DR #49 - Unique Audio Equipment - Phase Linear 1000 Service - Part 1

Duality Repair
132 followers
1 Video View·Mar 12, 2025
In part 1, I document the initial service of this Phase Linear 1000, which includes servicing and testing the main PCB and the peak unlimiter PCB.
Hey everyone, this is Nick. Welcome to Duality Repair, and this is a PhaseLinear 1000.
It's an autocorrelator noise reduction system and dynamic range recovery system. Quite the mouthful.
But this seems like a pretty interesting unit.
The owner tells me that he attempted to get this repaired about 5 or 10 years ago. He's not quite sure on the timeline.
He's also not quite sure what the actual problem was, but for whatever reason,
the repair shop that he sent it to was unable to get this repaired for him.
So, the plan for this is to open it up, go through each board, take a look at the circuitry,
figure out how it should work, and then get it working again.
Here's the first look inside, and let's make some observations before we get started.
Right on the top we have this piece of foam, and I believe its function is to just support these boards.
So you can see they're not supported at all in the back, in the front, or obviously from the top.
And so a lot of these are just kind of flapping around, and I do not like that, especially this one.
And so if we can get this unit functional again, I may want to do something to try and stabilize these if I can.
I notice there are a ton of electrolytic capacitors in here. Take a look at this left board, for example.
Look at all of the axial-leaded electrolytics down there.
So if we recap this entire thing, as I suspect we might, it's going to take a lot of effort.
I notice this LED on the top here, it's not mounted properly.
I suspect that that just came loose over time, so we'll get that mounted properly.
As far as the layout, we obviously have six boards on the top, mounted to the board on the bottom.
I think to start, I'm going to pull all six of those boards, and we'll take a look at that bottom board.
All right, we're looking at the main board, and PhaseLinear calls this the interconnect board.
There's a lot going on here.
On the far left, we have all of the input and output jacks.
On the right, we have three potentiometers and four switches.
And then electrically, there are basically two different circuits on this board.
Up here is the power supply section, so we can see the transformer, the several bulk electrolytic capacitors for smoothing.
There's four diodes hiding back here, acting as the bridge rectifier.
This heatsink-mounted transistor is acting as the series-pass transistor.
And then the reference for the entire power supply is based on this 33-volt steamer.
We'll take a look at this a little bit more in detail later by looking at the schematic.
The other circuit on this board is right here, all of these components surrounding this unit, which is an op-amp.
And this has to deal with buffering and amplifying the input and output signals.
So what kind of work do we need to do?
Well, we need to replace all of the electrolytics, and that includes the axial-leaded ones.
We will clean all three potentiometers and all four switches as well.
And then we'll see if we can do some testing to verify that this is functional.
Here's the serial tag and safety certification tag, tested in 1984.
I'm starting with capacitor removal and replacement.
And looking at this board, I'm going to have to be really careful.
These traces are extremely dainty.
There are no markings anywhere on this board that let you know what the polarity of a capacitor is.
So before I'm pulling any capacitor, I'm labeling the negative lead.
I'm all done recapping the entire interconnect board.
I think it turned out really nice.
Next, I'd like to test the functionality of the power supply section.
To see what we should be looking for, let's take a look at the schematic.
The power supply has three rails we need to verify, B plus 1, B plus 2, and B plus 3.
The entire supply is regulated by this series pass transistor.
And the reference at its base is generated by this 33 volt Zener.
So B plus 1 should be somewhere between 32 and 33 volts.
B plus 2 is sourced from B plus 1, but it's first split across this voltage divider.
100K ohm resistor, 100K ohm resistor, it's right in the middle, so it should be split in half.
So it should be somewhere between 16 and 17 volts DC.
And B plus 3 is also sourced from B plus 1, but it's first dropped across this 680 ohm resistor.
So this is going to be something less than 32 to 33 volts DC.
So let's see what we get.
I used the schematic and component overlay diagram to identify and label some convenient test points.
So we have B plus 1 here, B plus 2 here, and B plus 3 here.
The unit is on, and the LED in the front is on, which is great, but we do have a problem.
So let's just start with B plus 1.
This should, again, be 32 to 33 volts DC.
Take a look at what we have, 7.
That is no good.
My initial thought is the 33 volt Zener right here, but I'm not positive.
To be able to test that at all, we have to pull it out, so let's do that.
So how do we test a 33 volt Zener?
Well, you can't just use your multimeter.
It's not capable of supplying that type of voltage.
So what I have here is the Zener installed in a breadboard.
I have my power supply supplying 40 volts or so DC to the positive rail.
