NASA Workmanship Guide for wire splicing | Hacker News Reader

NASA Workmanship Guide for wire splicing

workmanship.nasa.gov

143 pointsasimpleusecase44 comments

Looks like the entire guide is here https://workmanship.nasa.gov/lib/insp/2%20books/frameset.htm...

Pretty darn cool little site, will have to mirror it.

Have used this site a lot over the years, really useful resource. I wish ESA was more willing to do this sort of guidance document rather than just the requirements.

NASA has some terrific workmanship videos as well. Here is the one on soldering techniques: https://www.youtube.com/watch?v=_RXugDd0xik

Even as a retired IBEW electrician, I learned several "workmanship tips" from this wonderful assembly guide:

Something that I definitely "did wrong" for decades was over-twisting stranded wires before inserting them into a crimp connector. According to NASA's guide, the correct way is to strip the wire and insert it without additional compaction. Their reasoning makes sense: leads to individual strand breakage and/or loose crimps.

Related, double-crimp terminals are designed to have one crimp over uninsulated wire, the second (closer to contact/space/barrel) mates with the stripped copper. Similarly, for twist-post connections, the first 1.5 twists should be of insulated wire, and then several more of stripped conductor.

I got trained in NASA soldering standards in the early 80s. Worked as a tech in college helping build test harnesses for an X-Ray telescope that was to fly on Challenger. Got delayed after the accident, but I think it eventually flew 20 years later.

They were very serious about this stuff, for obvious reasons. No rework possible in space.

I still use those skills on my home DIY hobbies.

> No rework possible in space.

I suddenly started imagining someone trying to strip some wire, twist the strands together, hold a soldering iron and the solder and the wires all while essentially having Mickey Mouse fingers from being in a space suit. If something ever deserved the No User Serviceable Parts Inside sticker...

I really hope someone stuck a "no user serviceable parts" or "warranty void if removed" sticker on JWST

That is why they made Wago lever nuts

Being unfamiliar with this, I had to look it up[0]. While these look interesting, I can't imagine where I'd use them. They are HUGE. Like bigger than that even. I can't imagine unbolting the side panels of a satellite in space, being able to use one of these, and then it being able to fit back into the original position.

[0] https://www.wago.com/us/discover-wire-and-splicing-connector...

They're _excellent_ for wiring devices into junction boxes in construction; particularly when connecting the higher gauge luminaire wires of an overhead light to romex. For some reason they're polarizing among electricians in the US (they carry all necessary listings for use in the US...). By contrast, they're widely accepted and used elsewhere.

> For some reason they're polarizing among electricians in the US

They’re 8x more expensive and take longer to install than a wire nut, and they take up more space inside the device box. They’re also difficult to unterminate without damaging the lever nut or conductor.

It’s not hard to splice #14 solid to #18 stranded with a wire nut, you just extend the smaller wire a bit beyond the end of the larger wire and twist.

I’d recommend wago lever nuts to homeowners doing their own electrical work because homeowners are terrible at electrical work and the less chances for electrical fires, the better.

I’m a commercial electrical project manager in the US.

> they're polarizing among electricians in the US

sounds like it needs a ground! /s

but seriously, what kind of junction boxes are used outside the US that these would fit?

as far as connecting a larger device wire to a smaller mains line seems odd to me. granted, i'm not an electrician, but it is pretty much opposite of all of my experiences with wiring commercial lighting and home lighting

I use them almost exclusively in my home wiring - they're so much more user friendly compared to wire nuts. They are not large at all - I find wiring that these take a lot less space in tight boxes than a wire nut since they're flat and you can push them to the back more easily:

https://imgur.com/a/tGTUJd9

They make stab Wagos as well which I don't recommend, but the lever ones are great. You can work 'hot' if you need to replace a switch or anything (don't work hot blah blah blah). Much of my existing wiring is crappy 50s era copper and in some areas, aluminum -- so to add GFCIs where useful, I use either the levers to join copper of shitty provenance or the ALCU junction connectors and a bit of the paste.

This 1958 NASA soldering training video still holds up pretty well (aside from the sociological aspects!).

https://youtu.be/_RXugDd0xik?si=Z9ToN_tMz0m6lHu3

At approximately 3:20 it explains that solder is a mix of tin and lead and through some quirk no-one understands, this causes it to melt at least 100c lower than either of those two elements on their own. We don't understand how, but we understand how to exploit that.

Do we now understand why that happens?

(edit: minor side-rant about high definition 16mm films being preserved at 480p mpeg.)

>They were very serious about this stuff, for obvious reasons. No rework possible in space.

There are other reasons that may be less apparent, like the fact that materials can have weird properties in space. One example is solder "whiskering"[1]. I believe NASA requires techs to be certified to the IPC-JSTD [2].

[1] https://nepp.nasa.gov/whisker/

[2] https://www.ipc.org/ipc-certifications

Note that tin whiskers happen in air here on Earth, too. Not a thing that only happens under vacuum.

Whisker growth rates from 0.03 to 9 mm/yr have been reported.

There are some elimination methods such as conformal coatings (very hazardous) and plating finishes (nickel/gold/palladium) that have been tested. I got certified by the IPC-J-STD-001H also soldering 201 surface mount components 0.6x0.3x0.25mm 0.05W handling.

