Subject:
|
Re: LEGO Quality Control
|
Newsgroups:
|
lugnet.ambassadors
|
Date:
|
Thu, 4 Sep 2008 20:37:59 GMT
|
Viewed:
|
18234 times
|
| |
|
|
In lugnet.ambassadors, Tommy Armstrong wrote:
| |
I actually had a pretty long and very interesting discussion with Bjarke
Schønwandt at BrickFair about the whole color issue. I am pretty much of an
expert in color quality control-it was my job for many many years to make
sure batch to batch color cosistancy of color in the paint we produced was
good. And our small factory achieved some very excellent results. In paint,
it is much more important than in say a plastic toy and thousands of hours of
paid labor are involved when colors drift and are inconsistant.
From what I gather, and this has been discussed before-probably by me, but
the general gist of the problem occurred when a paradigm shift in the
coloration of the parts was made. The way it was explained to me, and this is
from a bit muddled memory, was that now LEGO essentially mixes the colorant
at the machine. There are, if I understood him correctly, three hoppers --one
with the clear “base” ABS, one with ABS infused with concentrated colorant,
and the third hopper is composed of the scrap that occurs when the pieces are
cut from the molded piece. Essentially plastic necessary to create the piece
but not part of it.
Color variation can creep into the process from all these places. The ABS
base can have slightly different color dispersion characteristics, the
concentrated colored plastic will vary and because of the huge quantities TLG
uses they are bought from multiple vendors. If one did not do that, you would
have all your eggs in one basket and if one of the supplier’s factories went
down, your entire production would be compromised. The “scrap” will of course
will be only as consistant as the batches that produced it. One then weighs
an appropriate amount of base, concentrated colored plastic, and “scrap”.
There is a word for the “scrap” that I cannot recall. One needs to add enough
colorant to insure opacity and saturation of color. But the colorant portion
is the expensive part of the mix.So one needs to add just enough but not too
much. If one adds too much the strength and stability of the molded part is
compromised. This is very analogous to paint--one wants to add enough
colorant to achieve good coverage in the paint--but too much will weaken the
film because of the plasticizers and dispersants in the colorant.
In the past TLG bought huge quantities of pre-mixed ABS of a certain color.
The new system is much more flexible in that one only has to stock much
smaller quantites of the colored component and has only huge quantities of
the base. Again analogous to the paint store in which one has a base and can
mix an almost unlimited number of colors from that base.
I did ask him if they mixed various colors at the site of production to
achieve the desired color like is done in paint and he said no. In other
words they do not take a certain weight of one pellet say red, and a certain
weight of another pellet, say yellow and come out with orange. In their
process this is not possible and therefore the concentrate is one hue made
for one specific end color. But from multiple vendors.
The human eye can discern millions of different colors--it being far more
sensitive than the finest machine. And very small changes in any step of the
process will be detected by an observing eye.
To make matters even more conplicated, parts in kits come from many different
factories made from different supplier’s feed stocks so coordination of
quality across the entire company is necessary. They must all be on the same
page and having the same qc standards, etc. etc.
He insured me that they are acutely aware of this and have made great strides
addressing all these problems. But I know,there will always be problems in an
enterprise this large and with a process that is this critical and more
importantly easily discernable. The variables simply cannot be turned into
constants. He also insured me that when the fiasco of translucent pieces
became apparrent they went back and lookd at the whole broad range of
problems inherrent in this very difficult production process and reaccessed
what LEGO quality was.
I myself applaud TLC for achieving the quality that they do acheive in the
vast quantities that they deal with and with the thousands of parts that they
produce. I dare say few other companies do as good a job--and for sure no
other toy company of thier scale does.
Perfection with anything that has to do with color is simply not possible,
and one just strives to achieve a level of quality that in a great percentage
of situations is sufficient to achieve the goal. I really honestly think TLG
has done this for many, many years and has no intention sucumbing to lesser
standards. They did screw up a while back, but part of that is attributable
to a completely different way of production and perhaps the temptation to try
and cut a few colorant costs when economic times were dire. They know they
screwed up and are IMO well on the way to avoiding them in the future.
This whole color thing is really an incredibly complex task and most people
simply do not know how truly complex it is. Especially on the scale of TLG.
tommy Armstrong
|
Tommy,
The term for the scrap is more or less known as “regrind”. I’ve got some
injection molding plastics manufacturing in my background. When molding at the
machine, the material that’s not in the mold cavity that consists of the part
itself is known as the Sprue. In most operations, it’s commonplace to take the
sprue, toss it in a grinder, resulting in the sprue being ground to bits. Those
bits are called “regrind”. These bits can then be re-mixed in with new
material, usually at up to some percentage. The reason for the limit percentage
is that adding regrind essentially increases the material base for the color.
Being that the regrind already has some color in it, there can be potential for
some color variability, as you’ve stated. It’s a science with a huge amount of
variability, and a good chunk of it cannot be calculated beforehand. You’ve got
to run batches to see how the quality looks. Sometimes, a bad decision can be
made, due in part to local misunderstanding, or a lack of
communication/coordination from higher up. The only good thing to come out of
that is to fix the problem, and move on.
I liked the comment about the transparent parts becoming apparent problems..if
you think about that line, it’s kinda funny.
As with any venture--trying something doesn’t always guarantee a 100% success
rate. All you can do is try, try, again until you succeed. Unfortunately, the
toy market worldwide is very cutthroat and experiences low overall margins
(usually no more than 10%). Also interesting is that so much of TLG’s
sales/profits are heavily skewed to the fourth quarter, primarily due to
Christmas shopping season.
Scott
|
|
Message is in Reply To:
 | | Re: LEGO Quality Control
|
| I actually had a pretty long and very interesting discussion with Bjarke Schønwandt at BrickFair about the whole color issue. I am pretty much of an expert in color quality control-it was my job for many many years to make sure batch to batch color (...) (16 years ago, 4-Sep-08, to lugnet.ambassadors, FTX)
|
25 Messages in This Thread:
 
 
- Entire Thread on One Page:
- Nested:
All | Brief | Compact | Dots
Linear:
All | Brief | Compact
|
|
|
|
