Flawed Layouts Thread

[Canucker]

Ever come across a layout that didn't say "confirmed" or "unconfirmed" and you built it only to later discover something was wrong with it and not with your build?

Often something as simple as one component...installed backwards or of the wrong value.... I haven't seen this so I figured it was about time to put them all in one place so we could either give corrections, advise against building them or provide people with an alternate of the same project to build.

[Canucker]

http://www.aronnelson.com/gallery/main.php?g2_view=core.DownloadItem&g2_itemId=1503&g2_serialNumber=3 D6 should be the other way around in this layout.

[R.G.]

... and thereby starting the idea of the layout creator adding the word "confirmed" to every layout because they've checked it at least once.

[blackieNYC]

+1 on that, RG, but there are a lot of schematics out there with errors or omissions that could stand to be user catalogued in a central forum database, no? That backwards cap on the scrambler, the PT 2399 lockup prevention methods that work and don't work, the mods that fix problems in the Phase Royale that haven't worked their way into the kit instructions.  Hard lessons learned over and over again.  I think builders might be quite passionate about exposing their fixes to well established circuits.  A list to peruse before, or after, a build. 

[Canucker]

I've never used a layout creator cus I'm not at the level of creating....I'm just really good at bashing my head against the wall when something I build doesn't work and then bashing it even harder when through research I find out that the layout was wrong and not me....

if I recall correctly this SuperFuzz layout has two errors http://ustomp.com/?p=12 The Resistor at the input should be AFTER the Capacitor that follows it in the layout...not before it....and I believe one of the other caps is backwards...can someone confirm that? or which one? 

[duck_arse]

according to the circuit ^ referenced, there is no 1M, so it is in the wrong spot wherever. as is, 22k over 1M won't stop the circuit working. and the first 10uF cap is backwards as one end is pulled to ground and the other goes to the transistor base.

the resistor before the cap is already there after it, in the form of the 100k to ground.

[karbomusic]

Hard lessons learned

over and over again.

^ I sure hope the first prevents the second.  

More seriously, I do think there is extreme value in learning from making mistakes. Secondly, I'm a little new to this but I have only built one other person's layout (all the rest I have done myself) specifically because I wanted to be stronger so that when I did build someone else's I'd have a chance of being able to tell if the layout was right. Another thing of great value (again IMHO) is breadboarding it first even if I use someone else's layout because for every circuit I have spent a little extra time getting to know, it has worked perfectly on the first try. Makes troubleshooting anything that might go wrong much, much easier.

I know the above isn't always possible, and no reason this thread wouldn't be helpful to lots of people. It just reminded me of some ideas that might be worth passing along.

[R.G.]

A few things come to mind on reading these posts.

- I've done layouts since the 70s. I don't think I've ever looked at one I did where there isn't something that I would correct later.
- For things which will be made by hand, for which there is no written standard defining the operation. Does "correct" mean "makes noise" or "matches the original circuit" - if there was *one* version of the original circuit, and if the original circuit was not oddly sensitive to variations? Correctness it not binary if you don't have a written standard, with measurable limits.
- Databases are themselves things which are constructed by hand. They contain errors and need maintenance. Who either gets to or will continue to maintain the database and scrub it for errors in perpetuity? How do errors in the **database** get fixed?

It is very tempting to simply make lists of the right stuff on the internet, and this idea pops up over and over and over and over and... I understand the urge that beginners have to just have this one thing be perfect so they don't have to contend with bugs and errors. However, the fact is, the world is full of errors and incomplete or wrong information.  It is so much better to learn to be flexible and deal with errors - nay, EXPECT there to be errors, and defeat them - than to set up another layer of the world which is supposed to be perfect, but can't be.

Creating things without error is manufacturing. Creating unique, personal things is art. Art is seeing something in the future and shaping pieces of the world into that vision by finding and removing the things that are not part of the vision - removing the errors or differences from perfection.

It's better to expect error, learn to recognize it, and deal with it, than to be dependent on someone else to "remove" errors for you. That concept is why so much of Geofex is about learning how, not just "replicate this".

[armdnrdy]

I think that the intentions are good here but...

as we all know, there is no control as to what "pops up" on the internet.

If someone where to do a search for.....the Ibanez AD-80 for example, there is no guarantee that this "Flawed Layouts Thread" will pop up pointing out incorrect component values, component polarity, or connection.

I think all we can do is post corrected schematics on the net with the hope that someone researching a project, will stumble across the "good" drawing.

[Jdansti]

A few things come to mind on reading these posts.

