low power/current contacts

[zappedagain]

When is gold required? When is tin okay? Signal contacts? Power contacts?

I've had bad luck with tin contacts degrading under a continuous current of about two amps (see attached photo); that was fixed by changing to gold contacts. That surprised me.

I've seen info on switches that 10mA seems to be the threshold for reliable cleaning. Below that you need a connector specifically designed for low current operations.

I've read a lot on this site about silver vs. tin for high power/current applications. I'm surprised that the low power topic is so hard to find info on, especially with things moving to lower and lower power consumption these days.

Direct answers or references are welcome!

Z

 

[DRWeig]

Gee, zapped --

I've never gone to gold until I get down below 24V, 100 mA. I would think that 2A should give plenty of arc to do some cleaning unless there are odd contaminants in the air.

Idec had some good info but I can't find it right now. They recommended 50 mA 5V as a lower limit in their papers.

I pay attention to contact type in signal-level stuff, but to be honest I have never even looked at power stuff even though I design at a commercial / light industrial level.

Good on ya,

Goober Dave

Haven't see the forum policies? Do so now: Forum Policies

 

[IRstuff]

I think that your perspective is slightly incorrect. What generates damage in many systems is not the absolute current, but the current density and the compatibility of the contacting materials. Thus, for what appears to be an absolute low current may have relatively high current density. In your photo, thos types of pins are often only contacted by two shaped contactors that may have extremely little contact area, hence relatively large current densities.

Additionally, certain material combinations are recipes for failure. I've had miserable performance from 300VDC halogen lamps that are only running about 1/2 A, which is even lower than your example, with some sort of corrosion or electrolytic action.

TTFN
faq731-376

 

[zappedagain]

A bit more info... I'm dealing with low power signal connections (3.3V typ, 27mA for the digital signals or much lower currents for the analog signals), and all my power supplies are at or below 24V; my highest current draw is 5V/8A.

So when is tin a bad idea? I'm stuck using an Amp/Tyco/TE 5-292178-1 drawer connector that has gold plating over nickel on the main contact area that is designed for multiple connect/disconnect sequences. The back side of the connector is a 2mm header with a tin coating; this is the part that concerns me for low current signals. TE offers tin or gold plated contacts for the connector that mates with the header. Why would I put a gold contact on a tin post? When I want to fix less than half the problem?

The spec references 50mV and 10mA for some test cases; I asked TE to translate the spec from Japanese to English.

Does anyone know of any other drawer-connectors that are good for signal applications?

IR - I hadn't considered current density. Thanks for that tip. I know there are three sided contacts available for this type of situation.

Z

 

[IRstuff]

Is there anything usual about the environment? High temps or humidity can cause oddball problems. Tin, of course, is famous for growing dendrites, particularly when under stress, ala, pin contactors. There might also be some goofy eutectics that might be problematic, since gold is so good at alloying with a number of other metals, and semiconductors.

Gold and aluminum, for example, used to be famous for the "purple plague," which occurred with gold bonding wires on aluminum bond pads on ICs. I've personally never seen it under a scope, so can't really attest to whether it's actually purple. In any case, the eutectic does bad things to the contact resistance.

TTFN
faq731-376

 

[IRstuff]

Hmm, this guy seems to argue that there shouldn't be a problem at all:
thread240-316358

TTFN
faq731-376

 

[zappedagain]

Temps are industrial, 0-50C ambient, so I'm planning for 0-60C operating.

 

[DRWeig]

Ah, sorry for not bothering to look at the picture... I thought you were discussing relay or switch contacts that rely on some arcing to keep them clean. I haven't run into trouble on pin-and-socket and the like, tin plated brass.

Only exception in memory with signal-level stuff is when they get plugged and unplugged many, many times. I suppose they lose their contact pressure then, or perhaps a contaminant gets into the mix when unplugged.







Good on ya,

Goober Dave

Haven't see the forum policies? Do so now: Forum Policies

 

[itsmoked]

You need gold below 10V. Otherwise contact corrosion provides enough insulation to prevent conduction.

Keith Cress
kcress -

http://www.flaminsystems.com

 

[zappedagain]

Will tin/tin or tin/gold also corrode? (I'm assuming that's a yes)

 

[itsmoked]

Yep and the oxide tin forms takes more than 10V to reliably punch thru.

