Acceptable methods for connecting three or more wires

[Foswick]

If you're insured I would advise against solder.Insurers follow ABYC.... If you feel you absolutely must solder, because your crimp tool is low quality, crimp first then add some solder..

That is not about avoiding solder, it is based on two separate engineering concerns.

First, while solder is electrically conductive, it is mechanically weak and brittle. The guidance just says that you should not mechanically rely solely on solder to keep to conductors mechanically connected. But, you shouldn't really leave any wire unsupported - just hanging on the connection - regardless of connection method. Wires should generally be secured so they to not rely on their connectors to mechanically keep them in place. Solder is no exception.

Second, if you connect a wire to a rigid conductor, there is a concentrated stress point where the wire tends to get flexed back and forth and weaken - that point where the flexible meets the rigid. When the wire is not properly secured, it will tend to weaken at that point, over time. This has nothing to do with solder. It is just that plain soldered connections tend not to have any other support to spread out some of that stress.

If the wire is secured, like with any wire organization device (conduit, wire clamps, zip ties, whatever) such that the wire is not hanging solely on the connection (solder or otherwise), then you satisfy these concerns. But, when you shrink wrap a soldered connection with adhesive-backed heat-shrink, this also satisfies the condition. This primarily addresses the second engineering concern, but "technically" addresses the first ... sort of.

Bottom line: ABYC or not, wires should be secured so that they are mostly supported by some mechanical wire-routing. This is usually some combination of clamps, conduits, zip ties, etc. The connection is just about conductivity. It should not be relied upon to support the wire or the thing attached to the wire. That is the point of 11.5.3.8. It is not a recommendation against solder.

The recomendation to crimp, then solder, is often-repeated but controversial. The problem is that, if you solder such that solder wicks into the wire strands past the crimped connector, then you have created that concentrated stress point that 11.5.3.8 specifically cautions against. The fact that there is a crimped connector downwind of that stress point is irrelevant to that concern. And, a proper crimped connection should already create a gas-tight connection.

But, this concentrated stress point exists where the wire meets a crimped connection, just like it does on a soldered connection. It just tends to be somewhat more distributed. Yet, I don't think ABYC has a mirrored statement similar to 11.5.3.8 which addresses this for crimped connections. But, the standards are very clear and consistent on the general importance of:

• wire support near terminals to prevent flexing at the connection
• strain relief provisions for cables entering equipment and panels
• general requirements that connections be mechanically secured independent of the terminal itself

All this said, industries have pretty well settled on crimp-only connections over soldered connections. There are two reasons for this. First, crimped connections are just faster, and labor usually loses out to a part that cost a few cents. But, more importantly ...

Contrary to what a lot of people think, a properly soldered connection is not necessarily electrically superior to a properly crimped connection. And, improperly done, neither is adequate. The problem is that a crimped connection is easier to visually inspect. This is second reason industries prefer crimped connections.

None of this is a recommendation for or against soldered connections, though. Just that, if you solder, you need to do it properly and with due consideration - just like any method.

 

[Maine Sail]

That is not about avoiding solder, it is based on two separate engineering concerns.

First, while solder is electrically conductive, it is mechanically weak and brittle. The guidance just says that you should not mechanically rely solely on solder to keep to conductors mechanically connected. But, you shouldn't really leave any wire unsupported - just hanging on the connection - regardless of connection method. Wires should generally be secured so they to not rely on their connectors to mechanically keep them in place. Solder is no exception.

Second, if you connect a wire to a rigid conductor, there is a concentrated stress point where the wire tends to get flexed back and forth and weaken - that point where the flexible meets the rigid. When the wire is not properly secured, it will tend to weaken at that point, over time. This has nothing to do with solder. It is just that plain soldered connections tend not to have any other support to spread out some of that stress.

If the wire is secured, like with any wire organization device (conduit, wire clamps, zip ties, whatever) such that the wire is not hanging solely on the connection (solder or otherwise), then you satisfy these concerns. But, when you shrink wrap a soldered connection with adhesive-backed heat-shrink, this also satisfies the condition. This primarily addresses the second engineering concern, but "technically" addresses the first ... sort of.

"Shall not be the sole means of mechanical connection in any circuit" is very clear. Wire support is also part of E-11. Yes, this was arrived at based what, you stated but does not change the standard. Heat- shrink does not count as a mechanical connection. Mechanical connections must meet the pull-test standard.

