THE ENGINEER'S THUMB... At last, a better compressor!

[FlyingWild]

Thank you all your your help, I'm sort of relieved that which ever why the threshold pot goes in I can argue the case that it's the correct way!

Last night I had my first attempt at laying out the PCB, this was the first time I've done anything like this and I now have renewed respect for people who have produced beautifully laid out PCBs, it was daunting and frustrating at times. With a lot of vias and a number of tracks that go round three sides of the board to join components it did all join up in the end, but I plan to have another go tonight to see if I can do it better.

[digi2t]

A question regarding the Threshold pot. This one has always vexed me...

Would it be wrong to flip it around, i.e. pin 3 to the vref? That way, it would act as somewhat of a "sensitivity control" instead? Big signal inbound, turn it down, small signal inbound, turn it up?

Just looking at it from a different perspective I suppose, or am I out to lunch on this one?

[FlyingWild]

On Merlinb's schematic there are three 1uf electrolytic capacitors, do these have to be electrolytic caps?

The reason I ask is that I have some 1uf polyester box capacitors in stock from the last pedal I orders parts for, and I have some 1uf electrolytic caps on order, and I'm looking at the schematic trying to decide if electrolytic caps were chosen because they have properties that make them better suited?

I then took a sneak peek at Merlinb's web store, and saw he has some PCBs for sale and was looking for a bit of inspiration on my PCB layout, I realised there didn't seem to be enough round electrolytic caps on the silk screening... Then I spotted a couple of + signs next to another couple of capacitors, so I assume these missing electrolytic caps have been replaced by tantalum caps.

In trying to further my understanding of how electronics work, I'd be grateful to know if in the position where these 1uf caps are placed, is there something about them being polarized that is important?

[rankot]

Last night I had my first attempt at laying out the PCB, this was the first time I've done anything like this and I now have renewed respect for people who have produced beautifully laid out PCBs, it was daunting and frustrating at times. With a lot of vias and a number of tracks that go round three sides of the board to join components it did all join up in the end, but I plan to have another go tonight to see if I can do it better.

You can buy PCBs from Merlin. http://valvewizard.co.uk/shop.html

[FlyingWild]

Last night I had my first attempt at laying out the PCB, this was the first time I've done anything like this and I now have renewed respect for people who have produced beautifully laid out PCBs, it was daunting and frustrating at times. With a lot of vias and a number of tracks that go round three sides of the board to join components it did all join up in the end, but I plan to have another go tonight to see if I can do it better.

You can buy PCBs from Merlin. http://valvewizard.co.uk/shop.html

I know:

I then took a sneak peek at Merlinb's web store, and saw he has some PCBs for sale and was looking for a bit of inspiration on my PCB layout....

I built a compressor pedal many, many years ago from a kit that was featured in a UK guitar magazine, the compressor was call the "Paranormal" however there is something wrong with it now and it's eating the same 2N7000 FET each time I replace it. My knowledge of electronics is minimal and I decided I wanted to learn more, so in the last month I've built two over drive pedals, one from a kit, and one from a bought PCB, but it's still painting by numbers... So I'm currently in need of a compressor, and Merlinb's design seemed simple with regards to the number of parts and yet exciting as it's a different design, and I thought that trying to create my own PCB from his schematic would be one step further to gaining some understanding. I also have the advantage of having access to my dad's old etch tank and UV box.

[highwater]

On Merlinb's schematic there are three 1uf electrolytic capacitors, do these have to be electrolytic caps?

The reason I ask is that I have some 1uf polyester box capacitors in stock from the last pedal I orders parts for, and I have some 1uf electrolytic caps on order, and I'm looking at the schematic trying to decide if electrolytic caps were chosen because they have properties that make them better suited?

A 1uF electrolytic cap is going to be a LOT smaller and cheaper than a polyester cap. The polyester cap will be the same or better in just-about every other way.

[rankot]

I built a compressor pedal many, many years ago from a kit that was featured in a UK guitar magazine, the compressor was call the "Paranormal" however there is something wrong with it now and it's eating the same 2N7000 FET each time I replace it. My knowledge of electronics is minimal and I decided I wanted to learn more, so in the last month I've built two over drive pedals, one from a kit, and one from a bought PCB, but it's still painting by numbers... So I'm currently in need of a compressor, and Merlinb's design seemed simple with regards to the number of parts and yet exciting as it's a different design, and I thought that trying to create my own PCB from his schematic would be one step further to gaining some understanding. I also have the advantage of having access to my dad's old etch tank and UV box.

There is a thread about that compressor here: https://www.diystompboxes.com/smfforum/index.php?topic=52905.0, but lacks a schematic.

[merlinb]

On Merlinb's schematic there are three 1uf electrolytic capacitors, do these have to be electrolytic caps?

Tantalum, plastic/poly, or ceramic caps are preferred. Use electrolytic as a last resort.

[FlyingWild]

Thank you so much for making this schematic available Merlinb, this morning I etched my first PCB and just now I turned it on for the first time, and it worked! Tomorrow I'll spend some time playing around to see what it can do, but from a quick twiddle of the knobs and some very poor guitar playing I can already say I'm very happy.

