Reasons for/against circuit integrated with 3PDT PCB?

[CodeMonk]

I've never had a nut come loose on a 3PDT. I usually incorporate the switch onto the PCB, because it holds the board absolutely rock-steady and it cuts down on the rat's nest of wires a lot. No need for stand-offs, so no drilling or gluing for those at all. No chance of a short circuit to the enclosure.

All that said, I mill slots for the switch lugs. I've got a desktop milling machine with X/Y/Z wheels, so I do BNC instead of CNC (Brain Numerical Control ). This ensures zero insertion force when I push the switch into its slots - make sure it's perpendicular in both axes and then solder. Solid as a rock.

I also like to minimize off board as much as I can.
Although I still prefer not to mount jacks to the PCB. No particular reason for that other than I prefer it. Well, I do prefer to do top mounted jacks.


I mount stompswitches on the main PCB quite often.
I use this style of switch : http://www.smallbearelec.com/servlet/Detail?no=985
Not always, but more often than not.
A little Loctite goes a long way.
I also like to mount the pots to the PCB as well.

Never had a problem yet (knock on wood).

[GGBB]

So you have to take extra care to keep the stresses off any PCB the switch mounts to. The simplest way is to float the PCB, leaving it mounted only to the switch and not to other stuff. If you do that, the forces from stomping can act only on the button of the footswitch and the bushing, and do not stress the PCB or the solder joints.

Everything else you wrote makes complete sense, but on this I would have thought the opposite.  Wouldn't the hanging edges of the PCB be forever moving around - ever so slightly - due to gravity and vibrations, and therefore be constantly stressing the joints?  And the bigger/heavier the PCB, the greater the movement/force - worsening the problem.  It's not just stomping that we need to worry about - but any movement of the pedal as well.

[Arcane Analog]

No matter how well you plan out your pedal a stray foot to the PCB-mounted pots or DC jack is going to stress and/or crack the well planned PCB. I see it constantly. I do not care how well you have engineered your product. Shit happens on stage. Components like pots/jacks/switches/DC are much more likely to cause issues when they are attached to a PCB. Now you have a damaged part and a damaged PCB. Good times. That, and for typical part failure, fixing a PCB mounted component is more difficult.

[CodeMonk]

I solder my wires to wherever they go.
But there i one kind of connector where its pretty much impossible to remove the wire without tools.
Can't remember what they are called or any pictures of them (RG, I would bet you have seen or used these, help me  out here  ).
The connector has 2 holes/slots for each wire.
The wire just slides right into its hole.
The other hole is for removing the wire.
You insert a small flat head screw into the second hole, and twist the screw driver.
Once you insert the wire, the harder you pull on it, the more thay wire holds on

[Arcane Analog]

Everything else you wrote makes complete sense, but on this I would have thought the opposite.  Wouldn't the hanging edges of the PCB be forever moving around - ever so slightly - due to gravity and vibrations, and therefore be constantly stressing the joints?  And the bigger/heavier the PCB, the greater the movement/force - worsening the problem.  It's not just stomping that we need to worry about - but any movement of the pedal as well.

You are right - it does not make sense. In other discussions RG has frowned upon off board components that "float" yet here he recommends floating the entire PCB attached to something that gets repeatedly stepped on.

In the real world, a build that has the external hardware components - switches, jacks and pots - wired to the PCB versus everything PCB-mounted will be more durable and easier to repair. You can claim your "design" compensates for and considers all of the potential angles but in reality you still have PCB mounted components which will take any stress the box receives and transfers that stress to the PCB.

Mr. Musician has his front man accidentally step on his two prized pedals. Each sustained the same damage - a broken pot.

Pedal A has a broken pot. This pedal has the jacks, switches and pots mounted to the PCB. You need to remove every last nut and pull the entire circuit out of the pedal. Awesome. Then you need to desolder the pot from the PCB (which is never fun), replace it and put it all back together after you inspect the PCB/solder joints for cracks. Hopefully the PCB was undamaged. If it was, hopefully you can fix it. If the damage to the PCB is bad, the pedal will either be garbage or prone to further issues. You have to worry about adding stress tightening up all of those nuts and handling the PCB when fitting it back in the enclosure. Hopefully you do not have to manipulate the new pot or other components too much to fit it back in. Too much pressure on one can rack the old PCB or stress the joints. You have to love the ever so faint cracking sound of the board taking stress as parts adjust and the nuts tighten. Make sure you do not tighten them one at a time or it might not fit back in properly.  