Keith Cress
kcress -

http://www.flaminsystems.com

 

[zappedagain]

So here is the feedback from Amp/Tyco/TE regarding the tin finish on their 5-292178-1 connector:

==================================================================
In case of low voltage, engineer want to choose gold plating. But Product engineer consider product performance vs cost then designed our connector.

The current tin plating, designer consider the spring force over 100g and this mating force can removed oxidated material during connector mating. So, mating force become high to compare with Gold plating product because contact mating force are different.
Design concept of Gold plating, maybe 30-50g/contact.. Tin plating over 100g/contact.

Voltage is OK level because we evaluate this connector by Low level Voltage and current as Dry circuit condition.. You can see our measurement method if you check 108 spec attached.

Rather than, if your customer need the severe environmental application, we need to release gold plating product instead of Tin..

Our customer use this connector to business machine like Copy machine, LBP, and those product life is more than 5-10 years. Also, those customer are choose tin plating product thus we can propose to use this product.
=================================================================

SO I summarize that as saying high pressure, tin contacts are okay for long term (5-10 years) indoor environments with low voltage/current signals. But gold is preferred.


I think I need to dig up one of the old references on electrical reliability. This really seems like it shouldn't be a new topic, except for the lower and lower voltage ranges that electronics run at these days...

My search for other blind-mate connectors for low level signals continues.

 

[Skogsgurra]

zapped
"Amp/Tyco/TE 5-292178-1 drawer connector that has gold plating over nickel on the main contact area that is designed for multiple connect/disconnect sequences. The back side of the connector is a 2mm header with a tin coating"

This may be the reason. We were using a card with an edge connector many years ago. The mating (chassis) connector was gold plated. We had lots of problems with that equipment in certain locations and no problem at all in other locations.

We found out that there was a rather high chlorine concentration in the air on the sites where we had problems. We changed to tin on the mating contact and the problems disappeared. You could see some sort of white powder being formed between the gold and the tin.

Gunnar Englund

http://www.gke.org

--------------------------------------
Half full - Half empty? I don't mind. It's what in it that counts.

 

[itsmoked]

Thanks for providing that response. Confirms my thinking. From that butcheredlingguish I wonder where that guy's at.

Why don't you just go with gold and stop worrying about it? I often find gold to cost only a few percent more. In rarer cases up to maybe %40?

I see the problem as the spring tension diminishing over time and eventually getting intermittents which can be realllly costly compared to outright failures.

Keith Cress
kcress -

http://www.flaminsystems.com

 

[Skogsgurra]

Smoked
That butcheredlinguish is actually a Swede trying to express himself in a language that isn't his own - and obviously not yours, either. Sorry for that ;-)


Gunnar Englund

http://www.gke.org

--------------------------------------
Half full - Half empty? I don't mind. It's what in it that counts.

 

[Skogsgurra]

Oh, sorry. You meant the answer from Amp/Tyco/TE? I thought you meant mine

Gunnar Englund

http://www.gke.org

--------------------------------------
Half full - Half empty? I don't mind. It's what in it that counts.

 

[zappedagain]

Keith, I haven't found a blind-mate/drawer connector with gold contacts on the back side yet. Molex has some nice ones on their web-site, but no stock anywhere in the world. Positronics may have some but they aren't distributor stocked so I need to deal with them directly.

The spec for the 5-282179-1 is in Japanese; I cleaned up the response a bit before I posted it too!

Gunnar, they offer a gold socket contact to mate with the tin header. That didn't make any sense to me, and your chlorine story supports my suspicions.

 

[itsmoked]

Oh, sorry. You meant the answer from Amp/Tyco/TE? I thought you meant mine

Ha! See our post times? I actually thought the butcheredlinguish was very clear and informative. I sure didn't think it was of Swedish origin.

zappedagain; Dang. I hate hassles like that. At one time we had the same problem and had the connectors plated ourselves. It was surprisingly inexpensive.

Sounds like the Au/Sn thing might have some advantage anyway.



Keith Cress
kcress -

http://www.flaminsystems.com

 

[Skogsgurra]

Keith
I see. Photo-finish, I would say. No hard feelings.

Gunnar Englund

http://www.gke.org

--------------------------------------
Half full - Half empty? I don't mind. It's what in it that counts.