Crimp *and* solder a crimp connector? Good? Bad? Why?

[Psi]

I don't see how solder could improve a crimped joint. A properly crimped joint is already gas tight, and often there are even cold welds that develop, like wire wrap.

It improves your confidence that it's not going to pull out.

If you're worried about the wire pulling out, your crimps are no good. If you don't have complete confidence in your crimps, there's no point crimping. You should switch to some sort of solder cup style connector instead because that will be designed for soldering. The whole point of crimping is to make a gas tight, solid connection. If you only manage to make a crappy connection that "pulls out", or whatever, and you feel it needs to be soldered, isn't it better to use a connector that's designed to be soldered?

But that said, it's absolutely trivial to make a proper crimp with a crimper and the correct dies. None of this is particularly expensive, and you would have to be completely brain dead to mess it up. Just make a couple of test crimps and then yank on them to see what it really takes to compromise it. That should give you a bit of confidence in the technique.

You can be pretty sure the crimp is good if you use the manufacturers tool which usually cost in the $500-$1000 range. The cheap tools "seem" to work fine but are often not the locking ratchet type.

But the only way to be 100.000% sure the crimp was good is to pull it out and see how much force it takes which kind of ruins the joint.

[eneuro]

But the only way to be 100.000% sure the crimp was good is to pull it out and see how much force it takes

I think, you concentrate too much on mechanical strength of such connection, since in many cases, good electrical contact is the main purpose of making those connections 
I see no reasons for brute force mechanical tests of electrical copper wires which are not designed for mechanical loads.

So, to some degree those connections have to be not too loose, but I see no reason stress copper to its limits to see what wins wire or connection-at least all those forces have to be calculated and mechanical forse choosen corectly, not as much force as possible.

In high current welder application for me more important is to do not have power loses higher than in wire itself, so by increasing conductors contact area inside such connector, I see benefit of adding some solder to make perfect fits and remove any air from strand wire.

Watching those few NASA videos, conclusion might be, that we have to always try to clean connection from flux to avoid corrosion, so we want to have clean metals connected (soldered) .
Using additional thermal heatshrinks and antyvibration insulation we can easy limit any solder stres around connected wire.

[mzzj]

I don't see how solder could improve a crimped joint. A properly crimped joint is already gas tight, and often there are even cold welds that develop, like wire wrap.

It improves your confidence that it's not going to pull out.

If you're worried about the wire pulling out, your crimps are no good. If you don't have complete confidence in your crimps, there's no point crimping. You should switch to some sort of solder cup style connector instead because that will be designed for soldering. The whole point of crimping is to make a gas tight, solid connection. If you only manage to make a crappy connection that "pulls out", or whatever, and you feel it needs to be soldered, isn't it better to use a connector that's designed to be soldered?

But that said, it's absolutely trivial to make a proper crimp with a crimper and the correct dies. None of this is particularly expensive, and you would have to be completely brain dead to mess it up. Just make a couple of test crimps and then yank on them to see what it really takes to compromise it. That should give you a bit of confidence in the technique.

You can be pretty sure the crimp is good if you use the manufacturers tool which usually cost in the $500-$1000 range. The cheap tools "seem" to work fine but are often not the locking ratchet type.

But the only way to be 100.000% sure the crimp was good is to pull it out and see how much force it takes which kind of ruins the joint.

And pretty often I found that I have to join slightly different cable sizes. IE 0,75mm2 to 1mm2, 18AWG or 16 AWG. Car electronics seem like worst, gazillion of different wire gauges.

No-name crimp connectors and  assortement boxes are usually not any good, copper is too thin to provide sufficient strenght in the crimp. No-name 60/40 solder works just fine 

Proper crimp should hold more than 50% of the wire breaking strenght, ideally 100% so that the wire should break before cable slips out of the crimp.

[eneuro]

Proper crimp should hold more than 50% of the wire breaking strenght, ideally 100% so that the wire should break before cable slips out of the crimp.

No plans to stress too much my soldered 6mm in diameter strand wire into copper pipe of 8mm in OD and 1.5mm thickness, but when compare this "art" work with what I saw in local market with welding tools and saw ground welder wire crimped with connector in a way, that more than half of strand wire were broken and no chance to weld anything, I'm preaty confident this my connection made in protected with double thermal heatshribk in connector area will be much better. Strand wire insulation was cooled with wet (water) thermal mass close to connector during soldering (only a few seconds enougth to put inside decent amount of solder wire througth 1.5mm hole, than I've put connector from horizontal to vertical position to let solder go down to from connector during cool down).



After soldering cleared this connector with grain alcohol (95%) and after applying thermal heatshrinks it might be not so bad and a way better that this "commercial" investigated during quick teardown in local market with welding tools  It wasn't the cheapiest welder machine, so i was really  what I saw inside this ground connection wire.



