Securing AC and chassis ground points?

[Mike_J]

I am in the process of installing the safety earth bond and the circuit ground to the chassis of my 5f6a rebuild. Pdf64 has instilled in me a desire to make sure the ground is very secure. I figure while I am doing this I might as well make the circuit ground equally as secure.
 
Have attached a picture that includes a 10-32 screw, three different types of nuts, two different sizes of ring terminals and some permanent Loctite. I cut away the jacket on the ring terminals because it would be impossible to solder anything with it on so you will see it both ways. Needed two different sized ring terminals because the ground wire for the AC cord is quite a bit larger than the wire grounding the circuit board.

Two comments from different people I have discussed this matter with. One is my auto mechanic. Anyone who can charge two standard hours for something that takes one of his mechanics 15 minutes to do must be listened to. He says a crimped connection is better than a soldered connection. My guess is he can crimp a connection in a matter of seconds whereas soldering would probably involve more setup time and he is trying to get two hours of work done in 15 minutes. Anyway, I am going to both crimp the ring terminal and solder it as well.

The other is some kid at the hardware store. He says you get one chance with the nylon nut. If it needs to be loosened or removed it is finished. Would need to dispose of it and use a new nut.
 
Also including comments gleaned from writings of Merlin.

“The safety-earth bond should be made to a dedicated screw and connected to the chassis right at the point where the AC comes in with a short, heavy gauge wire. It should also be well bonded to the chassis, preferably soldered, so there is no chance of it coming loose.  A nyloc, shake-proof or star washer should be used, with two ordinary nuts, well tightened. The chassis area should be cleaned with emery paper.

To be safe it must be completely impossible for the metal chassis (and anything else the user might touch) to become live. Once the chassis is earthed it will be at the same potential as the person using it, and if any live wire were to touch the chassis it would immediately be shorted to earth and cannot shock the user.”

I am leaning toward using a keps nut with a regular nut tightly secured to it. The regular nut would get some permanent Loctite to increase the chances of it not coming loose and of course I would crimp and solder the ring terminals. Additionally, Merlin says to use a short ground wire. My ground wire will be at least two inches longer than the longest of the hot or neutral wires so if the wires are yanked out the ground wire will be the last to go. Also will use a star washer between the head of the bolt and the chassis.

Does anyone have an opinion concerning my method of securing these grounds?

[JB]

I crimp then solder.  Starwasher under the tag to bite into the chassis.  Nylock nut or some loctite.  And yes, keep it close to the inlet and longer than the live and neutral so it doesn't get pulled out. 

[Mike_J]

I crimp then solder.  Starwasher under the tag to bite into the chassis.  Nylock nut or some loctite.  And yes, keep it close to the inlet and longer than the live and neutral so it doesn't get pulled out.

Thanks JB. So you are in the nyloc camp instead of the keps camp. Any reason why? The starwasher under the tag is a completely new concept from my readings. Obviously you think it will give you a better ground connection than just applying the tag directly to the chassis. Presume you have never had any problems doing it that way.


Thanks
Mike

[Mike_J]

The reason I ask about the starwasher is I got something like 2,200 starwashers for something like $1.29. Pretty sure it is how the Chinese are getting rid of their used plutonium. Don't know how well they would conduct.

[Mike_J]

Have come up with a short-term solution until this issue is solved. Will use a keps nut only. No Loctite or nyloc, that can't be removed with destroying it, for now until a consensus is reached.


Thanks
Mike

[Mike_J]

Just read something on the internet concerning the nyloc. The Nyloc nut is an economic locking nut and ideal in applications where vibration or motion could loosen or undo the nut. Seems like that would be a plus for the nyloc, if it is true.

[JB]

Don't overthink it!  But yes, starwasher to bite through into the chassis.  We'd reject any mains powered kit not contsructed that way when doing type approval for broadcast engineering.

[Mike_J]

Don't overthink it!  But yes, starwasher to bite through into the chassis.  We'd reject any mains powered kit not contsructed that way when doing type approval for broadcast engineering.

You are significantly more qualified concerning this matter than I am. Learned something about thread galling. Have had that happen to me with stainless steel bolt and a stainless steel insert in an amp of mine that had the combination of them. Appears that it is not likely to be a problem if you go slowly and steady while putting the nut on. Of course I used a power drill most likely on the fastest speed when the problem I had occurred. Anyway, I will go slow and steady when installing the nyloc. Will probably still put another regular nut on top of it with some Loctite on it. Can't be too careful and will use a starwasher as you suggested. I overthink everything it is how I was made. I can't help it.


