Steps to take to improve longevity and reliability?

[80k]

I just wanted to gather some ideas for improving reliability and longevity of a stompbox. If you have time, attention, and $$$ to spare, and the goal was to make something as bombproof as possible and reduce expected future maintenance many years into the future, what would you do?

For me, I already do these:
- dabble hot glue on all the wires where they are attached to the board, to eliminate stress on joints (I've had wires break off at the board).
- replace all 1uF electrolytics with mylar/film caps to improve longevity.

I'd also consider the following:
- if willing to spend the money, perhaps replace larger value electrolytics with tantalums where it doesn't affect tone negatively, to improve longevity.
- dabble hot glue to secure larger caps to reduce stress on leads.
- add a series Schottky diode to protect from reverse polarity.
- take out all sockets and solder trannies and opamps directly to the board to reduce chance that they fall out. Or keep the sockets and secure them in place with hot glue.
- use removable loctite on all threads (i have already do this occasionally).

I'm curious whether anyone has other ideas? For instance, using Mil-spec hardware? Or other best practices (like using heated wire strippers to decrease chance of compromising wire strand integrity when stripping wires)? Even ridiculous ideas are welcome.

[brett]

Hi
use silicone sealant instead of PCB supports.  It'll reduce peak forces on the PCB and components by an order of magnitude (or more).
In fact, avoid rigidity wherever you can (unless the component is unbreakable AND doesn't fatigue).
cheers

[80k]

Interesting idea. So do you mean completely avoiding PCB standoffs and use sealant to to secure the board directly to the metal surface of the enclosure?

[head_spaz]

MIL-SPEC would require adding at least two RULES to our habits.
1.) ALWAYS incorprorate reverse biased diodes from the input to each supply rail to help prevent electrostatic shock from damaging sensitive semiconductor inputs.
2.) ALWAYS place small bypass caps across all IC powersupply pins to ground. This does wonders for reducing hissy noises and it helps to prevent unwanted oscillation.
I wish everyone would incorporate both the ideas listed above into every circuit and pcb they design. Almost every data sheet since the early 1970's have included references to both of these principles, but few are listening. Bad habits prevail.

Additional to the above, these are some of the things I do on every build:
After assembly and testing, I clean all solder flux from the circuit boards and then spray the complete board with a heavy coat of clear lacquer to help prevent moisture and corrosion.
I use high quality polypropylene film caps wherever possible and refuse to use tantalum types. Tantalums are even worse than electrolytics when it comes to life expectancy.
I use only tinned - teflon coated - stranded wires, securely attached to prevent any unnecessary motion, with a dab of silicone for added strain relief and corrosion protection.
I also add a drop of light machine oil to all stompswitch actuators.
And the most important way to protect the longevity of your pedals is to NEVER EVER loan them out.

[80k]

head_spaz,

great suggestions there. I didn't know about those MIL-SPEC rules you mentioned, and will look into doing that.

As far as spraying a heavy coat of clear lacquer, does this make it more difficult in the future to service the board?

Also, I was surprised to hear you say Tantalum life expectancy is worse than aluminum electrolytics. I thought they lasted much longer than aluminums? Or is it related to the risk of reverse biasing?

[head_spaz]

The lacquer is no problem at all. It melts super easy if you need to replace something.
It's not real obvious once you clear coat them, it just makes them look especially clean looking. They look like they're ready for the magazine covers.
I have several projects that I never housed and they collect a lot of dust out in the shop, but they clean up real easy when they've been clear coated. And they don't corrode in this humid climate.

Tantulum's have built up quite a reputation over the years. Probably half of the stereo gear I've accumlated over the years is a result of people selling them cheap, or throwing them out for being too hissy. And the first thing I do is go thru and replace all the tantulums with electros. And that fixes most of them. About the only good thing I can say about tantulums is that they have a low ESR when they're new. But I would be hard pressed to use one myself. YMMV.

