View Single Post
Old November 13th, 2010, 03:46   #28
Delierious Designer of Dastardly Detonations
MadMax's Avatar
Join Date: Dec 2001
Location: in the dark recesses of some metal chip filled machine shop
What's a TNC? In my small amount of experience designing products that were manufactured in China, I have learned that it's possible to make high quality products there if you're willing to make regular operations inspections or set up a permanent quality team there. If you're not willing to make the investment of regular inspections then you're opening yourself to a lot of screwups.

A designer has the knowledge of intent in a products design. A good one will pick high performance materials for components with difficult technical requirements (e.g. high strength heat treatable steel for trigger sears) and commodity materials like ABS or polypropylene for less physically demanding parts. There are many ways to appear to meet a drawings specifications while failing to meet the intent of a particular part. For instance I might specify a high grade glass filled nylon (GFRP) material for a part that needs to be resilient. We use virgin Zytel in our grenade shells which is a particularly pricey material made by Dupont. If say we made our shells somewhere we couldn't conveniently audit (china is a 24hr flight away) we might fail to observe that our injection moulder substituted our mat'l with a local brand of nylon GFRP. The local brand GFRP might have the same glass content (30% by weight) but their mean strand length might be much shorter, or more likely their plastic matrix may have higher shrinkage. Use material with very short mean strand length and you don't get the stiffness benefit of the glass fibre or the resilience of the Dupont product. Even worse, use a high percentage of regrind (sprues can be reground and remelted) and you'll lose a lot of toughness because the glass fibres break in the regrind and subsequent remelts of nylon reduce the mean chain length of molecules (less crack resistance).

It's easy to specify a material, make, and zero regrind, but if the plant runs out of Zytel because of a supply issue they might just grab something that's handy and not think it's much of an issue. You have to expect that there will be little understanding of why you have made your specifications.

In another product I worked on, we found a particular grease that provided exceptionally high cycle life on a heavily loaded worm drive made from pretty low cost materials. Good deal! Cheap parts with a measured dollop of expensive grease and you have a cheap but well performing total assembly. Our gearbox was driving a high torque driving a worm drive that pushed high force. With that particular grease, we could get good performance of the product (low noise, efficient drive) over about a hundred thousand cycles. It takes a f*cking long time to accumulate that many cycles even in a lifecycle testing rig that does actuations a thousand times higher than the expected daily usage rate. If you want to compare several greases, you have to repeat your lifecycle test for each goo. Anyhow, a container load of our much anticipated first run arrived and everything should have been fine. Our on site visit confirmed that they were using specified plastics and steel components and assembly procedures were being followed. After our team left, the plant decided to rush the production (I guess so they could start on someone else's contract) so they stopped applying the magical grease with the applicator we engineered for them. They applied the grease by hand (scoop out a blob and spread it) which consumed much more grease than anticipated. They blew through their supply early and applied something else that they had scads of and shipped the container without telling us what happened. The gunk they used was about a grade or two higher than bacon fat which is tasty but a poor lubricant so we had to design a squeegee to remove the crap and apply the proper goo. Almost every unit had to be disassembled and regreased before they could be sold.

If you can arrange a permanent inspection team or at least make visits for the duration of every production run then the situation is much different. For instance Ryobi tools are extremely well priced for their performance and features (except for their drill presses which are junk). I've got three of their drills and other tools that were very well priced and they're still kicking after over 5 years. They actually work quite well and on inspection (I love disassembling things) they've got high density powder metal gears and hardened components just where they're needed. From what I've heard, there is a permanent engineering team in China headed by a German tool maker who is their product design lead. With that kind of on site QC from a team, who really appreciates the design intent, it's possible to benefit from a very low cost of skilled labor and not be vulnerable to subtle screwups that make your well designed product a heap of junk. A strong engineering team in China can make very well engineered products produced at very economical prices.

Unfortunately AI isn't big enough to have an inspection team in China that appreciates the design intent of our products. We source a few items there, but their not difficult parts to make. Our spoon head castings are Chinese, but we intentionally make the mounting features undersized so we can precisely ream them so they slide onto a grenade just right.
Want nearly free GBB gas?

MadMax is offline   Reply With Quote