 

[Foswick]

"Shall not be the sole means of mechanical connection in any circuit" is very clear. Wire support is also part of E-11. Yes, this was arrived at based what, you stated but does not change the standard. Heat- shrink does not count as a mechanical connection. Mechanical connections must meet the pull-test standard.

To be clear, I was not saying that heat-shrink met the spirit of this standard. I just commented that it does, technically, meet the letter of it. (Which is true. It is not what I or 11.5.3.8 meant by "support" but it does satisfy the standard, as written.) And, indeed, it distributes stresses better than many crimped connectors. But, when I was talking about supporting wires, I was talking about fixing them to something solid with some hardware (clamps, zip ties, etc.)

E-11 is about minimum pull-out force for crimped terminals scaled by wire gauge. It does nothing to define a pull-test standard for evaluating the adequacy of secondary mechanical support of a soldered connection. It is true that a connection should survive the pull test, regardless of connection method. A properly soldered connection would pass this just fine. And, you are correct that heat-shrink tubing would do almost nothing to improve this. But, tensile force is not the cause of failure that 11.5.3.8 is addressing. That standard, as written, primarily addresses the flexural forces that break down a wire over time at this concentrated stress point. Mitigating those stresses is a solved engineering problem that often involves a strain-relief stiffener.

Hopefully I was clear that all wiring should be supported and not rely on the connector as its sole mechanical support. That is, a wire should not hang by its terminal connection. This is true of any connection. It doesn't matter what your connection method is, if the wire is not supported and that connection is repeatedly mechanically stressed, it will tend to fail at the point where the flexible wire meets the rigid body.

The ABYC standard does not advise against soldered connections. It just provides some guidance about how to do it safely. If ABYC meant to take a position against soldered connections it would say so, but it doesn't.

It is a misinterpretation to say that 11.5.3.8 suggests combining solder with a crimped connection. (I don't think you were saying that's what the standard suggests. I'm just saying that one could misinterpret it this way.) The standard is a response to failures from loose, unsupported wires being soldered to more rigid bodies, resulting in a failure over time at that concentrated stress point. The failure is from lack of support, not means of connection.

There are some good reasons to use properly crimped connections over soldered connections, but ABYC is not one of them, nor is insurance.

Respectfully, it is false to suggest that an insurer would have grounds to balk at soldered connections on the basis of the ABYC - at least not if they've read the standard.

 

[Garbonzo]

KEY WORD HERE, as Foswick stated is BRITTLE.. Soldered connections are brittle and prone to break at the solder edge from VIBRATION not just pulling. Same reason you don't use solid wire for 110 AC connections in boats and RV's while it is OK in a house. SO no soldered connections are best practice, period.

 

[Maine Sail]

To be clear, I was not saying that heat-shrink met the spirit of this standard. I just commented that it does, technically, meet the letter of it. (Which is true. It is not what I or 11.5.3.8 meant by "support" but it does satisfy the standard, as written.) And, indeed, it distributes stresses better than many crimped connectors. But, when I was talking about supporting wires, I was talking about fixing them to something solid with some hardware (clamps, zip ties, etc.)

E-11 is about minimum pull-out force for crimped terminals scaled by wire gauge. It does nothing to define a pull-test standard for evaluating the adequacy of secondary mechanical support of a soldered connection. It is true that a connection should survive the pull test, regardless of connection method. A properly soldered connection would pass this just fine. And, you are correct that heat-shrink tubing would do almost nothing to improve this. But, tensile force is not the cause of failure that 11.5.3.8 is addressing. That standard, as written, primarily addresses the flexural forces that break down a wire over time at this concentrated stress point. Mitigating those stresses is a solved engineering problem that often involves a strain-relief stiffener.

Hopefully I was clear that all wiring should be supported and not rely on the connector as its sole mechanical support. That is, a wire should not hang by its terminal connection. This is true of any connection. It doesn't matter what your connection method is, if the wire is not supported and that connection is repeatedly mechanically stressed, it will tend to fail at the point where the flexible wire meets the rigid body.

The ABYC standard does not advise against soldered connections. It just provides some guidance about how to do it safely. If ABYC meant to take a position against soldered connections it would say so, but it doesn't.