BTW: I cost me less than $0.30 to make this connector, since I've used 1.5mm thick 8mm copper pipe and drilled hole for M10 (10mm) screw after smashing half of this pipe in vise and with a little help of hammer   

We'll see during welding how it performs, but I've made connectors for spot welder in similar way and there were no problem what so ever so far 

[mikerj]

But the only way to be 100.000% sure the crimp was good is to pull it out and see how much force it takes which kind of ruins the joint.

Which is the entire purpose of a test crimp.  I used to work for a company that designed and manufactured railway signalling equipment, and a test crimp would be performed every day (per tool) and the result logged.  I don't recall a test ever failing whilst I was there.  Anyone soldering a crimped connection would have had their backsides well and truly kicked.

[eneuro]

Myth buster time comes 
I want to compare today two connection types: crimped and filled with solder to glue everything VS crimped only-both of them covered with thermal heatshrink 
Those connectors will be in series on huge welder transformer secondary  and slowly current will be raised until.... we'll  see strand wire insulation disapear and then lets see which connector fails first or pobably wire itself, I bet   

The solder will melt before the wire fails.

I can't wait to test this, while I do not believe in that 

Crimped vs crimped and soldered 


Batte field: spot welder high current output -no survivors? 


How this happened - raw footage from weak crimped soldered connector wires Myth Buster's scene 


As expected, connectors are fine - failed one of the wires... 


Update: Moment of wire explosion from step by step frames analysis 


Shortly before explosion-it looks like parts of copper flown around at high speed 

[Psi]

I think its more of a perception problem than a physical one.

It's hard to accept that crushing something together is a good way to join it. So we like to fall back to good old soldering.

[DanielS]

It's hard to accept that crushing something together is a good way to join it.

The really big difficulty with crimps is that you need to apply enough pressure to make a solid electrical and mechanical connection but not enough to compromise the cable's mechanical and electrical integrity. Depending on the type of wire and crimp connection, that sweet spot can be very narrow, hence the need for tools specifically designed for them to achieve consistent results.

[CatalinaWOW]

"Presumably"

At that point your argument becomes very much less convincing...

I am not arguing.  It is just an observation.  These are components in new products from a major manufacturer, which while not at the top of quality rankings, is not at the bottom either (and the range from top to bottom in that industry is actually not that broad).  While a presumption is not evidence of any kind, it is usually safe to assume that significant effort was taken to make sure that these things were done properly.   If nothing else the accountants drive it to keep the costs of warranty returns down.

Reading through this thread is almost like watching people discuss religion.  Both crimps and solder have their place.  Both have their strengths and weaknesses.  Both provide excellent service when applied properly.  Both can be botched.

My comment on soldering as a repair for the assembly in question is not an assertion of general superiority of solder.  But I will assert that soldering was the only practical rework process in this case, and suspect that if I did a teardown thread on the subject most would agree.  Fortunately for me, the rework process has proved at least as reliable as the original joint.  Which then leads to what is probably the real issue for the OP and others.  The vendor of many types of equipment will make many of the joints in question.  A proper crimping tool for the application costing $500 to $1000 will be amortized over these joints and is well worth it.  An individual or small shop doing repairs on such a device doesn't have the same economic base, and will look for an alternative that is either already capitalized or which requires no real investment.  Soldering is such an option.  The question then becomes:  Can it be done in such a way as to provide useful service?  Useful may not mean as good as original.  Merely safe, and lasting long enough not to be a bother may be enough.

Useful input from this thread would include care to limit how far up the stranded wire solder is wicked, providing mechanical support for the cable so that flex of the wire near the joint is prevented, and not attempting this type of repair when the application will lead to significant heating of the joint.

[Balaur]

As expected, connectors are fine - failed one of the wires... 

You've just proved that improperly made crimps (i.e. the clamp & wire contraction) failed before the other elements.

[eneuro]

You've just proved that improperly made crimps (i.e. the clamp & wire contraction) failed before the other elements.

Crimped connectors are in the middle, one of the copper wire failed, because of while there were series connection, some part of this circuit had slightly higher resistance than other, and maybe a few single wires lost from whole strand wire was enougth-it failed first, but crimped (and soldered) connectors has no chance to fail first, since inside connector wir eis supported by connector copper and in th ecase of soldered one, we have even lower resistance, so nope, wire must fail first, not connector

In spare time I will make slow motion version of this video, while things go so fast that it looks like nothing happends in the time of wire explosion, but one single frames I've found copper crap flying around 

Wire have no chance to survive with connector in series-because of proper connector (cross section area) with wire inside will always have smaller resistance and wire must fail first-if connector failed it could mean you choosen bad connector 

[Balaur]

You've just proved that improperly made crimps (i.e. the clamp & wire contraction) failed before the other elements.

