Delegate Compressor (Boneyard Edition) Distorts Sound

[du37]

I recently finished this build, but whenever I turn the sustain knob past around 9'o clock the sound starts to distort. Could this be from some of my other knob positions? Or, is something not working as it should?

 

[Erik S]

I don't know if it's right, but I definitely remember it was possible to get some distortion out of mine at certain settings.

 

[du37]

I'm not sure. If I remember correctly, the Boneyard edition was supposed to help with distortion. Most of my settings are pretty low, so that's why I'm also not sure that it's just my settings.

 

[andrewsrea]

IC1 pin voltages?

 

[BuddytheReow]

Is there distortion constantly or only for a brief period before going to "normal"? If the later, then there may just be too much signal hitting your circuit.

 

[du37]

Is there distortion constantly or only for a brief period before going to "normal"? If the later, then there may just be too much signal hitting your circuit.

It's constant, but worse with lower notes

 

[du37]

IC1 pin voltages?

What should it be at? I'm a college student so I won't have access to a multimeter until over the weekend.

 

[BuddytheReow]

Pics may help. What color is your LED?

 

[du37]

Yellow LED

 

[du37]

I think I might've figured it out. I have a 2N5089 in the spot labeled MPSA18. Is that the definitive source of the distortion, or is MPSA18 also a transistor?

 

[mybud]

Yes, the MPSA is also a transistor. In this case, worth getting the original and replacing the 5089 if possible. I don’t think they are equivalent and this may well be the cause of your distortion issue.

 

[andrewsrea]

What should it be at? I'm a college student so I won't have access to a multimeter until over the weekend.

Internet search an imige of the IC to determine orientation (pin location), referred to 'pinout'. Set your multimeter to DC volts, and set the range to 20v or higher. Send the common (black) probe to ground and the measurement lead (typically red) to the pins being measured.

All DC voltages should be + or - 1V of these:
Pin #1 = 8.7v Pin #8 = 17.5v
Pin #2 = 8.7v Pin #7 = 8.7v
Pin #3 = 8.7v Pin #6 = 8.7v
Pin #4 = 0v Pin #5 = 8.7v

8.7V is know as Vref and biases the AC component to 'zero out' or 'baseline' in between 0v and 17.5v (power supply, so that the AC signal does not get clipped asymmetrically or switched off, in extreme cases.

Pin voltages other than what is shown, will indicate a problem with the power supply Vcc (Pins #4 & #8), Vref (Pins which should be about half of Vcc), or an internal IC short (any Pins besides #8, which have Vcc voltages measured).

 

[andrewsrea]

I think I might've figured it out. I have a 2N5089 in the spot labeled MPSA18. Is that the definitive source of the distortion, or is MPSA18 also a transistor?

The 2N5089 and MPSA18 are both general purpose NPN transistors, with virtually identical 'on' and 'off' specs.

Additionally, those transistors control the compression ('Ratio') for LDR1, which acts basically like a volume pot which senses the AC signal intensity and leaks the signal to ground. It does not pass signal through to output. So, it is unlikely the cause of your distortion.

Other than the voltages I previously discussed; the other causes could be:
1.) The Ratio pot not being soldered well or not the correct value.
2.) The Sustain pot not being soldered well or not the correct value.
3.) Incorrect C6 or C20 values (from the picture, these look good).

The way this compressor works is:
IC 1.1 amplifies the signal according to the ratio and sustain settings (the more of each, the more amplification.
The more IC 1.1 amplifies, the more Q1 & Q2 sends current to the LED inside of LDR1, making that LED brighter.
Which causes the photoresistor inside of LDR of lose resistance and sends additional negative feedback to the IC 1.1
Which additional negative feedback lowers the amplification gain of the IC 1.1 circuit (same results as turning the Ratio down).

As you can see, this arrangement is constantly battling itself according to the amplification of the AC signal. As others have indicated, extreme clockwise settings of the Ratio and Sustain controls can cause distortion, as that 'battle' gets more intense. That battle can be adjusted by using different values for C6 and C20 (smaller cap values will make it less sensitive and larger ones will make it worse).

Distortion can also happen if Vref is not half of Vcc, as part of the signal waveform will get cut off when IC 1.1 amplifies. Ditto if the IC is shorted.

Note that IC 1.2 acts as an output buffer (unity gain - signal in = signal out) when the 'Level' is off (counterclockwise) and a small amplification as level is increased. The only way that can be creating the distortion would be Vref or a shorted IC.

Hope this helps.

 

[GizzWizzKing]

When I first built mine I had my LED installed backwards and I noticed that the saturation pot seemed to act a bit more like a gain control. There may have been other issues as well but maybe check your LED is in the property orientation. Good luck!

 

[du37]

So I finally got home for the weekend, and I swapped over the 2N5089 for a MPSA18, and it works perfectly now! I'm guessing the reason for the distortion was because the 2N5089 is not specified as a low noise transistior? Also, huge thank you to all of you for the help! As someone who knows little to nothing about how most of these little magic boxes work, the explanation from @andrewsrea was amazing!

 

[andrewsrea]

So I finally got home for the weekend, and I swapped over the 2N5089 for a MPSA18, and it works perfectly now! I'm guessing the reason for the distortion was because the 2N5089 is not specified as a low noise transistior? Also, huge thank you to all of you for the help! As someone who knows little to nothing about how most of these little magic boxes work, the explanation from @andrewsrea was amazing!

No. More like or more of your 2N5089's were out of tolerance / bad.

In my experiences and due to today's various manufacturers for ICs and transistors, the chances of getting an out of tolerance or defective part are rising. Many are using copies / clones of the etching masks, which only last for so many batches and some manufactures tend to use them past their lifecycle. For instance, my experience with components sourced through Amazon have been widely variable, from 'excellent' to 'not usable.'

 

[du37]

All of my 2N5089s were sourced from Digikey. Ironically, I was more worried about the MPSA18s because I got those off of Amazon.