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

[Majestic12]

Hey guys.

Can anyone offer any advice on how to go about calibrating the oscillator? I have a scope and sig gen, but not a lot of exp using them. A little more detail on the process would be awesome.

Thanks and cheers!

Hi Desol,

from your pictures it looks like you have not mounted the trimmers on the oscillator boards correctly. They seem to be aligned verticaly on the pcb holes which basically shorts out all pins.

 

[desol]

Majestic. Thanks so much for pointing that out!

 

[salomonander]

sorry i have to bump this one more time regarding transformers
i assume im fine with a 10k:10k input transformer
can i use a 150:600 on the output - or will impedance be messed up

sorry for the nnob questions i just cant answer it myseld. thanks

 

[shot]

Any chance someone has resistor values for 23 step switch?
Or just to point me to some calculator?

I'd really like to swap potentiometers to switches and I already have some spare 23 step ones!



Luka

 

[kosi]

Send Bruno2000 a pm, he has a spreadsheet for a 21step switch,
see  Reply #635

Any chance someone has resistor values for 23 step switch?
Or just to point me to some calculator?

I'd really like to swap potentiometers to switches and I already have some spare 23 step ones!



Luka

 

[desol]

Hey guys! Progress....I've managed to get both channels up and running and have been burning them in for the last day. The output sections seem to have gotten better/quieter from just burning in?, adjusting the levels, etc over the last 6 hrs. Fixed a few small problems. Basically both channels seem to be acting identically. Audio sounds good on both channels but the output signals are odd looking vs coming in(attached photo). I'm thinking this might be from the channels not being calibrated properly. Switchboards are functioning identically as well. Both my meters are showing half way on the scale. I think the original Pye's were like this. I'll need to figure out how to get them to zero.  Can someone help with the calibration procedure? I'm kinda new to this, but i'm having fun at it!

 

[bruno2000]

That looks like the 250kHz leaking through.  I see (?) that you have your scope connected to the board side of the output.  Try looking at the output side of the output transformer.
Best,
Bruno2000

 

[desol]

Hi Bruno! Yes, the sines look normal post output transformer. 

Now(i suppose) it's just a matter of getting the meters to sit right, doing the calibration and testing. Any advice with how to set the oscillator with the scope? Best place for the scope leads?

 

[Rob Flinn]

Both my meters are showing half way on the scale. I think the original Pye's were like this.

The original Pyes I deal with aren't like this.

 

[desol]

Hmph. Well, thanks for letting me know Rob.

 

[Patrick from Davis]

Out of curiosity, if you were to get different output transformers, what would you try?  It might be nice to get a list and some general details for future builders.  Not sure that there is a huge amount of interest, but still.
Patrick

 

[desol]

Don't know yet. I'd have to find out the impedances, etc...then choose a high quality, good sized unit based on that I think. I need to learn a few things first about measuring, etc. I think Joe put a whole separate output section and quality transformer on his...

Anyway, there has to be at least better than these edcors. (i'm assuming)

Here's something else had to do(for anyone building). Hole for meter adjust through frontpanel:

 

[desol]

(accidental erase of previous posts. Sorry to anyone who was looking!)

Well, DMM came in and I was pretty close. Channel 1 measured at 255khz. Channel 2 - 254.8khz.

Anyway, dialed them both down to 250. Still need to suss out an intermittent noise issue (I think at tr8) on right channel. Comes and goes, comes and goes. Got the vintage LED's in the meters and spec'ing out some white smaller knobs for it. This unit sounds exactly as I imagined it should sound. lol I'm still kinda blown away by it's character even a few weeks after having it here...with my other units. It's very powerful, the way it grabs and manipulates and in combination with the portico 5043(which i normally use to trail things) it's insane. Instant incredible sound.

Anybody have any tips for swapping out transistors, while doing minimal damage to the tiny pads/holes?

 

[mdainsd]

I sacrifice the transistor coming out. I snip its leads as close to the transistor as possible. Using tweezers i grab the lead and add heat to get it out of the hole, Then follow up with a solder sucker or wick to clear the hole.

Glad you are getting good results now. I haven't had much time to post of even play with stuff as the band I'm in is gaining traction and its consuming some serious time.

I will go back and find my notes on how I calibrated mine, but what i remember is I set the input to maximum sensitivity (10dB on the NEG side), then put the scope on the side chain where you see the DC control signal. I used a precision meter to verify the tone going in was at -10dB and adjusted R52 (I think) until the DC on the scope just barely started to drop (remember these things are ass-backwards with  their negative supply rail, so that DC level "dropping" is actually an increase in control. Once that was done I set the  input attenuator to 0dB and raised the input to 0dB and checked that the the control level was just starting downwards.