- I've done layouts since the 70s. I don't think I've ever looked at one I did where there isn't something that I would correct later.
- For things which will be made by hand, for which there is no written standard defining the operation. Does "correct" mean "makes noise" or "matches the original circuit" - if there was *one* version of the original circuit, and if the original circuit was not oddly sensitive to variations? Correctness it not binary if you don't have a written standard, with measurable limits.
- Databases are themselves things which are constructed by hand. They contain errors and need maintenance. Who either gets to or will continue to maintain the database and scrub it for errors in perpetuity? How do errors in the **database** get fixed?

It is very tempting to simply make lists of the right stuff on the internet, and this idea pops up over and over and over and over and... I understand the urge that beginners have to just have this one thing be perfect so they don't have to contend with bugs and errors. However, the fact is, the world is full of errors and incomplete or wrong information.  It is so much better to learn to be flexible and deal with errors - nay, EXPECT there to be errors, and defeat them - than to set up another layer of the world which is supposed to be perfect, but can't be.

Creating things without error is manufacturing. Creating unique, personal things is art. Art is seeing something in the future and shaping pieces of the world into that vision by finding and removing the things that are not part of the vision - removing the errors or differences from perfection.

It's better to expect error, learn to recognize it, and deal with it, than to be dependent on someone else to "remove" errors for you. That concept is why so much of Geofex is about learning how, not just "replicate this".

Lots of quotable quotes in that post. I've snagged one for myself. 

[Canucker]

my intentions with this post was to help people share info and experiences (like pretty much all of the threads on this board). It wasn't an attempt to promote laziness and prevent learning.

[CodeMonk]

Flawed layouts?
Me?
Never.
I've never made a mistake in a build or a layout. Ever.
I'm perfect when it comes to designing a layout and building it.

I'll be back to tell you all how awesome and perfect my work is later.
The Leprechaun family next door went to Narnia for vacation and I have to go feed their unicorns.

[stallik]

Lots of quotable quotes in that post. I've snagged one for myself. 

Agreed. There are at least 2 I'll use regularly. Thanks RG

[GGBB]

A few things come to mind on reading these posts.

- I've done layouts since the 70s. I don't think I've ever looked at one I did where there isn't something that I would correct later.
- For things which will be made by hand, for which there is no written standard defining the operation. Does "correct" mean "makes noise" or "matches the original circuit" - if there was *one* version of the original circuit, and if the original circuit was not oddly sensitive to variations? Correctness it not binary if you don't have a written standard, with measurable limits.
- Databases are themselves things which are constructed by hand. They contain errors and need maintenance. Who either gets to or will continue to maintain the database and scrub it for errors in perpetuity? How do errors in the **database** get fixed?

It is very tempting to simply make lists of the right stuff on the internet, and this idea pops up over and over and over and over and... I understand the urge that beginners have to just have this one thing be perfect so they don't have to contend with bugs and errors. However, the fact is, the world is full of errors and incomplete or wrong information.  It is so much better to learn to be flexible and deal with errors - nay, EXPECT there to be errors, and defeat them - than to set up another layer of the world which is supposed to be perfect, but can't be.

Creating things without error is manufacturing. Creating unique, personal things is art. Art is seeing something in the future and shaping pieces of the world into that vision by finding and removing the things that are not part of the vision - removing the errors or differences from perfection.

It's better to expect error, learn to recognize it, and deal with it, than to be dependent on someone else to "remove" errors for you. That concept is why so much of Geofex is about learning how, not just "replicate this".

All excellent points, but I like the OP's desire for outing faulty layouts, which I think the OP would define as a layout that wouldn't work if built as described.  Like with a backwards diode or polar cap, or an NPN transistor instead of PNP - many things that could merely be 'typo' type errors.  They might make for a good exercise for newbies and not-so-newbies in learning how to troubleshoot and fix, read schematics and layouts, understand how things work etc, but after you've been down that road many times you get to the point where sometimes you just want to build the dang thing and be done with it - no muss no fuss.  Not every pedal builder wants to be an electronics wizard.  We desperately need community QC.  Please continue this thread.

[armdnrdy]

... and thereby starting the idea of the layout creator adding the word "confirmed" to every layout because they've checked it at least once.

I think R.G. has made an excellent point here.

Who is going to verify all of the builds as working properly?

Do we take the first person to complete a "working" build at their word?

I know that I have built projects where I was confident that they worked perfectly, just like the original, only to find that I had a polarized cap oriented the wrong way. (I placed it on the schematic and the physical board the wrong way)
When I spun the cap.....the sensitivity control worked much better!

I posted the schematic and called the build good to go! Well......It turns out it wasn't.

