Welding copper wire to connectors

[splin]

To reduce thermal EMFs swaged/crimped connections are recommended but oxidation can make the connection much worse than a soldered joint. So can copper wire be welded to copper connectors? Would it create worse problems than it solved?

[sarepairman2]

keep in mind what happens to copper when you weld it.

It is a possibility of course, as is ultrasonic welding, if you weigh your concerns

a simple solution is to use a low thermal EMF solder (old cadmium based ones, I believe there is a new one out now)

[Galaxyrise]

A good crimp becomes a weld

sources:
https://www.eevblog.com/forum/projects/crimp-*and*-solder-a-crimp-connector-good-bad/
https://www.eevblog.com/forum/beginners/earthing-why-crimp-and-not-solder/

[Monkeh]

To reduce thermal EMFs swaged/crimped connections are recommended but oxidation can make the connection much worse than a soldered joint.

If oxidation can occur the crimp was improperly done.

[Richard Crowley]

To reduce thermal EMFs

What are you doing that "thermal EMFs" is significant?  You know that connecting to things like Thermocouple exotic alloys takes special handling to account for the cold-junction end.

swaged/crimped connections are recommended but oxidation can make the connection much worse than a soldered joint.

Please cite a reputable source for that statement. It does not agree with decades of high-reliability (including military and aerospace) standards.

So can copper wire be welded to copper connectors? Would it create worse problems than it solved?

I've seen spot-welded connections in certain applications (like mega-mass produced cheap consumer goods, etc.)
But conventional soldering or crimping is so much easier, and there is no particular advantage to welding.

We presume that you are using the word "welding" in the common western usage of melting solder.
Since "welding" and "soldering" are apparently interchangeable words in some eastern languages.

[rs20]

We presume that you are using the word "welding" in the common western usage of melting solder.

?? Did you mean "the common western usage of melting the base work material", as opposed to soldering, which is melting solder and leaving the base work material unmelted?

[calzap]

Exothermic welding of copper wire and connectors is fairly common for larger sizes.  A common brand of welding kits is Cadweld (www.erico.com), but there are others.

Mike in California

[acbern]

I wonder if anybody has ever had real EMF problems with a well done standard solder joint. I could never measure differences (with a 3458A) if things were properly done using standard solder compared to crimping. While crimping is good, it is not always possible, and in such cases soldering does a good job. Other effects (airflow, insufficient time for settling, dissimilar metals...) do have effects, but that is a question of proper measurement practice etc. to eliminate these.
So these discussions to me are really of theoretical nature, at least I havent seen differences. If anybody could come up with a practical example and measured values showing different results for different setups it would certainly be of interest.

[Vgkid]

It is funny that you mention this. In keithleys older nanovoltmeters they reccomend cadmium solder for joining wires to the  connector pins. Yet now they reccomend 90/10 solder.

[mikerj]

A good crimp becomes a weld

Exactly, it's a cold weld and done properly it will be a completely reliable connection.

[calexanian]

+1 for a good crimp.

[szhighstar]

We make transformer which current is big, we often use copper foil and heavy lead out wire, first we solder lead out wire, strip insulation lacquer first, keep heavy diameter wire flat, solder them using big power iron,enough solder, control soldering time and flux, can not fake soldering,soldering point is with enough tin and brightness,  pull force testing is necessary , in general, it is process technology question, thanks!

[RobertBG]

To reduce thermal EMFs swaged/crimped connections are recommended but oxidation can make the connection much worse than a soldered joint. So can copper wire be welded to copper connectors? Would it create worse problems than it solved?

Dont overthink this,just use a proper crimp and then coat in liquid electrical tape or a similar coating.Another option is dielectric grease (can be thinned with spirits and dipped),insulating varnish or self sealing heatshrink.With a proper crimp you only need to worry about surface corrosion. How you go about preventing surface corrosion depends on what you're working on and the environment it'll be subjected too but one of these should keep it simple and at bay.

[splin]

keep in mind what happens to copper when you weld it.

What does happen to copper when you weld it - I've no idea what you're suggesting?

a simple solution is to use a low thermal EMF solder (old cadmium based ones, I believe there is a new one out now)

Simple but dangerous. I don't know of any low thermal alternatives to cadmium although some silver based solders may be a bit better than tin/lead.