Have considered those standards in some safety critical, high vibration applications. NASA aren’t kidding. It was eye opening to me to consider crimps superior to soldering in terms other than convenience.

We never got out of the prototyping stage but it was clear that we wouldn’t use this approach for the parts of the system that weren’t vibrating. Of course if you’re building a space- or aircraft it’s worth doing this pretty much everywhere.

As I commented in the other thread where this comes from, not really current anymore.

> That is very old reference, and afaik not used at NASA anymore. More current ref would be https://s3vi.ndc.nasa.gov/ssri-kb/static/resources/nasa-std-... but also note that NASA has been moving away from having their own standards and instead adopting industry standards: https://nepp.nasa.gov/index.cfm/26139

https://news.ycombinator.com/item?id=39227420

I swear i recently was looking up torque recommendations per thread size into aluminum and NASA seemed like the only source to show up

Your first link doesnt work fyi.

Oh right, it got mangled from copying the comment over. Here it is full: https://s3vi.ndc.nasa.gov/ssri-kb/static/resources/nasa-std-...

From the second link:

The NASA Workmanship Standards Program has been working to adopt IPC/WHMA-A-620xS since 2008. The goal is to allow the use of IPC/WHMA-A-620B-S, Space Applications Electronic Hardware Addendum to IPC/WHMA-A-620B as an alternative to NASA-STD-8739.4, Crimping, Interconnecting Cables, Harnesses, and Wiring. This change facilitates a move by suppliers towards the industry voluntary consensus standard but does not force those suppliers who cannot accommodate the current IPC/WHMA-A-620B training options to do so.

Notice that the statement is undated but says they've been working on it "since 2008"; possibly that statement is old too.

I see that NASA is turning to space industry standards, not general electrical standards (e.g., for a commercial building). But I wonder why do it at all? I assumed NASA's internal standards met their quality and performance needs, but maybe it was just that there were no industry standards when NASA developed theirs?

I assume some of these standards might have their roots in Apollo days or something, and definitely there wouldn't have been many options back then. But why change now? The give clear explanation:

> In 1998, NASA established by internal mandate, the need for all NASA standards users and custodians, to comply with the 1995 Office of Management and Budget (OMB) Circular A-119 on Federal Participation in the Development and Use of Voluntary Consensus Standards and in Conformity Assessment Activities. This directive compels the NASA Workmanship Project to actively identify, enhance and adopt industry standards which can be used in lieu of the NASA-STD-8739 series documents.

Of course you might ask why that OMB was made, but tbh it just makes lot of sense imho, maintaining own set of standards is not cheap especially with all indirect effects that it implies

There's another guide [0] for wire wrap technique (solderless, cold welded connections to square posts [1]). Used extensively in telephony switching/patching in the past, as well as digital circuit construction like computers. Quite nice for prototyping, although qualitative sockets/posts can be hard to acquire at a reasonable price nowadays.

[0] https://workmanship.nasa.gov/lib/insp/2%20books/links/sectio... [1] https://en.wikipedia.org/wiki/Wire_wrap

Rebuilt a unit with over 7,000 wire wraps in it back in 2009 for F/A-18 flight computer testing. An electrical short had caused several runs to melt and basically run a hot knife through the internal wiring.

Had to rebuild it from original hand written point to point wrap instructions (of which there was only the original) coupled with all the changes in the 30 years since original build.

That was a hell of a project.

I want to compare this to the Boeing guide

When adjusting the bolts to the aircraft doors, adjust until finger tight

...if bolts are even present.

This hole left intentionally blank

*unintentionally

The characters "u" and "n" have been unintentionally left out.

We hope? With the info that has come to light about the processes and decision making within Boeing, I would not be surprised if some bean counter calculated that removing 0.5% of all bolts would reduce weight enough to improve fuel consumption. Of course this is totally made up conjecture on my part, but I would not be surprised by it one bit.

Pair the NASA documents with PACE soldering/rework videos to have a successful outcome each time, all the time.

For practical purposes, the lineman splice is probably the most useful one to learn. It's only moderately more difficult than the obvious "pair up and twist" approach, but the wires remain collinear.

Great guide, especially for beginners/laymen.

I consider these advanced splices. Someone fixing a power cord in their house need not go to such extremes.

Just a reminder to all homeowners that the other extreme is also bad; you can't just wire-tape two Romex cables together in an attic and call it a day!

I wish someone had told the person that did nearly this very thing to the low voltage control cable to my external AC compressor. At least in the attic it would not be exposed to weather and UV. <facepalmEmoji>

I always splice wires considering these good practices, regardless of what application. I just like doing things correctly and reliably.

I also have a little bit of a fetish. I really enjoy it when other people do their work correctly and do those small things that show their prowess. And I like to to think somebody someday sees what I did and will enjoy it the same.

That's why, when I sail, I always have all my lines neatly tied and organised. And use proper knots for their applications. And look at other yachts when docked to see if their owners know what they are doing or not.

Do you knoll when building Lego?

These days I leave Lego for my kids.

Ah yes, this has graced these halls many times. I use the SOLDER SLEEVE SPLICE INLINE as often as I can.

Reply on news.ycombinator.com