Pedal B has the same broken pot but uses wire leads. Unsolder the leads, remove the pot, replace it and wire it up. One component without potential PCB damage from stress of the accident and the repair.

[R.G.]

Everything else you wrote makes complete sense, but on this I would have thought the opposite.  Wouldn't the hanging edges of the PCB be forever moving around - ever so slightly - due to gravity and vibrations, and therefore be constantly stressing the joints?  And the bigger/heavier the PCB, the greater the movement/force - worsening the problem.  It's not just stomping that we need to worry about - but any movement of the pedal as well.

You are correct, in theory. That's what all that blathering I was doing about elastic versus non-elastic and creep was about. On about the first day of a statics and strengths class, you're presented with the idea of a stress-strain curve.

It's a matter of degree. The smaller the flexes of a joint (and any material) the more of them you can do without breaking it. At some point the expected lifetime before breaking goes to years, then decades, then centuries and humans quit caring. Yes, the bigger, heavier, and more especially how resonant or vibrated by other stuff there is, the worse solder cracking is.

Note that a wire running off a PCB is a hanging mass that flexes ITS joint by vibration, gravity, etc. too. 

It's a matter of degree - that know when to hold 'em thing.  I have run into the rare set of guidelines for how much overhang and support is needed for PCBs, but I don't have a reference now. I try to support a PCB no less often than about 2-3" between supports of some kind; I don't put heavy parts out on cantilevered sections of PCB without damping (!) to kill vibrations. Vibration isn't generally an issue in the type of stuff we do with pedals. PCBs mounted on motors or machinery - aye, there's some problems. PCBs mounted inside combo amps with multiplane supports and cantilevered parts... well, there are penalties that Mother Nature assesses, without mercy or appeal. 

Do it right and your PCBs work fine. Do it randomly, you get lucky sometimes, unlucky others. Do it consistently wrong and you go broke with warranty repairs.

No matter how well you plan out your pedal a stray foot to the PCB-mounted pots or DC jack is going to stress and/or crack the well planned PCB. I see it constantly. I do not care how well you have engineered your product. sh*t happens on stage. Components like pots/jacks/switches/DC are much more likely to cause issues when they are attached to a PCB.

We used to have a phenomena called "flat topping" that we got sometimes. A guitarist - typically wearing Doc Martins! - would run across stage, leap into the air, and land on the pedal, breaking off all of the pot shafts. We found that although the shafts were broken, most of the pots and PCBs were OK. Same repairs, of course; pull out the PCB, replace all the pots. But if the pot bushing is well fixed to the enclosure, the shaft breaks before the PCB is damaged. True, you can stomp the shaft axially hard enough to drive the shaft through the back of the pot and damage the PCB, but hey, they won't stand being stepped on by an elephant or crushed in a hydraulic press either. 

The flat-topping phenomena led us to experiment with cast enclosures with a swoop up in the top to protect the pots. We have video somewhere of driving a van's front tire over one our pedals with the swoop. It still worked after that - although that did not encourage us to guarantee the pedals against car traffic or train wrecks. 

I think you're missing the point. It's not that you can't damage a PCB mounted pot. Everything has limits. But you can design it so that for the vast majority of uses, it's fine and has a long lifetime. I do this all the time, and have to live with the results of my work. I know some of the problems because I've made many of the mistakes before. It's that "know when to hold 'em" thing.

Just to belabor the point, "Know when to hold 'em" is very different from "Here be dragons".

Now you have a damaged part and a damaged PCB. Good times. That, and for typical part failure, fixing a PCB mounted component is more difficult.

If it was done correctly in design, the majority of damage it to the pot/switch/etc.. If it wasn't done correctly in design - well, sigh. What you gonna do?

Fixing a PCB mount part is pretty simple if you'll clip off the leads at the PCB, take the body out, and then remove the leads one by one. Try it - I think you'll like it.

