Universal Audio 550-A Passive Filter

[dmp]

It's interesting that your unit is different than the pictures in the reverb listing. Yours has caps in parallel attached to the back side of the PCB. I wonder if that was a mod someone did later to move the frequency points a bit.
I'm curious what the exact structure of it is. Does it have input and output transformers so the internal circuit is unbalanced? And then a switched LC for each the high and low pass? That would be my guess, but hard to tell from the pictures.

 

[JMan]

I'm really hoping @MNJay will chime back in (and hopefully have some more pics and information).

As far as those differences between his pictures and the Reverb listing, I don't think they're mods. The individual caps in the unit listed on Reverb have incredibly specific values, and I would think they'd be tricky to purchase off the shelf nowadays. The one in MNJay's pictures has caps in parallel that are clearly trying to get as close as they can to those same values. I'd bet someone recapped MNJay's at some point (especially given the presumptive age of the unit and the slightly more, shall we say, "homegrown" look of the wiring job) and had to use standard modern values, hence the extra parallel caps.

I've drawn up a picture of everything I can see on that turret board, minus the wiring and inductors. Cap values below the board are the ones on MNJay's on the rearpanel-facing side, values directly above are MNJay's on the frontpanel-facing side (I believe I got this orientation right, but it doesn't matter too much). Numbers at the very top are the corresponding values that I could make out in the Reverb listing unit.

Seriously, if we can get more detailed photos and any meaningful observations, we should be able to figure out something that at least comes close. I suppose we are missing some information about the original, but why not give it a whack? It doesn't appear to be a terribly complicated circuit.

EDIT: Worth mentioning that I included the voltage ratings of any caps that I could see just for the sake of completeness, but given the nature of the circuit, I doubt that those are really minimum necessary ratings, but rather just what is typical of these types of caps. I'd think you could easily get away with caps rated at half of what these are.

 

[JMan]

In the UA catalog linked in the first post on the thread, on page 22, there is some useful information about the unit. First, this is a 600 ohm in/out device, so it's probably safe to assume that input and output transformers are in place, probably the usual 600:600 if I had to guess. Second, the filters are described as "Constant K 18dB/octave," with a corresponding graph on the page below. Note that they do not appear to be resonant filters. It says it is run unbalanced, which further supports the assumption that transformers are used on the input and output, since this is a standalone unit and not integrated into a console or something like that, so we're assuming it's balanced on the way in and on the way out, meaning we need the transformers to unbalance it inside the box.

Of additional-but-unrelated interest is the listing later in the catalog for the brand new offering from UA, the 1176 FET Limiter, which will be available this June September (love that handwritten edit!) for the low low price of $490. Gotta giggle a little at the promise of "severe limiting without added distortion."

 

[MNJay]

Any chance you could take some more photos? These are great, and with detail of the wiring, switches, and connectors, I'm fairly certain we could reverse engineer this piece (the inductor tap values should, I believe, just be a matter of math once everything else is clear). There's also one capacitor that can't be read - third image, far left cap. Can you get a value off of that?

Hey everyone, sorry for the delay! Attached are more photos. I'll be in the studio all day so if anyone wants more images lmk.

I attached the 620 ohm resister across the output and the unit performs quite differently. The filter is much more smooth! Without the resistor it sounded like it had a harsher peak, like a more resonant sound, which I'm guessing is "ringing" that people speak of when talking about this kind of stuff? Not sure on that. I definitely think with the resister on the output it sounds like the filter is behaving more "appropriately." Sounds really nice to me.

@JMan it was hard to get a clear focus on that one cap, it also seems to be rubbed out a bit which makes it hard to read. I think it reads .0068MFD

 

[PermO]

Nice filter unit !

 

[JMan]

Hey everyone, sorry for the delay! Attached are more photos. I'll be in the studio all day so if anyone wants more images lmk.

I attached the 620 ohm resister across the output and the unit performs quite differently. The filter is much more smooth! Without the resistor it sounded like it had a harsher peak, like a more resonant sound, which I'm guessing is "ringing" that people speak of when talking about this kind of stuff? Not sure on that. I definitely think with the resister on the output it sounds like the filter is behaving more "appropriately." Sounds really nice to me.

@JMan it was hard to get a clear focus on that one cap, it also seems to be rubbed out a bit which makes it hard to read. I think it reads .0068MFD

Thanks! Okay, I'll use the .0068MFD value for that cap for now. It should be interesting to see if I can get this to math out correctly. I'm by no means a whiz at this, but a little work should hopefully yield results. Will try to do it this weekend.

@MNJay , if you happen to have some time to get extra shots and extra angles on the wire routing, that could only help! No worries if you can't. Those switch connections are quite the little web of connections, more extensive than I would expect (surely we just need to be able to select one LC combination for each frequency point?), so it would be nice to try and suss out what's going on there if possible. Even with plenty of good photos, it may be tough to see.

If I'm not mistaken, it looks like the two larger toroids are the input and output transformers, and the two smaller ones are the inductors. Filter bypass switches must be two-position one deck (? number of poles needed), and frequency select switches are six position three deck (presumably two poles per deck? It appears that not all are used, though). This might all be fairly obvious, I'm just trying to compile information here.

 

[MNJay]

Thanks! Okay, I'll use the .0068MFD value for that cap for now. It should be interesting to see if I can get this to math out correctly. I'm by no means a whiz at this, but a little work should hopefully yield results. Will try to do it this weekend.

@MNJay , if you happen to have some time to get extra shots and extra angles on the wire routing, that could only help! No worries if you can't. Those switch connections are quite the little web of connections, more extensive than I would expect (surely we just need to be able to select one LC combination for each frequency point?), so it would be nice to try and suss out what's going on there if possible. Even with plenty of good photos, it may be tough to see.

If I'm not mistaken, it looks like the two larger toroids are the input and output transformers, and the two smaller ones are the inductors. Filter bypass switches must be two-position one deck (? number of poles needed), and frequency select switches are six position three deck (presumably two poles per deck? It appears that not all are used, though). This might all be fairly obvious, I'm just trying to compile information here.

Here are some more photos focusing on the wiring. The switches are very robust, they are 6 positions for each filter and two for In/Out. Extremely satisfying to use.

 

[MNJay]

one more of the wiring to the switches

 

[JMan]

These are great, thank you!

Upon closer inspection, I'm now feeling less confident about how I previously identified the inductors and transformers. Will give this a more proper look over the weekend!