Restoring an old Graphic Equalizer

[Spook of the lost]

Repost from Makestation, I'll post in both places as I update this Project Journal.

************************************************************************

Finally decided to make a new Project Journal since I came across what should be an interesting challenge. I recently got for free an ADC S-1 MKII, circa mid to late 70's, maybe early 80's based on initial appearance.

later found indicators of the former rather than the later was pretty beat up, but was intact otherwise. side note, this unit goes for anywhere between $40 and up to $70 working. 

It's not even a full featured one as it has only 5 rather than 10 sliders per channel but it'd be a fun restoration nonetheless. Here's the unit as I got it. 

  

Pretty beat up but the internals are hopeful and the potentiometers for the frequency adjustments slide smoothly, likely just need a good cleaning and a bit of silicone grease to lubricate them. 

part of my plan is to replace the wood part of the case by fabricating a new one, it's just a simple 3 sided box with a lip on one side to attach to, should be able to come up with something and either paint it or leave it with a nice wood grain finish. 

  

Also going to change a couple minor things to update it a bit and make it easier to service in the future. 

Here's the power supply which I'll likely need to replace due to it's age.

  

so a few things of note. several capacitors have started leaking though the traces are not on the capacitor side so that's good news, means there may not be any damaged traces. (edited, yes there's no obviously damaged traces, it's a single layer board, hand drawn too by the looks of it). 

At least 3 resistors are damaged, and all components are still available except for the transistors but I can get them pretty easily on ebay if needed. when I get the chance to go over everything I'm going to write up a component list and try to reverse engineer the circuit as I go along with my multimeter as well as note which components are faulty, that'll be the next update. I'm super glad I didn't plug it in to test it first given the state the components are in, but I can fix it for less than it's worth  oh! and it has all the slider knobs btw!

  

they just need a good soak in some warm soapy water and an air dry. 



The wires are fairly short so I'm going to need to note which ones go to which pins (which are numbered!!!!), De Solder them, Then Remove the board.

The great news is, it's a single layer board, the components are mounted opposite the traces so nothing was damaged by the leaky capacitors near as I can tell. 

Definitely 1970's. the traces are not electronically drawn, they're curved instead of straight angles which is a pretty solid indicator. 

The good news is, that makes reverse engineering it easier since being single layer and hand drawn means you can visually connect what pins on what components go where. the pins themselves are marked on the side they sit on so this should be pretty easy once I de solder the wires holding it in place. 

[Kyng]

Well, I guess 'Used' is one way of describing the condition... it appears to be in pretty bad shape . Still, I'm sure you'll be able to get it looking as good as new once you're done with it!

Good luck - although, I don't expect you'll need any .

[Spook of the lost]

So I finally had some time to sit down and start looking over this board and removing it from the wiring harness and front panel of the device.

and I learned something shocking.

looking at how the silkscreen was done on the component side of the board (circuit side and component side of a single sided board).
this was clearly intended to be assembled by a person and not a machine.

the wire sequence is easy to follow, the components are labeled by type and grouped by voltage/resistance/capacitance/etc, and the wires are color coded.

meaning that not only was this graphic equalizer's circuit drawn by a person rather than a computer, it was assembled by a person.

the last person to do anything to this device was a human not a machine, and to me that's huge!

the other takeaway is that the wires were wrapped around pins, not soldered to the board, which is something else I'd never encountered before, besides paper capacitors I mean.

Finally there's a patent listed on the back and I did some digging between my last post and now and well... it's expired, completely free to do anything with.

so long as I don't expect to make a profit from it I can totally just reverse engineer it and post a board derived from studying it so other hobbyists can build one themselves.

[Kyng]

Wow, that's quite an interesting discovery!

Do you actually intend to reverse-engineer it, though ?

[Spook of the lost]

@ Kyng sure, it's actually not that complicated a circuit, mostly filters and such to clean up the noise and help the frequency isolators on the front as well as amplifier chips to make sure the sound coming in is the same level as the sound coming out only with the modifications.

understanding it means I could in theory add more frequency isolation circuits or even take that knowledge to other projects like DIY audio switches and sound mixers and such a lot of which don't exist for analog circuits but rely on software based solutions these days.

to reverse engineer it I'm first making a chart of all the components and where they go before I de solder and test them.

then after I clean up the board given it's a single sided circuit board I'm thinking of scanning it in with a flatbed scanner then overlaying the silkscreen onto it mirrored so I can see where the components, jumpers, etc go relative to the circuit.

from there I can use a piece of circuit design software to see what a modern implementation of it would look like.

normally this wouldn't be the case but since it's not a layered PCB that makes this possible in the first place, it's a case of WYSIWYG (what you see is what you get or wizzy wig).

Edit: to decode stuff I've found that most components have writing or an indicator on them as to what they are, most capacitors have it written on the side for example, IC's typically have it on the top, etc.

for resistors though they use a color code.

I could spend forever trying to decode each one...
Luckily digikey has a nice little program to do that.

https://www.digikey.com/en/resources/con...color-code

just put in the color code on the resistor and it translates that into the correct value.

super handy to have when you're repairing/reverse engineering something.

edit 2: Couldn't figure out what the 8 diodes were without a magnifier... however they passed the diode test mode on my multimeter so I'm not going to worry about them right now, I know the power ones are IR D10's which are for AC/DC conversion.

the other 4 are too small so I need to get a magnifier or something to try and read them.

[Spook of the lost]

So I'm kinda stuck, I can fix it but there's one part of the circuit I don't quite understand.

This here is the filter circuit



there are 10 of them laid out mirrored along the two middle power leads with 5 chips straddling the power connections and each side of the chip handling a filter circuit.

each chip is an amplifier knows as a 4558, you can still get them and the diagram for this chip is as follows.


if you look the middle lines connect to pins 8 and 4 on the chip, considering this is the bottom of the board it makes sense if you invert the image.

so that all makes sense, where I'm struggling with is that these 10 circuits do not connect to the main circuit except by the wire going through the graphic equalizer, one per circuit, 10 wires in total.

how does the sound go in, get filtered, and come back out?

I'm very confused by this and am likely overthinking it.

here's how the components are laid out on each circuit



only thing I can figure is that the wire connecting to the two state potentiate/slider controls uses feedback some how which is a bit different from a more modern and standard setup as they often filter through then have the potentiometer's controlling the resistance for those frequencies instead and the circuit goes through the filter circuit and op amp rather than going in and out through the potentiometer controls.

on top of that the inputs and outputs are wired together on wires 1, 2, 3, and 4 on the board feeding into a 6th amplifier circuit which splits off the tape monitor and outputs to the output leads while the inputs go into that same circuit before somehow going through the filter circuits through a single wire?

just some weirdness.

[Kyng]

Yeah, that doesn't look easy .

Sadly, I can't help out with it - but, I hope you're able to either find assistance, or figure it out for yourself!

[Spook of the lost]

I did actually figure it out, I slept on it and came up with an answer

it's an inverse feedback generator circuit.

basically it's generating feedback on certain frequencies that counteract and oppose the one's intended and then strength of that signal is controlled by the slider potentiometers' on the front thus filtering  it out based on how much of that frequency is canceled out.

it seems weird but if you've ever looked at how mufflers, or helicopters reduce noise you'd get it, kinda fascinating really and very much analog in how it works.

[Kyng]

Not something I've ever come across before, but I guess it makes sense .

Glad you could figure it out!