I have this 910 ohm resistor here for current limiting,
and then I have our Zener here reverse biased as it should be to ground.
So let's check the input voltage.
40.2 volts DC.
Let's check the cathode of our Zener.
And there it is.
33 volts on the dot.
Now, this isn't a completely robust test.
This doesn't rule it out completely because it's such a low current application right now,
something like 10 milliamps or so.
And it may drop out at a higher current draw,
but there's certainly nothing that stands out about it, so I think we should move on.
Oh, boy. I did it again.
That's why it's so important to pay attention.
I found the problem.
These two electrolytics here, C16 and C17,
they are on either side of the Zener diode acting as filter capacitors.
They replace these, and maybe you can see where I made the mistake.
The black lines on these represent the positive leads.
A typical electrolytic, the black or white line represents the negative lead,
and that's what I was assuming.
I did not pay close enough attention.
And so when I marked the board, I marked the wrong side as the negative lead.
I got pretty lucky that those two electrolytics didn't pop.
I've had them pop before when I've installed them backwards,
but I'm sure I diminished the life expectancy,
so I'm going to replace them with brand new ones.
I have plenty of spares.
I'll put them in the correct orientation, and I bet this will work.
Let's give this another shot.
So I have new capacitors installed back here in the proper orientation.
I reinstalled the Zener, making sure that's also in the proper orientation.
Let's see what we get.
B plus 1 again should be somewhere between 32 and 33 volts.
Great. 32.7, that seems good.
Let's check B plus 2.
Again, that should be B plus 1 cut in half.
Here's B plus 2.
16.35, that looks correct.
And B plus 3.
That should be B plus 1 dropped across a 680-ohm resistor,
so something slightly less than B plus 1.
28.4.
Seems reasonable to me.
The power supply seems good. Let's move on.
Next, I want to clean all three potentiometers in the front,
and I want to see if I can clean these switches on the bottom as well.
I think the best way to get access to these is to pull the front faceplate,
so I'll do that now.
All three potentiometers have been cleaned.
I cleaned all four switches as well.
I also did some cosmetic cleaning, so I cleaned the front faceplate to the best of my ability.
I cleaned all three knobs, and I cleaned the actual buttons themselves.
I think it looks pretty good.
Before we move on to the cards,
I want to see if we can do a little bit more testing on this main board here.
So for that, let's go to the schematic.
Here's the full schematic for the interconnect board.
We've already taken a look at the power supply section down here.
Now I'd like to test the op amp and all of the components associated with the op amp.
These are all of our cards in the center here,
and we can get away with testing without any of the cards installed
because if we have all three switches out,
tape source, peak unlimiter, and autocorrelator,
if those are in their outward position, we bypass all of this.
You can see we have symmetrical circuits here.
The top is for the right channel. The bottom is for the left.
We'll take a look at the right channel.
We have our signal coming into the op amp here.
This is just a buffer circuit,
and so it's going to output exactly what it sees at its input.
The signal is going to be reduced to about one-fifth its level
by this voltage divider here.
Then since everything is bypassed,
it's going to come back into this op amp here,
this side of the op amp,
and this is an inverting amplifier with a gain of 5.
So we have the ratio of this resistor to this resistor, which is 5.
And so because our signal was reduced to about one-fifth its level here,
and then it's been amplified by about 5 here,
at our output, we should see the exact same signal that we fed in,
but inverted.
So let's see if that's what we get.
Let's take a look at the scope.
This is a 1 kHz, 100 mV sine wave at the left channel's input.
That's the top trace.
The bottom trace is the left channel's output,
and you can see that's just an inverted version of the input.
That's exactly what we'd expect.
So the left channel looks fine. Let's take a look at the right.
The right channel looks good, too.
So I think we've tested everything we possibly can on this interconnect board.
Let's move on and start diving into each card.
I'm going to start with the peak-unlimiter board, which is right here.
The reason I want to start with this is because this is the only board that's independent.
All these other boards, you can see correlator board, correlator board,
bandpass 1, bandpass 2, log amp,
these are all dependent on each other.
And so we can't test any of these until they're all done and reinstalled.
The peak-unlimiter board, however, once that one's done and installed,
we can test that by itself.
So what is the function of this peak-unlimit circuit?
Well, it's essentially a dynamic range expander.
It seeks to reverse a process called compression,
which is used in audio mixing and production.
Now, compression can be a good thing,
but there are some cases where you'd want to reverse that
so you can regain the authentic and original sound of a piece of audio.
Now, I'm no expert on compression or expansion,
but it seems like there are three things we need to do.
Number one, we need to set a threshold value.
On this unit, the threshold value is set using the peak-unlimit potentiometer on the front.