[head_spaz]

This is what I'm referring to:



The diodes are reversed bias. Typical input signal levels are undetered in any way, but if a large voltage appears, the diodes will conduct that voltage to either one or both of the voltage rails, whereby protecting the sensitive device's input. No matter how high the voltage, the input will see no more than .7V over the rail, because the whole device then floats with the rail. The "differential" is limited to the rail voltage... plus the .7 of forward voltage the diode will allow as it conducts.
Adding a 1K resistor in series with the input will increase the protection alot more as it becomes the dissipator of that energy.
Diodes are dirt cheap, and they don't take up a lot of space. This should be Standard Operating Procedure on ALL electronic devices that have external inputs. We should incorporate this option in every design, and let the individual builder decide if he wants to use it.

Bypass caps:
As ICs generate a sine wave output, they draw current in ALTERNATING PULSES from each of voltage rails. The alternating duty cycle creates a local current pulsation in each of the voltage rails because of resistance in the copper traces on the PCB.
If the voltage rails are not properly "stiffened" with local capacitance, this pulsated noise can ride rails throughout the entire circuit. This type of powersupply noise often interferes and modulates the power made available to other devices. It is a common source of unwanted noise and oscillation in sensitive circuits.
Caps are cheap, especially ceramic caps which are IDEAL for bypassing ICs. And they don't take up much space either.
I think we are so used to noisy pedals that we simply put up with more noise than we have to.
Is the cost of adding a dollar's worth of extra diodes and caps to our pedal projects so outrageous that it becomes a deal breaker? It shouldn't be. Especially when that dollar reaps huge rewards in building quieter pedals than we could buy anywhere else.
For me this a no-brainer.

And neither of these ideas are secret. Both are common practice and have been for many years in the electronic industry.
I'm actually surprised this stuff isn't frequently discussed here, because it's a well covered topic in most electronic forums.

Enjoy.

[R.G.]

I just wanted to gather some ideas for improving reliability and longevity of a stompbox. If you have time, attention, and $$$ to spare, and the goal was to make something as bombproof as possible and reduce expected future maintenance many years into the future, what would you do?

The simplest and most obvious is to build two, leaving the electro caps off the PCB on the backup unit, to be installed when it's needed. If I have something I absolutely need to be reliable, like the control electronics for the house water system, tank management stuff, I always build and test two to have a hot spare ready. It's costly, but the military has found out that having a hot spare is sometimes the only thing that will work.

- dabble hot glue on all the wires where they are attached to the board, to eliminate stress on joints (I've had wires break off at the board).

Good. Another way is to drill an extra hole in the PCB just outside the solder pad for the wire, stick the incoming wire, insulation and all up through the hole, then the stripped wire down into the pad and solder. The insulated part through the hole captures the wire and restrains the motion, while still letting it flex a little.

- replace all 1uF electrolytics with mylar/film caps to improve longevity.

That's good, but only as far as there are no other electros. All electros will one day die.

- if willing to spend the money, perhaps replace larger value electrolytics with tantalums where it doesn't affect tone negatively, to improve longevity.

Bad choice. Tantalums **are** electrolytic, just tantalum instead of aluminum. Tantalum pentoxide is different from aluminum oxide in some ways, but it will eventually degrade just like aluminum. And tantalum has some catastrophic failure modes that aluminum doesn't.

- dabble hot glue to secure larger caps to reduce stress on leads.

Good idea.

- add a series Schottky diode to protect from reverse polarity.

Also good if there is no reverse protection other ways.

- take out all sockets and solder trannies and opamps directly to the board to reduce chance that they fall out. Or keep the sockets and secure them in place with hot glue.

Good. I never use sockets for anything anymore.

- use removable loctite on all threads (i have already do this occasionally).

I'm bothered by this one. One of the best ways of making something last forever is making it easy to repair. This removes the failure of bolts and nuts from vibration but makes it harder to repair. The Japanese strategy is to use red glyptol insulating varnish on screws and nuts. Holds against minor vibrations, but breaks easily.

For instance, using Mil-spec hardware?

