Trouble at Mill!

[John FieldingParticipant @johnfielding34086]

The tale I will relate might have some relevance to all those who use CAD to make diagrams and parts for projects. If nothing else it makes us think before issuing a drawing!

Many years ago I was gainfully employed by a large concern that designed and manufactured military equipment. My forte was a RF Design Engineer and I had several projects under my care, both in development and in full production. Once you release a product onto the shop floor it becomes your personal "enfant terrible" and can haunt you for as long as it remains in production.

One day I received a call from a production operative on the shop floor giving me a heads up that there was "trouble at mill" brewing. When I asked which product was the cause of the current trouble he told me. But that product had been running for years and never gave any problems. After the initial teething troubles whilst the operators learned how to assemble and test the new animal things ran smoothly. So I was very surprised to hear there was now a problem.

My informant said it went deeper than that as there was a possibilty that all the production over the last few years might be affected. Horror thoughts like vehicle recalls to replace some parts at great expense went through my mind! Did I stuff up when I designed it?

Then my informant let me down gently, the new inspector on the line had rejected all the production for the last week and nothing was going out of the door and they wouldn't accept the word of a "nobody" to tell them to release the product for shipment.

Well, I put on my running shoes and made haste to the shop floor. My office was in a separate building removed from the factory and it took me a few minutes to get there. When I arrived a huddle of concerned people were gathered around a large table with the last few days production stacked on it. Each and everyone one of the finished items had a red label attached with string which bore the depressing words "Rejected".

I asked who had rejected the items and was shown to the door of the new inspector. He was a polite but firm gentleman who told me he couldn't pass the items as they did not conform to the drawings he was issued with to inspect said items. I was gobsmacked!

I asked him to point out what deviation there was between the finished items and the drawings. He informed me the screw slots on the items did not correspond with the assembly drawings and since the two differed they were not exactly to drawings, hence they failed! When I asked if the items actually worked and met the specification he informed me that apart from the screw slot problems everything else was perfectly in order.

When I told him the screw slot orientation was merely diagramatical and not meant to imply the screw slots had to be exactly as drawn, he agreed. But he also pointed out that the inspection drawings did not say this!

Apparently the others had made the same noises but he dug his heels in and wanted it from the "Design Authority", which would be yours truly. I told him he was perfectly correct, the items offered for inspection indeed did not comply with the supplied inspection drawings and I would immediately issue a temporary change note so the items may be passed for shipment. In that case he told me as soon as he received a copy of the works order informing him that a change note was to be issued he would release the items to the shipping department.

Although I could have, and was quite entitled to, give him a verbal b*llocking" for wasting our time, that would have been the wrong thing to do. He had made a very valid point, which if he had made the suggestion through the normal channels would have most likely been ignored. He showed us that there was a valid point, as the customers also were issued with the same inspection drawings so they could check goods delivered conformed to specification and if they had troubled to inspect with as much care as our man did would have been quite correct to reject the delivered goods. This would have been a highly embarrassing situation.

Although this is a silly and trivial occurrence it made me think about how we often take things for granted and don't give sufficient thought to the consequences. The upshot of this little "trouble at mill" was a directive issued by the Chief Engineer that in future all assembly and inspection drawings must add a note that "Screw head slots are purely for indication purposes unless noted otherwise". Problem solved!

[John FieldingParticipant @johnfielding34086]

Pitfalls for CAD operators

[Paul LousickParticipant @paullousick59116]

It is common practice that screws and nut orientation is only a representation of their finished position and it is virtually impossible to set a nut to an exact angle. The important thing is that it is tightened to the correct torque.

The problem falls back onto the inspector who has little experience and training and should not be in that position to start with.

[frank brownParticipant @frankbrown22225]

Reminds me of when I was an apprentice, when I was serving a period as an electronic inspector. Was passed a batch of assembled servo motor gearbox assemblies. They were OK but there was no evidence of the thread locking compound being used. Big stink, as was my luck I was supported by the whole of the inspection department including the head of inspection. So some one had to go out and buy the compound and then disassemble the drives etc etc. It turned out that NO thread locking compound had ever been used, despite this arrangement had been in production for at least fifteen years. I wonder if it was strictly necessary?

Frank

[MWParticipant @mw27036]

I don't blame the inspector myself, mainly because i've been in their shoes and it isn't worth letting things go if you're unsure of the job in question. I mean, i cant relate to this specific case but i did learn from my experience of "telltale" signs of what you'd call "general" tolerancing and "tightly held" tolerances, especially if they were not specifically referred to. But at the end of the day, you just don't know because you didn't work on the design of it, it's not worth your job to let it go when, as author correctly pointed out, it's much nicer to resolve a problem in house, even if you have spent a bit longer sorting it, than to let it go out, boomerangs back in and not only call into doubt the image of your product, but spend yet even longer still on it.

Michael W

[John HaineParticipant @johnhaine32865]

If it was an RF product they could have been tuning screws in which case they would definitely not be at predictable positions! I got a call like that from the shop floor at Marconi when they put an RF amplifier into production for a new TV transmitter. Unit didn't work at all at RF though all the bias levels were correct. Supercilious inspector eyes young development engineer. Quick glance reveals they'd soldered in the first transistor upside down, transposing collector and emitter. Some inspection, that!

[Martin ConnellyParticipant @martinconnelly55370]

I've had a conversation with an inspector that went:

The drawing states maximum gap of 0.5mm between the parts!.

They are touching so there is no gap.

There is a gap at the edge that is more than 0.5mm.

That is because the parts' faces are not flat, but they are touching so there is no gap.

There is a gap at the edge greater than 0.5mm.

Why is an assembly shop inspector looking at the part manufacturing drawing? What does the assembly drawing have on it?

Assemble the parts and check for leaks at pass off testing.

Then go and do that and when it fails that test come back and complain.

End of conversation.

He did not come back since the parts had already had a hydrostatic pressure test and had been passed off by the manufacturing department's own inspector. I think there are some inspectors who just love to justify their work and if they dig their heels in they take some shifting. I know one who I think was ready to claim that some stainless steel was no good because it didn't taste right when he licked it.

Martin

[Author]

Posts