[R.G.]

my intentions with this post was to help people share info and experiences (like pretty much all of the threads on this board). It wasn't an attempt to promote laziness and prevent learning.

Let me apologize if my comments seemed that way. I did not think you were attempting to promote laziness or prevent learning.

I *do* think that a thread that contains many references to "I built the XXYY circuit from the layout at this.web.site.xom and it worked great" is a good idea, as are the posts that say " I tried that layout and I could not get it to work." I think that's what you're saying now.

What I took issue with was the idea that any database of  confirmed layouts is practical. The idea is great- just maintain a list of what layouts are confirmed. However, I am a veteran of many such databases for all kinds of things, and I have seen firsthand the difficulty of setting it up, managing the users, keeping malicious users out, correcting mistakes in the database, and then keeping up the work on the database after the first flush of "Hey, this is GREAT!" wears off. These are serious issues, and they have killed many well intentioned efforts.

That is, I meant just exactly what I said, with no implications on you at all. I don't think you're lazy or unmotivated because of this. There are just some real problems in accomplishing it.

All excellent points, but I like the OP's desire for outing faulty layouts, which I think the OP would define as a layout that wouldn't work if built as described.

The continuous demonization of anyone who disagrees with someone else in the media has made "outing" a charged term.  I don't expect anyone ever really wants to leave a bad layout on the web. It's a very unfortunate term to choose that might start antagonism where there needn't be any.

Like with a backwards diode or polar cap, or an NPN transistor instead of PNP - many things that could merely be 'typo' type errors.  They might make for a good exercise for newbies and not-so-newbies in learning how to troubleshoot and fix, read schematics and layouts, understand how things work etc, but after you've been down that road many times you get to the point where sometimes you just want to build the dang thing and be done with it - no muss no fuss.  Not every pedal builder wants to be an electronics wizard.

To a certain extent, you're already living in the proposed database - this forum. Posts which start with "I built this effect and it doesn't work!" are one of the most common things here.

There are two possibilities at that juncture: either the person building the effect built it exactly per the information, but the description ("layout") was flawed, or the info was right, but the person made a mistake in building it. OK, there are three possibilties; both the layout could be bad as well as the builder making mistakes in building it.

When layouts get built, the probably of an unknown error being left in the layout declines dramatically with each one done. The more times a layout is built, the more errors get scrubbed out. By now, most of the layouts on the net would be getting pretty good - except for the swelling volume of first-time layout-makers with low experience. New layouts from new layers-out-ers and older layouts from experienced layout people are different.

We desperately need community QC.  Please continue this thread.

We not only desperately need it - we have it. It's right here, right now, this forum. All of it.

However, there is a continuous flow of new layouts that won't get tested without people building them. So there is a continuous amount of work testing new layouts that has to be done. At its heart, a layout-quality-certification database is the expression of hope that someone else found the bugs so one personally won't have to worry about them.

That's a nice state of affairs if you can get it. But as I noted, the real world isn't like that. Programmers have a good handle on this. They have a saying "There is always one more bug."

As a bit of rueful experience, I can tell you one almost certain scenario if the proposed Truth and Purity In Layouts database is set up.  The incidence of posts saying "I built this effect and it doesn't work! I know I did everything 100% correct, so the layout is bad." will decline for a while, then rise right back up to being one of the most common things here as new crops of newbies flood through. The immediate response on the part of the database maintainers will be to direct the new-newbies to the database. Then over time, the maintainers will get tired of it and go have a life, and the database quality will get worse.

And then one fine day, long from now, someone will say "Hey, why don't we put together a database of layouts that are known to be good?"

[armdnrdy]

What about starting a Flawed Builder Thread? 

1. armdnrdy
2.
3.

[R.G.]

I demand my place at the head of that list. I have made more boneheaded errors of commission and omission than any of you dag-nabb-ed young whippersnappers will get to for the next 20 years!

[armdnrdy]

I demand my place at the head of that list. I have made more boneheaded errors of commission and omission than any of you dag-nabb-ed young whippersnappers will get to for the next 20 years!

That may be but, you've made amends to the God of Perfection with all of the help and guidance that you've given us dag-nabb-ed young whippersnappers!

[GGBB]

It's a very unfortunate term to choose that might start antagonism where there needn't be any.

Well, I certainly wouldn't want to offend anyone.  I had no idea the use of the "O" word was so potentially explosive.  I was just using it the way I hear it used far more often than not - in a casual/joking manner.  My sincerest apologies to anyone who might have been offended.

Now lets get back to outing unmasking faulty layouts - after all, that's part of what this community is for.