0805 thermals

I suppose it's not the resistor itself that is the main issue. It's more what it's connected to.

Jeroen Belleman

That's not too bad of a temp rise. Remember all the goofy throug-hole parts specs where you can't cut the leads or can't let the resistor reach over 25C anywhere for "full" power ratings.

There were two camps as regards axial resistors:

Make the leads short.

Make the leads long.

I think I can go to 0603 resistors. I measured one at 120 K/W, not radically worse than an 0805.

First order, the hot-spot theta of a surface-mount resistor is independent of size.

Has anybody made a SMD strain guage to see what the push and pull forces are on a that rohs garbage solder? Nobody seems to publish data like "after went green, board failures were up 20% over 18 months" or anything like that.

For a period of time I was able to get better pricing on random products from companies by asking for the pre-lead free versions (if they had any) that could no longer be exported to commies. Most places didn't want to have stock of two items that could not be mixed up.

I think the big hazard is the typically higher reflow temperature.

Our manufacturing people want to go all lead-free no-clean for soldering. We have just finished a series of tests of electrical leakage vs flux vs humidity.

I'm sure many such analyses have been made, but are held close, for PC reasons.

But the US DoD forbids use of lead-free solders on equipment used in theater. The minimum is 5% tin (by weight?).

See S.E.D. threads "Re: USA and Australia vs. the World!" from August

That is one problem for sure, as is the increased brittleness leading to fractured joints.

Another is tin whiskers causing equipment to just fail. If this happens in high-power equipment arc-flash hazards and literal meltdown happen.

It's sobering to watch some of the arc-flash accident and test videos

- those pretty sparks flying this way and that are incandescent drops of boiling copper.

I'd suggest that manufacturing be looking for growth of tin whiskers as well. Lack of leakage is not a test for this - there is no leakage until something shorts. Some kinds of conformal coating can prevent tin whiskers. The NASA website cited in the December 2019 thread goes into this in some detail.

Joe Gwinn

I would be very interested in that data, most especially if a graph.

Thanks, R. Baer

There are no graphs, just initial and post-humidity leakage tests on various patterns soldered with various flux, hand and machine soldered. I'll see what we have that's postable. I initially tested a few boards myself, but it took a lot of time so I turned it over to C.

The Keithley electrometer takes about an hour to settle to 3e12 ohms, which was our upper limit on attention span.

Test board:

How did those tests go? What's the normal wash time for boards in a dishwasher? I came across some HVAC control boards that still had the remove after washing stickers over the buzzer openings. This is the nonsense that happens when plants move overseas and the experienced employees are all fired.

How would you have manufacturing inspect for tin whiskers? My understanding is they are rather microscopic and not visible to the naked eye. Until they cause a short there is virtually no way to find them other than scanning every square mm of leads, tinned vias and solder joints. That would seem to be rather impractical on a large scale. isn't this more of a problem that you must prevent by design?

John Larkin <jlarkin@highland_atwork_technology.com> wrote in news: snipped-for-privacy@4ax.com:

I remember we had some English (read UK) thru hole axial lead resistors in one of our designs and they were little tiny buggers (looked like our 1/8 watt jobs) but claimed to dissipate a whole watt!

John Larkin <jlarkin@highland_atwork_technology.com> wrote in news: snipped-for-privacy@4ax.com:

It was not "acceptable" by the IPC quality manual, but we elevated thru hole resistors off the board in many cases as we were making HV circuits and parts against FR4 makes great carbon trail pathways. We also elevated them for thermal purpose of not baking the board! If they get that hot, it is a bad design to have them mounted on a PCB to start with.

Cydrome Leader snipped-for-privacy@MUNGEpanix.com wrote in news:sj2mrn$ppv$ snipped-for-privacy@reader1.panix.com:

We noticed that potting with "Stycast" can shear SMD parts due to the thermal expansion properties it had and the fact that it polymerized into a very brittle medium. It would pop them out of their end terminations as well. They did sell a thermal version that looked like blue cake icing. We would mix silica shards in it to make it have even better thermal properties. All that just as RoHS was ushering in.

I think the lead free solders actually have better tensile and shear numbers... However...

The stuff (RoHS solder) just looks grainy and it took me a while to accept that as being the norm now for solder joint inspection. I'll take 63/37 any day.

Joe Gwinn snipped-for-privacy@comcast.net wrote in news: snipped-for-privacy@4ax.com:

For some things. It is an old rule and COTS pretty much shot that down. COTS use is so proliferant nowadays. Now for mission critical stuff like Honeywell's missile guidance circuits or Ratheon's stuff, sure. Things made here... in house. The problem with that is that a new design means getting a chip maker to do a custum run just for you with leaded solder balls on the chips. We had to send out tens of thousands of chips to get "re-balled". It is easier just to design with what you have, as in what is available. Not many J Std mil spec chips to be had any more. One has to get the maker to do custom runs just for you and you have to buy them all.

We (the DoD) are exempt and can design and use products which use leaded solders, however, if we want to sell our products overseas, they have to be RoHS compliant, even if it is a military product such as a

Copper pours, preferably on both sides of a multilayer board, can heat sink the leads a lot. A lot of inherited lore was from the days of

The vitreous enamel wirewound resistors have huge dissipation specs, because they can run almost red hot.

On a sunny day (Thu, 30 Sep 2021 21:09:04 -0700) it happened snipped-for-privacy@highlandsniptechnology.com wrote in snipped-for-privacy@4ax.com:

Old Philips color sets had vertical mounted ceramic resistors Did not some have a thermal auto-open solder joint on top? And you can get these:

Rick C snipped-for-privacy@gmail.com wrote in news: snipped-for-privacy@googlegroups.com:

And they take time to grow/form. A fresh solder job has NONE.

We use vertical ceramic resistors.

It was alligators; the subsonic mating call of the males.