[RobertBG]

keep in mind what happens to copper when you weld it.

What does happen to copper when you weld it - I've no idea what you're suggesting?

a simple solution is to use a low thermal EMF solder (old cadmium based ones, I believe there is a new one out now)

You can get a lot of contamination when welding it and that would most likely defeat the purpose of welding in the first place.

You can weld it but it requires a lot of power and the right gas to shield it.Having said that you're probably going to burn the wire up long before you actually get enough heat going on the terminal     He makes it look easy but I've only ever had luck using a purging chamber,otherwise it gets a massive amount of oxidation.

[Vgkid]

Did some reading in relation to my previous post. Keithley reccomends using pn# 2182-325A low thermal solder.
http://forum.keithley.com/phpBB3/viewtopic.php?t=130
Going back to the Keithley catalog, under the connectors section(pg368of 392) mentions that there is sn/Ag4 solder used in the low thermal shorting contacts. Furthermore between the 2 sources that care needs to be taken when soldering to the connectors.
http://www.testequipmentdepot.com/keithley/pdfs/keithley-2013-catalog.pdf
Looking that up leads me th Sn96/Ag4 solder(pricey stuff, maybe)

[plesa]

I know that it is more or less academic discussion. But I prefer soldering. Mainly for home use, where can be expensive to have good tooling for crimp.

[splin]

Thanks for all the responses. My interest is in building some nano-volt level instrumentation including nano-voltmeters and low thermal scanners. There are a lot of connections in such equipment and a few nano-volts at each point adds up. Hopefully with good thermal design many will cancel each other out - but that is a problem. It's not easy to model thermal flows accurately (for me at least) and very hard to measure thermal gradients of a few mK. Just 1mK temperature differential across a copper-solder-copper joint with thermal EMFs of 5uV/K is 5nV - not negligible. Of course many thermal EMFs will appear as offsets which can be nulled out but not always.

I'm not sure its easy to measure thermal EMFs at this level either - hand held probes will initially be at different temperatures to the test-points and thus create their own thermal EMFs which could take some time to equalize to mKs. The thermal environment will also be altered as providing access will change airflows compared to the PCB in its enclosure. Soldering on test leads on the other hand will create new solder joints with their attendant EMFs.

Low thermal solder using cadmium is not a route I'd want to take and I'm not sure that tin-silver is anywhere near as good, but  I'd be interested if anyone has any thermal EMF numbers for the latter. So using direct copper-copper connections, welded or crimped seems to be worth exploring.

I should have made it clear that I do understand that good crimped connections cold weld copper to copper so are almost ideal and reliable. The problem though is making quality crimped joints - my understanding is that they need to be done carefully using the right tool, the right pressure and the connector and wire size have to be closely matched. That's ok if you are doing a large batch of connections and can afford the right crimping tool and can get suitable copper connectors.

I don't have any crimping experience and am not sure I could produce reliable air-tight connections. And connections wouldn't be restricted to standard wire to board/terminal connections. For example, rather than soldering low-thermal SMD relays to a PCB, an alternative is to mount them dead bug style and interconnect them with wires crimped or welded to the (flat) lead-out pins. Crimping would require a small piece of copper of the right thickness to wrap around the relay pin and the wire. Maybe not too difficult to improvise but welding may be more reliable and easy to do using a small home-brew spot-welder.

Hence my question, which was more about the feasibility of welded connections and any deleterious consequences due to differences in the electrical properties in the interface between the bulk copper wire/component lead and the melted copper weld zone.

Splin

PS. I wonder if it's just google translate that equates soldering and welding in Chinese - English translation? I find it hard to believe the Chinese would use the same word for both.

[lm3baker]

I wonder if anybody has ever had real EMF problems with a well done standard solder joint.

Yes. On a current shunt, it created a DC offset that varied with thermal conditions. This was a problem when controlling DC injection (10's of mA on the amp) in an inverter. The solution was mostly a better layout, removing the temperature difference.

[targit]

Just stumbled across this topic when looking for something else, I can tell you that copper wires are often "spot welded" together in car wiring harnesses, usually to join multiple conductors. I still find this from time to time in newer cars but I think it used to be more common (or maybe I just used to unpick more wiring harnesses).