PYE Compressor/Limiter Thread *boards shipping* BOM up!

[marcn]

Lots of people have had trouble calibrating the Meter, mine doesnt sit at 0 either but it is also the cheapest 1ma meter i could get.
Like I said R52 seems to have an effect on the Meter and the release time as well, initially my needle would sit pretty far down and not move much, playing with R52 made the unit come alive and the meter now sits just below 0. Just play with it while feeding it some material like a drum loop, it will become pretty apparent what it should sound like.
As far as I understand Threshold is fixed at -10db and the Input pot just works as attenuator, I find that I have to run a pretty hot signal though to get things going even with the Input all the way up.
I'm running the sidechain transformer at 1:4 as well, 1:2 was a bit tamer. Lastly I hope you calibrated the oscillator. Mine runs dead quiet so far and its not even in a box yet.

 

[Spencerleehorton]

how do i calibrate the 555?

I have a scope and  MM.

 

[marcn]

Probe the Output of the 555 (pin3), set it with the trimmer to 250khz.

 

[Spencerleehorton]

Probe with what?

My scope? Multimeter? What do I set it to?

Sorry never done this before!!!!

Regards

Spence.

 

[shot]

Spence,

Are you doing a single channel or dual?
If you're going with dual, it would be smart to check first if audio path is performing well. Since this is feedback type of compressor, you would need to disconnect the sidechain to measure if everything is ok on the main audio path. To do that, just take out TR11 (or at least it's base leg). Then go and check voltages around transistors. Follow the pye14schb.pdf schematic - it's the upper part of the circuit. While you have TR11 disconnected, it's smart to check also in the sidechain for any significant drops.
Disconnect the 7pin "switchboard" connector and check with resistance meter if everything is ok between pins 1-2, pins 3-6, pins 4-5 and 4-7. Just follow the schematic of the switchboard (lower left corner). That way you'll know if your switchboard is wired properly.
In one earlier post you mentioned that your B- rail is -15v. That's a bit low! It should be from 18 to 20 volts. Trim it with VR1.
Also check what's the supply voltage on the sidechain. It should be close to -12v. If it's too off than it's probably bad zener D8.

That's few things I can think by just looking at schemo. Can't compare it to my unit since it's in the studio.



Luka

 

[Ilya]

Also check what's the supply voltage on the sidechain. It should be close to -12v. If it's too off than it's probably bad zener D8.

These zeners are 5% tolerance so even as low as 11.4V will be within spec.

 

[Spencerleehorton]

Ok thanks for the tips Luka, how do I setup 555?

 

[shot]

Ok thanks for the tips Luka, how do I setup 555?

You'll need an oscilloscope or at least DMM with the ability to measure frequency.
This actually pushed me to get my first oscilloscope - a cheap chinese Hantek

Oscillator board has four legs. One on the top and three on the bottom.
Attach your ground probe to the middle leg or right leg on the bottom and signal probe to the left bottom leg (the one next to D1 diode). Actually if you observe the schematic it should be logical - top pin is connected to the voltage supply and we have no interest in it now. Two signals are connected to ground - probe one of them with ground probe. All that you have left is the output signal!
On this leg you'll get a reading of the frequency. If you're using oscilloscope you should se nice asymetric square waveform (short tick). Digital oscilloscopes have "auto" function that should be able to find your waveform period, stabilize it on the screen and they read at what frequency it is. When you turn the trimmer on the oscillator board the frequency of this waveform should go up or down. Adjust to 250khz and you're done!

Oh, and as a sidenote - for this project I bought myself an oscilloscope. I went with the cheapest option avaliable. And I regret it now! It's full of bugs, software is hard to install and measurements are all jagged. Since then I got me one oldschool analog scope and it measures much better, but it's too big and it's a drag to read values by counting blocks on the screen. I got an oportunity to try PicoScope and this one is great! That will be my next buy. It's suitable for all the audio measurements we need (assuming you're not into digital multichannel stuff). That is if you're not into spending more for "real" standalone scope with screen.