Even the military is no longer using "Mil-spec". The military now buys to reliability standards for equipment, not to the old Mil spec stuff. If you look up the military electronics specifications that are what "mil-spec" refers to, they are all marked obsolete, used for advice only.

using heated wire strippers to decrease chance of compromising wire strand integrity when stripping wires

I have and use heated strippers. Heated strippers are one of the very few ways to strip wires without nicking the wires. However, heated strippers are very rare.

use silicone sealant instead of PCB supports.  It'll reduce peak forces on the PCB and components by an order of magnitude (or more).
In fact, avoid rigidity wherever you can (unless the component is unbreakable AND doesn't fatigue).

I guess I'd change that to say that nothing which is rigid can be rigidly supported at more than one point. It's a good idea through. My Workhorse amps used extra thick PCB stock, bolted to the chassis at sixteen places, and incorporated steel stiffeners onto the PCB itself to prevent flexing. The support points were all at least 12mm from active traces and heavy components were themselves gooped with soft curing silicone rubber compound. The inside of a combo amp is a hostile environment. Even there, the most important thing was getting the controls, switches and jacks off the PCB on flying wires. Every control mounted to a PCB amounts to a rigid mounting point, and you can only have one of those.

MIL-SPEC would require adding at least two RULES to our habits.

See above about Mil spec.

1.) ALWAYS incorprorate reverse biased diodes from the input to each supply rail to help prevent electrostatic shock from damaging sensitive semiconductor inputs.

Protecting inputs is a deeper subject. For instance, just having diodes will prevent electrostatic damage, but is not much help for lower voltage, higher current faults like connecting to another piece of equipment where the ground is moving around by 20Vac from a leaky transformer. This connection can supply enough current to first toast the diodes, then fry the input. If you're using protection diodes, you need some series resistance to limit the peak currents too.

2.) ALWAYS place small bypass caps across all IC powersupply pins to ground. This does wonders for reducing hissy noises and it helps to prevent unwanted oscillation.

The hissy noises *are* RF oscillation. This is another place where mostly people get away with it, so people who haven't yet had the problem keep telling everyone that it's not a real problem - just like negative ground on previously positive ground PNP devices. It's like smoking - people know that it kills other people, but it hasn't killed them (yet) so they think they're immune.

Bad habits prevail.

They do, don't they? 

This is what I'm referring to:

I think we are so used to noisy pedals that we simply put up with more noise than we have to.
Is the cost of adding a dollar's worth of extra diodes and caps to our pedal projects so outrageous that it becomes a deal breaker? It shouldn't be. Especially when that dollar reaps huge rewards in building quieter pedals than we could buy anywhere else.
For me this a no-brainer.

And neither of these ideas are secret. Both are common practice and have been for many years in the electronic industry.
I'm actually surprised this stuff isn't frequently discussed here, because it's a well covered topic in most electronic forums.

It's not unknown. See "What are all those parts for?" and "When good opamps go bad" at geofex. You're not the only voice crying in the wilderness.


The diodes are reversed bias. Typical input signal levels are undetered in any way, but if a large voltage appears, the diodes will conduct that voltage to either one or both of the voltage rails, whereby protecting the sensitive device's input. No matter how high the voltage, the input will see no more than .7V over the rail, because the whole device then floats with the rail. The "differential" is limited to the rail voltage... plus the .7 of forward voltage the diode will allow as it conducts.
Adding a 1K resistor in series with the input will increase the protection alot more as it becomes the dissipator of that energy.
Diodes are dirt cheap, and they don't take up a lot of space. This should be Standard Operating Procedure on ALL electronic devices that have external inputs. We should incorporate this option in every design, and let the individual builder decide if he wants to use it.

