5
\$\begingroup\$

I am designing a PCB that will have some Mill-Max pogo pins make direct contact with exposed copper pads on the PCB. This will have many connects and disconnects, think hundreds to thousands.

I know that it needs to be at least gold coated in order for this to be resistant to corrosion, but does that mean ENIG plating is enough? Or do I need selective hard gold plating?

\$\endgroup\$
  • \$\begingroup\$ You could also install SMT nickel pads. I have seen them, but never used them. If your pads are small, then installing nickel pads probably won't work. \$\endgroup\$ – mkeith Sep 7 '18 at 6:36
  • \$\begingroup\$ Also, mill max makes "target disks." mill-max.com/products/pin/1559 \$\endgroup\$ – mkeith Sep 7 '18 at 6:38
  • \$\begingroup\$ Harwin, too. They call them "contact pads." \$\endgroup\$ – mkeith Sep 7 '18 at 6:45
4
\$\begingroup\$

I have been putting solder on my pogo-pin pads. I apply the solder by hand on my prototype boards and have openings in the paste mask for those boards that have solder-paste applied with a stencil.

I find that the solder bumps have been far more reliable than bare pads for repeated use.

For those situations where the pogo-pin pads are used only occasionally, I find that whatever plating the board house uses is reliable. For me, this is both HASL and ENIG.

But: the pads will not stand up to repeated use. The cure is simple - just add a tiny bit of solder to each pad. A rounded bump is all that you need.

\$\endgroup\$
  • \$\begingroup\$ Interesting. If I can't find a less manual way I'll give this a shot. \$\endgroup\$ – Xavier Hubbard Anderson Aug 30 '18 at 16:15
  • \$\begingroup\$ @XavierHubbardAnderson You could add the pads to the paste mask layer, thereby having solder applied to the pads automatically in the reflow process. I've seen this used on test points for ICT and it seems to work well enough. \$\endgroup\$ – calcium3000 Aug 30 '18 at 16:17
  • 2
    \$\begingroup\$ In the context of a test fixture, where the pogo pins make contact repeatedly, but each board only goes through the fixture once, solder has not worked well for me AT ALL. For one thing, it is a bump, so the pins tend to deflect, leading to premature failure. For another thing, any flux residue builds up on the pogo pins and eventually they make poor contact. Obviously your situation must be different from mine. \$\endgroup\$ – mkeith Sep 7 '18 at 6:33
  • 1
    \$\begingroup\$ @mkeith - my situation is slightly different. Boards are washed (water-soluble flux), then tested. Pogo pins are either single sharp pin (tiny pad) or 4/5-point crown (larger pads). Our volumes are low: several hundred boards per month. \$\endgroup\$ – Dwayne Reid Sep 7 '18 at 18:15
  • 1
    \$\begingroup\$ High-current pins use large-diameter pins from Pylon. These pins are absolutely flat and make contact with the throug-hole pins on terminal blocks. Test current is about 20 Amps with these pins. \$\endgroup\$ – Dwayne Reid Sep 7 '18 at 18:17
7
\$\begingroup\$

I work in the semiconductor test industry. We've designed large boards, 40 layers, sometimes more. And typical pcbs too. We make a connection from our lager pcb to a probe card using pogo pins. These connection like yours are in the thousands.. Even 10k.. Because of the forces involved and the number of pogo pins making the connection. We use a selective hard gold plating, exactly: Minimum 30 micro inches gold over 150 micro inches nickel. With very good results over the years. Hope that helps..

\$\endgroup\$
  • \$\begingroup\$ Tho, I did end up going with just ENIG since the first draft of the boards will be low use. Do you think ENIG will be ok? \$\endgroup\$ – Xavier Hubbard Anderson Sep 7 '18 at 2:27
  • \$\begingroup\$ We did have a few issues using ENIG. The contact pads over time would become damaged. But in our case the PCB asserts a large force on the pogo pins. And are being compressed a few times a day. We needed a thick board with strong plating. And of course we needed the boards to last a long time. With the plating I stated before they do last for more than 10 years of heavy use. It might just be over kill for your situation, not sure. \$\endgroup\$ – Chris Williams Sep 7 '18 at 16:36
  • \$\begingroup\$ Excellent advice. I'll take it into account. I think we are pretty low force, so maybe I can get away with ENIG. \$\endgroup\$ – Xavier Hubbard Anderson Sep 7 '18 at 17:26
  • \$\begingroup\$ @XavierHubbardAnderson : Do please share how things turned out in terms of the pogo-pin connection reliability, in your test with ENIG plating! \$\endgroup\$ – boardbite Jun 24 at 0:38
2
\$\begingroup\$

That's a lot of connection cycles! In that case, in addition to the plating, it is important to choose the correct pogo pins.

A simple sharp point will eventually wear through the plating. The rounded, Mill-Max tips will last longer. But there are more available options. For example, these are a few of the different options from IDI (now Smiths Interconnect):

enter image description here

(web link to full document)

The H-style tip, especially, should always have a good point of contact regardless of wear. Not only will the probe find some high point on the contact, but it will rotate slightly over time and so find entirely new contact points.

If you couple that with @DwayneReid 's recommendation of putting solder over ENIG (or HASL) pads, it should last a very long time. This combination is particularly effective because the "soft" solder will provide a good landing, even as it is deformed by multiple connection cycles.

\$\endgroup\$
0
\$\begingroup\$

Is this just for occasional (testing), or one time (initial programming) use?

Hard gold would be great, but overkill. ENIG would be good, even the Hot Air thing (drawing a blank on the name) that iteadstudio uses as their base option would be good enough. I have iteadboards that I've had quite a while that look as good as when I received them, some are months old, some even older.

\$\endgroup\$
  • 1
    \$\begingroup\$ Hey! This would be for many repeated contacts. On the order of hundreds to thousands. Also, you're thinking of HASL? \$\endgroup\$ – Xavier Hubbard Anderson Aug 30 '18 at 15:58
  • \$\begingroup\$ Yes, HASL. For repeated contacts, I would think you'd want a harder surface to withstand the repeated bashing of the POGO pins into the board. \$\endgroup\$ – CrossRoads Aug 30 '18 at 16:05
  • \$\begingroup\$ Right. I agree, @CrossRoads. Do you thing ENIG would be enough or do I need selective hard gold plating? I haven't done this before, so I'm not sure what the industry standard is. \$\endgroup\$ – Xavier Hubbard Anderson Aug 30 '18 at 16:06
  • \$\begingroup\$ I can't say, I've only used HASL for installing parts one time, thru hole and surface mount. I've only ever had one pad lift, a really small one, when I tried to lift a pin to isolate the pin and rewire it. In the pad's defense, there wasn't a whole lot of surface area under the TSOP pin to help hold it down, and a 8 mil trace leading away. So other than that one instance, HASL has been all I've needed for assembling boards. HASL/ENIG with solder over it might be all you need. I haven't seen Tin/Lead solder get corroded either. \$\endgroup\$ – CrossRoads Aug 30 '18 at 16:10

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.