[AT] Getting water out of a gearbox/now bearing life

[rlgoss at twc.com]

Unfortunately, engineers don't always use good sense.  They suffer from the same problems the rest of us do, namely -- If a little bit is good, a lot must be better.  Over the years, several standards have been developed for specifying accuracy; from linear dimensions, to Geometric Dimensioning & Tolerancing, to Statistical Process Control.  Often. they end up specifying impossible conditions or conditions that are absolutely inappropriate and uncalled for because of the process or the material.  It doesn't help that the standards themselves end up being at odds with each other, and when all three styles of documentation appear on the same design internal arguments happen and things get put on the market that have major problems with them.  When you are dealing with a product as complex as an automobile where the supply, manufacture, and assembly can end up being International in nature, it's a wonder we end up with an end product that works at all!



Larry
---- Indiana Robinson <robinson46176 at gmail.com> wrote: 
> I have a stash of 1 5/16" line shafting, some of it from before 1900 out of
> very old shoe repair/making machinery.
> About 30 years ago before I had a lathe I had a machine shop owner I'm
> acquainted with make me some adapters from a piece of it so I could use
> some shoe repair devices on a Shopsmith woodworking machine. When I picked
> the stuff up the guy asked me where I got the shafting from. After I told
> him he said "That was some of the nicest stuff I have turned in a long
> time... extremely uniform". I hadn't really given it any thought, it was
> just mild steel shafting of the right size to me.  I guess they were not
> mixing in melted Pinto's or Yugo's back then  :-)
> 
> 
> 
> 
> On Mon, Jun 8, 2015 at 9:59 PM, <jtchall at nc.rr.com> wrote:
> 
> > Over toleranced parts is not an American problem. I daily work with German,
> > Dutch, and Japanese prints that are ridiculously toleranced. Some of it is
> > so closely engineered you need to manufacture in a tightly controlled
> > temperature setting and do the assembly in a clean room. The funny part is,
> > everybody wants crazy tolerances, but none want to pay for them. We've been
> > dealing with variables in raw materials for years. Anything from plastic to
> > stainless steel, you never know for certain what you are going to get. One
> > batch machines like it should, the next cuts horribly. At least at my work
> > we have been trending toward tighter tolerances and better finishes for
> > quite some time---a lot of it is absolutely unnecessary.
> >
> > John Hall
> >
> >
> > -----Original Message-----
> > From: macowboy at comcast.net
> > Sent: Monday, June 08, 2015 8:59 AM
> > To: Antique tractor email discussion group
> > Subject: Re: [AT] Getting water out of a gearbox/now bearing life
> >
> > I have been reading this topic with great interest. For the past 15 years I
> > have been involved either on the supplier side providing components to the
> > big automotive assembly lines or on the OEM side for the medical device
> > industry. I have been to most of the major automotive assembly plant across
> > the country and a few of the big truck manufacturers too. One things that
> > was made clear to me that todays vehicles are designed to last 150,000
> > miles. This is straight from the engineers at the assembly plants. I don't
> > know where this number came from but this is what I was told. The Japanese
> > do a much better job with design and assembly. They use a practice called
> > Probabilistic Design where they take the variability of each component in
> > an
> > assembly and determine what the process capability(defect rate) will be. If
> > it is less than a zero defects, back the drawing board. I have found that
> > the American manufacturers tend to design in the electronic world(CAD) and
> > do not take into consideration the manufacturing capability of the process
> > they have selected. What happens is that when a certain combination of
> > tolerances occur in a assembly, you will have failures. The design guys
> > will
> > tend to put very unrealistic tolerances on their drawing as it makes their
> > jobs much easier. The DFMEA that was talked about earlier is more of a
> > paper
> > work/regulations requirement rather than a good tool to use. Most engineers
> > that do this conduct a top down DFMEA where they look at the assembly as a
> > whole. A better way to do this is a bottom up DFMEA where you look at each
> > requirement on each component. One engineering manager did this and his
> > final product was as close to flawless as possible. BTW, as far as
> > components go I can say the electronics(PCB's) are the worst followed
> > closely by raw materials/chemicals for failures.
> >
> >
> > Jim Thomson
> > macowboy at comcast.net
> >
> >
> >
> > _______________________________________________
> > AT mailing list
> > http://www.antique-tractor.com/mailman/listinfo/at
> >
> 
> 
> 
> -- 
> -- 
> 
> Francis Robinson
> aka "farmer"
> Central Indiana USA
> robinson46176 at gmail.com
> _______________________________________________
> AT mailing list
> http://www.antique-tractor.com/mailman/listinfo/at