It is a misinterpretation to say that 11.5.3.8 suggests combining solder with a crimped connection. (I don't think you were saying that's what the standard suggests. I'm just saying that one could misinterpret it this way.) The standard is a response to failures from loose, unsupported wires being soldered to more rigid bodies, resulting in a failure over time at that concentrated stress point. The failure is from lack of support, not means of connection.

There are some good reasons to use properly crimped connections over soldered connections, but ABYC is not one of them, nor is insurance.

Respectfully, it is false to suggest that an insurer would have grounds to balk at soldered connections on the basis of the ABYC - at least not if they've read the standard.

So, you've done the training, attended the standards meetings, voted on changes, passed the certification etc.? The standard is clear, solder "solser shall not be the sole means of mechanical connection in any circuit."

You can solder, as I have said, but there needs to be a mechanical connection first eg: crimp. I actually understand exactly how this was arrived at, I'm a member of the Electrical PTC. What you miss by just reading the standards is how in-depth it is to arrive at a full blown standard... A major part of this was companies such as AMP, Molex, T&B etc. stating that soldering already crimped connectors is not advised. ABYC still allows it after crimping, but the first sentence is clear. When ABYC uses statements like shall not there is no wiggle room. Yes, many a good surveyor has flagged solder only terminations.

It is how a one of our customers picked up their boat for a song. it was a Sabre 34 MK I, the owner, an EE, had soldered every termination on the boat on the boat leaving inadequate wire to cut/re-terminate.. He was asking 32K and our customer picked it up for 8K because it needed so much electrical work. It went beyond solder, there was nothing safe about the electrical system. Decks were dry, everything else in good shape, but it needed a complete electrical re-fit.Two other marine electricians quoted correcting the deficiencies and their quotes exceeded the asking price..We told our customer if he re-ran all the circuit wires we could terminate them in a day..It took him 1000 hours but the next spring he was ready.. Solder was actually a benefit to Jim but not to the original owner..

 

[colemj]

the owner, an EE...

....It went beyond solder, there was nothing safe about the electrical system.

LMAO! Forums like this are full of EE's saying things they are completely convinced are correct, while demanding people accept them because they are an EE.

I find many of these EE's lead their posts with "I'm an EE...".

Mark

 

[Maine Sail]

LMAO! Forums like this are full of EE's saying things they are completely convinced are correct, while demanding people accept them because they are an EE.

I find many of these EE's lead their posts with "I'm an EE...".

Mark

Bingo! EE's are some of the worst DIY's, we worked with. It's like the minute they got out of school, and into their cube farm, they forgot Ohm's law, Kirchoff's, common sense, etc....

 

[JamesG161]

The Square Root of 1 = EE EE EE EE !!

Jim...

PS: EE believe in imaginary numbers

 

[colemj]

I am not an EE...

I understand the question in this thread is about largish size wires, where solder is an improper connection. However, when it comes to small wires like NMEA0183 connections, where one needs to extend a wire, I personally have no problem soldering these - ABYC be damned.

I put heat shrink on the individual wires, then a heat shrink over the wire bundle, so that it just becomes an extension of the existing wire with a little bump in it.

I also like the scotchlok's, even though they are inappropriate for stranded wire. But the problem with them is the two wires being connected are side by side afterward, and if connecting several wires in a bundle like an NMEA cable, it gets to be quite the mess. I have some that connect 3 wires, and those are useful at times.

I've never had either of the above fail.

Mark

 

[Project_Mayhem]

I put heat shrink on the individual wires, then a heat shrink over the wire bundle, so that it just becomes an extension of the existing wire with a little bump in it.

I do the same for many projects. You can make connections last a lot longer with proper support

I also like the scotchlok's, even though they are inappropriate for stranded wire. But the problem with them is the two wires being connected are side by side afterward, and if connecting several wires in a bundle like an NMEA cable, it gets to be quite the mess. I have some that connect 3 wires, and those are useful at times.

I love scotchloks as well and have a fairly low failure rate with them. The last failure I had generated quite a bit of heat. The clear insulation was black. It became evident to me that there simply wasn't enough surface contact between the single blade and wire. They have their time and place but I try not to use them these days.

For small wires I've had good luck with lever connectors. They are of course not great for applications where corrosion is likely as there is no seal. I have not tried dielectric grease with them