Bypass caps:
As ICs generate a sine wave output, they draw current in ALTERNATING PULSES from each of voltage rails. The alternating duty cycle creates a local current pulsation in each of the voltage rails because of resistance in the copper traces on the PCB.
If the voltage rails are not properly "stiffened" with local capacitance, this pulsated noise can ride rails throughout the entire circuit. This type of powersupply noise often interferes and modulates the power made available to other devices. It is a common source of unwanted noise and oscillation in sensitive circuits.
Caps are cheap, especially ceramic caps which are IDEAL for bypassing ICs. And they don't take up much space either.
I think we are so used to noisy pedals that we simply put up with more noise than we have to.
Is the cost of adding a dollar's worth of extra diodes and caps to our pedal projects so outrageous that it becomes a deal breaker? It shouldn't be. Especially when that dollar reaps huge rewards in building quieter pedals than we could buy anywhere else.
For me this a no-brainer.

And neither of these ideas are secret. Both are common practice and have been for many years in the electronic industry.
I'm actually surprised this stuff isn't frequently discussed here, because it's a well covered topic in most electronic forums.

Enjoy.

This is what I'm referring to:



The diodes are reversed bias. Typical input signal levels are undetered in any way, but if a large voltage appears, the diodes will conduct that voltage to either one or both of the voltage rails, whereby protecting the sensitive device's input. No matter how high the voltage, the input will see no more than .7V over the rail, because the whole device then floats with the rail. The "differential" is limited to the rail voltage... plus the .7 of forward voltage the diode will allow as it conducts.
Adding a 1K resistor in series with the input will increase the protection alot more as it becomes the dissipator of that energy.
Diodes are dirt cheap, and they don't take up a lot of space. This should be Standard Operating Procedure on ALL electronic devices that have external inputs. We should incorporate this option in every design, and let the individual builder decide if he wants to use it.

Bypass caps:
As ICs generate a sine wave output, they draw current in ALTERNATING PULSES from each of voltage rails. The alternating duty cycle creates a local current pulsation in each of the voltage rails because of resistance in the copper traces on the PCB.
If the voltage rails are not properly "stiffened" with local capacitance, this pulsated noise can ride rails throughout the entire circuit. This type of powersupply noise often interferes and modulates the power made available to other devices. It is a common source of unwanted noise and oscillation in sensitive circuits.
Caps are cheap, especially ceramic caps which are IDEAL for bypassing ICs. And they don't take up much space either.
I think we are so used to noisy pedals that we simply put up with more noise than we have to.
Is the cost of adding a dollar's worth of extra diodes and caps to our pedal projects so outrageous that it becomes a deal breaker? It shouldn't be. Especially when that dollar reaps huge rewards in building quieter pedals than we could buy anywhere else.
For me this a no-brainer.

And neither of these ideas are secret. Both are common practice and have been for many years in the electronic industry.
I'm actually surprised this stuff isn't frequently discussed here, because it's a well covered topic in most electronic forums.

Enjoy.

[Mark Hammer]

These days, I almost always put a piece of heat shrink over any wire connection, whether solder lugs on a switch or a pot or a jack.  Even power jacks.  The shrink shold cover the lugs itself and a couple of millitmeters of the wire, too.  I do so for two reasons.  First, wires can fracture with bending, so the shrink provides added strain relief.  Second, unless you take exceptional steps to secure all chassis mounted controls/jacks or provide lots of space around them, there is often a risk that a toggle or pot or jack will be rotated juuuuuust enough that a lug shorts out against something else near it, like the chassis, or the back of another pot, or even the lug of an adjacent pot.  So, insulating the wire connection along its entire length (including the lug) is a good practice I find.

I used to hot glue the base of wire leads at the PCB.  Then I tried modding a circuit.  Bleccchhhh!!!   Probably far better to use silicon caulk that can either be pulled off, or at least won't melt if you attempt to reflow solder joints or change a part post-build.

I know that more folks are using expertnal power supplies on their pedal-board but not everyone does all the time, so battery connectors, and especially the flimsy wires they come with, are often a source of breakdown.  So, a number of years ago, I started making my own from the tops and bottoms of dead 9v batteries and decent wire.  I have an illustrated tutorial at my website (http://hammer.ampage.org/?cmd=lt&xid=&fid=&ex=&pg=7).  Unfortunately, there was a server crash, and the file is inaccessible at the moment.  From what I understand there are other such tutorials around, so do some googling.

