s197.co.uk

Soldered or crimped connectors

Be as specific as possible, include as much information as possible and be patient for advice - we're all new to the S197 but there's a lot of knowledge out there.

by Montana » Fri Mar 07, 2008 11:24 pm

I keep seeing this come up on this board, especially concerning SVA work.

Within the professional motorsport industry and the OEM car loom manufacturing, no-one joins wires with solder.

A poor quality pre-insulated crimp (the likes of which come from halfords) is indeed a bad thing, and soldering would be preferable to those.

However, the fact remains, two metals of similar hardness will weld/join if enough force is applied. For example, you can produce an actual friction weld of two metals with a aluminium spade connector and copper wire with a proper crimp tool.

Because of this, and the fact it then only produces one stress point is the reason critical joints are done this way.

With soldering, you have to strip wire, removing insulation, then join using a third party compound and then re-insulate. Not only this, but solder can oxidise and corrode very easily leading to cracking. Solder is also solid, producing at least two points of stress on any given joint.

As such, given reasonable equipment (decent crimps and tools are expensive) a crimp is ALWAYS better than soldering.

Obviously in regard to components, such as relays and diodes, they have to be soldered, but the pressure is taken away as they are generally mounted.

Now obviously you dont have to take my word for it, I may be lying ;) but recently PPC magazine did an article, so here are the relevant parts:

Image
Image
Image

Bear this in mind for any SVA work.................
Montana
 
Posts: 181
Joined: Tue Feb 05, 2008 9:38 am

by LV51FER » Sat Mar 08, 2008 9:05 am

Point noted but I have to say that getting a good crimped joint is harder for the amateur who hasn't always got access to all this posh stuff and for my purposes, the minimal solder application will always make me feel safer. Besides, I've got a soldering iron and I don't have a crimping tool. I have seen the crimped joints on production cars and know I could never get a join like that in a million years.
User avatar
LV51FER
 
Posts: 6378
Joined: Tue Aug 28, 2007 4:46 pm
Location: Caerphilly Medieval Settlement
Name: Wayne
Year Of Mustang: 2005

by Montana » Sat Mar 08, 2008 2:45 pm

A crimping tool that will do crimps like that is about 30GBP, a proper crimping tool is 70GBP, so its not a bad investment...
Montana
 
Posts: 181
Joined: Tue Feb 05, 2008 9:38 am

by Robbed666 » Sat Mar 08, 2008 3:01 pm

Montana wrote:I keep seeing this come up on this board, especially concerning SVA work.

Within the professional motorsport industry and the OEM car loom manufacturing, no-one joins wires with solder.


Because it is time consuming, costly, and requires higher level skill than crimping. It also and has health and enviromental 'issues'.

A poor quality pre-insulated crimp (the likes of which come from halfords) is indeed a bad thing, and soldering would be preferable to those.

Winding the bare ends together and wrapping in Cellotape would be preferable!

However, the fact remains, two metals of similar hardness will weld/join if enough force is applied. For example, you can produce an actual friction weld of two metals with a aluminium spade connector and copper wire with a proper crimp tool.

No crimptool will apply enough pressure to fusion weld dissimilar materials. I have never seen an electrical crimptool that has achived a friction/fusion weld.

We weld aluminum to stainless steel for oil well head heat exchangers. It requires a massive explosive force to bond the two. Yes I'm talking a real detonation!
The process was found by accident in the Vietnam war. Fused aluminium and steels at the point of weapon impact on tanks aircraft etc.

Aluminium has an extremely tough oxide layer (which reforms very quickly even after mechanical removal). So much so, it is used as as abrasive, Aluminium oxide grit etc.
When actually welding aluminmum, as in a TIG process. The AC waveform positive half cycles are used to strip the oxide layer off, to allow molecular bonding/welding to take place. and requires a level of energy that is an order of magnitude that it would be impossible to achive with a hand crimptool.

Apart from the fact, I know of no method without using an 'alloy' (in between) to join copper to aluminium.

The commercially available spade/crimp connectors are tinned copper/copper alloy. Not aluminium.

The Americans use aluminium conductors/connectors often with dire results! ;)




Because of this, and the fact it then only produces one stress point is the reason critical joints are done this way

With soldering, you have to strip wire, removing insulation, then join using a third party compound and then re-insulate. Not only this, but solder can oxidise and corrode very easily leading to cracking. Solder is also solid, producing at least two points of stress on any given joint.



