Minisågverket: A simple metal distortion project

Started by Fancy Lime, August 17, 2019, 02:41:45 PM

Previous topic - Next topic

jfrabat

Quote from: duck_arse on June 19, 2020, 12:09:14 PM
I can't see it good enough, but I thimk you have one of your jacks wired signal to sleeve. check resistance between the sleeve and the other sleeve.

Sh!t!  Was SO focused on the board, did not even see that one!  LOL!  You are absolutely right!  Well, that fixed the problem!!!

Layout is verified.  Just one thing, the LOUD pot should be wired backwards (assuming I am looking at the back of the pot) so that clockwise is louder.  But it certainly works.
I build.  I fix.  I fix again.  And again.  And yet again.  (sometimes again once more).  Then I have something that works! (Most of the time!).

Fancy Lime

Quote from: jfrabat on June 19, 2020, 01:25:03 PM
Quote from: duck_arse on June 19, 2020, 12:09:14 PM
I can't see it good enough, but I thimk you have one of your jacks wired signal to sleeve. check resistance between the sleeve and the other sleeve.

Sh!t!  Was SO focused on the board, did not even see that one!  LOL!  You are absolutely right!  Well, that fixed the problem!!!

Layout is verified.  Just one thing, the LOUD pot should be wired backwards (assuming I am looking at the back of the pot) so that clockwise is louder.  But it certainly works.

Great, thanks for checking!

Are you sure about that Loud pot wiring, though? I am tired so my brain might deceive me here but I think it's correct on the schematic.


Quote from: Mark Hammer on June 18, 2020, 01:37:57 PM
RE: changing the opamps.

I've increasingly come to take the view that, in some instances, what we are hearing is the sound of an op-amp that has simply run out of headroom, subsequently getting re-clipped with diodes.  Given that the "Chug" control can theoretically increase gain by as much as 1000x, even very modest settings of the control exceed the headroom of the op-amp used, when powered with 9VDC.  So, while a superficial glance would suggest that "all" it does is crossover distortion, in fact the crossover distortion from the 1N4007 pair is simply supplementing the hard clipping produced as higher gain settings run the signal smack up against the ceiling on voltage swing.

All of which would suggest that, yes, in an application like this, varying the op-amp type can make an audible difference.

Yes, all the clipping comes from the opamp hitting the rails so different opamps sound quite different. That is kind of the whole point of this thing. Not so much difference at maximum gain but quite noticeable at "moderate" gains. With the 2M2 bias resistor, an NE5532 is actually very starved for bias current, which leads to an asymmetrical working point and audible "sag" when the s#!t hits the rails, so to speak. With 4M7 that is even more prominent. Misusing opamps is fun... A regular old 4558 with 470k or 1M bias should sound a bit more civilized, if someone who is after that actually considers building this thing here. Making the coupling cap C2 larger also makes it more civilized and less noisy, making it smaller produces more sag. Lots of fun to be had. Sticking antiparallel diodes in parallel with R6 makes it distort a lot more but obviously drops the volume and changes the sound. May be a good switchable option.

Andy
My dry, sweaty foot had become the source of one of the most disturbing cases of chemical-based crime within my home country.

A cider a day keeps the lobster away, bucko!

jfrabat

#22
Quote from: Fancy Lime on June 19, 2020, 04:09:31 PM
Great, thanks for checking!

Are you sure about that Loud pot wiring, though? I am tired so my brain might deceive me here but I think it's correct on the schematic.

Ah, seems like my dyslexia got the better of me again.  I was the one who had it backwards!  Your diagram is correct.  Sorry for the confusion!
I build.  I fix.  I fix again.  And again.  And yet again.  (sometimes again once more).  Then I have something that works! (Most of the time!).

jfrabat

Boxed it up.  Here is the final version:



And here is the mandatory clip (chug at about 2 o'clock, home-made Les Paul clone with Seymour Duncan Jazz pickup, using bridge pickup, through the Minisågverket to a Vox AC-5 clean and recorded with Cell Phone):

https://drive.google.com/file/d/1LEKPTjAXXnT4AWgrEAMxFiD11IvKDzhp/view?usp=sharing

I tried to do a riff similar to the one you posted, to compare (a not so accurate reproduction, but similar!)
I build.  I fix.  I fix again.  And again.  And yet again.  (sometimes again once more).  Then I have something that works! (Most of the time!).

