soc.history.war.world-war-ii

"Louis C" wrote in message
news:86fdb2a5-5f4e-4695-b1b3-c84be43fd5b2@d4g2000prg.googlegroups.com...
> Geoffrey Sinclair wrote:
>> Pretty much, the original contracts for the Hurricane and Spitfire
>> were from before things became urgent. So 600 Hurricanes and
>> 310 Spitfires and the designs were years old before production
>> began. Similar for the early designs of all air forces.
>
> Yes, but that wasn't what I questioned.
>
> Never mind, as you address the issue below.
>
> And indeed, it seems that French and US aircraft manufacturers
> believed until early 1939 that they might be experiencing a
> speculating bubble in aircraft construction, which made them reluctant
> to invest in mass-production equipment.

Commercial firms making rational commercial decisions.

In Germany what happened to Hugo Junkers was a powerful lesson
for the aviation industry to heed.

The new German government wanted lots of warplanes and were
prepared to finance the expansions much earlier than other
governments.

> The various incentives that the US government had to take to make
> American industrials invest are well-documented in a variety of works,
> most of which are now online, and that was after Britain and France
> had paid directly for some of the additional capacity.

Not surprising, the depression made for a spend little mindset.

The British ended up with the shadow factory, basically paid for
by the government but run by a firm that was supposed to know
what it was doing.

By the way, which of the US histories are you thinking of?

> The story in Britain and France was the same, and I assume Italy, too.
> Not sure about the Soviet Union though that would likely not be all
> that well-documented. What about Japan?

A shooting war starting in 1937 would be both a help and hindrance,
the war cost money but required more aircraft. The IJAAF had its
1939 shock of meeting the Red Air Force.

Italy had a government that wanted a large military earlier than
Germany, with a government that was willing to spend the money
and then discovered how good the equipment park was in 1939.
And how difficult it was to make lots of better equipment.

> (snip more agreement about 1937-38 being essentially "armed peace"
> production orders)
>
>> Then came the urgency as the bomber would always get through
>> and the Germans were ahead. Which mandated numbers and
>> second or more sources of supply.
>
> It seems that the Germans being ahead was more important a factor than
> the bomber always getting through, after all if the latter had been so
> important air rearmament would have focused on masses of bombers for
> an early version of the balance of terrror.

As far as I know that's what was going on in Britain, the problem being
the bombers cost more and took longer to put into full production.
The situation is complicated by the way the RAF liked light bombers,
with fond memories of some WWI successes.

In 1938 there were 160 medium and 538 light bombers to 371 fighters
built, in 1939 758 medium and 1,079 light bombers to 1,324 fighters.
Fighter Command's increasing confidence helped move the production
mix but I doubt it had much effect on the pre war production. The way
fighters were cheaper helped, their earlier arrival helped the numbers game.

Germany was about the same ratio, the USSBS says 1,541 Bf109,
315 Bf110 to 557 single engined and 2,314 twin engined bombers in
1939, the difference being the Germans were building bigger bombers
on average.

>> As the main powers rearmed so export orders were generated
>> as other countries felt the need for more aircraft.
>
> Agreed, though the export market remained too small to constitute a
> shift in production. Now, it was large enough that national producers
> couldn't fully supply it, but compared to national orders export
> orders were small: 50 aircraft here, 100 there, can't think of orders
> larger than 200 at the moment.

Agreed, I was thinking in terms of mind set, it was giving the industry
more confidence with full order books and still more people wanting
to buy. So after various requests for aircraft in a memo dated
November 1938 the British Foreign Office noted its preferences for
foreign Spitfire sales, in order,

France, Belgium, Estonia, Turkey, Romania, Portugal, Switzerland,
Yugoslavia, Holland, Greece, Bulgaria, Iran and Lithuania. Most
wanted about a squadron's worth, the biggest order was Turkey's
one for 60. All up it comes to 220 to 292 aircraft, near doubling
the initial RAF order.

As well by August 1938 Switzerland, Belgium, Holland and
Yugoslavia had applied for production licences.

In 1939 came China, Norway and Finland

>> Ah, I see, what I need to do is separate out the orders for mass
>> production and the efforts built into the designs to make them
>> easy to produce.
>
> See the paragraph of yours that I quoted in my previous post. I
> understood it to mean that Germany moved to mass-production earlier
> than other powers because the latter were still in peace-time mode.

If you define peace time mode as expecting an order for hundreds
of the new type after which another design would be produced to an
order of hundreds, then I agree.

As noted the rapid evolution of aircraft performance in the late
1930's seemed to mandate new designs, but the need for numbers
now mandated building what was already in production.