Luka

 

[Spencerleehorton]

I have old anologue scope which only has gnd and tip on each channel and I have two channels.
It sounds like your saying there is 3 connections I need to make?
Sorry I'm still not understanding how to connect and how to set my scope and how to read 250khz?

 

[Spencerleehorton]

I've taken the lorlins off now and i've drawn out how they should have been wired and rewired them, now none of them function!!!

I have 1mA meters from hairball i think, not joking when they say this build is hard!!

anyone have any info on how to set this thing up correctly please?

 

[shot]

I have old anologue scope which only has gnd and tip on each channel and I have two channels.
It sounds like your saying there is 3 connections I need to make?
Sorry I'm still not understanding how to connect and how to set my scope and how to read 250khz?

Spence,
You just need to probe one signal! That is the output from the oscillator board (lower left pin on schematic). But to have scope show you the correct measurement, it needs ground also. Usually scope probes have small wire with alligator clip on them for ground connection. Take the ground from the nearest possible pin (usually lower mid pin).
It takes just one channel of your oscilloscope!
On an analog scope, you have to manually set the TIME to be small enough to show you the waveform. If you convert frequency to period, 250kHz is 4 uS (microseconds). Set the time on your scope to a close value and adjust volts/div to scale it vertically. And that would do it.
If you want to be sure your scope is working properly, you can test it with a regular audio signal first! Just use some testgenerator plugin to send a sinus signal from your audio interface output and probe the tip and ground of the output cable with your scope. You should see a nice sinus on your scope, and you'll get a grip how to set time and volts/div parameters.
Oh, and be sure to set your scope to AC signal measurement!

L

 

[shot]

I've taken the lorlins off now and i've drawn out how they should have been wired and rewired them, now none of them function!!!

Don't just test it in your unit!
You must check if you get correct resistances referring to the schematic.
That is the easiest way to check if everything is wired correctly

 

[Spencerleehorton]

Ok I think I understand how to set the timer now!! Thanks very much for explaining.

I took off the lorlin Pcb and checked it with MM and everything is per schematic.

The only trimmers so far that do anything are the 100R trimmer but I've put in 200R as that the smallest I have.
The other trimmer that does something is the 5Meg but I have put in a 2Meg.
The other trimmers do very little to nothing?

 

[marcn]

I'd leave all the trimmers alone for now, however if you have tweaked them I'd advise you to set them at the following values as a starting point before further troubleshooting:
R23 =68R
R48 = 1m8 (you can install a fixed resistor here)
R42 = 5k1 (fine tunes the meter adjust, leave it alone until last i suggest)
R44 =220k
R52 = not indicated on schematic, i measured on mine to get good results approx 900R with junction of R51 (and consequently 100R with junction of 4148)

Now as mentioned earlier you need to tweak VR1 in your PSU to get anywhere between -18 to -22V as your B-. Do this first.
Then I suggest you go measure if all your Ratios are behaving right, Ilya pointed this out 2 pages back.
For this connect your switchboard and measure the voltage at the cathodes of D5/D6 under each of the ratio settings.
Typical results are mentioned on the 2 previous pages.

 

[Spencerleehorton]

Ok sounds like a plan! Will report back once I've done.

 

[Spencerleehorton]

Have set everything as you have ask and get nice signal through but no compression?
Also my meter is all the way over to the right?
Must have it wired backwards?

 

[Spencerleehorton]

Incidentally the input is 2 seg per div and the output is barely 3/4 of a segment.
I've changed the meter now so it sits at the correct side but controls do nothing apart from input and output pot?

 

[Spencerleehorton]

Adjusting R48 to lower value brings back a good sized waveform which matches input signal.

 

[Spencerleehorton]

Oscillator isn't oscillating, that's my main problem I think!!!

 

[shot]

Oscillator isn't oscillating, that's my main problem I think!!!

Try to measure AC (and DC while you're at it) on the output of the oscillator with DMM. If it works, it should show some value.
If it's idle it should output zero.