Thank you for the great suggestions,
Mike

[Mike_J]

I crimp then solder.  Starwasher under the tag to bite into the chassis.  Nylock nut or some loctite.  And yes, keep it close to the inlet and longer than the live and neutral so it doesn't get pulled out.

Last question, hopefully. Would you using an internal starwasher or an external starwasher?


Thanks
Mike

[JB]

Whichever I find in my hand when I dip into the box!  Probably internal though...

[Mike_J]

Whichever I find in my hand when I dip into the box!  Probably internal though...

Thanks again.

[Mike_J]

Thought if I went slow and steady there wouldn't be a problem with the bolt and the nyloc nut. Oh how wrong can I be as shown by the picture of the broke nut. Put a little anti-seize on the next one and had no problems. Have heard paraffin will work as well. Couldn't find my paraffin but knew where the anti-seize was.

[Mike_J]

Here are pictures of the extra long AC ground wire. Made it at least two inches longer than the hot or neutral wires. Another picture shows the ground connection with the nyloc nut. With the bolt I used there wouldn't be any room for another nut. Guess I can take the Loctite back and get another bolt and nyloc nut.

[dude]

I'd be more worried about the melt on the wires, watch your soldering, push all wires away from your soldering iron. Just take your time and be careful what your soldering iron touches.

[acheld]

I would NOT use Loctite on nuts you may need to undo.  Over time, even the blue stuff can make a mess.

Keps are cheap.  Nylocs are way better.  My buddies use them all the time in kit aircraft building.  They don't come loose with vibration and they are reusable until the nylon wears out.

For a kick, I went to https://www.aircraftspruce.com/catalog/hapages/an365_2.php  and had to laugh because I am paying more in my local ACE Hardware for their (unidentified) locknuts than certified versions through Aircraft Spruce.  Plus shipping of course. 

[Mike_J]

I'd be more worried about the melt on the wires, watch your soldering, push all wires away from your soldering iron. Just take your time and be careful what your soldering iron touches.

There is a reason I use nothing but Teflon anywhere I can. If they made PTs, chokes and OTs with Teflon wire I would buy them.


Thanks
Mike

[Soulfetish]

I think it might be helpful to clear up some things.
First, a properly crimped terminal connection gets zero benefit from the addition of solder. If fact if the crimp connection is done correctly, the addition of solder only increases the risk of the connection failing. Here's why:
A correctly terminated crimp connection requires a specific tool (often specific die) which is determined by each type of connector. In addition, it requires the connection by executed by a technician trained in how to do it. A crimp connector and termination is designed to create an airtight mechanical and electrical bond, which is much more resilient to mechanical stresses, and temperature fluctuation than a traditional solder joint.
While a proper solder joint creates a mechanical and electrical bond as well, it also makes the wire strands brittle at the junction of the soldered connection and can lead to breakage if subject to mechanical stresses or vibrations.
Having said all that, if you don't have the right tools, or don't know how to properly crimp the terminal your using, it's very easy to end up with a bad crimp. If this is the case, then soldering the connector after crimping will greatly improve a bad crimp job (A bad solder joint is just as problematic). But, if you don't have what you need to make a good crimp connection (either tooling or experience), then you're better off using a terminal which is designed to be soldered in the first place.
UL requirements for safety earth termination in the last iteration that I had access to forbid the crimp terminal from being soldered at all, either from tin soldering before crimping or after. They did allow for the earth conductor to be soldered directly to the chassis, with the provision that the earth wire have its own dedicated strain relief to prevent the wire from movement and wire fatigue at the solder joint. If you're chassis is steel, and you have a soldering iron that can heat up the chassis hot enough for solder to flow, then you might find it easier to use that approach, and end up with a safer connection. Here is an example of how it can be done using a vintage Fender as an example...
In the first photo, the wire was stripped and tinned for preparation. Then I used a magnet to secure the wire and hold it flush against the chassis. To prepare the chassis, I used a scribe and emory board to remove any dirt and oxidization in combination with 99.9% Isopropyl alcohol for cleaning. I then applied some liquid rosin flux to the area being soldered.
For the soldering I tin the soldering iron tip and put the flat part of the tip right at the junction of one side of the wire and chassis, and after a second or two, it's hot enough to begin to flow solder at the junction of the other side of the wire and chassis. (Don't just melt the solder over the hot soldering iron tip and let it flow to the chassis that way.)
After I'm done applying enough solder, I clean the area, and use a screw on cable tie to secure the single wire and hold it in place.

[Soulfetish]

Also, count me on team nyloc with acheld. 