> My understanding is that in many cases (originally "most" though that
> changed) the aircraft industry was simply not capable of mass
> production so that was a secondary concern.

Agreed.

> Metal planes were still a
> relatively new technique, with the pros and cons of metal vs wooden
> construction not clearly being in favor of the former.

Agreed. Though I would say stressed skin versus fabric covering
rather than wood.

> Metal plane
> construction was more expensive and only became effective if really
> large orders were placed. Wooden plane construction was quicker (for
> small series), easier to set up and less expensive. For the purpose of
> the discussion, I'm deliberately ignoring nitpicks about no aircraft
> being entirely of wooden construction etc.

Not sure about wood being cheaper to build, the presses DeHaviland
used to build Mosquito fuselages do not look cheap. The few indications
I have about Mosquitoes are they were "hard" to build

I do agree all metal was new and the industry was having enough
problems moving to the new standard, it was a lot to expect them
to also add ease of manufacture of the new standard at the same
time especially given years of small orders.

>> Since of course different designs had different
>> levels of production difficulties and judgements had to be made
>> about whether the complexity was worth it.
>
> Precisely.
>
> My understanding - which could be wrong - is that ease of production
> didn't become an issue until fairly late, exceptions being the two
> star projects in Germany.

Correct, it had to be "second generation", with the evidence of the
problems of building the first generation in numbers coupled with
the clear evidence orders were going to be in the thousands. Then
there had to be some sort of encouragement to make the effort.

> Everywhere else, the industry was playing
> catch up so when a promising type appeared it was ordered in
> quantities that were too small to make true mass production an issue
> though they would often be large enough to severely strain the
> manufacturer's capacity.

As far as I know the sudden requirements for massive increases in
output strained every private firm. In Germany the government was
willing to pay, and pay well, much earlier, and also quite ruthless
about what it wanted. Junkers was officially worth 130 million marks
in 1936, and effectively owned by the Government according to Antony
Kay.

In England the aircraft output was 893 in 1935, it doubled to 1,830
in 1936, grew to 2,218 in 1937, then 2,837 in 1938 then 7,938 in
1939, compared to the wartime peak of 26,461 in 1944. With
production being 20,074 in 1941. The sort of investment to do this,
even allowing for multiple shifts using the same machinery, was simply
beyond the private firms.

>> The Hurricane was rated as easy to produce, much more so than
>> the Spitfire.
>
> Two caveats in the figures that follow. First, they're from memory and
> I didn't bother to look them up (I do have something more precise that
> I can look up if this becomes an issue). Second, the usual warning
> about what different organizations put in nominally similar
> indicators.
>
> Man-hours to build
> Spitfire: 13,000
> Hurricane: 8,000
> Bf 109: 5,000

I will add the other warning, they need to be from the same point
in the manufacturing cycle. For example in 1940 the Gloster production
line of Hurricanes was coming on line, 32 built in 1939, 1,211 in 1940.
The Bf109 production in 1939 was supposed to be around 1,541
according to the USSBS versus something like 569 Hurricanes and
431 Spitfires.

The 1940 figures are something like 1,870 Bf109 to 2,521 Hurricanes
and 1,249 Spitfires despite the Castle Bromwich problems and the
bombings.

As another example, sticking to fighters, in the 1939/1941 period the
new P-39 cost about $77,159 each, the P-40 $60,562, by 1944
they cost $50,666 and $44,892 respectively. And that must reflect
a lowering of man hours per aircraft.

In 1939 the official export price for a fully equipped Bf109E
was set between 162,000 and 130,000 marks depending on
the quantity ordered, the higher the order the lower the unit cost.
Just the airframe alone would cost 74,000 to 70,000 marks.

The exchange rate was about 10 marks to the pound. So divide the
above by 10 when comparing the following.

Two Spitfire mark I, sold to Turkey in 1940 were priced at 11,700
pounds with export profit margin, the original price had been
13,000 pounds for a complete Spitfire.

The Spitfire sent to France pre war cost 16,436 pounds including
a spare engine. The Greek order for Spitfire I was at 9,335 pounds
per aircraft.

Of course what we do not know is the profit margins but it appears
the Germans were quite willing sellers as they wanted the foreign
exchange. I note the airframe costs above versus those below.

I do not now how accurate the following is, and it involves the
Germans estimating UK costs, from Jukka I Seppänen

"here are projected costs for the 1,000th a/c of each series,
excluding engine(s) and propeller unit(s) based on a GL Ltd.Chef-Ing.
report of 24.4.41.

