Manufacturing Quality During the War

[mrbill]

WWII movies, novels, and even historical books or documentaries rarely
have any mention of the poor manufactured quality of WWII era
equipment. In movies guns almost always fire, engines always start,
radio sets don't work when they've been shot or other wise damaged, and
almost nothing breaks down. The truth is much different. WWII
manufactured quality was abysmal by our standards and had a significant
affect on the outcome of battles, maybe even campaigns. All countries
suffered from the effects of trying to push too much stuff out under
trying circumstances. Some thoughts on the relative quality of
manufactured goods during the war:

A) Allies
1) U.S.
Probably had the best quality of mass produced material although there
were notable problems such as the Packard built Merlins and Curtiss
Helldivers that had major teething problems. The U.S. was best suited
for modern mass production and didn't face bombing raids or major
dislocation. A bigger problem was the social disruption with men in
uniform and people moving all over to take jobs.

2) Britain
British standards were relaxed during the war to increase production.
Also, industrial cities were bombed and supply lines disrupted by the
Germans and Japanese to some extent. British expertise has
historically been in design and mass production resulted in poorer
quality.

3) USSR
Russia equipment was generally crude but they produced lots of it. If
a T-34 broke down there were 5 more to replace it. Soviet quality
(probably never great) suffered by the huge disruption and massive
relocation east.

B) Axis
1) Germany
German quality has always been excellent but it took a nose dive as the
war went on. In particular, employing slave labour with a vested
interest in sabotaging the production had a profound affect on quality.


2) Italy
Il Duce had no business sending men off to fight with the equipment
they had. It's a bit surprising to me as Italian machine tool equipment
is top notch at present.

3) Japan
Until the 60s the Japanese had a reputation for horrible quality. As
their supply lines were cut and their cities bombed their quality could
only become worse. I suspect that few aircraft even came close to the
performance figures typically cited.

Perhaps this is one aspect of WWII that still requires study.

[Don Phillipson]

WWII manufactured quality was abysmal by our standards and had
a significant
affect on the outcome of battles, maybe even campaigns. All countries
suffered from the effects of trying to push too much stuff out under
trying circumstances. Some thoughts on the relative quality of
manufactured goods during the war . . .
1) U.S.
Probably had the best quality of mass produced material although there
were notable problems such as the Packard built Merlins and Curtiss
Helldivers that had major teething problems. The U.S. was best suited
for modern mass production and didn't face bombing raids or major
dislocation. A bigger problem was the social disruption with men in
uniform and people moving all over to take jobs.

1. The two named "notable problems" are quite dissimilar.
British Merlin engines were proven efficient before their
mass-production in the USA was ordered. So the only
problems were those of redesign (to suit US materials,
tools and fasteners) and output quality as mentioned.
But the Helldiver was a weapons system, requiring
integration of its components (airframe, engine, weapons)
into a useable weapon of the USN -- within which the
quality of fabrication was relatively minor.

2. Social problems of training or relocating factory
manpower are only distantly related to quality control.

2) Britain
British standards were relaxed during the war to increase production.
Also, industrial cities were bombed and supply lines disrupted by the
Germans and Japanese to some extent. British expertise has
historically been in design and mass production resulted in poorer
quality.

These remarks appear to concern only weapons. The post
ignores the vast governmental "utility" control scheme for
essential goods for civilian consumption (clothing, furniture,
etc.) which (1) abolished through rationing the second most
expensive tier of the market (2) apparently raised quality
standards for everything else. The Utility label persisted
as long as did rationing (early 1950s.)

In the case of weapons, no evidence seems to suggest
building larger numbers of weapons (e.g. Spitfires, Lee-
Enfield rifles, submarines etc.) "resulted in poorer quality."
We know about various notoriously dud designs, e.g.
Avro Manchester or Westland Whirlwind: but the topic is
the low quality of manufactures and no one says these
aircraft were duds because poorly fabricated.

2) Italy
Il Duce had no business sending men off to fight with the equipment
they had. It's a bit surprising to me as Italian machine tool equipment
is top notch at present.

This condemnation without evidence does not advance
our understanding. The topic is manufacturing quality,
i.e. the poster here asserts that:
(a) Prewar Italian manufacturing quality was good
(b) Wartime weapons were bad
(c) Defective manufacture was the reason wartime
weapons were bad (rather than design etc.)

This is not a logical sequence, merely a set of assertions
without evidence and with no interconnection. E.g. war in
North Africa suggested Italian tanks as obvious candidates
for "inferior weapons:" most writers suggest this was
because they were obsolete for combat and ill-designed for
desert conditions; but the suggestion here is that they
were shoddily manufactured because mass-produced,
albeit in "top notch" factories (and the British engineers
who assessed captured enemy equipment somehow
failed to notice this.) I do not believe it.

If any of us believed "WWII manufactured quality was abysmal by
our standards and had a significant affect on the outcome of
battles, maybe even campaigns" we ought to be able to name
one battle lost because of manufacturing defects in the
weapons available. I doubt that this poster can.

[james]

Post by Don Phillipson

WWII manufactured quality was abysmal by our standards and had
a significant
affect on the outcome of battles, maybe even campaigns. All countries
suffered from the effects of trying to push too much stuff out under
trying circumstances. Some thoughts on the relative quality of
manufactured goods during the war . . .
1) U.S.
Probably had the best quality of mass produced material although there
were notable problems such as the Packard built Merlins and Curtiss
Helldivers that had major teething problems. The U.S. was best suited
for modern mass production and didn't face bombing raids or major
dislocation. A bigger problem was the social disruption with men in
uniform and people moving all over to take jobs.

1. The two named "notable problems" are quite dissimilar.
British Merlin engines were proven efficient before their
mass-production in the USA was ordered. So the only
problems were those of redesign (to suit US materials,
tools and fasteners) and output quality as mentioned.
But the Helldiver was a weapons system, requiring
integration of its components (airframe, engine, weapons)
into a useable weapon of the USN -- within which the
quality of fabrication was relatively minor.

Indeed I watched a documentary on the woman who ran the production
facility in Thunder Bay Ontario which produced first Hurricanes, then
Helldivers.

They apparently had no difficulties at all producing high quality
Hurricanes, even through many redeseigns and modifications. Things
changed when they switched to the Helldiver, an admittedly more complex
aircraft. They striggled and fought with Curtiss, and eventually the
woman was fired. Further analsyis pointed the finger at poor design.

James

[mrbill]

Post by james
Indeed I watched a documentary on the woman who ran the production
facility in Thunder Bay Ontario which produced first Hurricanes, then
Helldivers.
They apparently had no difficulties at all producing high quality
Hurricanes, even through many redeseigns and modifications. Things
changed when they switched to the Helldiver, an admittedly more complex
aircraft. They striggled and fought with Curtiss, and eventually the
woman was fired. Further analsyis pointed the finger at poor design.

Manufacturing quality is often related to poor design. The old adage
that you can't make a silk purse out of a sow's ear applies.

[Diogenes]

Post by james

Post by Don Phillipson

WWII manufactured quality was abysmal by our standards and had
a significant
affect on the outcome of battles, maybe even campaigns. All countries
suffered from the effects of trying to push too much stuff out under
trying circumstances. Some thoughts on the relative quality of
manufactured goods during the war . . .
1) U.S.
Probably had the best quality of mass produced material although there
were notable problems such as the Packard built Merlins and Curtiss
Helldivers that had major teething problems. The U.S. was best suited
for modern mass production and didn't face bombing raids or major
dislocation. A bigger problem was the social disruption with men in
uniform and people moving all over to take jobs.

1. The two named "notable problems" are quite dissimilar.
British Merlin engines were proven efficient before their
mass-production in the USA was ordered. So the only
problems were those of redesign (to suit US materials,
tools and fasteners) and output quality as mentioned.
But the Helldiver was a weapons system, requiring
integration of its components (airframe, engine, weapons)
into a useable weapon of the USN -- within which the
quality of fabrication was relatively minor.

Indeed I watched a documentary on the woman who ran the production
facility in Thunder Bay Ontario which produced first Hurricanes, then
Helldivers.
They apparently had no difficulties at all producing high quality
Hurricanes, even through many redeseigns and modifications. Things
changed when they switched to the Helldiver, an admittedly more complex
aircraft. They striggled and fought with Curtiss, and eventually the
woman was fired. Further analsyis pointed the finger at poor design.

My father served as US Navy fighter pilot in WWII, but had occasion to
fly the Helldiver a few times as a ferry pilot. He pronounced it the
worst aircraft he he ever had the misfortune to fly. The official Navy
designation was SB2C, for Scout Bomber, 2nd model, Curtiss.

Dad insisted that it actually stood for Son-of-a-Bitch, 2nd Class.
----
Diogenes (***@hotmail.com)

The wars are long, the peace is frail
The madmen come again . . . .

[mrbill]

Post by Don Phillipson
1. The two named "notable problems" are quite dissimilar.
British Merlin engines were proven efficient before their
mass-production in the USA was ordered. So the only
problems were those of redesign (to suit US materials,
tools and fasteners) and output quality as mentioned.
But the Helldiver was a weapons system, requiring
integration of its components (airframe, engine, weapons)
into a useable weapon of the USN -- within which the
quality of fabrication was relatively minor.

