
Diode Debunker
Information
Posts: 744
Joined: 20 Aug 2007, 18:24
Has thanked: 7 times
Been thanked: 61 times

Post by The Rotagilla »

I've got a question and I may be talking out my ass on this one but here it goes...

On the way into work today I was thinking about the lightshield that covers the cells. Do the reflective properties of the shield play a part in the sound? Could the size, shape or how polished the inside faces are have any affect on the overall sound?
The television will not be revolutionized.

User avatar
R.G.
Resistor Ronker
Information
Posts: 337
Joined: 22 Sep 2007, 02:24
Been thanked: 39 times

Post by R.G. »

amp_surgeon wrote:The Perkin Elmer app notes say that the standard light source is a tungsten filament lamp with a color temperature of 2850K with a known candlepower output at a specific voltage and current. While these can be obtained from NIST at considerable cost, they suggest a suitable substitute is a conventional frosted 100W incandescent bulb. They say a bulb of this type should produce about 90cp at 120VAC, with a color temperature between 2700K and 2800K.

I'm putting a light meter sensor next to the LDR socket just to confirm the illumination. I don't plan to vary the voltage to the lamp, other than to switch it on and off for the rise/fall time tests, because running it at a lower voltage will change the light spectrum. I'm making the lamp moveable instead. PE recommends a distance of at least 5X the bulb diameter, so I'm making it adjustable out to 10X the diameter.
Yep, they do recommend that. Got that same app note.

What they don't tell you is how to turn on and off a 100W incandescent bulb in 'way less than a millisecond so you can avoid polluting the measurement of the LDR rise/fall times with the rise/fall of the emittance and color temp of the bulb as it comes on and off. I went with LEDs, even in the face of the incandescent being more standard in some ways because I could start and stop the light much, much faster than the LDR could respond, so I could ignore issues of startup/tail off of the light source. I could never come up with better than some kind of d@mned fast mechanical shutter mechanism to to on/off faster than the LDR, so I went with the LED.
To test the calibration of the box I plan on producing curves from some Perkin Elmer LDR's and comparing my charts with the ones Perkin Elmer published. If my charts are pretty close then I'll go ahead and produce charts for the four LDR's in my Uni-Vibe and average them out. This should give me a starting point for tracking down a suitable substitute. I'm not expecting to find an LDR with an exact match over the entire curve, but that's not really necessary. It just needs to match fairly closely in the portion of the curve where the LDR's will be operating in the Uni-Vibe circuit, or I'll look at making adjustments to the light driver circuit to see if I can get an off-the-shelf LDR working within the right portion of it's operating curve.
Good luck to you in your search. I went through that for much the same reasons back when I was messing most actively with the 'vibe. I finally determined that an LDR you can actually get is worth ten of the perfect ones that may only exist in 10th century Spain or on one of Jupiter's moons. What that rapidly telescopes down to is to build it with LDRs you can actually get, and if you aren't happy with the sound, keep looking, meanwhile playing the ones you have. Love the ones you're with, but keep your eyes open, if you will.
Mike Fuller and Bob Sweet purportedly went to the extreme of having custom LDR's fabricated. I'm hoping this isn't necessary. Frankly, I wouldn't know where to go to have that done. Perkin Elmer isn't making LDR's anymore, apparently.
Fuller did. I have a quartet of his that he sent me. Didn't know about Sweet.

In reality, the LDR is a dying technology. The RoHS legislations that are being proposed and put in place all over the world will phase out all cadmium bearing anything, and the "CD" in CdS stands for Cadmium. It's ironic that the folks who tell us what we may do because it's good for us to do only what they tell us to are also phasing out the incandescent light bulb. Better buy a stock of those 100W incandescent reference bulbs. :shock:

User avatar
amp_surgeon
Information
Posts: 18
Joined: 05 Jul 2011, 18:59
my favorite amplifier: 1955 Gibson GA30
Been thanked: 32 times
Contact:

Post by amp_surgeon »

R.G. wrote:Yep, they do recommend that. Got that same app note.

What they don't tell you is how to turn on and off a 100W incandescent bulb in 'way less than a millisecond so you can avoid polluting the measurement of the LDR rise/fall times with the rise/fall of the emittance and color temp of the bulb as it comes on and off. I went with LEDs, even in the face of the incandescent being more standard in some ways because I could start and stop the light much, much faster than the LDR could respond, so I could ignore issues of startup/tail off of the light source. I could never come up with better than some kind of d@mned fast mechanical shutter mechanism to to on/off faster than the LDR, so I went with the LED.
Yeah, that one has me a little stumped, too. I've seen LCD shutters that are pretty fast - up to a couple of milliseconds, but they cost more than I'm willing to spend at the moment. I've got some thermal inertia charts from an old Sylvania incandescent light manual. I might just be able to just flip a switch on the bulb and drop the data from the TDS1000B into a spreadsheet containing the numbers from the Sylvania manual and see if it corrects my readings enough to duplicate the charts published by Perkin Elmer. If that doesn't work then I might consider getting the LCD shutter. I really want to avoid having to go with LED's, if at all possible. Yes, they do switch on nearly instantly, but they aren't the same spectrum as incandescent lights, which will mean my charts won't match PE's charts, or probably anyone elses.

The rise and fall times for the LDR are much faster at high light intensities, and this is where I'm likely to have the greatest amount of error in my readings. Fortunately, this isn't the range that the LDR's in the Uni-Vibe light tank operate at.