[Arcane Analog]

Fixing a PCB mount part is pretty simple if you'll clip off the leads at the PCB, take the body out, and then remove the leads one by one. Try it - I think you'll like it.

That is how I generally do it - but you are still left with

You need to remove every last nut and pull the entire circuit out of the pedal. Awesome. .... Hopefully the PCB was undamaged. If it was, hopefully you can fix it. If the damage to the PCB is bad, the pedal will either be garbage or prone to further issues. You have to worry about adding stress tightening up all of those nuts and handling the PCB when fitting it back in the enclosure. Hopefully you do not have to manipulate the new pot or other components too much to fit it back in. Too much pressure on one can rack the old PCB or stress the joints. You have to love the ever so faint cracking sound of the board taking stress as parts adjust and the nuts tighten. Make sure you do not tighten them one at a time or it might not fit back in properly.  

which you did not address. None of those are issues with a well constructed (read wired) pedal. The first thing I ask when someone wants me to repair a pedal I am not familiar with is whether it is PCB mounted our not. Chances are I decline the work if it is and I kindly advise that they purchase a pedal constructed differently in the future.

The real world issues are key, RG. Cutting costs and maximizing profits are your #1 goal. Fair enough. That is a distant second to me. My priority is to build with durability and ease of repair in mind. If durability/repair were placed over costs/profits you would not be PCB mounting anything.

[Mark Hammer]

No matter how well you plan out your pedal a stray foot to the PCB-mounted pots or DC jack is going to stress and/or crack the well planned PCB. I see it constantly. I do not care how well you have engineered your product. sh*t happens on stage. Components like pots/jacks/switches/DC are much more likely to cause issues when they are attached to a PCB. Now you have a damaged part and a damaged PCB. Good times. That, and for typical part failure, fixing a PCB mounted component is more difficult.

If one is using a folded metal chassis, and what is holding the stompswitch in place is a white plastic washer over a too-big hole, with a hex nut over top, AND the board is secured at the other end by one or more long-lugged pots that are secured to the chassis and PCB, then yes, there CAN be differential pressure applied to the board, via the stompswitch, although it may be in the realm is a few mils of downward movement at one end while the board is securely held at the other.

Of course, a proper sized hole, and less pliable washers above and below the hole for the switch, OR anchoring the board at only one point, via the stompswitch, such as RG suggests, can avoid that differential downward pressure on the board.

Board/trace fracturing is not all that hard to prevent.

Incidentally, there is nothing that precludes one from using copper-clad board that is a little sturdier, with traces that are a little more resilient.

[R.G.]

You need to remove every last nut and pull the entire circuit out of the pedal. Awesome. .... Hopefully the PCB was undamaged. If it was, hopefully you can fix it. If the damage to the PCB is bad, the pedal will either be garbage or prone to further issues. You have to worry about adding stress tightening up all of those nuts and handling the PCB when fitting it back in the enclosure. Hopefully you do not have to manipulate the new pot or other components too much to fit it back in. Too much pressure on one can rack the old PCB or stress the joints. You have to love the ever so faint cracking sound of the board taking stress as parts adjust and the nuts tighten. Make sure you do not tighten them one at a time or it might not fit back in properly.  

There are many ways to do things wrong. That's what building education and experience is all about - learning to do them right. For instance, it took a long time to learn to do electrical power distribution without killing people and burning things down. I don't think I'd argue that since we still get a few electrical fires and electrocutions that we ought to stop using electrical power. Would you?

It's a lot like going to a doctor and saying "Doc, it hurts when I do this," and having the doctor say "Well, don't do that."  

As a bit of a side issue, if it's hard to get the board into the box and tighten the nuts down without cracking the board or sequencing tightening, then someone hasn't designed the pots, nuts and hole positions and tolerances correctly. Remember - there are lots of ways to get things wrong.

There are lots of ways to get hurt riding horses. Should we ban horses? Or perhaps enforce a waiting period and certified training in riding before we will allow people to buy hay?   Wouldn't it be better to tell them to go get some education in how to ride and care for horses?

which you did not address.

Frankly, it never occurred to me to design pedals where it's hard to get the pot shafts and bushings into the holes. And going with a one-plane fixing to the enclosure helps a lot, too. In fact, I suspect that there are many, many ways not to get it right that I didn't address.  