You will note that the subject title is "Manufacturing Quality" not
"quality of fabrication". The fact that Packard Built Merlins were not
as reliable as their Rolls Royce counterparts and that the Helldiver
had serious design flaws that delayed an effective launch are both very
much manufacturing quality issues.

Post by Don Phillipson
2. Social problems of training or relocating factory
manpower are only distantly related to quality control.

They are very directly related in the following ways:

- large numbers of people were hired on mass, most of whom,
particularly the women, had never worked in factories before. Proper
industrial engineering and training was crucial for these people to do
their jobs effectively
- factory manpower was not relocated. Hundreds of thousands of people
moved to take jobs away from where they lived. New jobs, new cities,
no family support. This kind of stress will negatively affect people's
work performance.
- the standard three shift rotation was started during WWII to increase
production. It's since been found to cause problems among workers.
The affect during the war would have been even more significant as
people were unprepared for shift work at all. Tired workers make
mistakes.
- cases of women leaving their children in movie theatres for their
shifts have been cited. Can someone concentrate on their jobs and not
make mistakes while worrying about their kids?

Poor employee morale negatively impacts quality. It's been
demonstrated. It's not distantly related.

Post by Don Phillipson

Post by mrbill
2) Britain
British standards were relaxed during the war to increase production.
Also, industrial cities were bombed and supply lines disrupted by the
Germans and Japanese to some extent. British expertise has
historically been in design and mass production resulted in poorer
quality.

These remarks appear to concern only weapons. The post
ignores the vast governmental "utility" control scheme for
essential goods for civilian consumption (clothing, furniture,
etc.) which (1) abolished through rationing the second most
expensive tier of the market (2) apparently raised quality
standards for everything else.

Based on what evidence? Ceasing production of second rate goods
doesn't make the first rate goods better.

Post by Don Phillipson
The Utility label persisted
as long as did rationing (early 1950s.)
In the case of weapons, no evidence seems to suggest
building larger numbers of weapons (e.g. Spitfires, Lee-
Enfield rifles, submarines etc.) "resulted in poorer quality."
We know about various notoriously dud designs, e.g.
Avro Manchester or Westland Whirlwind: but the topic is
the low quality of manufactures and no one says these
aircraft were duds because poorly fabricated.

Actually, I think there is plenty of evidence which hasn't been
published. I remember my father who was a mechanic during the war
tellling me that they had to rebuild every truck that they received.
In his words 'the engines were just dumped in'.

Years ago I looked at the 1941 edition of the "Machinery's Handbook"
and it held a special notation titled "War Emergency Tolerances".
These tolerances were not tighter but considerably looser.

Post by Don Phillipson

Post by mrbill
2) Italy
Il Duce had no business sending men off to fight with the equipment
they had. It's a bit surprising to me as Italian machine tool equipment
is top notch at present.

This condemnation without evidence does not advance
our understanding. The topic is manufacturing quality,
(a) Prewar Italian manufacturing quality was good
(b) Wartime weapons were bad
(c) Defective manufacture was the reason wartime
weapons were bad (rather than design etc.)
This is not a logical sequence, merely a set of assertions
without evidence and with no interconnection. E.g. war in
North Africa suggested Italian tanks as obvious candidates
for "inferior weapons:" most writers suggest this was
because they were obsolete for combat and ill-designed for
desert conditions; but the suggestion here is that they
were shoddily manufactured because mass-produced,
albeit in "top notch" factories (and the British engineers
who assessed captured enemy equipment somehow
failed to notice this.) I do not believe it.

Another poster to this NG posted an example some time ago of an Italian
tank breaking down during the Greek campaign. You have postulated that
the design of equipment is distinct from manufacturing quality. They
are in fact integral. A bad design can not be manufactured well
(Helldiver) and a good design will make manufacturing relatively
problem free (Lee Enfield).

Post by Don Phillipson
If any of us believed "WWII manufactured quality was abysmal by
our standards and had a significant affect on the outcome of
battles, maybe even campaigns" we ought to be able to name
one battle lost because of manufacturing defects in the
weapons available. I doubt that this poster can.
--

I'll rise to that challenge. First, let me address your implication
that WWII manufactured quality was somehow comparable to today. Modern
world class manufacturing generally requires that the producer have a
defect level below 50 parts per million (PPM). Following the war the
US published Military Standard 105-D which established among other
things sampling frequencies for Acceptable Quality Levels (AQLs) to
improve the quality of materials by the US armed forces from war and
post-war levels. Stringent AQLs for this period were on the order of
2.5 to 5% or 25,000 to 50,000 PPM. So, the defect rate in the last 60
years has dropped by a magnitude of 1,000; probably more in some cases.

Specific Battles:
1) The Normandy Campaign- Blackburn cites in "The Guns of Normandy" the
high percentage of duds in the German shells. Would the Germans have
won if they hadn't had so many duds? Probably not, but Allied
casualties would have been significantly higher.
2) The Air War over Japan- One of the reasons that the Japanese
couldn't mount an effective air defense against the B-29s was that the
performance of their fighters varied so greatly. If Japanese fighters
had decimated that the high flying B-29s would the USAAF have still
switched to the low flying raids that eventually produced results?
3) Kursk- The hastily introduced Panthers suffered from a very high
rate of breakdowns. The Germans wouldn't have won anyway but having
most of their best tanks out of action didn't help.

[nightjar]

"mrbill" <***@hotmail.com> wrote in message news:***@e3g2000cwe.googlegroups.com...
....

Post by mrbill
- large numbers of people were hired on mass, most of whom,
particularly the women, had never worked in factories before. Proper
industrial engineering and training was crucial for these people to do
their jobs effectively...

I'm in engineering and I've trained school leavers to do simple repetitive
tasks - typical of production line work- in a matter of weeks. Most would be
up to full production rates in under a month. It is only training workers to
be flexible enough to undertake multiple tasks without supervision that
takes time.

Post by mrbill
- the standard three shift rotation was started during WWII to increase
production. It's since been found to cause problems among workers...

The only problems I ever found running three shifts was that productivity
levels dropped on the night shift, sometimes to half that of the day shift.
As night work carries a cost premium, that often makes it uneconomic,
although soemtimes it can be the only way to achieve the output required.
Some of my night workers actually preferred it to working during the day.
Quality was never affected.

....

Post by mrbill
Poor employee morale negatively impacts quality. It's been
demonstrated. It's not distantly related.

Simply working for the war effort was often a great boost to morale, so
peacetime experiences are not neccessarily relevant.

....

Post by mrbill
I'll rise to that challenge. First, let me address your implication
that WWII manufactured quality was somehow comparable to today. Modern
world class manufacturing generally requires that the producer have a
defect level below 50 parts per million (PPM). Following the war the
US published Military Standard 105-D which established among other
things sampling frequencies for Acceptable Quality Levels (AQLs) to
improve the quality of materials by the US armed forces from war and
post-war levels. Stringent AQLs for this period were on the order of
2.5 to 5% or 25,000 to 50,000 PPM. So, the defect rate in the last 60
years has dropped by a magnitude of 1,000; probably more in some cases.

I was supplying government specified goods to an AQL of 1% up to last year.

Colin Bignell

[Timothy J. Lee]

Post by nightjar

Post by mrbill
Poor employee morale negatively impacts quality.

Simply working for the war effort was often a great boost to morale,

Also, simply working at all in a job that was likely to be around for
more than a few months may have been a great boost to morale for people
who were used to being unemployed a large percentage of the time in the
last decade or so.

[mrbill]

Post by nightjar
I was supplying government specified goods to an AQL of 1% up to last year.

You would have been dropped from any automotive supplier list 40 years
ago for supplying to anything less than 0 defects.

20 years ago when I was a QC manager I had one of our production
supervisors sort through about 1,000 parts after an inspector found a
bad part in an audit. I checked with him to see if he had sorted
through them so that we could do a re-audit. He said, "Yeah, we sorted
them although I don't know why. We only found 12 bad ones." This view
was a carry over from the war and post-war era in which some level of
defects in shipments were expected and acceptable.

Nowadays if you ship one bad part then you are expected to institute
permanent irreversible corrective action. Governments sometimes tend
to lag.

[Don Phillipson]

Post by mrbill
20 years ago when I was a QC manager I had one of our production
supervisors sort through about 1,000 parts after an inspector found a
bad part in an audit. I checked with him to see if he had sorted
through them so that we could do a re-audit. He said, "Yeah, we sorted
them although I don't know why. We only found 12 bad ones." This view
was a carry over from the war and post-war era in which some level of
defects in shipments were expected and acceptable.

I do not think a case from industry in the 1980s can
by itself tell us anything about industrial practice in
the 1940s. But the anecdote suggests a discrepancy
between production standards and operational standards
which was obviously known to the public in the 1940s cf.
(1) the rumour that one or more all-welded Liberty ships
(not riveted together, the norm in the 1930s) had simply
split apart and sunk instantly;
(2) the stage play All My Sons by Arthur Miller, about
a manufacturer who supplies defective engines to the
USAAC. One of them kills his aviator son. This play
was on the New York stage in about 1947.

