Plating on connectors

[tkamiya]

I think this topic best fits in this category

I've been shopping lots (and lots) of connectors and adapters.  Then a question arose..... plating.

In 80s and 90s, silver plated connectors were pretty common, especially higher ends.  Now-a-days, they are mostly not available.

Now, we ave stainless steel (not plated), brass + gold plated, brass + zinc plated, and few other exotics.  I can see why stainless steel is popular.  Strong and does not tarnish.  Gold plated one does not tarnish and amount of gold used does not adversely affect pricing.  Zinc is just inexpensive.

But performance wise, and use-ability wise, how do they differ?  Only thing I know is stainless steel tend to gal sometimes. 

[RandallMcRee]

In metrology, thermocouple effects are a primary concern.

Keithley talks about this in the Low Level Measurement Handbook (you can google it) Added link.
https://download.tek.com/document/LowLevelHandbook_7Ed.pdf

Generally, copper-copper contacts are preferable, but cleaning may be necessary before use. For that reason, gold is usually the best compromise. It depends on your goals and error tolerances.

From the handbook, sect 3.2.2:
TABLE 3-1: Seebeck Coefficients
Paired Materials* Seebeck Coefficient, QAB
Cu - Cu ≤0.2 μV/°C
Cu - Ag 0.3 μV/°C
Cu - Au 0.3 μV/°C
Cu - Pb/Sn 1–3 μV/°C
Cu - Si 400 μV/°C
Cu - Kovar ~40–75 μV/°C
Cu - CuO ~1000 μV/°C
* Ag = silver Au = gold Cu = copper CuO = copper oxide
Pb = lead Si = silicon Sn = tin

[martinr33]

Those stainless steel connectors are most likely chrome-plated brass.
The next step up is some measure of gold on nickel on brass.

on eBay, "brass" translates to "copper", so don't expect copper. Good for most  things, but not ultra precision voltage work.

Of all the typical banana suppliers on eBay, only Viborg products are plated copper. They use gold over silver over copper, which is nice.

There are a few suppliers of plated copper connectors designed for low thermal offset work. Plating is a relativley minor issue compared to having a brass body.

Or, you can go hardcore and use bare clean copper. Coax cable is a good choice, because it brings shieldiing with it.

[tkamiya]

I am most suspicious of eBay GOLD plating.  I wonder what they actually are....?

[martinr33]

Ebay connector gold plating is usually that. Just enough to look gold and hold off corrosion. It is good enough - the metrology issue comes from the brass underneath.

I’ve not run in to pot metal or zinc banana plugs, but it would not surprise me to find some.

[tkamiya]

So the seller wasn't lying when he said "REAL GOLD!"   

[mansaxel]

From a durability standpoint (not really metrology then) silver is preferable to gold as plating, if mating cycles are frequent. It might very well be so that an often mated gold connector is worse than a silver one.  When I built audio systems we had gold plated XLR inside cabinets, where things were installed and sat still, but connection panels etc got silver plated connectors.

[eplpwr]

Another view on plating: how about doing the plating yourself?

Conrad used to have small plating kits (they still do but these are missing from their .se site). Maybe it would be possible to replace crimping/soldering with plating [Cu] connector and wire together, and finish it with a gold/silver plate?

[GeoffreyF]

There are metals more conductive than Gold actually.  However, there are none which are as immune to oxidation than gold.  *THAT* is the value of the plating.  The gold only needs to be there and strong enough not to be scratched off.  The cost of the gold is literally minuscule. Steel is also fairly resistant to oxidation. Copper and Silver though better conductors oxidize, get tarnished and thus don't contact as well.

[IconicPCB]

Copper can be electropolished o a mirror finish. We use electropolishing of laminate in between hole wall activation and electrolytic plating of holes.

We also use immersion silver plating.

I wonder how a polished copper lug would behave with silver plating?

[tkamiya]

There are metals more conductive than Gold actually.  However, there are none which are as immune to oxidation than gold.  *THAT* is the value of the plating.  The gold only needs to be there and strong enough not to be scratched off.  The cost of the gold is literally minuscule. Steel is also fairly resistant to oxidation. Copper and Silver though better conductors oxidize, get tarnished and thus don't contact as well.

I have some "gold plated" RCA connectors that "gold" is not even providing 100% coverage.  It's so thin that it looks like patch work.  For my own use, I'm not too concerned about thermo-couple effect.  (although I will keep this in mind)  I guess any time two dis-similar metals meet, there is TC effect, so in some cases, one connection has FOUR sources in one connection?  Basemetal vs plating vs plating vs basemetal.  Argh!

As far as I know, Brass is 1/3 Zinc and rest is copper.  Is it so bad to have Brass?

[GregDunn]

Red Brass is more like 80% copper (often used in plumbing drains) and Cartridge Brass (plumbing fixtures) may be 75% copper.  So, it depends on where you acquire the alloy.  I worked in a brass foundry during school and we had to be careful not to mix the scrap from various areas of the manufacturing floor, so that the people running the furnaces didn't have to worry much about the composition of the pour.