None of those are issues with a well constructed (read wired) pedal.

Ah. The binary implication of English. I'm sorry. Well constructed is not equal to "flying wires on the controls". There are a lot of other dimensions to the issue. But nice try at oversimplification in favor of your opinion.

The first thing I ask when someone wants me to repair a pedal I am not familiar with is whether it is PCB mounted our not. Chances are I decline the work if it is and I kindly advise that they purchase a pedal constructed differently in the future.

That's probably a good thing for you to do if you can afford to turn down the work.  

The real world issues are key, RG.

We completely agree on that one!

Cutting costs and maximizing profits are your #1 goal.

Sorry. You have no clue what my #1 goal is, or how I get there.

But ad hominem attacks like this are one of the classically recognized logical fallacies in discussion techniques.

That is a distant second to me. My priority is to build with durability and ease of repair in mind. If durability/repair were placed over costs/profits you would not be PCB mounting anything.

Ah. The evil, dispicable, money-grubbing things are not important to you. Kewl, dude.  

Your priorities are your own, and I hope you do well with them.

But the statement about durability/repair and cost/profit issues are not logically related. I was forced to attend huge amount of formal training on quality, durability, and low warranty costs (which amount to investing money up front so you don't get forced to pay it back later) methods back in the "why can't we do it like the Japanese?" era. I'm pretty sure I still have the handbooks showing the relationships between design for quality/reliability and the eventual costs of warranty repairs, and how it's better to do it right first and avoid the unknownable and probably unendurable costs of field repairs.

To make that argument stick, you'd need not only the repairs you do, the numbers for the repairs you *haven't* done, the ones that worked and that you never saw as failures because they didn't come into your shop. Then you'd have to compare that against the stats for *all* PCB mounted stuff, and back out all the ones which didn't have decent design as I've hinted at briefly; and then show in a statistically valid way that the failure rates for all PCB mounted stuff was the same to show that well designed and poorly designed PCB mounting had the same failure rate and that therefore all PCB mounting is equally unreliable.

You may have an opinion that all PCB mounted stuff is junk. Fine. You get to have your opinion. I have the opinion, based on my experience, that there are better and worse ways to do PCB mounting, and that the difference is always in the details.

And nice (albeit overused) argumentative gambit. Make the sideways implication that because the other guy doesn't agree with you, then he is malicious, reckless, or uninformed. It works often enough that it's used in a huge amount of political speech, but it's not logically supportable.

[Arcane Analog]

That post is simply too painful to try to address. You need to learn how to quote properly.

That said, your use of anecdotes and analogies are of little help in the debate which is simply your way of deflecting arguments you cannot best. They add nothing of substance but make it appear that you have addressed the statement.

I call bullshit on your use of board mounted components not being related to cutting manufacturing costs and maximizing profit. There is nothing wrong with that at all but be honest. You design for a company that has the bottom line in mind. If your company had the desire to make the most durable product possible you would not be using the PCB mounted layouts. I do not care if you ever admit that wired leads are better in that they are more durable. You know it and so do I. You can at least own up to the reality of your situation. You design gear to maximize profit. Your company would not be as profitable if they built to a higher standard. Plain and simple.

[gjcamann]

At the risk of splitting hairs

When soldering on a PCB, I feel that the solder joints are more consistent and reliable. Compared to off-board pots that require twice the number of solder joints and I find the final soldering to the floating pot not all that easy to consistently make.

Have you ever taken apart a Fulltone pedal? That is one overly constructed pedal. I really like how they mount the pots on the PCB. BTW they use discrete wired footswitches. (But I still put mine on a daugheter board).

[Arcane Analog]

Done properly, you should have a much better mechanical connection to a pot lug over a PCB trace. Sticking the wire into a lug and soldering it is not good technique. You can make a very secure connection to a pot by tining the wire, putting the wire through the lug and then "wrapping" it around and securing it. I can pull on my wire to lug connections before apply solder and have a difficult time removing them. They are so strong that post soldering it is sometimes easier to cut and remove than to simply remove.

Edit: Not my work but a quick example. PCB connections are not going to take the abuse this will.



I repaired an OCD a few months ago. Not a lot of fun to repair.  

[R.G.]