The point is that production standards for some goods
may accept that one item in a thousand (or one in a
million) should not work as designed. But it depends
what sort of a thing we are talking about:
(a) Veterans of WW1 knew that X per cent of artillery
shells were duds, i.e. could be fired but did not
explode on arrival. Numbers varied from 5 to 25 per
cent. Perhaps this did not seem to matter much
in the intensity of WW1 barrages from 1915 onwards.
(b) Maintenance of aircraft and submarines cannot
accept targets (norms) of less than 100 per cent
perfection. If X per cent of aircraft will not fly or
Y per cent of submarines will not surface on
command, the mission cannot be completed and
the whole crew is probably lost.

Case (b) was examined in detail by the RAF Coastal
Command operational research unit of approx. 1941
which changed RAF squadron work methods valuably.
The old tradition was that a particular small team of
servicemen (one engine mechanic, one airframe
mechanic, one armourer) looked after a particular
aircraft. Repairs and servicing went on every day,
and each squadron reported every day its "serviceability,"
the number of aircraft certified by inspection as ready
for use. Serviceability was never 100 per cent (the
purpose of inspection was to detect unserviceable
aircraft and send them for repairs, not out on operations)
but could be quantified. Coastal Command found that
serviceability increased (using the same numbers of
repairmen) when centralized on a squadron basis
rather than done separately for each aircraft. E.g.
a unit with 20 aircraft might average 18 serviceable
every day instead of the former 15.

This is by itself no statement of manufacturing quality,
but it emphasizes that weapons functioned in a single
continuous system, from the factory exit door at one
end to the enemy's guns at the other end. The forces
needed to send to the enemy's coast the maximum
number of ships, aircraft, tanks, etc., all fit to fight:
which was done more by planning than by prayer.

[Louis C]

Post by Don Phillipson
(1) the rumour that one or more all-welded Liberty ships
(not riveted together, the norm in the 1930s) had simply
split apart and sunk instantly;

I read an account of a French oiler doing just that in 1939, and wonder
how frequents such accidents were anyway.

LC

[Brad Meyer]

Post by Louis C

Post by Don Phillipson
(1) the rumour that one or more all-welded Liberty ships
(not riveted together, the norm in the 1930s) had simply
split apart and sunk instantly;

I read an account of a French oiler doing just that in 1939, and wonder
how frequents such accidents were anyway.
LC

We studied an incident where a Liberty ship suffered brittle fracture
from one side of the hull to the other whiled tied up to a pier. I
forget the details but that incident and the great molassas flood were
our introduction into fracture mechanics and brittle fracture during
nuke school.

[Andrew Venor]

Post by Brad Meyer

Post by Louis C

Post by Don Phillipson
(1) the rumour that one or more all-welded Liberty ships
(not riveted together, the norm in the 1930s) had simply
split apart and sunk instantly;

I read an account of a French oiler doing just that in 1939, and wonder
how frequents such accidents were anyway.
LC

We studied an incident where a Liberty ship suffered brittle fracture
from one side of the hull to the other whiled tied up to a pier. I
forget the details but that incident and the great molassas flood were
our introduction into fracture mechanics and brittle fracture during
nuke school.

I remember the broken ship example from Nuke School too. I also saw it
again later in college. Basically what happened was that when one plate
fractured the crack would propagated across the welds until the entire
hull broke in two. In an older riveted ship the fracture would
propagate to the edge of the plate and stop.

Not to long ago C SPAN 2 reran a program featuring author of the book
Dark Tide: The Great Boston Molasses Flood of 1919. I suppose I should
check the book out one of these days. Or just add it to the pile I have
in the living room of books I'll get to one day.

ALV

[Brad Meyer]

Post by Andrew Venor
Not to long ago C SPAN 2 reran a program featuring author of the book
Dark Tide: The Great Boston Molasses Flood of 1919.

I went to prototype at S3G during a time when the Rx vessel was very
badly embrittled (they were talking of anealing it on the next
refueling) and we had 5 degree by 3 psi "window" we had to thread our
way through doing a plant heat up. I always though about that molasses
tank when we were doing that.

[Jim]

..

Post by Louis C
I read an account of a French oiler doing just
that in 1939, and wonder how frequents such
accidents were anyway.
LC

The father of a girlfriend of mine in the '70's was a WWII USN vet. He
served on LST's as the navigating officer, IIRC. One night he told me
the tale of setting out from an east coast port bound for England,
pre-D-Day... It took _three_ tries to get there. I've heard (and told
;-)) some sea stories; this one had none of the usual exaggerated
feeling of TINS. I think it was true.

IIRC, he said that the first time, they suffered irrepairable failures
of one and then the other main engine, and wound up aground and wrecked
in the Azores.

On the second try in another LST, they hit a moderate storm. The ship
started cracking in front of the superstructure, first on the main deck,
then down the sides of the hull. They used the existing bitts to rig
chains and turnbuckles across the crack, and then welded on improvised
anchor points for more of the same on the tank deck.

But the hull eventually broke into two pieces, and they abandoned her
(onto another LST, IIRC). They returned to the States, were assigned to
yet another LST, and made an uneventful third voyage.

So yes, ships broke in half. Cracking of welded hulls was a problem,
AIUI, as were other quality problems. I have no idea of the frequency
of such problems, tho'.

But I served in the USN myself in VN in the late '60's, and there were
some what I think were the same class of LST's still in service then,
tho' most served in relatively static roles as barracks or repair ships
for riverine forces.

Jim, "Entropy never sleeps; do y'all?"

[Brad Meyer]

Post by Jim
But I served in the USN myself in VN in the late '60's, and there were
some what I think were the same class of LST's still in service then,
tho' most served in relatively static roles as barracks or repair ships
for riverine forces.

They were also used as rocket launching platforms. HCU-1 occupied one
for a portion of its time in country.

[h***@bbs.cpcn.com]

Post by Don Phillipson
The point is that production standards for some goods
may accept that one item in a thousand (or one in a
million) should not work as designed.

As you say, production is only part of the picture. Goods may (and
often were) damaged in transit. The field operators of the goods might
not be properly trained or an emergency substitute in a combat
situation.

Post by Don Phillipson
This is by itself no statement of manufacturing quality,
but it emphasizes that weapons functioned in a single
continuous system, from the factory exit door at one
end to the enemy's guns at the other end. The forces
needed to send to the enemy's coast the maximum
which was done more by planning than by prayer.

Yes, there was considerable planning efforts. Out of wartime research
to improve production (both volume and quality to the resolve
breakdowns fast) was new statistical quality control techniques and
what today is called "Operations Reseach". Planners had to juggle a
great many variables on which factory should make what component. Lots
of tradeoffs. They developed math and analytical tools to solve these
problems.

Getting a working howitzer and shells out of raw steel ore and minerals
overseas to the front was not a trivial task. A lot of people worked
very hard to make that happen. Such workers did not get the glories
accorded to those in combat but were vital just the same.

[nightjar]

Post by mrbill

Post by nightjar
I was supplying government specified goods to an AQL of 1% up to last year.

You would have been dropped from any automotive supplier list 40 years
ago for supplying to anything less than 0 defects.

That is not an achieveable target without 100% fully automated inspection,
something that was not even remotely possible in WW2. Without machines doing
the inspection, you can only achieve a certain level of probability of zero
defects.

....

Post by mrbill
Nowadays if you ship one bad part then you are expected to institute
permanent irreversible corrective action.

That level of specification is rare in general industry.

Colin Bignell

[mrbill]

Post by nightjar

Post by mrbill

Post by nightjar
I was supplying government specified goods to an AQL of 1% up to last year.

You would have been dropped from any automotive supplier list 40 years
ago for supplying to anything less than 0 defects.

That is not an achieveable target without 100% fully automated inspection,
something that was not even remotely possible in WW2. Without machines doing
the inspection, you can only achieve a certain level of probability of zero
defects.

Actually it is. Current practice usually incorporates statistical
process control and Poke Yoke techniques. SPC was still in its infancy
during WWII and wasn't widely practiced as far as I can tell. The term
Poke Yoke certainly hadn't been coined but the idea of using locating
holes to ensure that parts couldn't be incorrectly assembled was as old
as time.
Current requirements for process capability are +/- six sigma (6
standard deviations) which works out to about 5 in a million (I don't
have any charts that actually go to 6 sigma but it's that order of
magnitude). +/- 3 sigma (<3 in 1000) was the standard 25 years ago and
that predated a lot of the computerized information systems that we
rely on today.

The problem is that during the WWII era manufacturers were still relyin
on inspectors to inspect quality in. As the pressure to produce
increases the pressure to drop inspection or make it less stringent
increases as well.

Post by nightjar
....

Post by mrbill
Nowadays if you ship one bad part then you are expected to institute
permanent irreversible corrective action.

That level of specification is rare in general industry.

True but most of our consumer goods are far better than they were 50
years ago. I'm always impressed by photcopiers; incredibly complex
pieces of equipment with hundreds of extremely precise parts and
generally they go on like the Energizer Bunny. We wouldn't enjoy that
level of quality and reliability without very stringent programs to
maintain them.