Mine sits with the meters at zero with no compression and I did nothing with the values of those restores around the meter.

The weak area of this build is specing the output transformer and also the input transformer for the side chain...but...I found these damned things worked so well for what I needed, I quit fiddling with it and just use it.

 

[desol]

Hi Mdainsd! Thanks so much for the reply. Best of luck on your guys band endeavors!

I'm a little confused about the input control; how it's supposed to work. The front panel silkscreen(as on the originals) connote that the input(threshold) should be at zero all the time...and then you turn it into the negative(like other compressors) to get gain reduction. Easy. However on mine, depending on mic pre gain...it starts to compress as I turn UP the input from let's say zero at the bottom (if the pot were turned fully counter clockwise). Maybe I have the pot wired backwards?

I did realize that something wasn't making sense with the operation of input control on my unit, but the unit seemed to be working fine so I hadn't looked into it too much at that point.

In the manual on the basic control circuit operation, I'm assuming that we want to probe the 'end' of the control signal which is at the base of tr11.

 

[Rob Flinn]

Made some more progress on this after it being on the shelf for the last 2 years (yikes).

Turns out my noise on the output was the jensen transformer I was using, or how I'd hooked it up possibly.  I need to get the data sheets again.  I subbes in a 600:10k Gardeners octal & all is well.  Although like Bruno says it looks like it could do with another roughly 6dB of gain. I notice that on the attached pic of an alternative version of the PYE it seems to have a rather more beefy output stage.

I'm using a pair of OEP transformers for the sidechain & Gardeners octals for the inputs.

I still wish we knew more about these trimmers.

 

[mdainsd]

Hi Mdainsd! Thanks so much for the reply. Best of luck on your guys band endeavors!

I'm a little confused about the input control; how it's supposed to work. The front panel silkscreen(as on the originals) connote that the input(threshold) should be at zero all the time...and then you turn it into the negative(like other compressors) to get gain reduction. Easy. However on mine, depending on mic pre gain...it starts to compress as I turn UP the input from let's say zero at the bottom (if the pot were turned fully counter clockwise). Maybe I have the pot wired backwards?

I did realize that something wasn't making sense with the operation of input control on my unit, but the unit seemed to be working fine so I hadn't looked into it too much at that point.

In the manual on the basic control circuit operation, I'm assuming that we want to probe the 'end' of the control signal which is at the base of tr11.

you may want to revisit your input "threshold wiring".
The lower you turn it (counter-clockwise), the lower the knee of compression. So if it is set to say -10dB. As soon as your input goes above that number it start to compress anything above that at the ratio you have selected.

Note: If you have the threshold on -10dB and the ratio selected for "limit", then anything above -2dB is limited. There is an offset on limit of 8dB.

 

[Rob Flinn]

you may want to revisit your input "threshold wiring".
The lower you turn it (counter-clockwise), the lower the knee of compression. So if it is set to say -10dB. As soon as your input goes above that number it start to compress anything above that at the ratio you have selected.

Note: If you have the threshold on -10dB and the ratio selected for "limit", then anything above -2dB is limited. There is an offset on limit of 8dB.

Not really getting your explanation here. Is the threshold level not fixed & you just wind more or less signal into it to get it compressing at different levels of signal.  The input pot is just a level control surely ?

 

[mdainsd]

Not really getting your explanation here. Is the threshold level not fixed & you just wind more or less signal into it to get it compressing at different levels of signal.  The input pot is just a level control surely ?

The input or threshold is an attenuator. When it is turned fully ccw (-10dB), there is no attenuation to the input signal at all, straight through into the compressor. The base compressor circuitry uses this -10dB as the knee if adjusted correctly. The output gain is also set at minimum (in fact some original PYEs have these two controls on the same shaft).

As you turn the threshold CW you are attenuating the signal going into the compressor portion. The output or makeup gain would be increased the same amount.

So, to reiterate: threshold set to -10dB, no attenuation, compression starts at -10dB input level. No output make up gain, so signal leaves compressor at same levels it came in albeit with compression at levels above -10dB.

Threshold set to 0dB: Input (lets say 0dB) is attenuated down to -10dB where the compressor does its work. Then the output or makeup gain is set for 10dB, boosting the output back up to the same as the input, again albeit the compression above 0dB.

This works so much easier when the threshold and makeup gain are on the same control.

I hope this helps.

 

[Rob Flinn]

So what you are saying is that the input control should be reverse log (if it were an individual pot) so that as you turn it clockwise you turn the signal entering the compressor down & therefore you get less compression because your peaks over -10dB won't be as much.  The threshold (electronically speaking) is fixed & remains at -10dB regardless ?