Finally - and I know some of you are tired of hearing this - when I get new pots, I apply a few droplets of a contact-enhancing substance, made right here in Canada, called "Stabilant" (www.stabilant.com).  NOT a contact cleaner, it provides what is essentially a liquid solder joint between the resistive strip and wiper that does not dry out.  It provides a kind of lubricating function as well.  While it can be used as an adjunct to cleaning the resistive strip of an errant well-used pot, it is best used right up front so that the pot never accumulates the residue of too many wiper-to-strip rubs.  Keeps 'em noise-free for a long time, and since the stuff is chemically stable, one application is usually enough for the life of the product.

[Processaurus]

Even there, the most important thing was getting the controls, switches and jacks off the PCB on flying wires. Every control mounted to a PCB amounts to a rigid mounting point, and you can only have one of those.

Out of curiosity, how does the factory do the wiring in an efficient way? Is it just a matter of having bins of wires, cut to the right length and stripped already, for each connection, and a jig to hold the controls whilst soldering?  Or ribbon cable?

[80k]

Great responses everyone. head_spaz, thanks for the explanation on input protection and bypass caps. I am going to try to incorporate those into my future builds, for sure.

R.G., thanks for setting me straight on the Tantalums. For some reason, i thought life expectancy on tantalums were far greater than aluminums. I'm glad I haven't gone out and bought any expensive tantalums yet.

Mark Hammer,
I will look into silicon caulk instead of hot glue, and also heatshrink tubing, thanks for the suggestions. I have never tried to service a circuit in an area that had hot glue so never experienced the messiness that might arise from doing so. On the other hand, I'm generally careful not to put hot glue near areas of likely modding or parts that might need to be changed out (like electrolytics). My assumption is that if my solder joints are good, that hot gluing sensitive areas (like wires, larger caps, or multiple suspended resistors hooked in series or parallel) will make the board last virtually forever, as long as the electrolytics are still accessible to swap out.

[nelson]

Even there, the most important thing was getting the controls, switches and jacks off the PCB on flying wires. Every control mounted to a PCB amounts to a rigid mounting point, and you can only have one of those.

Out of curiosity, how does the factory do the wiring in an efficient way? Is it just a matter of having bins of wires, cut to the right length and stripped already, for each connection, and a jig to hold the controls whilst soldering?  Or ribbon cable?

Not to be annoying, but isn't nearly everything in consumer electronics PCB mounted?

I will defer to your experience, what kind of failure rate would you associate with PCB mounting everything, pots, switches jacks etc?

[snap]

Not to be annoying, but isn't nearly everything in consumer electronics PCB mounted?

thats 1 of the problems!

[R.G.]

Even there, the most important thing was getting the controls, switches and jacks off the PCB on flying wires. Every control mounted to a PCB amounts to a rigid mounting point, and you can only have one of those.

Out of curiosity, how does the factory do the wiring in an efficient way? Is it just a matter of having bins of wires, cut to the right length and stripped already, for each connection, and a jig to hold the controls whilst soldering?  Or ribbon cable?

When you know you're going to do a lot of anything, you can set up to do it efficiently. First you set up your wire-chopping machine (even if that's a person) to make a zillion of each color wire cut to the right length. Then you take the resulting buckets of wire pieces and strip them. There are machines for this, but you can also afford to make a machine with a heated "V" of nichrome strip which heat-cuts the insulation as you lay it in the V and twist, then pull. With buckets of stripped wires (or with a person right after the Nichrome-V person) you momentarily dip each stripped end in a solder pot to tin it. Then you solder three wires into each pot, two or three into each jack, etc. And with the resulting buckets of pre-wired controls, you can insert leads into PCB holes and solder quickly, especially if one person places and holds while another solders. When this is fully set up and running, the actual time per connection or control is quite small, as humans get very good at the movements as they practice. Jigs, fixtures (which are different!) and other setup items help you do this even more efficiently. Ribbon cable is usable only if you have insulation displacement connectors for it. Otherwise, the error rate at cutting, stripping, and tinning is too high.