[helius]

I have some "gold plated" RCA connectors that "gold" is not even providing 100% coverage.  It's so thin that it looks like patch work.

Less than a microinch gold is considered "gold flash", which is mainly for looks. It does have some benefit against corrosion during shelf storage, very similar to ENIG plating on circuit boards. It is not intended to survive mating; for that, you need 15 to 30 microinches of hard gold (gold with nickel or cobalt impurities).

I guess any time two dis-similar metals meet, there is TC effect, so in some cases, one connection has FOUR sources in one connection?  Basemetal vs plating vs plating vs basemetal.  Argh!

This is a common misconception. The Seebeck effect is an electric field caused by differences in temperature: it does not require dissimilar metal contact. In the common scenario of thermocouple wires, the different electric fields caused by the temperature change in two different materials lead to a measurable voltage difference. This is true whether or not the wires are in contact.

As far as I know, Brass is 1/3 Zinc and rest is copper.  Is it so bad to have Brass?

Brass is used as a hard, light weight, anti-magnetic material for connector shells. Contacts are more commonly copper or bronze.

[radiolistener]

These "gold plated" SMA connectors from aliexpress is not gold, this is just some yellow metal and it's resistance is not good. The worse thing is that this "gold" peeled off and contaminate other connectors. These "nickel-plated" SMA from aliexpress much better 

I don't have golden plated RF connectors, so I cannot say if it's good.
But I have golden plated probes for multimeter and it works great.

Also I have many silver plated RF connector and they are very good. Yes, they are oxidized when you store it for a long time, but it is good even when oxidized. I'm clean it with alcohol and rubber. They are the best and I like it.

[SilverSolder]

I guess any time two dis-similar metals meet, there is TC effect, so in some cases, one connection has FOUR sources in one connection?  Basemetal vs plating vs plating vs basemetal.  Argh!

This is a common misconception. The Seebeck effect is an electric field caused by differences in temperature: it does not require dissimilar metal contact.

Why does it matter, then, what the plating on the connectors is? (from the Thermal EMF perspective).  -  As long as the wires that go from the meter to the DUT are made of the same material, would the Seebeck voltages not just cancel out, irrespective of the plating material?

[helius]

Why does it matter, then, what the plating on the connectors is? (from the Thermal EMF perspective).  -  As long as the wires that go from the meter to the DUT are made of the same material, would the Seebeck voltages not just cancel out, irrespective of the plating material?

To the millivolt level, absolutely. But what if you're measuring nanovolts?
When you plug the connectors together, current doesn't move through them uniformly. There's a contact patch where the surface oxides are scraped off and current moves around those in an irregular way to get to the underlying cable. Where there's current there's heat; where there's heat there's convection (air currents); and the temperature of the contact area will change over time. Minimizing these effects requires low thermoelectric power materials for plating.

[Wendy_Preston]

I have some "gold plated" RCA connectors that "gold" is not even providing 100% coverage.  It's so thin that it looks like patch work.

Less than a microinch gold is considered "gold flash", which is mainly for looks. It does have some benefit against corrosion during shelf storage, very similar to ENIG plating on circuit boards. It is not intended to survive mating; for that, you need 15 to 30 microinches of hard gold (gold with nickel or cobalt impurities).

Not trying to be argumentative here at all, but just thought I'd give you some insight on what we supply on our connector range, for comparison.
- Our "Industry standard" products are either hard acid gold flash (around that 1 micro-inch level) over nickel, or tin (40-80 microinches) over nickel. Even though the gold is pretty thin, it is still a superior coating to the tin for wear (look up tin fretting) - such that gold flash connectors are often rated for as many as 300 operations, whilst tin only for 50.
 - Our "High-Reliability" products are generally hard acid gold plated (again over nickel) with about 10 to 12 microinches - more than enough for 500 operations on one range, 1000 operations on another range.
- The use of 15 to 30 microinches gold is OK for the contact area (but can get expensive with little improvement in performance), but definitely not recommended on the solder tails - the effect of gold embrittlement kicks in, where gold molecules mix with and weaken the solder joint over time.

All the contacts are copper alloys - brass, phosphor bronze or beryllium copper - giving different levels of spring force and long-term performance, and will be used as appropriate depending on the rest of the specifications.

Generally our connectors are used for in-equipment connections, and to be fair many seem to end up in the fit-and-forget types of applications. But the thinner platings with gold are still perfectly good for function. However, the standard of the manufacturer is key - to ensure that you have connectors with good plating adhesion that will not flake off, as other members have encountered.

[GeoffreyF]

The main difference is oxidizing /  tarnishing, as you intimate.   At the end of the day, how resistive is the connection and is that an issue for whatever signal is transferring through the connection?   In many cases, it doesn't matter.  Not all environments are as challenging to avoid oxidation or tarnishing - that's another factor.

In my experience, people often are overly obsessing and overly spending out of concern for the quality of cables and connectors.    That said, if you feel better with something better, that's worth something if your goal is to feel better.