Fancy Lime

I just now realized that the original schematic got downsampled somehow when I uploaded it, making it pretty much unreadable. Here is another try:



OK, looks good in the preview now. Let me know if you guys still have the issue with the low resolution. Really strange, never happened to me before...

Andy
My dry, sweaty foot had become the source of one of the most disturbing cases of chemical-based crime within my home country.

A cider a day keeps the lobster away, bucko!

jfrabat

Quote from: Fancy Lime on June 21, 2020, 01:30:55 PM
I just now realized that the original schematic got downsampled somehow when I uploaded it, making it pretty much unreadable. Here is another try:



OK, looks good in the preview now. Let me know if you guys still have the issue with the low resolution. Really strange, never happened to me before...

Andy

Wish I had had this before killing my eyes with the other one!  LOL!  No kidding, I had a blown up version (for the board), and a smaller one on which I used the phone's magnifying glass to read the numbers (they were too soft on the blown up one).
I build.  I fix.  I fix again.  And again.  And yet again.  (sometimes again once more).  Then I have something that works! (Most of the time!).

Liaztraht

#26
Sorry for the resurrection. I have been wanting to try my hand at etching boards, and have been able to visually verify some basic layouts, but really wanted to do this circuit.

I wanted to make a layout without jumpers, and with all components laying flat.

If anyone, especially Fancy would be willing to look it over, it would be most appreciated.

If it checks out, you all are free to use it. Won't fit smaller than a 1590B, but OCD kicked in for me.

Made some corrections.


Fancy Lime

Hi Liaztraht,

Thanks for the layout, looks great! I cannot find any errors. However, the two 1u film caps could probably do with a bit more space. These things are usually about twice as wide as in your layout. Not really a problem, the will fit with a bit of bending of some leads but if you give them more space, the assembled circuit will look a lot better. Also, the 100n cap C13 should sit as close as possible to pin 8 of the opamp. It's pretty close already, almost certainly close enough, but my OCD says it could be a little closer still :icon_wink:

Cheers and thanks,
Andy

P.s., just the other day I thought I should try making a PCB layout for this thing but I am so darn slow at that... Really appreciate you taking the time and putting in the effort!
My dry, sweaty foot had become the source of one of the most disturbing cases of chemical-based crime within my home country.

A cider a day keeps the lobster away, bucko!

antonis

Just realized Andy's schematic flaw.. :icon_wink:
Pin 4 (and not 3) should be grounded..
Pin 3 should be biased to +4.5V via 2M2 resistor..

or am I missing something relative to mis-bias distortion..??
"I'm getting older while being taught all the time" Solon the Athenian..
"I don't mind  being taught all the time but I do mind a lot getting old" Antonis the Thessalonian..

Mark Hammer

One of the things to remember about crossover distortion is that all the clipping wants is a signal that exceeds the forward voltage of the diodes.  Unlike clipping diodes going to ground, that provide harsher sound the harder you pick, crossover distortion occurs when you reach threshold, and doesn't increase appreciably beyond that.  However, your signal dynamics WILL be preserved.  In other words, strum harder and you'll be louder, rather than simply more distorted.

That said, the use of op-amp gain that greatly exceeds the headroom of the chip, using that supply voltage, means you'll still get some clipping and compression from the op-amp.  Seems to me that, unlike setups using a pair of diodes in the feedback loop of an op-amp, like your Novice Driver, ( https://www.diystompboxes.com/smfforum/index.php?topic=126297.0 ), that varying the supply voltage of the circuit will yield different feels and degrees of headroom-related clipping.  Looks to me like a job for a charge-pump.

Liaztraht

Quote from: Fancy Lime on February 01, 2021, 05:03:01 PM
Hi Liaztraht,

Thanks for the layout, looks great! I cannot find any errors. However, the two 1u film caps could probably do with a bit more space. These things are usually about twice as wide as in your layout. Not really a problem, the will fit with a bit of bending of some leads but if you give them more space, the assembled circuit will look a lot better. Also, the 100n cap C13 should sit as close as possible to pin 8 of the opamp. It's pretty close already, almost certainly close enough, but my OCD says it could be a little closer still :icon_wink:

Cheers and thanks,
Andy

P.s., just the other day I thought I should try making a PCB layout for this thing but I am so darn slow at that... Really appreciate you taking the time and putting in the effort!

I will make some room on it, and see about getting C13 closer. Also noticed the pads on the trace side are too small.  So it shouldn't be used just yet. A drill would just cut them out  :icon_lol:

Fancy Lime

Antonia is right, as usual. The grounded pin is 4.