Not to be annoying, but isn't nearly everything in consumer electronics PCB mounted?
I will defer to your experience, what kind of failure rate would you associate with PCB mounting everything, pots, switches jacks etc?

This is not an obvious point. All you really have to do to get a full understanding of why to put controls on flying wires is to talk to an experienced amp tech. I went to several and asked them about what not to do, in addition to having my own ideas.

Everything is PCB mounted to wring the last penny of labor cost and manufacturing time out of it. This is yet another facet of MBA Disease.  Most consumer electronics lasts for a year or two, and is then thrown away. All it really has to do is last longer than the warranty and not so short that you get a bad reputation; generally customers move on to something new before wear-out starts anyway. The general consumer design market is NOT ready for musicians who insist their stuff should last for several decades.

Mounting more than one control to a PCB and then mounting the PCB to a chassis is forcing the PCB to be tied to several mutually immovable points, then letting temperature and time move the PCB and controls around. Eventually something cracks, and it's usually the solder joint on the control leg because solder has not only low mechanical strength but also a low fatigue life. You can reinforce around this by making legs compliant and adding reinforcing legs and holders to take some of the stress off the signal pins of the pots, etc. You also have to design the chassis holes to have lots of extra room so the tolerances of placing and wave soldering controls doesn't make bolting the controls to the chassis put the PCB under continuous stress. The failure rate of one stressed solder joint is multiplied by the number of stressed joints; Marshall has some amps which are roundly cursed by any tech working on them because they have up to two dozen pots and jacks on a signal control PCB. To do any work on this at all, you remove 24 knobs, then 24 bushing nuts, then you pull out the PCB, and now you can start troubleshooting.

Like everything else, there are ways to do this right. We use PCB mounted pots and controls in our effects, but the distances are small (lower thermal movement over time) the thermal changes are low, the boards are designed not to flex, and the jacks are themselves on flying wires so as to not telegraph in any cable movement to further stress the PCB and pot joints. Jacks on PCBs are a bad idea for reliability because they have to be wiggled around manually to do their job and are prone to being stepped on.

But inside a hot tube combo amp, everything goes on flying wires to avoid making the heat and vibration issues any worse.

[Paul Marossy]

Use 24 gage silver stranded teflon wire. That's some very sturdy wire.

[Processaurus]

Thanks RG, that's an interesting peek into how it'd possible to manufacture music equipment affordably without PCB mount controls.  The solder pot for tinning wires sounds especially handy.

A compromise I was making in a beginning boutique businessman  pedal run was PCB mounted controls, but hand wired 1/4" jacks and DC jack (having had to fix SO many PCB mount jacks on mine and other people's stuff, I could never in good conscience do the same) and little daughter PCBs for the 3PDT footswitches.  The pots would all get mounted to their pedal first, and then (hopefully) the board drops right on and gets soldered, so there isn't any constant stress on the pot connections.  The pot's  will use the little locator tabs so if the nut comes loose, hopefully the stress is mostly on that rather than the legs going down to the PCB. 

Getting things good and tight, and using lockwashers helps reliability (so when they come loose non tech people don't end up spinning them around and twisting the wires off when re-tightening them), a deep socket wrench to tighten the jacks and footswitches down is handy.  It is a shame that the standard alpha pots are made out of aluminum and strip fairly easy, hard to get them tight tight...

I like the hot glued wires too, though the silicon sounds like it might be easier to repair.  Both are better than the awful glue boss uses in their pedals.

[Earthscum]

- use removable loctite on all threads (i have already do this occasionally).

I'm bothered by this one. One of the best ways of making something last forever is making it easy to repair. This removes the failure of bolts and nuts from vibration but makes it harder to repair. The Japanese strategy is to use red glyptol insulating varnish on screws and nuts. Holds against minor vibrations, but breaks easily.