Bf 109E = 15,000 RM (the Bf 109F would be the same)
Do 17Z = 41,000 RM
He 111H = 71,000 RM
Ju 88A = 73,000 RM

This was a report comparing the manufacturing times, etc. of German vs.
British a/c in the same category, and using the same criteria.
Spitfire I = 12,500 RM
Blenheim = 26,000 RM
Wellington = 38,000 RM"

>> The point being given the production numbers expected
>> there seemed little need to design the types for mass production.
>
> Agreed. My point is whether Germany anticipating a truly large and
> lasting rearmament effort was a factor in terms of adding a
> requirement for mass production to aircraft design, as opposed to it
> being an inspired guess by a couple of manufacturers.

I would lean to it being somewhat more than an inspired guess,
as it was based on experience, but ease of manufacture does not
seem to be part of any tender documents I have seen, so it was
a more informal requirement.

On what I have picked up I would say the ease of production
came from the manufacturers as a selling point more than built
to a government requirement. Governments wanted top speed,
range etc. Sort of like so many aircraft enthusiasts. Messerschmitt
was an outsider, Martin Baker even more so which makes them
look for more selling points.

>> For a start there is the question of whether the firms understood
>> what they needed to do after a decade or more of 100 aircraft
>> being a major order, Hawkers were the best given their successes
>> as far as the British were concerned.
>
> Additionally, the construction techniques of modern types had evolved.

Agreed.

> On the one hand they lent themselves better to mass production
> techniques, on the other hand it required a massive effort before
> economies of scale kicked in so I'm not sure that being late in
> rearming was such a factor.

Agreed metal parts are better suited to mass production than wooden
ones, though the aircraft industry was largely all metal airframes by the
mid 1930's, the new thing was metal stressed skin, versus fabric
covering of the aircraft.

As far as I am aware there are two factors, firstly having the information
to design an aircraft to be easy to build, previous knowledge was being
over taken by the new construction techniques, which were a struggle
to master let alone make easy to use. Secondly the firm had to decide
it was worth trying for given government was not strongly thinking in such
terms when most of the wartime designs were actually being designed.
As war grew closer and aircraft number requirements grew more attention
was paid to the problem.

As noted before take a look at how many WWII aircraft were around
pre war.

I will add another point it was probably more efficient to build an
aircraft with as few parts as possible. However given combat aircraft
take lots of damage I suspect a few big parts would be a real problem
for the maintenance people.

And another point, the per aircraft costs came down anyway once
production runs were lengthened as development and tooling costs
were spread over a larger production. The accountants could see
cheaper aircraft by the method of ordering more.

The same technique applied to even the large warships.

> For what it's worth, the British and French rearmament efforts weren't
> all that far behind the German one, at least chronologically.

Would this apply to monies made available for factory expansion? My
understanding is the Germans received such monies before their rivals.
Years rather than months.

>> For the British the culmination was the MB5, see the mechanics
>> and as far as I know the production engineers reports on the
>> aircraft, the design made things that much easier for them.
>
> Yes, at this point it does seem that design was influenced by
> production concerns as well. The US aircraft industry had been largely
> built by people with experience of rearmament.

Is this referring to the WWI surge? The US had quite a shock when
it realised how far it had fallen behind in 1917 and 1918.

As far as I know the US aviation industry was not thinking in terms
of mass production in the 1930's, the introduction of the vehicle
industry into the aviation industry was a real culture shock.

To put it another way the DC-3 was supposed to be a major pre
war success, in production between 1935 and 1947 some 10,654
were built, but 10,323 of that total was January 1940 to August
1945. And it was 115 built in 1940 and 165 in 1941. Given some
made post war it would appear the major success was something
like 300 sales in 4 years. No real incentive to do mass production.

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 of manufacture, came to around 76,000 aircraft.
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 alone the US manufactured 1/3
of the cumulative all time total US production from 1903 onwards.

>> The idea the one design could be pushed
>> the way the Spitfire was pushed was not considered.
>
> Agreed, though the Bf 109 was still designed with ease of production
> in mind, like the narrow undercarriage that made landings so dangerous
> also meant that the wings could be removed and IIRC engine change was
> much simpler than on contemporary fighters.

Are you sure the undercarriage was arranged like that for the
maintenance requirements or because it helped keep the wing
thinner and eliminated the need for a strong wing structure
around the landing gear to absorb the landing stresses?

Also the Messerschmitt standard for undercarriage was
fixed to the fuselage, it was the general standard for
small aircraft.

It is interesting to see the bulges the Bf109 needed in 1944
and 1945 to accommodate the tyres its weight gains mandated.

> Not sure that this was because Messerschmitt anticipated to build
> 35,000 of them.