A crimp will certainly put the conductor strands under 'pressure'.

Solder will oxidise on its surface. Where it is in contact with the conductor strands it has actually excludes the air/oxygen and PREVENTS corrosion !

Lead is, has been, and still remain an material with an extremely high impermeability, so much so it blocks harmful radition, let alone oxygen/water ;) Yes modern solders arn't lead, but the synthetic equivlent is almost as good. ;)

Even a soldered lapped joint has a far better positive contact area than a crimp. The major problem with soldered joints failing through flex fatigue, is caused by 'wicking'. The main cause of this, is poor soldering technique. Crimps will equally, if not more so, at the crimp pinch point, if mechanical support is not given to the wire.





As such, given reasonable equipment (decent crimps and tools are expensive) a crimp is ALWAYS better than soldering.


'Horses for courses' A blanket statement, as above, is not true.



Obviously in regard to components, such as relays and diodes, they have to be soldered, but the pressure is taken away as they are generally mounted.


On a PCB ? Or by mounting lugs?

On a PCB, relays are normally only secured by their soldered joints, and the conformal coating may help stick them down (a bit). As for diodes, transistors, large electrolytic capacitors etc are sercured normally by their leads.
Strange when I've seen components 'shocked' from PCB etc it very unlikely at the soldered joint, but at the lead to body connection.


Now obviously you dont have to take my word for it, I may be lying ;) but recently PPC magazine did an article, so here are the relevant parts:

Image
Image
Image

Bear this in mind for any SVA work.................


I bet those '80% failures' are the result of badly made 'mechanical electrical connections' i.e. terminal blocks, scotchlocks, and crimps!
But NOT Soldered joints! ;)

The chances of either failing in your Mustang, if done reasonably well is highly unlikely

My knowledge is from 'field' obtained experience, and almost 30 year as a trained engineer.
(Senior QA test, Senior design engineer Marconi (GEC) , then technical Director of a bespoke industrial equipment supplier.)

Here the current project we're involved in:
Image


P.s We normally crimp connections for mechanical placement, then solder to to prevent oxidization and lower connection resistance, then heat shrinkable adhesive lined connector shells.

I 'll dig out some commercially made (not be us) failed crimps and post the pics. This from an environment which isn't all that harsh :mrgreen:
User avatar
Robbed666
 
Posts: 1567
Joined: Tue Aug 28, 2007 4:54 pm
Year Of Mustang: 2005

by Montana » Sat Mar 08, 2008 3:39 pm

Wow, you have a LOT of knowledge!

My point of view is purely from an automotive point of view, hence it should be relevant to work 'we' carry out. Plus its what I've been told from more senior engineers at work, so it may be incorrect but all knowledge is gained from a more authoratative source, so the assumption is for motorsport, thats how it works.

However, your input is far more valuable as you have 'harsh conditions' knowledge.

Ta for that!

P.S. I may now PM you for crimping/joining type advice for a project i have coming up ;)
Montana
 
Posts: 181
Joined: Tue Feb 05, 2008 9:38 am

by Robbed666 » Sat Mar 08, 2008 3:54 pm

Don't get me wrong, I'm a great fan of crimps.

In a manufacturing environment, there's less chance of making a bad connection i.e repeatability.
The process is far more environmentally friendly. When you've got a bulkhead connector with a 100+ connections. The last thing you want to see is solder buckets! With all that soldering, sleeving etc.
Crimp, then insert! wins every time ;)

I hate any type of 'free' inline join, but sometimes you've no choice :roll:
User avatar
Robbed666
 
Posts: 1567
Joined: Tue Aug 28, 2007 4:54 pm
Year Of Mustang: 2005

by hertfordtyres » Mon Mar 10, 2008 2:18 pm

soldered !

= less wire risistance (ohms)

= less chance of the sjb going wrong

Dave
User avatar
hertfordtyres
 
Posts: 3989
Joined: Thu Mar 06, 2008 7:45 am
Location: Hertford, sawbridgeworth, Hertfordshire
Name: Gill + Dave
Year Of Mustang: 2006

by Robbed666 » Mon Mar 10, 2008 6:36 pm

Rho ( P, resistivity) for Solder is 16.5 X ((10^(-8), for copper it's 1.68 X ((10^(-8).