That post is simply too painful to try to address. You need to learn how to quote properly.

Gosh, you know how it is with us beginners. We almost never get it right. There, I fixed it. Better now?   

That said, your use of anecdotes and analogies are of little help in the debate which is simply your way of deflecting arguments you cannot best. They add nothing of substance but make it appear that you have addressed the statement.

Pots and kettles come to mind. Let me simply say that:
Your use of anecdotes and analogies are of little help in the debate which is simply your way of deflecting arguments you cannot best. They add nothing of substance but make it appear that you have addressed the statement.

Over.

I call bullsh*t on your use of board mounted components not being related to cutting manufacturing costs and maximizing profit.

I call bullsi*t on your trying to make it seem like I was saying something I wasn't. I never said board mounted was not related to cutting costs and maximizing profits. I said, and if you'll go back and re-read my stuff - and I guote (properly...    )
You said:

Cutting costs and maximizing profits are your #1 goal.

And I said:

Sorry. You have no clue what my #1 goal is, or how I get there.
But ad hominem attacks like this are one of the classically recognized logical fallacies in discussion techniques.

So I call bullsh*t on your calling bullsh*t. Read first, then bluster. 
And while we're at it, let's talk about trying to deflect arguements you can't best.   

There is nothing wrong with that at all but be honest. You design for a company that has the bottom line in mind. 

Well, let's look at that. Yes, I design for a company that has the bottom line in mind. But - as you say - be honest. Companies that don't have the bottom line in mind don't exist for long, do they? Be honest, now. And while we're at it, do YOU have the bottom line in mind? 

If your company had the desire to make the most durable product possible you would not be using the PCB mounted layouts.

Who said anything about making the most durable product possible? I certainly didn't. I said that simply using PCB mounted parts could be done better or worse, and that there were ways to do it well. In this case "well" means "fits the needs of amateur and professional guitarists"; this implies that I think that there is a limit where more durability doesn't add much to the value of the pedal to users.

Given any pedal, I suspect that I could come up with ways to make it more durable. As a practical matter, the "most durable product possible" hasn't been done. Mmmm - let's think; I suspect most any pedal on the market could be made more durable if you replaced every single part accessible from the outside with 1/4" thick hardened and heat treated high-alloy steels. Solid gold wiring would arguably me more reliable than copper wires, and spot welded component connections would be more durable than soldered. Shoot, even hard silver-solder (not 2% silver, the 40% and up stuff) or brazing the components would be more durable than solder at all.

So getting back to "most durable possible", that cannot be an unlimited criteria, can it? Are you instead talking about the most durable product that's reasonably feasible to make? Who gets to call "reasonable", then? Seems like there must be some criteria for what's good enough, yes?  (And be careful how you answer that one, there is a fishhook in it.  ) Do players really need pedals that can withstand a nuclear attack, or being submersed in aqua regia?

I do not care if you ever admit that wired leads are better in that they are more durable. You know it and so do I. You can at least own up to the reality of your situation. You design gear to maximize profit. Your company would not be as profitable if they built to a higher standard. Plain and simple.

In fact, I don't care if you admit that you simply disagree with me about where the line for "good enough" is, or that "good enough" does not arbitrarily exclude PCB mount whatevers. You're trying to divert a discussion on engineering matters into an accusation of corporate greed.

Have a look here: http://owl.english.purdue.edu/owl/resource/659/03/. I can't completely stay free of them, but I know they exist and try really hard to not pull them.

And while we're on corporate greed, have a look a what the the boss did while I wasn't looking. He recently upped the warranty on products from a year to lifetime *retroactively* to all previous pedals in our next earlier line. See http://visualsound.net/2012/warranty/  I think that means he's willing to put his money on the line for durability.

[Arcane Analog]

That was exactly the post I was hoping you would make.

You are incorrect on the corporate greed - I specifically said there was nothing wrong with the goal of minimizing costs and maximizing profits. You too need to read more carefully. I said you designed with that in mind and use that build technique for that very reason. Please clarify if I am wrong on that point.

For the record, I am not picking on your pedals at all. I have only had one on my bench. I declined two subsequent requests to fix two other units. I am not saying they are poorly built, cheap nor crappy in comparison to the industry average. I am making no claims as to the quality of your design or the companies pedals.