Agreed. Remember in basic design the Bf109 was in many ways a
single seater Bf108, the first flight of the Bf108 was in June 1934,
the Bf109 in September 1935. Making the fighter the second all
metal type produced by the company.

Though the Bf108 looked destined for a small production run
until the Luftwaffe decided to make it a standard light communication
type. I am not sure of that order date and it could be well after the
Bf109 order.

So I am not claiming the Bf108 told Messerschmitt how to design the
Bf109 for ease of manufacture. I do note some 341 Bf109 B models
were built, 58 C and 647 D with manufacture of the E model beginning
in late 1938. So the initial contract was for at least a similar number
to the early Spitfire order.

>> The Germans seem to have been unable to switch to good enough
>> arrangements, instead completing parts etc. to tolerances that
>> were not needed.
>
> Part of the problem is that when you have been ahead it takes longer
> for the realization to that you now need to catch up to sink in.

It is also the overnight success factor.

In July 1943 the Luftwaffe knew it was in trouble, then in one month
it was shot out of the skies over Sicily, failed to achieve air superiority
over Kursk and watched Hamburg burn. Then in August came the
Schwienfurt and Penemunde raids, and Jeschonnek's suicide. All
of s sudden it was very obvious things were going very wrong.

In late 1943 and early 1944 the USAAF was desperately scratching
around for high performance fighters to send to Europe. A few
months later the allies have air supremacy in the theatre and all the
other theatres suddenly find they can have all the fighters they want.

The gap between perception of what is going on and what is really
going on and the old it worked last time so try it again.

>> As far as I can tell in the late 1930's Germany was ahead in
>> terms of "knowing" there would be large orders for aircraft
>> coupled with the fact it was ahead in terms of production from
>> earlier orders. This meant more mass production experience
>> and more attention to the problem as the next generation of
>> aircraft were prepared for production.
>
> The Germans would definitely have more mass production experience,
> though the RAF should be catching up in 1940 and yet the two weren't
> organized remotely alike.

Perhaps it would be best to list what you consider the differences
to be since I seem to be missing the point at times.

Both industries were heavily reliant on government money to expand
production, either direct investment in plant or high prices for output.
The RAF was putting its first generation monoplanes in production,
with its 4 engined types in prototype form, the Germans were starting
on their second generation, with things like the Ju88 and Do217.

The RAF had the situation the 4 engined types had to overcome the
problems of being much bigger aircraft versus the current bombers,
the Ju88 and Do217 were more evolution than large change.

> What I wonder is how much of that experienced translated into
> different design requirements.

The answer is the histories do not directly cover it. Things like
ordering multiple prototypes cost more but also save by allowing
more testing earlier is an obvious example from experience. I have
note seen any tender document that had ease of manufacture as
a main point, instead performance figures were mandated, the rest
to follow.

However the air forces had long experience with keeping
aircraft flying and there seem to have been standard terms
regarding this in specifications.

For example section 9 "Miscellaneous" in the RAF F7/30
specification paragraph (j) states the design should allow easy and
rapid maintenance in the field, using standard equipment. Paragraph
(k) notes parts requiring frequent repair or replacement must be easy
to inspect and repair/replace. Paragraph (a) notes the aircraft should
be designed to have quickly detachable units for ease of transport
and storage. This would help wartime repairs as well.

So requirements were there, but later in the list, not first.

One of the sections of the first Spitfire order mandates the wing tips
and end fuselage bay must be easily detachable and they and the wings
must comply with "current interchangeability requirements".

>> I should have been more precise which would have meant making
>> the ambiguities clearer. :-)
>
> This is actually a nice illustration of the problem: you didn't
> anticipate needing the large production run of added precision, sort
> of like myself when Rich accused me of being too vague, unsuspecting
> that I was that this was going to be argued over.

I think to try another attempt,

The ease of production parts of the various designs were more than
accidents or guesses but seem to have come more from industry, which
after all had the experience of building the aircraft, rather than in
response
to government pressure. Which explains why different companies had
different results, but it required experience in the new construction
techniques and a belief the extra design work would pay off due to
large orders to push the ideas along.

Governments seem to have decided they wanted performance,
then ease of maintenance, then ease of manufacture. Though
clearly ease of manufacture means cheaper aircraft. As noted
before the ordering of large production runs did cut the
individual aircraft cost, and that gave Governments the appearance
of manufacturing efficiency without them having to do very much.
The aircraft might be hard to build but the machines needed to
build it are being written off over more aircraft.