You wouldn't happen to know any test engineers who used to work at Perkin Elmer's optoelectronics division before they shuttered, would you? :hmmm:
R.G. wrote: Good luck to you in your search. I went through that for much the same reasons back when I was messing most actively with the 'vibe. I finally determined that an LDR you can actually get is worth ten of the perfect ones that may only exist in 10th century Spain or on one of Jupiter's moons. What that rapidly telescopes down to is to build it with LDRs you can actually get, and if you aren't happy with the sound, keep looking, meanwhile playing the ones you have. Love the ones you're with, but keep your eyes open, if you will.
Yeah, understood. I just feel better having that perfect match described exactly on a sheet of white paper. I trust specs a lot more than I trust my ears. Sign of age, I guess. :oops:
R.G. wrote: In reality, the LDR is a dying technology. The RoHS legislations that are being proposed and put in place all over the world will phase out all cadmium bearing anything, and the "CD" in CdS stands for Cadmium. It's ironic that the folks who tell us what we may do because it's good for us to do only what they tell us to are also phasing out the incandescent light bulb. Better buy a stock of those 100W incandescent reference bulbs. :shock:
Yes, I'm already considering alternatives. I'm planning on the light tank and driver being a plug-in module. I'll make something that approximates the same functionality using strictly solid state technology for use in the EU. I can make a kit version of the LDR light tank module for sale in the EU, since CdS cells are permitted in electronic kits and for hobby use. You just can't sell finished products that use them.

Personally, I think the RoHS regulations are a little extreme. The amount of cadmium in a CdS cell is really pretty low, but the RoHS regulations forbid anything over 100ppm in any homogenous material. Even if the CdS film were microscopically small it would still violate the RoHS regulations.

There are also other semiconductor materials that can be used to make LDR's. Cadmium just happens to be the least expensive for visible wavelengths. Selenium is probably one of the earliest materials used. Germanium can also be used. In fact, GeCu sensors are commonly used in infrared astronomy. I don't know if it would be possible to fabricate a GeCu cell that would work adequately with visible light. If I managed to come up with one I'm sure Marshall and Fender would be in line to buy them for optoisolators. :idea:

I'm not planning on needing this test box any longer than it takes to produce my charts, which is one of the reasons I don't want to spend too much money on it. Maybe I'll stock up on 100W "soft white" bulbs, and in another 20 years I'll make a little cash selling NOS light bulbs for vintage Easy Bake ovens. :lol:

User avatar
ramaclaproth
Information
Posts: 25
Joined: 22 Mar 2013, 01:51
Has thanked: 8 times
Been thanked: 20 times

Post by ramaclaproth »

There are differences between the 1968 UniVibe and the 1969 UniVibe that we all should know and try it out!

Look again at the layout from Turretboard ---> http://www.turretboard.org/wp-content/u ... _names.pdf

The differences are the caps on C13 and C16 are switched the other way around!

EARLY UniVibes (double rubber version) had C13 for 4,7nF, and C16 for 470pF.

The late UniVibes (single rubber version) had C13 for 470pF and C16 for 4,7nF.

EARLY UniVibes (double rubber version) had C6 (-) negative goes to LDR.

The late UniVibes (single rubber version) had C6 (+) positive goes to LDR.

EARLY UniVibes (double rubber version) had C17 (-) negative goes to R32 R33 R34.

The late UniVibes (single rubber version) had C17's polarity orientation the same way like the early UniVibes had it too, BUT most clone builders follow the "915" factory schematic, the RG schematic or the Turretboard layout which they got it WRONG for C17's polarity orientation.

Most EARLY UniVibes (double rubber version) had R37 for 47K but also some early ones from the same year had R37 for 100K.

Most late UniVibes (single rubber version) had R37 for 100K.

Here are the layout of BOTH the 1968 and 1969 UniVibes:

https://imageshack.us/a/img703/2669/uv1968.jpg

https://imageshack.us/a/img22/9903/uv1969.jpg

Note: The ones that use the 10nF and the 6nF for C13 and C16 (which were mentioned by Turretboard as the "Wattson" tracing) they were mistakes made by Shin Ei BUT were made very limited that way.

Keep in mind the 1968 UniVibes were NOT made by Shin Ei, they were made by a company named Honey. Shin Ei took over Honey in 1969 and years later Korg took over Shin Ei. Honey was the original company which first produced UniVibe and put UniVibe on sale in 1967. Unfortunately, Honey went bankrupt in 1969 and Shin Ei became in charge of the production of UniVibe and as a result, UniVibe after 1969 experienced several changes in specification such as the design of logo and the circuit and hence the change in the sound occurred.

Jimi NEVER used the single rubber version UniVibes, Jimi used the double rubber version UniVibes.

User avatar
The Rotagilla
Diode Debunker
Information
Posts: 744
Joined: 20 Aug 2007, 18:24
Has thanked: 7 times
Been thanked: 61 times

Post by The Rotagilla »

Unfortunately the Image Shack pics will not load.
The television will not be revolutionized.

User avatar
The Rotagilla
Diode Debunker
Information
Posts: 744
Joined: 20 Aug 2007, 18:24
Has thanked: 7 times
Been thanked: 61 times

Post by The Rotagilla »

Did anyone save the pics from ramaclaproth's post? If so, could you message me? Thanks!
The television will not be revolutionized.

Post Reply