[Steven Schmid]

Post by mrbill

Post by nightjar
I was supplying government specified goods to an AQL of 1% up to last year.

You would have been dropped from any automotive supplier list 40 years
ago for supplying to anything less than 0 defects.
20 years ago when I was a QC manager I had one of our production
supervisors sort through about 1,000 parts after an inspector found a
bad part in an audit. I checked with him to see if he had sorted
through them so that we could do a re-audit. He said, "Yeah, we sorted
them although I don't know why. We only found 12 bad ones." This view
was a carry over from the war and post-war era in which some level of
defects in shipments were expected and acceptable.
Nowadays if you ship one bad part then you are expected to institute
permanent irreversible corrective action. Governments sometimes tend
to lag.

If you're a QC manager, you must know that zero defects is not possible. Six
sigma practice, which is essentially a universal standard these days, still
doesn't eliminate defects.

The automotive industry doesn't require such quality, either. That's not part
of ISO 9000 or anyone's practical experience. It may have been quoted as an
excuse in order to dump a supplier.

But we're way off topic. In WWII, production was the key. If you could
deliver a desired quantity of parts quickly, a few defects would be forgiven.
Commonly, extras would be ordered in order to plan for defects within a lot.
The world has changed a lot.

I still think there were lots of defects, lots of in-field repairs and
modifications, lots of wasted resources. But there was probably not a real
advantage with any one country. The US just had a lot more resources to throw
at the opposition.

Best Regards,

Steven Schmid

[mrbill]

Post by Steven Schmid
If you're a QC manager, you must know that zero defects is not possible.

Used to be a QC Mgr.

Post by Steven Schmid
Six
sigma practice, which is essentially a universal standard these days, still
doesn't eliminate defects.

6 sigma does give world class manufacturing level quality as I
indicated. The difference 30 or 40 years ago was in how the 0 defect
concept was applied versus AQLs. Depending on the sample size a lot
with defects in the sample can be shipped with 10, 5, or even 1% AQL
levels. In the automotive industry after Ralph Nader it was no longer
acceptable to knowingly ship defective parts.

Post by Steven Schmid
The automotive industry doesn't require such quality, either. That's not part
of ISO 9000 or anyone's practical experience. It may have been quoted as an
excuse in order to dump a supplier.

I worked for 3 different automotive companies during the 70s, 80s, and
90s and now consult with several others. This level of quality is
definitely expected at least in certain areas. Safety items and
emissions items absolutely demand this level of quality.

Post by Steven Schmid
But we're way off topic. In WWII, production was the key. If you could
deliver a desired quantity of parts quickly, a few defects would be forgiven.
Commonly, extras would be ordered in order to plan for defects within a lot.
The world has changed a lot.

That's true. Of course, the issue with spares is that they are
generally no better than the original production. For example, if 10%
of the OE distributor caps were defective then 10% of the spares would
have been defective as well which means that some of the in-field
repairs would not have worked.

Post by Steven Schmid
I still think there were lots of defects, lots of in-field repairs and
modifications, lots of wasted resources. But there was probably not a real
advantage with any one country. The US just had a lot more resources to throw
at the opposition.

The US, Canada, Australia and other significant producers outside of
the war zones would not have had to face the disruption of bombings or
serious supply interruptions such as the other Allies and Axis faced.
The quality was probably better based on the standards of the time.

[Andrew Clark]

Post by mrbill
The US, Canada, Australia and other significant producers outside of
the war zones would not have had to face the disruption of bombings or
serious supply interruptions such as the other Allies and Axis faced.

Britain's industry was not much troubled by bombing after spring 1941. The
Baedeker raids targeted non-industrial areas and the V1/V2 attacks in 1944
were too light and random to have much systematic impact on industrial
output.

The main constraints on British industrial and manufacturing output in WW2
were import stringency and labour volatility.

[Jim Lillie]

Did the Italians have a serious problem with their grenade fuses?
I remember reading that after a battle many grenades were lying around
that had not detonated - yet. But would detonate if moved.

Jim Lillie

[mrbill]

Post by nightjar
The only problems I ever found running three shifts was that productivity
levels dropped on the night shift, sometimes to half that of the day shift.
As night work carries a cost premium, that often makes it uneconomic,
although soemtimes it can be the only way to achieve the output required.
Some of my night workers actually preferred it to working during the day.
Quality was never affected.

Actually the problem is not so much running three shifts as the way the
shifts rotate which is opposite to the way people adapt. None of this
was known during the war and 3 shift rotation has continued in the same
manor since then. Some people like working at night and there
generally isn't a problem if they stay on nights.

[Don Phillipson]

Post by mrbill

Post by Don Phillipson
These remarks appear to concern only weapons. The post
ignores the vast governmental "utility" control scheme for
essential goods for civilian consumption (clothing, furniture,
etc.) which (1) abolished through rationing the second most
expensive tier of the market (2) apparently raised quality
standards for everything else.

Based on what evidence? Ceasing production of second rate goods
doesn't make the first rate goods better.

Exactly. The effect in Britain was to put better
quality goods into the hands of the lower 75 per cent
of the market, those who before "utility" would have
been able to buy only shoddier fabrics, chairs that
broke too easily, and boots that let in the rain.
There is a large volume of literature on the permanent
benefits to the Britain of the wartime utility system
immediately in the quality of goods in stores and, in
the long run, in factory design teams.

Only the second top class of purchasers suffered.
The very richest civilians could get the prewar quality
they were used to (mostly eked-out prewar stocks.)
All goods were rationed (including furniture and fabrics)
but the share available to middle and working classes
was better quality than they were used to. Quantitatively,
the quality of British consumer goods increased in
wartime although the quantity decreased. In the case
of civilian consumer goods, the British utility system
appears to contradict the thesis that wartime production
means lower quality.

[Andrew Robert Breen]

Post by Don Phillipson

Post by mrbill

Post by Don Phillipson
These remarks appear to concern only weapons. The post
ignores the vast governmental "utility" control scheme for
essential goods for civilian consumption (clothing, furniture,
etc.) which (1) abolished through rationing the second most
expensive tier of the market (2) apparently raised quality
standards for everything else.

Based on what evidence? Ceasing production of second rate goods
doesn't make the first rate goods better.

Exactly. The effect in Britain was to put better
quality goods into the hands of the lower 75 per cent
of the market, those who before "utility" would have
been able to buy only shoddier fabrics, chairs that
broke too easily, and boots that let in the rain.
There is a large volume of literature on the permanent
benefits to the Britain of the wartime utility system
immediately in the quality of goods in stores and, in
the long run, in factory design teams.

Utility goods were often of a very good standard
- not, perhaps, up to the standards of the top end
of the market, but in materials and design often
superior to a great deal of current manufacturing.
I've got several bits of utility furnature in my
house: the design is very respectable and the materials
are good - solid timber, nicely jointed. It's only
at the back that ply is used, and even that is good-
grade material. The stains and varnishes used made
them look dowdy to modern eyes, but that was the
style of the time. They can be stripped back to
reveal very nice light oak. I shudder to think
what you'd have to pay for that quality of design
and manufacture now..

[Andrew Clark]

Post by Don Phillipson
In the case
of civilian consumer goods, the British utility system
appears to contradict the thesis that wartime production
means lower quality.

I agree; this has been the subject of several authoritative studies.

I have a number of 1930-41 era Utility blankets that were handed out free to
householders for use in air raid shelters. They are thick wool with woven
seams, of a quality even after 60+ years which exceeds that of most high-end
department store products today.

[Cubdriver]

On Thu, 26 Oct 2006 12:27:44 -0400, Don Phillipson

Post by Don Phillipson
If any of us believed "WWII manufactured quality was abysmal by
our standards and had a significant affect on the outcome of
battles, maybe even campaigns" we ought to be able to name
one battle lost

I dont' see that that follows, given that the problem was worldwide,
i.e. German defects offset by American defects.

Anyone who had to swap his Motorola cell phone for the WW2
walkie-talkie shown in the advertisements ('What did you do in the
war, daddy?') would certainly grouse about the quality of WW2
equipment. Whether that was manufacture or the state of the art is
another question.

Every veteran (whom I've talked to) who met the Heer in combat came
away with the belief that if there was any problem with German
equipment, it came at the opposite end of the spectrum: it was
manufactured to such high tolerances than it was vulnerable to damage
by sand or mud.

[Dwilma]

Post by mrbill
WWII movies, novels, and even historical books or documentaries rarely
have any mention of the poor manufactured quality of WWII era
equipment. >
Probably had the best quality of mass produced material although there
were notable problems such as the Packard built Merlins and Curtiss
Helldivers that had major teething problems. The U.S. was best suited
for modern mass production and didn't face bombing raids or major
dislocation. A bigger problem was the social disruption with men in
uniform and people moving all over to take jobs.

