| FORUM

FEDEVEL
Platform forum

Test Points and Flying Probe Testing

Tom Yunghans , 10-30-2022, 09:57 PM
Hi Robert,

I have seen a few threads on this topic over the years in your forum. In those threads you stated that you had never used a flying probe testing approach on any of your boards. Just wanted to make a few points and see if you have looked into this any further...

Flying probe testers need a place to probe, and generally they need at least one place to probe for each net. These testers are useful on low volume boards at both the fabrication level (to test the bare board) and at the assembly level (to test a fully assembled board). With no parts installed, the flying probe can easily probe the component pads. However, once components are soldered on the board, the pads are not really accessible, as they are mostly covered by the component itself (e.g. BGA, chip devices) or the pins of the component (e.g. QFP, SOIC). I have also heard that probing on top of the pins is undesirable.

I understand that the newer flying probe machines are quite accurate and if the pad is long enough the tester can make contact with the pad without touching the component pins. I have seen one recommendation from a flying probe tester vendor that all you need is 12 mils of metal to be probed. Therefore, if the pad is 12 mils longer than the end of the pin, it can be probed. This may require that the footprint pad be made a little longer than the IPC recommendations.

Vias can also be probed, if they are large enough and not covered by solder mask. Most vias pad diameters are larger than 12mils, but my understanding is that you really don't want the flying probe to go into the via hole, which would therefore require the via be filled and capped (extra cost). However, if you have a fairly low density board, it seems like you could make the via pads annular ring 12 mils and then the flying probe could then land on the pad and avoid the hole, so you can skip the extra cost of filling and capping the vias.

Can you comment on this?

Thank you!
qdrives , 10-31-2022, 02:49 PM
If you have a "fairly low density board" I would say it will not be a problem to place specific test pads.
That is more of a problem with high density boards.
Do consider that not all contract manufacturers have that high precision flying probe testers. It is best to keep the pads at least 0.8mm (30mil) in diameter.
If you can put all test points on the bottom, they can also be used for in circuit testing when producing on a large scale.
robertferanec , 11-02-2022, 08:48 AM
Important question is, how are you planning to test all the signals? In many boards, it is not necessary (maybe not even possible) to test every single signal during production test. It can be super difficult, time consuming and costly to measure every signal and try to determine if it is a correct signal or not.

If I believe a signal is important I place there a test point (unless it can be measured on the connector e.g. through hole connectors can be probed by pogo pins or a simple breakout board can be designed specifically for testing). Often, this way, there will be no many testpoints.
Use our interactive Discord forum to reply or ask new questions.
Discord invite
Discord forum link (after invitation)

Didn't find what you were looking for?