What I am saying the only reason companies use board-mounted components is because it is faster and cheaper to manufacture. Period. It is not because they are more durable than the lead alternative. Would you agree with that?

My point, and you know exactly what I am driving at as we have debated this before, is that wire leads are more durable and are easier to repair than board mounted components. If I built one of your pedals with PCB mounted components and a very similar pedal designed by the same company but setup up for wired leads the latter is the more durable unit. Would you agree with that?

[R.G.]

What I am saying the only reason companies use board-mounted components is because it is faster and cheaper to manufacture. Period. It is not because they are more durable than the lead alternative. Would you agree with that?

No. I believe this may be part of the issue with you. It's not that simple. The cost of a product is never as little as the cost of parts and labor to manufacture.

While the cost of labor and parts may be part of the equation, companies that have survived growing up a bit know that using the cheapest parts and labor can bite them HARD later. Long term, the products that make the most money are the ones which are durable *enough* for the needs of the people that buy them, and cheap *enough* for the customers to buy, and cheap *enough* to build to let the company keep going. Your insistence on the idea that board mounted is done for cheapness *only* may be limiting your outlook. The product has to fit the market and the market's ability to pay, and has to be of sufficient quality to meet the needs of the market, and has to be profitable enough to keep going. Miss any of those in any significant way, and the product, possibly the company too is toast.

Customers are not clueless sheep that have to buy whatever cheap junk is foisted off on them. They're pretty smart about what's good enough and what's not. Customers have the ability to avoid a company's products in droves. If you don't meet their needs - in the several dimensions of needs - not just in price or durability, or color or smell, or whatever else, they'll go elsewhere. Durability is only one of the things to look for. Durability can even be a bad thing - toilet paper comes to mind.

So no, with thought and reasoning behind the opinion, I disagree fully with the idea that the *only * reason companies use board mounted is to make them faster and cheaper. It is a gross oversimplification of a complex problem to state that it is.

My point, and you know exactly what I am driving at as we have debated this before, is that wire leads are more durable and are easier to repair than board mounted components.

I do know exactly what you're driving at, and it's misguided and oversimplified in several ways in my opinion.

If I built one of your pedals with PCB mounted components and a very similar pedal designed by the same company but setup up for wired leads the latter is the more durable unit. Would you agree with that?

No. It's a gross oversimplification of a fairly complex problem.

I'll have to go back and look up our earlier discussion, but I'd be willing to bet I mentioned Mencken and Einstein.

"For every complex problem there is an answer that is clear, simple, and wrong."
H. L. Mencken

"Make everything as simple as possible, but no simpler."
Which is attributed to Albert Einstein, but probably a simplification (!) of things he actually did say.

[Arcane Analog]

And that was the response I expected from you. Two hundred words where ten are needed. You say I oversimplify. I say you muddy the waters at every opportunity. 

For a fellow that likes to 'protest' fallacies in other's posts you sure love to use them alot.

[R.G.]

And that was the response I expected from you. Two hundred words where ten are needed. You say I oversimplify. I say you muddy the waters at every opportunity. 

For a fellow that likes to 'protest' fallacies in other's posts you sure love to use them alot.

Ad hominem
And that was the response I expected from you.

[SmoothAction]

Hey guys. It was not my intention to make waves. The issues concerning the pros/cons of on board 3pdt's have been explicitly expressed on both fronts, and I very much appreciate all views on the subject. There is no reason to degrade the integrity of the forum or each other.

[R.G.]

Hey guys. It was not my intention to make waves. The issues concerning the pros/cons of on board 3pdt's have been explicitly expressed on both fronts, and I very much appreciate all views on the subject. There is no reason to degrade the integrity of the forum or each other.

Not  your fault. It's an old, old issue. Sorry if we muddied up the waters for YOU.

[Arcane Analog]

Pointing out your hypocrisy is not an ad hominem argument. It is a simple observation. You have thrown out ad hominenm, circular arguments and formal falacies all over the place. I just choose to not point them out because you already have a good idea that you are employing them and you will simply add a few hundred words to blurr the issue more. I find it strange that you cry fallacy when you use them in almost every post.

Good night RG.