Although this is anectodal, my father piloted new U.S. aircraft from
factory to field in '43 and '44. They learned to be very very cautious
about the quality control in the planes put together by many men
imported from rural and impoverished backgrounds. Brand new planes
crashed with fatal results due to factory errors (examples available
upon request) To be sure this was a better situation than equipment
assembled by slave laborers, but there is something of a myth to the
Industrial Miracle of U.S. production.

David Wilma, Deputy Director
www.HistoryLink.org

[David Thornley]

Post by mrbill
3) USSR
Russia equipment was generally crude but they produced lots of it. If
a T-34 broke down there were 5 more to replace it. Soviet quality
(probably never great) suffered by the huge disruption and massive
relocation east.

Another thing to consider is the Soviet attitude.

In the Soviet Union, a tank only had to last six months. If, by accident,
it survived that long, it could always be replaced. The Soviets were
pretty ruthless in their design, to try to make functional weapons and
vehicles and such with minimum resources.

Of course, the quota system characteristic of crude centrally planned
economies didn't help. If a factory was tasked with producing five
hundred trucks, then five hundred trucks would be produced, whether of
usable quality or not.

Post by mrbill
1) Germany
German quality has always been excellent but it took a nose dive as the
war went on. In particular, employing slave labour with a vested
interest in sabotaging the production had a profound affect on quality.

Much of the quality was excessive. An engineer friend of mine looked
over German equipment, and found that much of it was of the highest
standards, with a good many resources used to give them a very nice,
but militarily useless, finish. (US manufacturing came between German
and Soviet in this regard, and as the US had more and better tools to
begin with, the US could produce somewhat higher-quality stuff without
undue labor or expense.)

Of course, some things were low-quality to start with. German torpedos
had serious problems early in the war, and Bismarck fired apparently
low-quality shells. She hit Prince of Wales something like ten times
without a full shell detonation.

Post by mrbill
2) Italy
Il Duce had no business sending men off to fight with the equipment
they had. It's a bit surprising to me as Italian machine tool equipment
is top notch at present.

Some Italian weapons were ridiculous, some well-designed. Most of it,
in 1940, was obsolescent at best. Late war Italian fighters, with German
engines, were excellent.

Italy was waging war beyond its means, and when quantity goes up too
much, quality suffers.

Post by mrbill
3) Japan
Until the 60s the Japanese had a reputation for horrible quality. As
their supply lines were cut and their cities bombed their quality could
only become worse. I suspect that few aircraft even came close to the
performance figures typically cited.

The Japanese had other problems.

The honorable thing to do was to fight; specifically, to attack. This
meant that manufacturing war materiel was not a particularly honorable
or highly thought-of activity. Any attempt at mass-production would
generally be undermined by feature change requests from the front.
After all, if the warriors want something, the mere workers should
comply.

It wasn't until well after WWII that the Japanese finally realized
that making things could be important.


--
David H. Thornley | If you want my opinion, ask.
***@thornley.net | If you don't, flee.
http://www.thornley.net/~thornley/david/ | O-

[Steven Schmid]

Gentlemen:

An entertaining discussion so far. Let me add a little to the discussion. I
think I may have some qualifications that are unique here - I co-author a
pretty popular manufacturing textbook, and quality is part of Chapter 16. ;)

The modern concept of quality is pretty advanced, but remember that the WWII
generation did not benefit from the ideas of Deming or Taguchi. The engineers
at the time did have some tools like control charts and the like (but not
often), and nothing like six sigma or anything. They didn't need it. The key
to manufacturing in the 1940s was production, not quality. (An emphasis that
persisted well into the 1970s.) All countries tried hard to mass produce
goods. No one tried to have quality programs like we think of them today. So
at the very beginning we have to recognize that we're imposing our worldview,
and quality expectations we have didn't exist at the time.

Production in the 1940s was *hard*. No CNC machine tools, no computer
controls, no continuous casters, limited polymer and ceramics options, etc.
(Parsons didn't do his work until the 1950s. Ernst and Merchant were 1950's
too. Vinodagrova was 1949. Manufacturing science was not developed. Don't
take for granted that everyone knew how to mass produce products, even what
we'd call low quality products.)

The other problem is that people are confusing quality with performance or
cutting-edge technology. Performance is easier to quantify (which plane was
faster, which tank had better armor, etc.). Arguing which designs are
technically advanced is fun, but not related to manufacturing quality.
Reliability could be quantified statistically (but no one here is trying to
do so, only reporting on heresay that probably relates more to performance).

Also, quality is more than just reliability or the number of defects
produced. A high quality product is one that (Taguchi philosophy):
e Delights the customer
e Is available in the quantities desired when needed
e Is low-cost
e Functions reliably from birth to death
e Has no variation in its performance (all members of a population should
function well, not some great and others badly).
e Is robust (not sensitive to the surrounding environment)
e Is upgradeable
e Looks good

If you think about your favorite products, they probably satisfy more of
these requirements than your less-favored products. But a Porsche 911 and a
Honda Accord can both be classified as high quality, which makes sense. Hey,
Taguchi is a sharp guy.

It is rare that a product (or weapons system) fulfills all of these
requirements. Some weapons were better than others. For example, the German
jet fighters certainly delighted the pilots, but were never available in the
quantities desired when needed. The Sherman tanks arguably fit the definition
better, even if they didn't compete with Tigers (when they functioned) in 1:1
combat.

Maybe the original query didn't really intend to discuss manufacturing
quality. Maybe I'm reading too much into this. But I think people are talking
about technology when the question was about quality.

As for WWII production on a national basis, I think it would be really
fascinating if some historian tried to analyze this. Apply methods like the
loss function, quality function development, house of quality, etc., to
whatever data can be found. I think such data may not exist anymore, but it
would be fascinating if the data could be found. Right now, we just can give
(entertaining) arguments about what we think was coolest.

I suspect that the highest quality weapons system was the atomic bomb; it
fulfilled just about all of the quality requirements. But the sample space is
small; hard to have confidence in the reliability statistics. ;)

My hunch, though, is that by modern standards, quality for every country was
abysmal. Just think of your favorite WWII tank - which of these did it
satisfy? Now think of the M-1. Maybe it's not cheap enough, but it does
better than the WWII era tanks at fulfilling all of the quality criterion.

Personally, I'd love to keep reading opinions about quality in WWII. I of
course can't force you to restrict your definition, I'm just hoping that some
of you can think of quality as we understand it today to make your posts that
much more enjoyable to me.

Thanks in advance, and Best Regards,

Steven R. Schmid

Post by David Thornley

Post by mrbill
3) USSR
Russia equipment was generally crude but they produced lots of it. If
a T-34 broke down there were 5 more to replace it. Soviet quality
(probably never great) suffered by the huge disruption and massive
relocation east.

Another thing to consider is the Soviet attitude.
In the Soviet Union, a tank only had to last six months. If, by accident,
it survived that long, it could always be replaced. The Soviets were
pretty ruthless in their design, to try to make functional weapons and
vehicles and such with minimum resources.
Of course, the quota system characteristic of crude centrally planned
economies didn't help. If a factory was tasked with producing five
hundred trucks, then five hundred trucks would be produced, whether of
usable quality or not.

Post by mrbill
1) Germany
German quality has always been excellent but it took a nose dive as the
war went on. In particular, employing slave labour with a vested
interest in sabotaging the production had a profound affect on quality.

Much of the quality was excessive. An engineer friend of mine looked
over German equipment, and found that much of it was of the highest
standards, with a good many resources used to give them a very nice,
but militarily useless, finish. (US manufacturing came between German
and Soviet in this regard, and as the US had more and better tools to
begin with, the US could produce somewhat higher-quality stuff without
undue labor or expense.)
Of course, some things were low-quality to start with. German torpedos
had serious problems early in the war, and Bismarck fired apparently
low-quality shells. She hit Prince of Wales something like ten times
without a full shell detonation.

Post by mrbill
2) Italy
Il Duce had no business sending men off to fight with the equipment
they had. It's a bit surprising to me as Italian machine tool equipment
is top notch at present.

Some Italian weapons were ridiculous, some well-designed. Most of it,
in 1940, was obsolescent at best. Late war Italian fighters, with German
engines, were excellent.
Italy was waging war beyond its means, and when quantity goes up too
much, quality suffers.

Post by mrbill
3) Japan
Until the 60s the Japanese had a reputation for horrible quality. As
their supply lines were cut and their cities bombed their quality could
only become worse. I suspect that few aircraft even came close to the
performance figures typically cited.

The Japanese had other problems.
The honorable thing to do was to fight; specifically, to attack. This
meant that manufacturing war materiel was not a particularly honorable
or highly thought-of activity. Any attempt at mass-production would
generally be undermined by feature change requests from the front.
After all, if the warriors want something, the mere workers should
comply.
It wasn't until well after WWII that the Japanese finally realized
that making things could be important.
--
David H. Thornley | If you want my opinion, ask.
http://www.thornley.net/~thornley/david/ | O-

[h***@bbs.cpcn.com]

Post by Steven Schmid
The modern concept of quality is pretty advanced, but remember that the WWII
generation did not benefit from the ideas of Deming or Taguchi. The engineers
at the time did have some tools like control charts and the like (but not
often), and nothing like six sigma or anything.

Some basics of statistical quality control were developed directly out
of WW II.