The circuit relies on the opamp itself clipping, so a charge pump would decrease the clipping. It was originally designed with a bit opamp like the njm2068 or NE5532 in mind. These have relatively high bias currents, meaning the 2M2 bias resistor cannot bias them "properly". As a result, the clipping is quite asymmetrical and gets even more so, the higher the gain pot is set. This gives an interesting sagging component to the sound that disappears when an opamp with higher input impedance like the TL072 is used. That one sounds tighter in this circuit. Basically the whole thing is a study in what happens if you use an opamp wrong :icon_wink:

Andy
My dry, sweaty foot had become the source of one of the most disturbing cases of chemical-based crime within my home country.

A cider a day keeps the lobster away, bucko!

antonis

"I'm getting older while being taught all the time" Solon the Athenian..
"I don't mind  being taught all the time but I do mind a lot getting old" Antonis the Thessalonian..

Liaztraht

Quote from: Fancy Lime on February 01, 2021, 06:00:43 PM
Antonia is right, as usual. The grounded pin is 4.
I noticed that while laying out the board. Kept referencing your mini layout. Ran around in circles for a bit! It threw off this noob.

Mark Hammer

But as I may not have conveyed clearly enough before, increasing the supply voltage will have an impact on the op-amp clipping, but NOT on the crossover distortion.  So one should be able to vary the relative contribution of the two types of distortion, and get ONLY X-over by increasing the headroom and setting the gain for just enough to exceed the forward voltage of the series diode, but not so much that you crash into headroom limits of the op-amp.

I tried to do this (i.e., get different proportions of X-over and clipping) with the Contrafrizz, however: 1) I used a "normal" 9V supply,  2) varied the clipping, using clipping diodes, and 3) varied the contribution of crossover distortion by allowing greater and lesser amounts of that distortion to pass in parallel with the clipping.

Marcos - Munky

Nice layout, Liaztraht. Just a small tip, it's a nice idea to avoid those traces with right angle curves. While they do work, they are easier to be damaged and removed out from the board by overheat. Not an issue if you have some soldering skills, but anyway.

Liaztraht

Quote from: Marcos - Munky on February 01, 2021, 07:00:51 PM
Nice layout, Liaztraht. Just a small tip, it's a nice idea to avoid those traces with right angle curves. While they do work, they are easier to be damaged and removed out from the board by overheat. Not an issue if you have some soldering skills, but anyway.
Thanks for the tip! This is only my third attempt at trying to create a layout. I am tempted to go over everything with curved traces for a neater look.

Should I be bothering with keeping the ground fill out of the center? Or just not bother? It's a pain redoing it after making room for the 1uF caps  :icon_eek:

Liaztraht

Quote from: Fancy Lime on February 01, 2021, 05:03:01 PM
However, the two 1u film caps could probably do with a bit more space. These things are usually about twice as wide as in your layout. Not really a problem, the will fit with a bit of bending of some leads but if you give them more space, the assembled circuit will look a lot better. Also, the 100n cap C13 should sit as close as possible to pin 8 of the opamp.

I was able to move both 1u, but C13 is staying. Would have to move a lot of things to get it to work.

Also thickened the traces and pads a bit. Should be more durable for my first attempts at etching.

Updated layout in my first post.

Fancy Lime

Quote from: antonis on February 01, 2021, 06:15:16 PM
Quote from: Fancy Lime on February 01, 2021, 06:00:43 PM
Antonia

I definately have to re-count my Y chromosomes .. :icon_redface:
Ah, well, we need do something about the gender imbalance on this forum anyway. Damn autocorrect does not know greek names. I keep accidentally calling my friend Petros "Pedro" in text messages.

@Mark, ah, now I get it. Adding a pot for variable voltage, say 4-18V may indeed be interesting. But the way it is set up, you already get very prominent crossover distortion with most pickups if you stay below 10 o'clock or so on the gain pot. I really like pure crossover distortion for bass. Extra harmonics without loosing punch.

Cheers,
Andy
My dry, sweaty foot had become the source of one of the most disturbing cases of chemical-based crime within my home country.

A cider a day keeps the lobster away, bucko!

Marcos - Munky

Quote from: Liaztraht on February 01, 2021, 08:58:58 PM
Should I be bothering with keeping the ground fill out of the center? Or just not bother?
I'd just won't bother with that.