Actually, there are several "grades" of LocTite. The classic "red" loctite is the permanent stuff you find in older Hondas (car). The blue loctite does exactly what you said. It is meant as a TEMPORARY (until broken free) security. As an ex-mechanic, I can confirm that the blue loctite is definitly a good idea if you are worried about loosing bolts. Also, a drop on the threads of your pot will help keep that nut tight, but still serviceable.

Another thing for thought is if you have a fairly thick housing, you can, instead of cutting off the location pin from your pots, grind it down to a small nub and use a bit to make a dimple in the back side. When you tighten the pot down, the nub will sit inside the dimple and keep it from rotating if it's loose enough and you get overzealous trying to crank it to the mythical "11". You have a solid pot that doesn't spin, and no unnecessary holes in your enclosures.

[ClinchFX]

I've seen some good quality gear where there are slots routed in the PCB most of the way around jacks, switches, etc.  This allows the small section of board with the jack or whatever to flex independently of the main portion of the PCB.  I still use flexible wire in my pedals.

As for stomp switches, if you have a stomp switch that's not soldered in to a PCB, try this test.  Mount the switch in a pedal enclosure or something that will hold it solidly.  Lightly hold a finger on the switch terminals while operating the switch.  With most switches, you will feel the terminals move very slightly as the switch operates.  This tiny shock can create dry joints when the switch is soldered into a PCB.  It's really important to design the PCB so that there's very little free space in the holes where the switch terminals fit.  Plated through holes will be even better. 

If I decide to use a rigid PCB for switches, I'll be ordering switches with PCB pins instead of the wire terminals, so that the hole in the PCB can be very small.

[R.G.]

Actually, there are several "grades" of LocTite. The classic "red" loctite is the permanent stuff you find in older Hondas (car). The blue loctite does exactly what you said. It is meant as a TEMPORARY (until broken free) security. As an ex-mechanic, I can confirm that the blue loctite is definitly a good idea if you are worried about loosing bolts. Also, a drop on the threads of your pot will help keep that nut tight, but still serviceable.

Yes, there are many grades and formulations of the cynoacrylate thread lockers, of which LocTite is one trade name. Machinists even use the stuff for gluing metals while they're being machined in some circumstances. I'm really familiar with the stuff, which is at least one reason what I said was correct.

The gook on the run of the mill Japanese stereo stuff from a decade (and more) ago is not one of the LocTite analogs. It's a quick drying electrical varnish used as a glue. It just happened to be red. I don't doubt that there are some places where thread lockers are used in the genre, but it's not the most common stuff in consumer equipment.

Another thing for thought is if you have a fairly thick housing, you can, instead of cutting off the location pin from your pots, grind it down to a small nub and use a bit to make a dimple in the back side. When you tighten the pot down, the nub will sit inside the dimple and keep it from rotating if it's loose enough and you get overzealous trying to crank it to the mythical "11". You have a solid pot that doesn't spin, and no unnecessary holes in your enclosures.

It can be done that way. But using a larger knob to cover the extra hole is a lot less demanding machining.

[Earthscum]

R.G.

I wasn't comparing it to the old varnish stuff, I was just saying (since nobody mentioned it) that there are grades of loctite that aren't permanent. Honda motors is notorious for using loctite (or something similarly strong) to help longevity in their vehicles. If you have parts (faceplate, for example) that aren't going to ever need removed, and you have them screwed down, go permanent. If you just don't want a screw falling out, try the temp stuff... I don't think anyone should be discouraged from using something like this, if they feel that it could be a solution to a problem. Personally, I'll probably start using a dab on my pots (and my guitarist's volume knob... he keeps twisting it off, so may grab the hot glue gun while I'm at it, lol).

And the "extra machining"... you clip the nib off and usually have enough sticking up to seat in a dimple. Just a quick spin from a bit. Basically, when you lock the nut down, it has just enough grabbing to keep the pot from spinning and possibly causing more problems. AND you can use micro knobs. Please don't shoot me down on this, it's something I've done in the past on repairs and it works awesomely. I just figured I'd give my 2c.