BGA's in space applications are a relatively recent thing. The problem was that conventionally every solder joint has to be visible for inspection, something that is impossible with a BGA other than by X-ray. That is less of a problem with space-qual stuff because it has to be rad-hard anyway.
Even the shrink wrap that is used in the spacecraft wiring has to be transparent so that wiring splices and connector joints can be inspected. Different world.
Even the shrink wrap that is used in the spacecraft wiring has to be transparent so that wiring splices and connector joints can be inspected. Different world.
X-ray inspection is automatically added by most assembly houses when doing BGAs now. Never had X-ray inspection cause an issue for a chip that I've personally seen, but I guess it's certainly possible.
https://www.cypress.com/file/202516/download
https://www.cypress.com/file/202516/download
Last edited:
Interesting. The BGA device that caused inspection problems with ESA was the processor in Digital Processing Unit of MIXS (of which I was project manager). ESA had to approve an inspection protocol for that device.
MIXS and 11 other instruments were launched in October '18, did its first gravity assist past Earth in April '20 and in Oct '20 its first assist past Venus. Due to arrive in Mercury orbit in 2025.
MIXS and 11 other instruments were launched in October '18, did its first gravity assist past Earth in April '20 and in Oct '20 its first assist past Venus. Due to arrive in Mercury orbit in 2025.
Yeah, but I bet ESA has not really speeded up their approvals of new technology much in the decades since I last worked with them. But then again I still think micro-bga are the work of the devil. At least space doesn't have lead free solder to worry about 🙂
Have you checked the SSM2166, depends on your noise / gain requirements.
SSM2166 Datasheet and Product Info | Analog Devices
TCD
Yes, that is absolutely true. Things move glacially slowly in the space agencies, and certainly in ESA. Every electronic part in the machine apart from the BGA processor was entirely "conventional" - gull wing multi-pin devices, circuit boards with no solder resist or silk screen etc. The only thing that is different to consumer electronics is several orders greater part price for space qualified parts. As an example the FPGAs alone were £30k each. Two launched and one in the Flight Spare.
And of course no lead-free, in common with defence and hospital equipment; IOW anything that is reliability critical.
Craig
Actually, medical is almost all RoHS 2 compliant now even if there are currently some exemptions. I worked on a few devices that are technically exempt but the manufacturer wanted the electronics to comply anyway. The expectation is that the EU will eventually rescind it and they don’t want to do work later if they have to resubmit for whatever reason. Of course, everything still out there is fine.
Lead-free is just where everyone has gone. The commercial BGAs are all specified for lead free now, so if you deviate and have problems it’s going to end up your issue I’d imagine.
Is no solder mask due to thermal issues, or radiation?
Lead-free is just where everyone has gone. The commercial BGAs are all specified for lead free now, so if you deviate and have problems it’s going to end up your issue I’d imagine.
Is no solder mask due to thermal issues, or radiation?
Last edited:
IIRC whisker grow at higher altitudes was a problem with leadfree solder. If that wasn't solved in the recent two years or so ....
Sure, but what does that have to do with solder mask or medical devices?
I don’t think anyone implied lead free solder is going to be used everywhere in aerospace.
I don’t think anyone implied lead free solder is going to be used everywhere in aerospace.
Sorry,my bad, I simply misread Billshurv's sentence about "don't have _the_ problem" as "don't have _a_ problem" .
Yes. Solder mask and silk screen printing degrade in space due to particle flux and outgas forever. So if the instrument has any optics, it can be rendered blind by outgassing junk condensing on the surfaces and detectors.
In addition circuit boards can have no exposed tracks on the front and back. And typically they have many layers for signals, power and ground, with vias, micovias and buried vias by the hundreds. They also have a thermal layer; since there is no convection in space, any heat needs to be conducted to the instrument and spacecraft structure.
There was a similar problem with use of silicone grease for lubrication of any moving parts in space vehicles. It migrates in vacuum (so it was found), and made at least one UV instrument blind by creeping over the optics. ESA issued a directive to remove any silicone grease in any instrument, or overall spacecraft, destined to launch.
Stripping down flight hardware is not something to do lightly. Particularly since the only way to clean silicone grease completely needs plasma etching. Since any reassembled flight hardware has to go through stringent inspection and approval, the silicone debacle added at least 6 months, and often much longer, to hardware delivery and launch.
The current allowed lubricants are creep free, UHV mineral greases.
I used to like looking at the solar array bearings for 3-axis stabilised satellites All hogged from solid Be as it doesn't need lube. Space is fun for Tribologists.
Anyone remember the flap the semi industry went into over whiskers about 15-20 yrs ago when we went lead free?
The whiskers thing seemed the biggest problem and then all the process changes, customer hand holding (especially auto!)
Fun times.
(Nice to see supply shortages have now doubled stock prices. NXP trading at $194 last I looked up from $110 a year ago - TI similar with mkt cap now at $180 billion. )
The whiskers thing seemed the biggest problem and then all the process changes, customer hand holding (especially auto!)
Fun times.
(Nice to see supply shortages have now doubled stock prices. NXP trading at $194 last I looked up from $110 a year ago - TI similar with mkt cap now at $180 billion. )
Last edited:
You can't get everything with lead, and what you get does not
withstand the lead free solder profiles. That forces you to completely lead free.
We also used solder masks, in spite of having an optical SPAD.
Some use Mupsil or Solitane as conformal coating, IIRC.
We did not get the solder procedures of contemporary FPGAs, and
much less chips, so we had to use original Virtex from the stone age.
Stuff for ISS, and before DT.
Last edited:
The beryllium alloy I am most familiar with for bearing use is CuBe, but I like learning new things! I have not attended a STLE meeting in a while and things move on...and tribology is one of the most fascinating aspects of physics I have found!
Howie
STLE#55984
I guess I am the frictional element in this discussion....Hey, ever try to trademark a lubricant? I have, and the sex lube industry has pretty much sewn up all the good names...
Cheers!
Howie
Cheers!
Howie
I've always liked the name 'Mr Zoggs sex wax' that surfers use. Although that's apparantly for stick rather than slip 🙂
I've just gone down a rabbit hole looking at advances in BAPTA and APM technology in the last 30 years. Looks like ceramics are heavily used now, which makes a lot of sense. Will research more...
I've just gone down a rabbit hole looking at advances in BAPTA and APM technology in the last 30 years. Looks like ceramics are heavily used now, which makes a lot of sense. Will research more...
Conformal coating, usually vacuum applied, is needed in space applications before launch. Most launch sites are near coasts, and near the equator. So humidity and salt atmosphere are a big thing, both for circuit boards and stress corrosion cracking of structural elements.
For scientific missions and other satellites, conformal coating is superfluous once on orbit.
Of course manned space flight, and in crew quarters on the ISS humidity is again a thing, so also need conformal coating on electronics systems.