Post by Steven Schmid
They didn't need it. The key
to manufacturing in the 1940s was production, not quality.

They sure DID need quality. Weapons that fail, planes that crash are
all worthless and dangerous to the troops as well as being a waste of
resources. Many industries pressed into war production had to adapt to
new production techniques and standards.

The auto industry was used to make aircraft. There are many
differences between mass production of automobiles and of aircraft,
indeed, aircraft until then were more custom built in limited
quantities. It is well documented how the production engineers of the
two industries "butt heads" until they worked out their differences.

Post by Steven Schmid
Don't
take for granted that everyone knew how to mass produce products, even what
we'd call low quality products.)

There were a number of industries in the U.S. that mass produced
products or sub-components of products. Lionel toy trains, for
example, and Lionel was used to make tiny engines for servo mechanisms
using its experience of model engines. Household goods from clocks to
refrigerators to radios were all mass produced by large companies.

Post by Steven Schmid
As for WWII production on a national basis, I think it would be really
fascinating if some historian tried to analyze this.

As mentioned in another post, there are obscure books on this under the
subject heading of "economic mobilization" of the war. I've read
some, but can't recall their titles. Some books might have a
bibliography that will refer to other books. Older books on
"operations research" may have some war history included. (When I had
that in college they told us it was developed out of the war along with
numerous management techniques.)

The Groueff (sp?) book on the Manhattan Project goes into great detail
about the industrial supply side. There were numerous small industries
that contributed skills and parts and the author details them. I would
recommend that book. (The author does do some cheerleading, but his
info is not available elsewhere.)

I know Life, Newsweek, and Time magazines had articles during the war
on production, as well as detailed advertisements on production.
(Those magazines back then tended to be for the elite and ads were
targeted toward such people who would appreciate them.) I would
suspect Fortune magazine would have detailed articles on production
issues. All of those should be available on microfilm in larger
college libraries and I recommend reading them.

[Dave Smith]

Post by h***@bbs.cpcn.com
They sure DID need quality. Weapons that fail, planes that crash are
all worthless and dangerous to the troops as well as being a waste of
resources. Many industries pressed into war production had to adapt to
new production techniques and standards.

There was a definite need for quality, but war time was seen by
some as an opportunity for profit, and some companies lobbied
hard for sales. Recent posts on "precision bombing" reflect on
one of them. The B17 with the Norden bomb site was sold to the
air force as an incredibly accurate bombing system that failed to
measure up to expectations. It did not perform anywhere near as
well in under under normal weather conditions as it had under the
ideal weather conditions.

Then there was the issue of ease of manufacture. The Sherman tank
was an example of that. It was a pretty good tank, but not a
great one. It's major advantages included being mass produced in
very large numbers and with standardized parts. It may not have
been a match for Germany's heaviest tanks, but there were enough
of them to overwhelm the enemy and win a war of attrition.

[Geoffrey Sinclair]

Post by Dave Smith

Post by h***@bbs.cpcn.com
They sure DID need quality. Weapons that fail, planes that crash are
all worthless and dangerous to the troops as well as being a waste of
resources. Many industries pressed into war production had to adapt to
new production techniques and standards.

There was a definite need for quality, but war time was seen by
some as an opportunity for profit, and some companies lobbied
hard for sales.
Recent posts on "precision bombing" reflect on
one of them. The B17 with the Norden bomb site was sold to the
air force as an incredibly accurate bombing system that failed to
measure up to expectations. It did not perform anywhere near as
well in under under normal weather conditions as it had under the
ideal weather conditions.

There was another factor, to produce so many bomb sights tolerances
had to be relaxed, the wartime Norden sites had something like 5 to
6 times the aiming errors than the pre war version.

See America's pursuit of precision bombing by McFarland.

Ironically once the 8th Air Force went to bomb on the leader many
bombers left their sights at home to save weight.

Other factors, (US day raids)

100 pound bombs had around 5 times the average error of 2,000 pound
bombs, 500 pound bombs around 3 times the average error of a 2,000
pound bomb.

Weight of attack, given all aircraft could not bomb together, the 5th
box of bombers to attack had around 1/3 the accuracy of the first box.
If the time between boxes was increased from 2 to 25 minutes the 5th
box accuracy improved by 178%. Table is box, normal accuracy (boxes
that put at least 10% of their bombs within 1,000 feet of the aiming point),
increase in accuracy going from 2 to 25 minutes between attacks, all figures
are percentages.

1 / 82 / 0
2 / 60 / 32
3 / 48 / 58
4 / 47 / 105
5 / 30 / 178

It explains why it was hard to obliterate a large target.

Overall at 10,000 feet going from 3 to 15 boxes of bombers increased the
expected error from 570 to 765 feet, at 29,000 feet where altitude errors
are more important the error went from 1,605 to 1,700 feet.

Post by Dave Smith
Then there was the issue of ease of manufacture. The Sherman tank
was an example of that. It was a pretty good tank, but not a
great one. It's major advantages included being mass produced in
very large numbers and with standardized parts. It may not have
been a match for Germany's heaviest tanks, but there were enough
of them to overwhelm the enemy and win a war of attrition.

If I understand things correctly the Panther was designed for mass
production much more than the Panzer IV and it well may have been
cheaper despite being 50% heavier.

Generally the mid war designs tended to take into account ease of
manufacture more, particularly for aircraft and tanks, the relatively
new kids on the engineering block.

Quality in this particular thread really has three components,

1) How good was the particular design in relation to the best in its
field, and for how long, everything ultimately became obsolete.
2) How reliable was the design, how easy was it to break or simply
end up under performing.
3) How easy was it to manufacture.

The Spitfire is a good case of point 1 coming out well in front of
point 3. The Ju88 seems to have been well represented in both
points 1 and 3. Both types seem well represented in point 2.

The US aircraft industry was hit hard by the depression, in 1933 some
1,324 military and civilian aircraft were built, in 1935 the total was
1,710,
1936 was 3,010, 1937 was 3,923, 1938 was 3,623. In 1944 it was 95,272.

The cumulative total US aircraft production from 1903 to the end of 1938,
that is around 36 years was around 76,000. By the end of 1939 it had
reached
around 102,000, by the end of 1940 the total was around 151,000. So in 1940
the US manufactured 1/3 of the cumulative total US production from 1903
onwards. Not surprisingly the vehicle manufacturers had plenty to tell the
aviation industry about mass production and also not a surprise was the
aircraft industry planning for relatively short production runs. After all
with technology evolving so fast in the 1930's aircraft designs were
becoming
obsolete much quicker than before.

By the looks of it Douglas produced 10,654 DC-3, 430 "pre war" from
1935, for the civilian market, then another 149 originally civilian orders
were impressed into military service, there were 28 DC-3D models
produced post war for civilian aviation, the rest, some 10,047 were
military orders. Some 10,323 DC-3 were built between January 1940
and August 1945, leaving something around 300 built in the 5 years 1935
to 1939.

Geoffrey Sinclair
Remove the nb for email.

[Brad Meyer]

Post by h***@bbs.cpcn.com
They sure DID need quality.

Not as much a quantity though.

Post by h***@bbs.cpcn.com
Weapons that fail, planes that crash are
all worthless and dangerous to the troops as well as being a waste of
resources.

OTOH, I would rather have a thousand of something of indifferent
quality theyn I would a dozen of something of first class quality.
Quantity has a quality of its own.

Post by h***@bbs.cpcn.com
The auto industry was used to make aircraft. There are many
differences between mass production of automobiles and of aircraft . . .

??? Actually, very few and even leass back then when both had frame
and skin construction.

Post by h***@bbs.cpcn.com
indeed, aircraft until then were more custom built in limited
quantities.

Nonesense. Boing and a few other manufacturers had been building
serial quantites of their products for over a decade.

Post by h***@bbs.cpcn.com
It is well documented how the production engineers of the
two industries "butt heads" until they worked out their differences.

It is also well documented that almost all of those differences were
personal rather then professional. Not unlike the "patent wars"
nonsense that was going on during WW I.

[Rich Rostrom]

Post by h***@bbs.cpcn.com

Post by Steven Schmid
They didn't need it. The key
to manufacturing in the 1940s was production, not quality.

They sure DID need quality. Weapons that fail, planes that crash are
all worthless and dangerous to the troops as well as being a waste of
resources.

There had to be trade-offs between quality
and quantity.

For instance, which is more useful to an
artillery battery?

500 shells, 20% of which are duds

or

200 shells, 1% of which are duds

Unless it is important that individual shells
go off, the higher dud rate is acceptable, in
return for more than twice as much ammunition.
The effective ammunition supply is 400 shells
versus 198 shells.

Which is more useful to an armored brigade?

50 tanks, of which 50% will break down every 200 km

or

10 tanks, of which 10% will break down every 1,000 km

The first group of tanks would be severely attritted by
mechanical failures, and could never sustain a deep
breakthrough. OTOH, in the opening stages of a battle,
there would be 50 tanks fighting. The second group
could never have more than 10 tanks fighting, insufficient
strength to make a breakthrough of any kind.


Which is more useful to a company of infantry?

