OTA + single LDR based phaser?

[LeftyStrat]

I've built several univibe clones over the years, and finding and matching suitable LDR's is always a pain.

I started thinking about an OTA-based phaser that used lamp and single LDR to provide the characteristic wobble of the univibe.

Two questions, has anyone done this, and if not, is it worth pursuing, or is the mismatch of LDR's part of the mojo of the 'vibe?

[top028]

no experience yet, but electronics goldmine has a great selection of LDRs. I hope to build one soon

[Mark Hammer]

Unlike FETs in a phaser, the LDRs do not need to be matched for a phaser or Uni-Vibe.

The matching of FETs is normally essential to having all FETs continue to change resistance together throughout the entire sweep cycle. They may not all have exactly the same resistance, but are changing in unison, and none of them stops short before the end of the cycle at either the positive or negative end of the sweep.

Unless you are using halogen headlights as your light source, LDRs will nearly always have a little more "headroom" in their 'on' resistance, and unless you have a perfectly lighttight enclosure will hardly ever reach their maximum 'off' resistance. This means they will all be able to keep changing resistance in unison across the sweep cycle.

One of the many reasons why photocell-based phasers and vibes sound so damn good.

[LeftyStrat]

Thanks for the info. Maybe I'm just a victim of mis-information. I had read in some article (I believe from Roger Mayer or Mike Fuller), that the reason some original univibes sound good and others don't had to do with mismatched LDRs. Perhaps it was just marketing.

I also read that you're looking for 5k min resistance and 500k max. Is this another myth?

[Mark Hammer]

Well, given that Uni-Vibes are essentially 4-stage phase-shifters that create broad shallow dips instead of focussed notches, you still want the dips to be situated in the appropriate spot in the spectrum, and you need the resistance of the LDR, in tandem with the selected cap values, to result in the dips being in that same preferred location every time.

The 5k-500k range is pretty common. I imagine that if you just grabbed 4 LDRs from a bin at a surplus place, it might BE possible that one would give you 1k-50k, another 5k-500k, and another 70k-1.5M with the same illumination. Given that the cap values normally used for vibes are different for each stage, juggling different-value caps and different-value LDRs would give so many combinations and sequences that no two units could ever sound alike, even with trimpot tweaking. So there is a need for a certain degree of consistency.

Would the 4 LDRs that Mayer/Fuller/whomever all read the exact same resistance with the same illumination? Nah. Highly unlikely. Even when dealing with a sealed optoisolator (one LED and one LDR per package in a lighttight epoxy module), you can't guarantee that feeding the identical current to the LED in each yields the same exact resistance in the LDR. And the LDR arrangements in a great many vibe units have the LDRs mounted on the board and facing a single light source. The amount of light that results in one of them reading 53k might produce 68k in another, 46.9k in another, and 62k7 in the 4th. But they are all not THAT far off from each other, and all keeping changing resistance in a ballpark range in unison. And when it comes to phasers, that's what matters.

So let's say that what you read is not "wrong", but perhaps "matched" was too strong a word to use and misleads the reader into thinking about LDRs the same way that, say, transistors are closely "matched" in high-quality VCAs and VCOs.