One other reason can be seen from the following table,

Effort in man hours, Spitfire production, mark / design / jigging
and tooling

I / 339,400 / 800,000
II / 9,267 / unknown
III / 91,120 / 75,000
V / 90,000 / 105,000
VI 14,340 / 50,000
IX 43,830 / 30,000
XII / 27,210 / 16,000
VII / 86,150 / 150,000
VIII / 24,970 / 250,000
XIV / 26,120 / 17,000
21 / 168,500 / unknown
PR XI / 12,415 / unknown
Seafire I / 10,130 / 18,000
Seafire II / 3,685 / 40,000
Seafire III / 8,938 / 9,000
Seafire XV / 9,150 / unknown
Spitfire on floats 22,260 / 35,000

The Spitfire 21 introduced an entirely new wing.

> (snip again)
>
>> Could some French examples be supplied?
>
> They wouldn't be very indicative because the two largest French
> production runs were only a little over 1,000 each for the Morane 406
> and the Potez 63-11, though the Dewoitine 520 was certainly set up for
> a larger run and so were modern bombers.

I suppose for this discussion, what is the information on how easy
the various types were to build? What indicators are present that
designs were deliberately made easier to build, not just learning
from the mistakes of the first generation but looking towards big
production order.

> Potez had a good reputation with the air force as a reliable, no-
> nonsense manufacturer that would usually provide no-problem planes.
> That they were also unremarkable is something more apparent from
> people looking at specs years later than to people operating them, so
> call it a French Hawker equivalent.

Ah the reliable one that produces reliable aircraft with a slightly out
of date performance penalty.

Hawkers have the Typhoon of course, though really Napiers had the
Sabre would be fairer.

> Morane-Saulnier was the exact opposite, it was a small workshop with
> an absurdly inefficient production organization, originally the
> company was specialized in carpenting.

I should put in some jokes here, with that recommendation.

> Pulling figures from the top of
> my head as I'm away from the relevant documentation at the moment, the
> first few aircraft came in at 25,000+ man-hours. A ruthless effort to
> rationalize production brought that figure down and half of the
> production was around 8,000 man-hours per aircraft. The plane was
> definitely not designed for ease of production, it was needlessly
> complex but had been bought as the sole modern fighter available at
> the time.

I presume this is the MS-406? With its "plymax" wood and aluminium
skin? Deliveries started in late 1938 with about 1,000 built when
production ended in March 1940? Six per day built by April 1939?

They did better than Supermarine when it came to numbers built.

> The Dewoitine 520 average was about 10,000 man hours so 8,400 per ton,
> higher than the late figure for the Morane and higher than the Potez
> 63-11 (7,300 man hours per ton), though this is an average for all the
> aircraft built pre-armistice and incorporates the low scores of early
> production. I believe that the last batches were cheaper, though I
> don't think I have the figures.

Supermarine were paid 8,783 pounds per aircraft for the first 49
Spitfires, 5,782 each for the next 26, 5,768 and a half pounds
each for the next 31 then 5,696 pounds each for the remainder
of the initial order of 310 aircraft. The Government needed to
supply additional equipment before the aircraft was operational.

Those French fighters seem rather light, the Spitfire I came in
at nearly 2 tons empty weight. The D520 weight I have is similar
to the Spitfire, or is the above man hours for airframe only, with
another ton of engine, radio and weapons etc, to be added?

> Late-war US production was by far the most efficient of all the
> figures I remember when assessed in terms of man-hours per ton of
> aircraft manufactured, regardless of the manufacturer.

Not surprising given the 1945 US aircraft production mix was
largely designs that had been in full production for years and the
US could organise for maximum output, not with one eye on
bombing resistance.

>> The C-47 cost $85,000 in 1945, versus the $117,000 for a B-25
>> and $83,000 or a P-47. A 1944 P-38 came in at $97,000. The
>> monetary costs of the late war production should be a reasonable
>> indicator of "production ease" but of course then there are the
>> trade offs between better performance and cost of manufacture.
>
> Absolutely.
>
> The Curtiss H-75 fighters bought by the French cost about twice as
> much as contemporary French fighters of comparable performance (Morane
> 406, Bloch 151) but in exchange for paying premium prices the
> government got 200 extra fighters before the war started, which was
> more than local manufacturers could deliver.

This seems to be more a trade off between quantity and time, rather
than individual aircraft costs.

> Some of the German aircraft production was sub-contracted in occupied
> Europe, with absolutely horrible efficiency figures. On the other
> hand, this was essentially free additional capacity so why not use it?

Agreed. I think the captured shipbuilding industry was even less efficient,
apparently there were things happening like workers would go back at
night and undo their day's work. I will add though ship building steel was
in short supply so there were opportunity costs. And there were probably
some similar costs in the aircraft industry.

Geoffrey Sinclair
Remove the nb for email.