200 rifles of which

100 are accurate to within 10 cm at 200 meters

50 are accurate to within 10 cm at 400 meters

50 are accurate to within 10 cm at 800 meters

or 150 rifles of which

100 are accurate to within 5 cm at 400 meters

50 are accurate to within 5 cm at 800 meters

(accurate defined as 10 of 10 rounds hitting a target
of that size at that range when fired from a test stand).

Few infantrymen are good enough shots to achieve even
the lowest accuracy rating listed here. Most would be
of little more effectiveness with the best rifle; and
choosing the second group means 50 men are left unarmed.

[Andrew Clark]

Post by Rich Rostrom
There had to be trade-offs between quality
and quantity.

I have snipped your useful examples purely for brevity.

Your thesis is that better reliability could only be achieved with a major
sacrifice to quality. Is that thesis justified?

For example, some (by the standards of the time) reliable weapons systems
like the Spitfire or Sherman were produced in dizzying-large quantity. The
Lee Enfield rifle, I understand, was also very reliable and produced in
quantity, while the notoriously unreliable early model Sten was actually not
produced in volume compared to the sufficiently reliable later versions
which were produced by the million.

It seems to me that the tradeoffs between quality and quantity is not so
marked as you suggest, or at least that the consequences could be
ameliorated by good design and manufacture.

[Brad Meyer]

On Wed, 01 Nov 2006 06:15:37 -0500, "Andrew Clark"

Post by Andrew Clark

Post by Rich Rostrom
There had to be trade-offs between quality
and quantity.

I have snipped your useful examples purely for brevity.
Your thesis is that better reliability could only be achieved with a major
sacrifice to quality. Is that thesis justified?

Most definately!!

Post by Andrew Clark
It seems to me that the tradeoffs between quality and quantity is not so
marked as you suggest, or at least that the consequences could be
ameliorated by good design and manufacture.

Look at your sten example. At some point there had to be a shutdown in
production in order to incorperate changes on the line. That is a
quantity hit of the worst sort.

If we are building 1500 widgets a month and can substantially increase
reliability at the cost of three months production someone has to
decide if we can afford that loss in order to get the increase. German
a/c production was regularly caught in this sort of bind. Are 1500
indifferent widgets today more or less valuable then 1500 fist class
widgets sometime next quarter?

[Rich Rostrom]

Post by Andrew Clark

Post by Rich Rostrom
There had to be trade-offs between quality
and quantity.

I have snipped your useful examples purely for brevity.
Your thesis is that better reliability could only be achieved with a major
sacrifice to quality. Is that thesis justified?

I make no such claim. I just suggested some
completely imaginary scenarios in which a
trade-off of quality for quantity would be
justified.

There are other cases where it would be much
less justified. For instance, a 20% dud rate
in naval torpedos could be disastrous, and
the supply requirement for torpedoes was low.

Also, the difficulty of getting a torpedo
into firing position was such as to exceed
greatly the cost of the shot, and in general
one got only one shot.

Post by Andrew Clark
It seems to me that the tradeoffs between quality and quantity is not so
marked as you suggest, or at least that the consequences could be
ameliorated by good design and manufacture.

One must always optimize design and manufacture
to get the best results in both areas. Actually
there are three areas: quality, quantity, and
availability (when is the stuff delivered?).

Software developers call this the two out of
three rule. End users want the software to
work well, to cost not much, and to be ready
quickly. Good, cheap, fast - pick any two.

In the question of war equipment production,
there may be issues like: the equipment _must_
be available in quantity in six months so that
men can train with it for an operation that
is planned for a year away. But the present
design has manufacturing problems that result
in a high defect rate. The engineers say they
can rework the design to eliminate these problems,
but it will take three months. That's three
months before tooling can be produced,
production workers trained, and supplier
arrangements finalized. The first units
will come out three months after that.

If the delay for redesign is accepted, the
equipment won't be available in six months
for training...

[BernardZ]

Post by mrbill
WWII movies, novels, and even historical books or documentaries rarely
have any mention of the poor manufactured quality of WWII era
equipment.

I have seen a read a few books that do discuss this issue.

The items were often tested hopefully before battle and if faulty, were
taken to what most sides had plenty of brilliant mechanics.

Of course some items could not be tested like US grenades that did have
a history of blowing up before the correct time, so killing the
throwers.

[Andrew Clark]

Post by mrbill
WWII movies, novels, and even historical books or documentaries rarely
have any mention of the poor manufactured quality of WWII era
equipment. In movies guns almost always fire, engines always start,
radio sets don't work when they've been shot or other wise damaged, and
almost nothing breaks down.

Well, one big-budget movie which does at least capture some of the issues is
a Bridge Too Far, which correctly records that hardly any of the British
backpack tactical radios used by 1st AB actually worked for very long in the
field, due to poor manufacturing quality.

[mrbill]

Post by Andrew Clark

Post by mrbill
WWII movies, novels, and even historical books or documentaries rarely
have any mention of the poor manufactured quality of WWII era
equipment. In movies guns almost always fire, engines always start,
radio sets don't work when they've been shot or other wise damaged, and
almost nothing breaks down.

Well, one big-budget movie which does at least capture some of the issues is
a Bridge Too Far, which correctly records that hardly any of the British
backpack tactical radios used by 1st AB actually worked for very long in the
field, due to poor manufacturing quality.

Interesting, I didn't pick that up from the movie. I'd understood that
the problem was due to the frequencies used by the British radios.
That certainly had a significant impact on the way the battle at Arnhem
played out.

[james]

Post by Andrew Clark

Post by mrbill
WWII movies, novels, and even historical books or documentaries rarely
have any mention of the poor manufactured quality of WWII era
equipment. In movies guns almost always fire, engines always start,
radio sets don't work when they've been shot or other wise damaged, and
almost nothing breaks down.

Well, one big-budget movie which does at least capture some of the issues is
a Bridge Too Far, which correctly records that hardly any of the British
backpack tactical radios used by 1st AB actually worked for very long in the
field, due to poor manufacturing quality.

Funnily enough, I saw one of those Batttlefield Detectives episodes
recently that went into this very issue. They actually took both modern
and those particular sets in question to the landing field and through
experimentation, discovered that the radios functioned, until they
reached a certain range, which may be the difference between the ideal
in lab tested range, and the practical range when you add trees,
buildings, atmospheric interference and other factors. The new sets
functioned better of course but were not immune. They even employed
WWII vets to operate the sets. So it wasn't manufacturing, it was
design- they didn't have enough range.

James

[Louis C]

So it wasn't manufacturing, it was design- they didn't have enough range.

Anyone using radios will know that the theoretical range is only that -
theoretical.

I had a vehicle-mounted radio with a theoretical range of 10km. My
platoon's radios were supposed to be the same. 10km is what we would
usually get on level terrain (plains with a few trees and habitations),
though generally it was closer to 8. In hilly terrain, or in built-up
areas, the range could drop to about half. Stationing one's vehicle on
the top of a bald hill and transmitting with fully-charged batteries
(i.e. the engine had been running for a while and you hadn't been
transmitting too much), one could exceed the theoretical range - I have
in the back of my mind a personal record of 14km.

In practice, we were usually spread out on maneuvers (recon platoon)
and I considered the effective range to be 5-7km i.e. I would expect to
be out of touch with parts of my platoon if I sent them further (i.e. I
would plan meeting points rather than rely on the radio to tell them
where to meet).

The infantry had similar sets, except that their performance was
inferior because the vehicle-mounted sets practically never operated
with depleted batteries. Higher HQs obviously had more powerful sets.

At Arnhem, we are talking about WWII-era man-portable sets. These are
not very powerful sets and their batteries will quickly run out.
There's absolutely nothing surprising in the airborne getting out of
touch. Had 82nd Airborne been dropped there instead of 1st Airbourne,
things would have been the same. This wasn't a problem with
manufacturing quality.

LC

[Andrew Clark]

Post by james
Funnily enough, I saw one of those Batttlefield Detectives episodes
recently that went into this very issue. They actually took both modern
and those particular sets in question to the landing field and through
experimentation, discovered that the radios functioned, until they
reached a certain range, which may be the difference between the ideal
in lab tested range, and the practical range when you add trees,
buildings, atmospheric interference and other factors. The new sets
functioned better of course but were not immune. They even employed
WWII vets to operate the sets. So it wasn't manufacturing, it was
design- they didn't have enough range.

Yes, I recall seeing that episode too.

As I recall, 1st AB at Arnhem found that any moderate thump or jolt threw
the radio settings and dislodged the crystals etc, meaning lots of resetting
and repairs. Many never subsequently worked. At the operational inquest,
this was found to be due to various technical defects in manufacture.

Of course, even when the portable sets did work, range of transmission and
reception was a big issue due to the lack of high aerials. In addition, the
big long-range VHF radio sets, mounted on special jeeps or Veeps, were
damaged on arrival.

[h***@bbs.cpcn.com]

Post by Andrew Clark
As I recall, 1st AB at Arnhem found that any moderate thump or jolt threw
the radio settings and dislodged the crystals etc, meaning lots of resetting
and repairs. Many never subsequently worked. At the operational inquest,
this was found to be due to various technical defects in manufacture.