[pdf64]

Yes, for a crimped joint to qualify as better than one that's soldered, for any specific crimp, the crimping tool approved for it by the manufacturer must be used, the tool must be within its period of calibration validity, and the operative must be competency assessed.
Those things are a world away from even a regular amp repair shop, nevermind a DIYer's workbench
Some random generic crimp that's done using a hardware store tool by a regular Joe isn't going to cut the mustard.

For our purposes, a soldered but mechanically unsecured joint to a lug isn't sufficient, as in the event of a fault, the solder will rapidly melt and the wire become detached or the connection otherwise fail. Hence crimped then soldered seems the best way option.

[terminalgs]

I use allen-hex screws and nuts with the captive star washer for anything between 4-40 and 10-24.  #1 it is easier to torque than a phillips-head and it won't cam over and slip, #2 and you can really torque them) and #3 musicians rarely carry allen keys that large so they are unlikely to attempt a DIY repair and innocently loosen a screw they ought not.  phillips-heads for regular service screws (like chassis to cabinet, or on a guitar pedal, the battery cover).


nylon locks certainly work and if I used them it would be more the reason to use allen-hex because the wrong type profile of a phillips-head screw driver will make a mess of the head of a phillips.   I've parted out a dozen 1960's cars for which torqued to spec fasteners that use captive nuts never loosen in all those 200-300K hard-life miles.  nylon lock is good, but captive washers (star or flat) are just as good if properly installed.


and +1 for crimp + solder.

[Soulfetish]

Yes, for a crimped joint to qualify as better than one that's soldered, for any specific crimp, the crimping tool approved for it by the manufacturer must be used, the tool must be within its period of calibration validity, and the operative must be competency assessed.
Those things are a world away from even a regular amp repair shop, nevermind a DIYer's workbench
Some random generic crimp that's done using a hardware store tool by a regular Joe isn't going to cut the mustard.

For our purposes, a soldered but mechanically unsecured joint to a lug isn't sufficient, as in the event of a fault, the solder will rapidly melt and the wire become detached or the connection otherwise fail. Hence crimped then soldered seems the best way option.

Fair enough, but level of compliance you describe protects you against someones lawyer more than electrical/mechanical failure

[pdf64]

Yes, for a crimped joint to qualify as better than one that's soldered...

Fair enough, but level of compliance you describe protects you against someones lawyer more than electrical/mechanical failure

Confused, please could you explain your thinking?

[JB]

Too much overthinking going on.  This isn't about high speed high frequency data where crimps, crimping tools and the method used are criticial and connections that look correct and measure ok on a dc loopback can nevertheless be faulty.  Many £/$K's worth of test kit required to diagnose.  I get involved in that at work, what we're talking about here is literally the opposite end of the spectrum.  This is all about passing fault current at 50/60Hz.  The barrel of the eyelet needs to be a snug fit and then compressed with the crimp tool onto the stranded cable to mazimise the surface area in contact.  A big pair of pliers squeezed hard will do the job.  The addition of solder to where the strands poke out of the barrel toward the eyelet can help add strength and a tiny bit more electrical conductivity.   

If in doubt then a Class 1 PAT test can be employed to measure the earth continuity with fault current, ideally with a tester with the higher current option, 10amps.
Not unusual for venues here in the UK to insist on seeing PAT test records for a bands equipment before they can play, or even be booked to play.

[Soulfetish]

Confused, please could you explain your thinking?

Sure, what I was trying to say was that in order for a crimp connection to qualify as better than a soldered connection, one didn't have to check off all those levels of compliance you laid out. Certainly, what you described would be considered best practices.
But what you described was the kind of process that a company would implement in order to shield themselves from any kind of liability for negligence, in case of litigation resulting from product failure.
I understand that the average diy amp builder probably doesn't have access to the kind of tools which would be found in a production or repair shop. So, I don't want to be overly pedantic about the crimp then solder technique. I think those who promote it, can make a compelling case in it's favor. Particularly, in this application where the termination is permanent and in a fixed location. But, my position is that "crimp-then-solder" it is an acceptable work-around.
Probably the safest work-around IMO, but I haven't done any testing on it.
But, for the professional technician, or anyone building and commercially selling Class I equipment. This should not be an acceptable practice. A quality ratcheting crimper is one of the most valued tools I own. I rarely go a day without using it, and I often use it several times a day. I chose one which included a die for 14-22AWG insulated terminals, because I would probably get the most use out of that. (Then gradually added dies for uninsulated terminals, open-barrel connectors, and wire ferrules.) A tool like this, used with a quality crimp terminal, is going to create a better termination without soldering than if you added solder after crimping, with the added benefit of being more efficient and meeting IEC and UL standards.