Rebuilding Altec 1588B and C Microphone Input Transformers

The Altec 1589B preamp is rebuilt, and I now need to rebuild the input transformers.  There are 5 electrolytic capacitors that need to be changed out.

Altec made a number of input transformers that could be used with their preamps or mixers.  There are microphone transformers (high gain), phono transformers (higher gain), and line transformers (more moderate gain).  There were different variations within those categories - generally circuit changes as the transformer type evolved over time.

The transformers all were mounted on octal (8-pin) tube bases so they could be interchanged easily - just unplug a transformer and put another in its place.

On the right are two 1588C microphone transformers I'm going to recap.  There were 1588A, 1588B, and 1588C models made.  I also have 5 of the 1588B transformers.

I wanted to show the back panel of the 1589B.  On the right, you can see the inputs.  There are two female XLR (aka Cannon) inputs, and then the respective octal socket beside it.  I have one of the microphone transformers plugged into the number one input to show how you'd use on in practice.  Simple.

On the left of the panel, you can see the output terminal strips - here you can wire a 150 or 600 ohm output to the power amp, mixer, or whatever you're running the 1589B into.

The 'can' of the transformer is easy to open.  There are 4 tabs holding it on to the chassis - just pry them back and the innards slide out.

Altec put foam over the electronics - the foam on all of mine has deteriorated over the years.  I think it was just to keep the boards in place - I'm not bothering to replace it.

Here's the 'gut shot.'

There are two sections - the circuit board on the top, and the transformer itself underneath.  The grey fiber board goes between the pcb and the transformer to insulate them from each other.  You have to have an insulator there - there are 4 connections on the back of the transformer that would short against the bottom of the pcb otherwise.

Here's the difference between 1588B and 1588C transformers.

The 1588C is on the top in the picture.  It has the pcb and a bigger transformer mounted below it as we saw above.

The 1588B has one pcb, and the transformer is right on the board - it's the small can.  The B and C circuits are entirely different.  (My understanding is that the 1588A and B are identical).

I didn't draw schematics for them, but there are 5 electrolytics on the C, and 3 on the B.  My guess is that the C will sound better, given its larger transformer.  But I can't speak to that just yet.

Closeup of the 1588B electronics.  This will be a lot easier to recap for sure, compared to the C version.

Top view of the 1588C.

You can see the 5 electrolytics - the small silver cylinders.  I'm not sure, but I suspect they're tantalum.  They're unusually small for an older piece of gear - the modern caps I'm replacing them with are physically larger.  Which will make recapping this thing interesting.

Another angle of the 1588C pcb.

Note the large rectangular-shaped component.  It was bent over to fit in the can, and I bent it back up to access the board.

That's a Couplate, or P.E.C., made by Centralab.  "P.E.C." stands for "Printed Electronic Circuit." It's basically an early version of an integrated circuit (IC).  It's a circuit unto itself - small resistors and capacitors sealed into an enclosure (in the shot above you can see the bumps where components are).

I've seen them in lower-level Hallicrafters radios (S-120, TW-1000).  They tend to be reliable over time due to the way they were packaged.  As an aside, here's how to translate the date code on this one - 1347735M.  "134" is Centralab's EIA code, and the "77" is the year, and "35" is the week of the year.  The "M" is most likely a shift code - e.g. day or night shift.

I found two interesting links about P.E.C./Couplates: this page has some description about the devices, and here are some Centralab ads for them, and some schematics.

Back to the recapping.  I used Elna Silmic II capacitors, and one Nichicon FG to replace the old ones.  You can see the new caps in this shot on the right.  A rare instance where the modern caps are physically larger than the old ones - the old and the new are both 25v rated caps.

It took some head-scratching to get them all to fit.  The location was important since it all has to go back into the can!

I was more experienced for the second transformer, so it went quicker than the first, and I learned something about better positioning the caps.  Now you know why I said the 1588Bs will be easier - much more straightforward to work on.  (I haven't done those yet).

I mounted one cap sideways between two transistors (green arrow) and I stacked two caps on top of each other (red arrow).  This let the couplate bend over a tad more and made it easier to get the assembly into the can.

You can also see that I used red (positive) and black (negative) shrink tubing on the leads - to insulate them but also to identify them as I put them in.

Reassembly is the reverse of disassembly.  Make sure you put the fiber insulator back in between the pcb and the transformer.

Then pry the tabs back down to hold the can in place.

I put the 1589B back together and plugged the 1588Cs in it.  Ran the output of a radio into it (see the goofy clip leads) and connected the output into my little Vox workbench amp. It works!

I'm leaving the preamp running that way for a couple days to make sure nothing blows up (after 30 hours so far it's fine).

There is a claim/myth on the interwebs that the Silmic II caps "sound best after a 100 hour break in period."  I am not sure how true that is; after all, I don't understand how the materials inside a capacitor can change.  And Elna themselves make no mention of this in their literature.  And, think about it "sounds better...," isn't that totally subjective?  Where are some scope images?  Hmmm.

At any rate, I'll leave it connected and running for a few days, it certainly won't hurt anything.

Now to look for some other trouble to get in to.