Crimped connectors are in the middle, one of the copper wire failed, because of while there were series connection, some part of this circuit had slightly higher resistance than other, and maybe a few single wires lost from whole strand wire was enougth-it failed first, but crimped (and soldered) connectors has no chance to fail first, since inside connector wir eis supported by connector copper and in th ecase of soldered one, we have even lower resistance, so nope, wire must fail first, not connector

In spare time I will make slow motion version of this video, while things go so fast that it looks like nothing happends in the time of wire explosion, but one single frames I've found copper crap flying around 

Wire have no chance to survive with connector in series-because of proper connector (cross section area) with wire inside will always have smaller resistance and wire must fail first-if connector failed it could mean you choosen bad connector 

I was half joking:

The wire failed close to the clamp, and I mean by that the ring clamp you used to connect the wire to the spot welder electrodes.
Technically it's a crimp, as it relies on mechanical pressure to create a (cold-weld?) connection.
Since this contraction caused the wire to fail very close to it, creating a particular mode of failure, it's less than ideal.

[Tandy]

What I do when crimping stranded wire is to remove more insulation than needed and tin the end of the wire. Then trim it down so that the solder just acts like a little end cap on the end of the wire. It will help stop moisture getting in between the strands from the ends and also makes pushing the wire into the connector easier as the strands are held together. It can be fiddly if you push the wire in and don't line it up perfect and a couple of strands end up outside the the crimp area.

You don't want solder on the conductor where the crimp goes as the solder would actually act as a sort of lubricant making it easier for the wire to pull out of the crimp joint.

You will find this method used a lot in UK mains plug screw terminals.

[eneuro]

Since this contraction caused the wire to fail very close to it, creating a particular mode of failure, it's less than ideal.

When we analyse where this wire could fail, so it is clear than due to huge thermal capacity of spot welder 15mm in diameter copper pipe, where I've connected those tested wires (pressed from both sides by good conductors, but better from pipe copper side), when this lowest resistance part of circuit started to heat up at high current it was cooled from spot welder connection itself and by rest of the wire to connectors side, so somewhere in the middle (depending on thermal capacity at the edges of the wire) it will fail not at the edge but somewhere between since huge heat gradient concentrates in the middle, not at the ends)

Update: Yep, highest temperature as expected in the middle in one of 30fps frames 



At slower fps speed, even if this is not the greates video quality (old Pentax digital camera) before wire melted its cleary visible hot air around wire.
When we tried pure wire without any connectors in the middle probably it will fail somewhere in the middle due to mentioned higher thermal capacity at the edges connected to spot welder copper pipes.
I will try tomorow brake the same wire without connectors just to see where it fails and make slow motion video instead to see more details when it breaks 

[eneuro]

I've today another experiment using (spot) welder 11Voc and this wire insulation quickly gone into flames and copper wire broken 



I've used my spot welder which has programmed 2 seconds delay to start welding after pressing start/stop button, to be able press hard spot welded pieces or like in this experiment I had 2 seconds to run awa as quickly as possible_i've programmed weld duration for 10 seconds 

Anyway, no special effect this time-it was mounted to welder wires using those crimped and soldered connectors, which as we can see are best surrivors in this disaster-wire insulation quickly changed into fire ball and smoke after broken at high temperature (copper melting point temperature is around 1000*C).

So, another experiment which debunks theory of bad quality of crimped and soldered connectors 

What can I say-NASA rules 

[smile]

I think this would be good addition.

https://hackaday.com/2016/11/03/specifications-you-should-read-the-nasa-workmanship-standards/#comment-4190294

See the guide http://workmanship.nasa.gov/lib/insp/2%20books/frameset.html

The interesting part is they show themselves crimped solder tinned wires.

Here is shot from this manual:



yet state that: "Crimping of ...stranded wire that has been solder-tinned, is prohibited.NASA-STD-8739.4 [4.3.4]"



https://workmanship.nasa.gov/lib/insp/2%20books/links/sections/207%20Butt%20Splices.html

So it means all that marine cable is prohibited from crimping? When infact it's the only right cable type to use for crimping since oxidation is less for such crimps. All cable in washing machines is "solder plated cable", boats too use it. So was NASA really in space with such rules?

[Siwastaja]

By "solder tinned", they must mean wire manually tinned using solder, either in a solder pot or using an iron and solder wire.

This joins the strands together and forms a very brittle structure. Try to crimp that and you'll agree that it's not reliable.

Chemically separately tin coated copper strands is completely different and I don't think that NASA spec refers to that.

[smile]

@Siwastaja "they must mean wire manually tinned using solder, either in a solder pot or using an iron and solder wire."

I wonder what stopped them from saying "manually solder tinned" then?

[smile]

@Siwastaja "This joins the strands together and forms a very brittle structure."