Was the weakness to bumping really a "manufacturing defect" or a design
"limitation"?

To use an old analogy, we could build automobiles far safer and long
lasting than the ones now, except they'd be so expensive few could
afford them. Today's autos are a compromise between safety features,
lifespan, and cost.

So, would it have been practical to mass produce a radio set in WW II
so durable that it could take a large amount of physical abuse without
dislodging the crystals, AND, be not so big that a man could carry it?
I'm not sure that was technically possible back then. Likewise with
range--given the batteries available then and antenna in combat, how
much range was realistically possible in a hand carried device? Given
all the research conducted in radios by the Signal Corps and
manufacturers (RCA, Bell Labs, etc), I suspect the field radios
delivered were indeed state of the art for their time.

Note--the green book Signal Corp history discusses this, as does the
Bell Labs History "Service in War & Peace".

[Andrew Clark]

Post by h***@bbs.cpcn.com
Was the weakness to bumping really a "manufacturing defect" or a design
"limitation"?

(snip interesting commentary)

As I recall (it was some time ago), the wireless sets had some sort of
rubber buffer system to help prevent jarring and dislocation of components,
but this had been badly fitted. So the issue was manufacturing quality, not
design limitation.

Dimly, I think I remember the fitting issue was similar to one solved by the
US when they took over the Bombe manufacture.

[h***@bbs.cpcn.com]

Post by mrbill
WWII movies, novels, and even historical books or documentaries rarely
have any mention of the poor manufactured quality of WWII era
equipment. In movies guns almost always fire, engines always start,
radio sets don't work when they've been shot or other wise damaged, and
almost nothing breaks down.

Movies make great use of something called "poetic license". Having
broken equipment when you need it the most adds to the dramatic effect.
Movies are not a reliable source of information in a question such as
this.

Post by mrbill
The truth is much different. WWII
manufactured quality was abysmal by our standards and had a significant
affect on the outcome of battles, maybe even campaigns.

What is the basis (source?) for your assertion? What exactly do you
mean "by our standards"?

If you mean standards of today, of course things will be different.
Technology has changed radically over the last the 60 years. For
example, radios back then used vacuum tubes and soldered circuitry.
Radios were not designed to be bounced around on beaches and filled
with sand and salt water. Presumably military radios were "hardened"
to the best extent possible, but still there were glass tubes and
circuits and they can take only so much shock.

If you mean standards of the war years, I would strongly question your
assertion. As far as I know, US built military equipment performed
equally or better than comparative civilian equipment of the time. If
it was new technology developed fresh for the war, then naturally there
will be teething pains as experience is gained.

For example, the jeep was beloved by soldiers and the C-47 airplane
was very rugged.

Post by mrbill
Perhaps this is one aspect of WWII that still requires study.

There are a number of books and essay papers discussing the production
issues, some papers go into considerable detail. However, this
material is obscure and difficult to find and quite dry. The National
Archives undoubtedly has quite a bit of the paperwork on this issue but
it would take much tedious study to go through it and properly
summarize and assign weights to the statements shown.

The official Army green books on the Quartermaster and Transportation
Corps discuss some of the issues, including the challenges they faced
and how they met them.

One challenge was the long shipping route. Goods were built in the
U.S., shipped to a port, waited for a ship, then shipped overseas, then
shipped again by rail or truck to the front. A lot could--and
did--happen to materiel on the way. Waterproofing from salt air was a
big challenge, the common home product Saran Wrap was developed for
that, to give one example.

[Cubdriver]

Post by h***@bbs.cpcn.com
For
example, radios back then used vacuum tubes and soldered circuitry.

We tend to forget how comparatively primitive 1940s gear was. Robert
Mikesh published a book, Japanese Aircraft Equipment 1940-1945
(Schiffer, of course) with color photos and data on altimeters,
compasses, guns and sights, etc. I've spent hours with those
photographs. The better you realize how they worked, and how they
fitted together in the cockpit or fuselage, the more astonishing it
seems that men actually *flew* those things, let alone *fought* in
them.

On the other hand, primitive also has its virtues. I fly a 1946 Piper
Cub with the same instruments as the plane had when it came out of the
factory at Lock Haven sixty years ago (I do have a GPS
suction-cup-mounted to the port window, and a hand-held radio tucked
in the seat-back pocket in front of me). Everything works as it should
except the off-on switch for gasoline: we are told never to turn it
off, because the valves are too expensive to replace. (In a Cub, for
example, the fuel-quantity gauge consists of a varnished cork with a
wire piercing it; the wire projects through a hole in the fuel cap
and, in theory, tells you how much fuel remains.)

[Brad Meyer]

On Tue, 31 Oct 2006 11:15:34 -0500, Cubdriver

Post by Cubdriver

Post by h***@bbs.cpcn.com
For
example, radios back then used vacuum tubes and soldered circuitry.

We tend to forget how comparatively primitive 1940s gear was. Robert
Mikesh published a book, Japanese Aircraft Equipment 1940-1945
(Schiffer, of course) with color photos and data on altimeters,
compasses, guns and sights, etc. I've spent hours with those
photographs. The better you realize how they worked, and how they
fitted together in the cockpit or fuselage, the more astonishing it
seems that men actually *flew* those things, let alone *fought* in
them.
On the other hand, primitive also has its virtues. I fly a 1946 Piper
Cub with the same instruments as the plane had when it came out of the
factory at Lock Haven sixty years ago . . .

This also points up one of the (IMO undervalued) virtues of slide rule
design. Everything turned out to be more robust then it had to be in
order to make certain it was as robust as it had to be mostly because
of the limits of accuracy of slide rules and the length of time needed
to perform some of the calculations involved.

I think sometimes people forget (or never consider) that today
Opal-Anne Office Helper spins up her Excel and does things in a
handful of keystrokes that would have taken a team of accountants a
week to do by hand 60 years ago, and also quite possibly does
something that could not be done with paper and pencil in any
practical amount of time by any amount of people.

Say a body was designing a bridge that had to be able to hold X amount
of weight. He does his calculations and makes sure all his errors are
off in the same direction (what we Nukes called RadCon math) and ends
up with a bridge that will hold at least X but is unable to say
accurately how much more then X it will hold. Nowdays, with the aid of
computers the errors are minimized and one end up with a design that
will hold X, plus or minus 7.235 micrograms.

[Cubdriver]

Post by h***@bbs.cpcn.com
For example, the jeep was beloved by soldiers and the C-47 airplane
was very rugged.

The C-47 however was not designed for the military. It was a DC-3
commercial airliner with the interior ripped out, an olive-drab paint
job, and (I think) a wider door.

I've always thought it said something about America that its stellar
WW2 equipment--the jeep and the C-47--were both essentially civilian
items.

[Dave Smith]

Post by Cubdriver
The C-47 however was not designed for the military. It was a DC-3
commercial airliner with the interior ripped out, an olive-drab paint
job, and (I think) a wider door.
I've always thought it said something about America that its stellar
WW2 equipment--the jeep and the C-47--were both essentially civilian
items.

The only snag there is that the jeep was not originally a
civilian vehicle. It started off as a vehicle manufactured by the
America Bantam Car Company for the US army in 1940. The US army
has contracted for 70 of them. The US Army did not think the
company was big enough to supply the vehicle in the numbers they
would be needed. Willys won the contract with there vehicle,
which was similar to the Jeep but better. Ford got a contract for
a similar vehicle.

After the end of the war the jeeps became more common as civilian
vehicles. Not only had they proved themselves through their
performance in the war, but the market had been flooded with army
surplus jeeps.

[mrbill]

Post by h***@bbs.cpcn.com

Post by mrbill
The truth is much different. WWII
manufactured quality was abysmal by our standards and had a significant
affect on the outcome of battles, maybe even campaigns.

What is the basis (source?) for your assertion? What exactly do you
mean "by our standards"?

Actually I posted that in my original email. World Class Manufacturing
now has an expectation of defects at less than 50 parts per million.
The 5 and 10% AQL levels that followed WWII were an improvement over
the inconsistent quality of the war years. So, the defect level is
somewhere around 1,000 times less than it was postwar and during the
war.

Post by h***@bbs.cpcn.com
If you mean standards of today, of course things will be different.
Technology has changed radically over the last the 60 years. For
example, radios back then used vacuum tubes and soldered circuitry.
Radios were not designed to be bounced around on beaches and filled
with sand and salt water. Presumably military radios were "hardened"
to the best extent possible, but still there were glass tubes and
circuits and they can take only so much shock.

Damage in transit is a quality problem as inadequate packaging is a
quality issue.

Post by h***@bbs.cpcn.com
If you mean standards of the war years, I would strongly question your
assertion. As far as I know, US built military equipment performed
equally or better than comparative civilian equipment of the time. If
it was new technology developed fresh for the war, then naturally there
will be teething pains as experience is gained.
For example, the jeep was beloved by soldiers and the C-47 airplane
was very rugged.

Many of these products were successful because they were massively
overdesigned not because they were well manufactured.

By the standards of the day they were well manufactured.