[POTL]

Hi Merlin.
I collected information about compressors some time ago and came across your project again, I want to ask a couple of questions.
1) When comparing with the MXR, you indicate a number of advantages, including feedforward, what is its advantage over feedback?
2) In the side chain you use a half-wave rectifier, I saw a lot of circuits and they use a full-wave rectifier, why did you decide to restrict yourself to only half a wave?
3) Maybe earlier I asked this question, but still I will clarify, at the circuit input there is a capacitor with a small capacity, only 10 nanofarads, will this not be too small a value?

[merlinb]

1) When comparing with the MXR, you indicate a number of advantages, including feedforward, what is its advantage over feedback?

As explained on the webpage:
Feedback compression is simpler to arrange but technically inferior as it carries the massive disadvantage that you get overcompression on long attack times. In other words, the sound goes unnaturally quiet just after a loud transient, and then slowly returns, creating annoying pumping effects. This is why most pedal compressors don't let you control the attack- it is set permanently fast. With feedforward compression you don't get this problem, so you can have any attack time you want.

2) In the side chain you use a half-wave rectifier, I saw a lot of circuits and they use a full-wave rectifier, why did you decide to restrict yourself to only half a wave?

Full-wave requires an extra opamp.

3) Maybe earlier I asked this question, but still I will clarify, at the circuit input there is a capacitor with a small capacity, only 10 nanofarads, will this not be too small a value?

Too small for what? Does your guitar go down to 16Hz?

[POTL]

1) Thanks again
2) Yes, it would require another operational amplifier, but wouldn't it be the best solution? Indeed, many modern solutions have a full-wave rectifier.
3) Yes, you are right, I did not carry out a calculation, I just got used to seeing that all the schemes use 47 nanofarads or more.

[PRR]

> I did not carry out a calculation, I just got used to seeing that all the schemes use 47 nanofarads or more.

Why do "all" use 47nFd? Because they do not carry out the calculation!

What is it driving? 1 Meg plus two FET inputs (>100Meg). I don't even calculate that; I "know" 0.01uFd (10nFd) into 1Meg is 17Hz. Which is often good enough for wide-range High Fidelity. And certainly ample for guitar.

A fancier rectifier. I'm a big fan of excellent rectification in studio and concert-hall recording limiters. My experience in AM transmitting showed that some sounds are asymmetrical. But mostly male speech. There are other advantages. But half-wave gets you 90% of the benefits for less than half the cost.

Cost can be several things. The main thing I see looking at the ET is how few chips it uses. And how that simplifies the build. You know that half the people here would struggle with anything which had more parts. Builder time and struggle is part of the cost.

There is NO END of ways to "improve a limiter with more parts". It is a bottomless pit. On the one hand you can do a limiter with 5 parts around a power amp. At the other end, well, the EMT 266X comes very close to falling in the pit. Blesser spent over a year putting features into it. The box is stuffed with boards stuffed with parts. The knobs adjust every maybe-useful aspect of dynamics. Nobody has tried all settings. More to the point: it was very expensive to build, and unsurprisingly it has been hard to keep working over the decades.
https://www.bn1studio.co.uk/wordpress/wp-content/uploads/2018/11/0151680.jpg

Another over-kill limiter is the Fairchild 660. Weighs more than most large guitar rigs.

You have to find your market point. Do you want a Kia Trio? A Toyota Corolla? A Bentley? A Lamborghini? Yeah, the Lambo is cool, but often the Toyota is a fine choice.

[garcho]

Yeah but the Fairchild has big knobs, big knobs = better

[merlinb]

2) Yes, it would require another operational amplifier, but wouldn't it be the best solution? Indeed, many modern solutions have a full-wave rectifier.

[POTL]

I agree
Each design has its advantages and disadvantages.
Many vintage schemes, such as the Univibe and 1176 nowadays can be constructed using fewer components.
This is a way of obtaining knowledge, theoretical and practical, as well as a variety of experiments.
I am always happy to receive information from experienced people and it is most pleasant to receive it from the author of the scheme in order to understand how he is guided when making certain decisions, I learn and are happy to receive new information, including the one that does not always lie on the surface, it only appears with experience and it's wonderful that many experienced people are willing to share knowledge here.

[Ben N]

Yeah but the Fairchild has big knobs, big knobs = better

QED!

Ed: On second thought, is this argument some kind of portal back to the Burst Box thread?

[deadastronaut]

hi guys, just built the ET....cool comp. nice work merlin. 

just asking if anyone put the ''comp led indicator'' led in theirs and it worked ok.

as in PRR's led mod here.
https://www.diystompboxes.com/smfforum/index.php?topic=97123.msg1031906#msg1031906


cheers rob .

[rankot]

I tried that LED indicator, but wasn't too satisfied with that. Can't remember all the reasons, it was long time ago.

[deadastronaut]

Cheers man, I,ll just tack it on and try it .....see how it goes.