Using 60/40 solder and wrapping a thin copper wire strand around then the wire is solder tinned it, makes a very reliable connection.


It can be used in screw terminal. This is exactly what was done in all electrical outlets in lamps in Europe and in Russia in 80's-90's and it worked OK. AFAIK there were no ferrules back then.

Only now days everyone likes to use these spring action crimper like


But there will be no joint that is "gas free" like everyone point's out with spring loaded tool.

[Siwastaja]

"Solder tinned" is very well understandable IMO. It already contains the "solder" word for extra unambiguity.

Wire you buy with separately chemically tinned strands is not tinned by solder: https://en.wikipedia.org/wiki/Solder

But misunderstanding is always possible and happens no matter how careful you are.

[smile]

"Solder tinned" is very well understandable IMO. It already contains the "solder" word for extra unambiguity.

Wire you buy with separately chemically tinned strands is not tinned by solder: https://en.wikipedia.org/wiki/Solder

But misunderstanding is always possible and happens no matter how careful you are.

Nice info about solder types, so what is used in "chemically tinned strands" if it ain't solder then?
What is used for ferules is it same thing as "chemically tinned strands"?

[Siwastaja]

Using 60/40 solder and wrapping a thin copper wire strand around then the wire is solder tinned it, makes a very reliable connection.

A reliable connection, until you compress it under the screw or crimp; even if the connection between the two resistors remains ok, the connection to the terminal block or the crimped connection will be severely compromised.

It can be used in screw terminal.

No. Never. Ever. Except temporarily, at your own risk .

This is exactly what was done in all electrical outlets in lamps in Europe and in Russia in 80's-90's and it worked OK. AFAIK there were no ferrules back then.

Was done to some limited extend in 1980's yes, and by some DIYers still in the late 1990's, whether it worked "OK" or not depends on what's your definition on "OK". It was outlawed for a very good reason. Does cause fires.

(If you don't have ferrules or crimps available and must put a stranded wire under a screw, then you just do it as-is, trying to be careful to get as many strands under the screw as possible. You do not tin the wire.)

Of course, most likely OK, like most electrical mishaps. If all you do is 10 connections in your entire life, they'll most likely be all ok, and even if one goes bad, it's very unlikely it causes a fire. Just some sparking and a bit of smoke.

It's all about risk management, and the modern electrical code is based on minimizing the risk to very low levels, so that while millions of people die on cancer, and hundreds of thousands in traffic, it's "too much" if one dies in electric shock or electrical fire.

And NASA cares about reliability for obvious reasons.

Solder connection shown in the image is very reliable. When crimped or compressed under a screw, the reliability is orders of magnitude worse.

[Siwastaja]

Meaning of "solder", from Wikipedia:

Solder (/?so?ld?r/,[1] /?s?ld?r/[1] or in North America /?s?d?r/[2]) is a fusible metal alloy used to create a permanent bond between metal workpieces. The word solder comes from the Middle English word soudur, via Old French solduree and soulder, from the Latin solidare, meaning "to make solid".[3] In fact, solder must be melted in order to adhere to and connect the pieces together


So no, factory tinning on strands is not "solder", and has nothing to do with "solder".

But you can always misunderstand. Please just go on.

[smile]

Well but crimping with spring loaded tool like the expensive over 100Eur KNIPEX

does not make gas tight ferrule crimps. Nobody measures wires and crimped result with micrometers against a standard like they should, yet everyone pretends crimp is better then soldered, or crimped then soldered is even worse. Where is only logical that to insert the connector pin into the block you must crimp it first and then solder, not the other way around.

The multitool 97 43 200 A KNIPEX is 130Eur for handle and anywhere between 40Eur and 300Eur for a die set.

Luckily I found that a company in Taiwan makes this similar tool:

KY8161-0010 10pcs Professional Smartcrimper tool kit

http://www.kyptool.com/front/bin/ptdetail.phtml?Part=KY8161-0010&Rcg=5

I wonder who copied who, and if KNIPEX dies would fit it, but at least for most popular set you pay like 90Eur and get 6 dies total where with KNIPEX same set would cost about 500Eur.

The tool is made by few companies

http://www.mactools.com/en-us/Pliers/Wiring-Repair-Pliers/TCT747-KIT/Master-Terminal-Crimper-Tool-Kit-

[wraper]

Cheap ratcheting tools tend to suck. They look OK but crimps they make usually are crap.

I wonder who copied who, and if KNIPEX dies would fit it, but at least for most popular set you pay like 90Eur and get 6 dies total where with KNIPEX same set would cost about 500Eur.

These tools are very different. Except the general shape, I barely see anything common in them.
Crimp made in the end of the video looks like crap. Crimps the same connector I make with knipex multicrimp look much neater. And connector got bent along, it's no longer straight.

The tool is made by few companies

No, it is a cheap tool made by one company and rebadged with different brands.