Therefore solder (60/40 'eutectic type') has a resistance, '10 times' that of copper per unit lenght/cross sectional area :ugeek: ;)
User avatar
Robbed666
 
Posts: 1567
Joined: Tue Aug 28, 2007 4:54 pm
Year Of Mustang: 2005

by Montana » Mon Mar 10, 2008 7:09 pm

What are crimps made from?

Aluminium, how does that compare? I would have thought modern solder would always have more resistance than any crimp due to mass/density as well as flux being non conductive. Not sure though...
Montana
 
Posts: 181
Joined: Tue Feb 05, 2008 9:38 am

by Robbed666 » Mon Mar 10, 2008 8:23 pm

Crimps are 'typically' tinned copper. Sample: http://www.maplin.co.uk/module.aspx?ModuleNo=42711&doy=23m7.

'Quality' crimped pins and sockets in connectors are sometimes gold plated copper alloy, or silver plated, over nickle, over a copper alloy.

Aluminium’s resistivity varies considerably, As there are a plehora of Aluminium types. Though typically from 1.6 to 3.5 times that of copper. The best electrical conductor is Silver (NOT Gold, as sometimes quoted!)


The flux is normally burnt off during soldering, it's there to improve 'wetting' and bonding by stripping oxides.
Any left then migrates to the surface and outside edges of the joint.

Solder has a higher resistance than copper, but as most joints have some sort overlapping/interwinding of the copper conductors (when splicing wires), and therefore cause an increase in crossectional area at the joint .
Unless very loose, there should not be a resistance increase in the area of the joint, when compared to the rest of the conductor within the cable as a whole.

Haven't done this for a long time, but tomorrow I'll setup a couple of little experiments. :mrgreen:

A typical solded lapped wire splice and a commercial crimped joint, say in 2.0mm^2 cable.
I'll do a PD drop test and a temperature rise test at a say a current of 10 amps DC. ;)
User avatar
Robbed666
 
Posts: 1567
Joined: Tue Aug 28, 2007 4:54 pm
Year Of Mustang: 2005

by LV51FER » Tue Mar 11, 2008 8:23 am

Rob, I've tried similar experiments to this. Switch, bulb, does it work, yes, experiment over. :lol:

Isn't that really what counts?
User avatar
LV51FER
 
Posts: 6378
Joined: Tue Aug 28, 2007 4:46 pm
Location: Caerphilly Medieval Settlement
Name: Wayne
Year Of Mustang: 2005

by Robbed666 » Tue Mar 11, 2008 2:38 pm

Rob, I've tried similar experiments to this. Switch, bulb, does it work, yes, experiment over.

Isn't that really what counts?


.....and I'm sure that's the same attitude, and depth of knowledge/investigation that a 'dodgy' SVA conversion company applied to the s197 Mustang!!!

"Worked when it left here pal ! It's only a bulb and a bit wire after all!" ;)





From someone that spends his time arguing that 'one shade of gray', is lighter, or darker than another. I don't know :roll: :roll: :mrgreen: :mrgreen:
User avatar
Robbed666
 
Posts: 1567
Joined: Tue Aug 28, 2007 4:54 pm
Year Of Mustang: 2005

by LV51FER » Tue Mar 11, 2008 3:45 pm

Damn your logic, damn your lies as Stevie Nicks once almost sang. :D
User avatar
LV51FER
 
Posts: 6378
Joined: Tue Aug 28, 2007 4:46 pm
Location: Caerphilly Medieval Settlement
Name: Wayne
Year Of Mustang: 2005

by hertfordtyres » Wed Mar 12, 2008 6:40 am

hi robbed
have you done your experiments yet ??

im really interested to find out the results

dave
User avatar
hertfordtyres
 
Posts: 3989
Joined: Thu Mar 06, 2008 7:45 am
Location: Hertford, sawbridgeworth, Hertfordshire
Name: Gill + Dave
Year Of Mustang: 2006

by Robbed666 » Wed Mar 12, 2008 7:43 am

Not Yet Dave,

As on most days, What's planned, goes out the window within 5 minutes of getting into work! :mrgreen:

Hopefully today!
User avatar
Robbed666
 
Posts: 1567
Joined: Tue Aug 28, 2007 4:54 pm
Year Of Mustang: 2005

Next

Return to The Garage - Engine/Bodywork/Electrics/Audio questions