• Welcome to our site! Electro Tech is an online community (with over 170,000 members) who enjoy talking about and building electronic circuits, projects and gadgets. To participate you need to register. Registration is free. Click here to register now.

All that glitters is not ...

granddad

Active Member
Gold .. Anybody know what the 'gold' metal used on connectors is, okay some board pads and expensive audio may be gold, but PCB connector pin headers, I assume something else.…Rhodium ? To make a short story long .. Having chased many 'stupid' faults during my service engineer career I came across a couple that were due to that connector plating … ( this is one such ) back in the 1980's A savings bank had a batch of new (386) PC's for their front counters , the user requested a longer vga cable for a small 10” colour monitor. 'Sales' had some made and fitted ( an internal 10 pin connector) .. after a few months at one or two sites the customer complained monitor had reverted to monochrome ! Engineer swaps out monitor … Oh new monitor cable too short... so swaps old cable back... good repair all working again ...! No a month or so later monitor is mono again.. also customer is having same problem at more sites (sometimes a pink display )..after several monitor swaps .. customer was a 'bit' annoyed.. I was tasked to find cause.. as rework could not get a PC / Monitor setup to show the fault. Okay. The new long cable had cheap 'tin' tags attached to the monitors 'gold' pins . And with time and some sweaty hands of an bench engineer the two metals reacted and caused some nasty black coating...the new PC's had a new type of Cirrus vga chip that could tell if it was connected to a colour or mono monitor and the correct driver then installed . The higher resistance caused by the connector contamination , caused it to select the wrong one.. I have not learned …. I regularly use 'tin' tags on 'gold' pins .
 

KeepItSimpleStupid

Well-Known Member
Most Helpful Member
I'll make a couple of comments from experience.

A company mad some Berilium copper wafer probes hat turned nasty black at high temperature. They were looked at under an SEM and we tried to EDAX it. No luck.

The company used Sn96 or 60/40 solder to solder the needles. Aside: they changed from welding the ceramic to the spring support to 60/40 solder which created another headache down the road. they would solder with Sn96 for us, but could not go back to the weld process.

So, it was determined to be unremoved flux which the company would not fix. Our fix was to boil 10 minutes in Baking Soda and Electrolys Gold plate. They would not plate anything soldered, they did gold plate other parts of the assembly.

In a High School Explorer's post at Hewlett Packard when they did gold plated boards, you learned of the layer of Nickle that needs to be put down first. So, it's copper, nickel and gold for the PCB.

The same was done for electron-beam evaporated materials where I eventually worked.

You also learn with plumbing and silver soldering the importance of removing the flux. it's easier to remove when it's fresh.

When silver soldering AC refrigerent lines, you have to flow nitrogen in the inside of the tube while soldering. That's an extra step. When I put the AC in my car from a "big box", I used techniques borrowed from UHV (Ultra High Vacuum) No fingerprints anywhere. Lubercae all O-rings. You change gloves for a few reasons. You might have to change the glove because you contaminated it.

Try reading some medical rules with gloves that prevent contamination.

If you really care about purity such as i had to when sealing semiconductor synthesized materials in quartz, you have to etch the quartz in pure HF (Hydrofloric Acid). You don;t want to get near that stuff. A drop on your skin might kill you. The torch used was pure Oxygen and Pure Hydrogen. Again to reduce contaminants.

When working in High Vaccum environments, NO fingerprints are allowed. Then there are wacky thngs called virtual leaks that you have to eliminate when building the system. Screws (Vented screws) have to be used which have a hole down the center so you can pump out the air between the threads.

I've seen plenty of PCB board damaged when the flux wasn't removed.

For soldered pins, I generally used a resistance soldering tool, so the pins would not get solder on it.

I had contamination issues with brand new solenoid valves. brand new valves would buzz if operated off of dry air and 24 VAC. As received, they had to be taken apart, cleaned and re-assembled.

The extra steps were required and I never had issues.

So, for your cables, resistance soldering and flux removal ought to have worked.

For cables that I made at work for use at work, I generally used gold flash pins. These were typically circular connectors with 4, 9, xx and 37 pins.

For electrical connections that used screws, loctite 222.
 

gophert

Well-Known Member
Most Helpful Member
Gold can be plated in extremely thin layers and still achieve corrosion resistance of the underlying copper (this means that even some inexpensive connector pins can be plated in gold at 0.5um). Thin is good enough for connectors that are connected/disconnected only a few times in their lifetime.

You would think thicker gold plating would be a good thing - more is better. In general, more is better, unless it is not.

In the unusual case where gold is crunched onto a tin surface, you can get intermetallic migration that forms a gold/tin alloy at the surface (Au(Sn)4). It is brittle and dark - one type of purple plague. To avoid, make sure thick gold plated pins are not mated to dull tin finish. There is not generally enough gold to cause a brittle gold/tin alloy cleavage issue when thin gold plated pins are used.

Cheers.
 

granddad

Active Member
In my 'tale' I don't think there was any solder / flux involved , just a 'gold' female connector replaced by a 'tin' version . ( Bench engineer in gloves :D ) Still not sure if this M/F pairing is okay ?

gtot.jpg
 

gophert

Well-Known Member
Most Helpful Member
In the unusual case where gold is crunched onto a tin surface
Sorry I was unclear but I did not mean soldered when I said crunched. I meant forceful physical contact causes slow, intermetallic migration - weeks to months time scale.
 

shortbus=

Well-Known Member
Those terminals shown marked as 'tin' look very similar to the ones we made at Delphi Packard Electric, for use in car wiring. And some of them had gold plating on them. But they were only plated where they made conact with the mating terminal. The plating was actually done before all of the stamping and forming of the terminals. Never got to see how the plating was done on just small parts of a strip of metal that was much wider. The process was done in a locked room with restricted access do to the gold.
 

Latest threads

EE World Online Articles

Loading

 
Top