Second thing we need to do is to detect peaks of our audio signal
so that we can compare them to the threshold value.
And then number three, if the peak is at or above the threshold value,
we need to increase the gain.
The circuit basically has three sections.
This small section on the bottom is the power supply.
This section in the middle is our peak detection section,
and this is used to create control voltages, A and A prime, B and B prime, C and C prime,
which will be used up in this third section.
If these control voltages are at or above the threshold value that we set,
they'll be used to modify the gain of our audio signals up here.
So you can see A and A prime in the feedback network,
B and B prime, and C and C prime, again, in the feedback network.
So that's how we modify the gain of the signal if a peak is at or above a certain threshold.
So what kind of work do we need to do on here?
Obviously, tons of components, tons of electrolytics.
I am going to pull and replace these five because they're part of the power supply section,
and I do have them in stock.
I think all of the rest of the components, including the axial-leaded electrolytics,
I'm just going to spot check and only replace as needed.
There's more capacitors with these positive marked leads.
Not going to fool me again, Shigma.
I'm all done replacing those five electrolytics.
Before I move on to spot checking the other components,
when I was soldering these on, I did notice a suspicious area on the back of the board.
Let's take a look at that.
So down here in the corner, you can see a large area of flux.
This is nowhere near where I was soldering the new electrolytics on,
and so it looks like someone's done some work here in the past.
This is a capacitor.
See this film capacitor, and you can see how loose it is.
Take a look at that bar trace right there as I wiggle it.
Very loose.
And then hopefully you can also see this trace is missing.
It's a resistor pad down here.
It should be connected to that capacitor, but it's completely missing.
I told you these traces were flimsy, so we'll have to address this before we move on.
Electrolytic spot check number one.
This should be 2.2 microfarads.
It's reading 2.6. That's fine. Let's check ESR.
It should be less than 8 ohms. We're reading 7.1.
This one's fine.
Spot check number two. Another 2.2 microfarad at 50 volts.
Reading 3.1. Looks okay.
The ESR is not okay, though. 16 ohms.
That's going in the wrong direction.
I'll replace this one.
Here's another 2.2 microfarad at 50 volts with the same manufacturer's last one.
Reading 3.1 for capacitance. That's fine.
But how does the ESR look?
The ESR for this one is too high as well. 14 ohms.
It might just be that all of these black ones, the IEC branded ones, have failed or are starting to fail.
I'll probably pull one or two more. We'll check those.
If they all look bad, I'll just replace all of these.
Sure enough, I pulled a third 2.2 microfarad at 50 volt IEC capacitor.
And the capacitance was reading fine, but the ESR was much too high, just like the other two.
For reference, let's take a look at the ESR of a brand new 2.2 microfarad at 50 volt axial leaded capacitor.
Reading 6 ohms.
So I'll use these as replacements.
I finished replacing all of the black IEC branded electrolytics.
I also pulled and inspected a few more of the original blue electrolytics.
These are actually Nichicon branded, good high quality brand.
So not surprisingly, they're still reading fine for capacitance and ESR.
So I'm going to leave those in.
I'm going to start spot checking a few other components now, some diodes and resistors.
If those all check out, I'm ready to reinstall.
The resistors and diodes that I measured checked out just fine.
So I reinstalled the board and we're ready to test the circuit now.
Per the service manual, I have a 250 millivolt at 2 kilohertz sine wave at each input.
And the scope is monitoring the output from each channel.
Now this isn't a perfect measurement.
It's not a calibrated scope, but it should be good enough for what we're doing here.
So you can see the top channel is measuring around 240 millivolts.
The bottom, pretty similar, right around 240, 245 millivolts.
I have the peak and limit knob rotated to its counterclockwise position.
So when I engage the circuit, we should see a pretty significant attenuation or reduction in each signal.
So let's see if that's what we get.
So you can see them both drop pretty significantly.
Top channel is now around 130 to 140 millivolts, and the bottom, similar, around 140 millivolts.
So that part of the circuit is working.
I'll disengage it again.
We'll rotate the knob all the way to its clockwise position.
And it may be hard to see, but that LED is actually on now, which means that the circuit is active, which is good.
So now, when I engage the circuit by pushing in the switch, we should see a gain or an increase in both signals.
And we do.
So the top channel, right around 290 millivolts.
The bottom, right around 300 millivolts.
Fantastic.
So I would say the peak and limit circuit is definitely working.
I'm pretty happy with the progress I've made so far on this PhaseLinear 1000 noise reduction system.
I think I'll save the rest of this fun for Part 2, so stay tuned for that.
Thanks for watching.
