Mass Producing the Merlin

[MIflyer]

Interesting little article from the March 1946 issue of Flying magazine.

 

[Builder 2010]

What about the aircraft parts required so many machining hours. I understand that fit and finish is higher, but running surfaces are running surfaces. Can someone shed some light on the subject?

 

[pbehn]

What about the aircraft parts required so many machining hours. I understand that fit and finish is higher, but running surfaces are running surfaces. Can someone shed some light on the subject?

To start with they are bigger, much bigger.

 

[MIflyer]

Weight is critical. They will go to remarkable ends, do additional machining, to reduce weight by a few ounces. When I first started working with aircraft pneumatic equipment I would marvel at the milling that was done for no obvious reason. It was done due to weight reduction requirements.

The Merlin was characterized by a great many more parts than the Allison V-1710. There is a story that Henry Ford looked at the carburetor for the Model T his engineers had designed, saw it used something like 23 screws, and told them to redesign it. He made them get down to one screw. I do not think that process occurred very much in England.

Packard introduced the Bendix Pressure Carburetor to the Merlin to replace the float type carb originally used and which was the cause of the lean cut out/rich cut out when Merlin equipped aircraft shoved over and created negative G.

 

[pbehn]

Much of this is simply discussing the economies of scale. Packard would not accept an order of less than 5,000. When you have an order for tens of thousands you arrange things differently. The Merlin started as a private venture with no guaranteed orders and actually no certainty of orders or even a war to use it. To gear up for mass production was a "given" in 1940 because there was a war going on. If you set up to produce thousands and only produce 380 as with the Peregrine then you go bust, very quickly.

 

[stona]

Much of this is simply discussing the economies of scale. Packard would not accept an order of less than 5,000. When you have an order for tens of thousands you arrange things differently. The Merlin started as a private venture with no guaranteed orders and actually no certainty of orders or even a war to use it. To gear up for mass production was a "given" in 1940 because there was a war going on. If you set up to produce thousands and only produce 380 as with the Peregrine then you go bust, very quickly.

Exactly.

A major problem in the mid to late 1930s was that there was a huge gulf between the short term engine requirements for the various rearmament schemes and the longer term estimates for wartime requirements. The government wanted to create enough industrial capacity to meet wartime requirements, estimated at 10,000 Merlins per annum in 1936, but did not want to commit itself to more engines than were required for the peace time programmes.

As an example, Scheme L in 1938 increased the RAF's front line fighter strength from 420 to 608 aircraft, significantly increasing engine demand. Rolls Royce had to find increased capacity; in fact it failed to do so, and the new Crewe factory was agreed, but nobody wanted to pay for it. The company's total capital expenditure since the start of re-armament exceeded £750,000*, in 1930s pounds, an it was unwilling to spend more. Rolls-Royce was a business in a competitive market and even the outbreak of the war did not change this. It was not some sort of national charity, willing to sacrifice all for the greater good.

This bears no resemblance to a war time situation of guaranteed demand, guaranteed contracts and guaranteed government financing.

Cheers

Steve

*Difficult to convert as there are various measures, but certainly more than £30,000,000 in today's money.

 

[Ivan1GFP]

Much of this is simply discussing the economies of scale. Packard would not accept an order of less than 5,000. When you have an order for tens of thousands you arrange things differently. The Merlin started as a private venture with no guaranteed orders and actually no certainty of orders or even a war to use it. To gear up for mass production was a "given" in 1940 because there was a war going on. If you set up to produce thousands and only produce 380 as with the Peregrine then you go bust, very quickly.

Hello PBehn,

Do you believe that this influenced the idea of the Rolls Royce Merlin requiring parts to be chosen and fit to a particular engine without a great concern for parts interchangeability? This is not intended as an argument but REALLY is a question. Redrawing for proper tolerances for parts interchange obviously took a fair amount of time for Packard and I wonder if that was an unnecessary cost for a short production run.
Many prototypes need to be modified in some way for actual mass production.

Did Rolls Royce change their methods to adjust to an obvious need for mass production when the time came?

- Ivan.

 

[pbehn]

Hello PBehn,

Do you believe that this influenced the idea of the Rolls Royce Merlin requiring parts to be chosen and fit to a particular engine without a great concern for parts interchangeability? This is not intended as an argument but REALLY is a question. Redrawing for proper tolerances for parts interchange obviously took a fair amount of time for Packard and I wonder if that was an unnecessary cost for a short production run.
Many prototypes need to be modified in some way for actual mass production.

Did Rolls Royce change their methods to adjust to an obvious need for mass production when the time came?

- Ivan.

This has been discussed many times here. There is a strange idea that Rolls Royce didn't mass produce Merlins until Packard rode in to save the day. All the changes were agreed with Rolls Royce and many of them were already in hand when Rolls and Packard started working together.

 

[Deleted member 68059]

The Merlin was characterized by a great many more parts than the Allison V-1710. There is a story that Henry Ford looked at the carburetor for the Model T his engineers had designed, saw it used something like 23 screws, and told them to redesign it. He made them get down to one screw. I do not think that process occurred very much in England..

I would be very interested to know if anyone actually did a full analysis of that statistic, and showed exactly WHERE all those parts were actually situated...

Its obviously not in any of the major parts, as it has - for obvious reasons- exactly the same number of cranks, rods, pistons, valves, props, reduction gears etc etc etc as any other 4-valve per cylinder aero engine.

 

[kmccutcheon]

I would be very interested to know if anyone actually did a full analysis of that statistic, and showed exactly WHERE all those parts were actually situated...

Mostly fasteners, but mustn't forget the three-part connecting rods.

 

[Ivan1GFP]

This has been discussed many times here. There is a strange idea that Rolls Royce didn't mass produce Merlins until Packard rode in to save the day. All the changes were agreed with Rolls Royce and many of them were already in hand when Rolls and Packard started working together.

Hello PBehn,

You answered a question that I wasn't asking. It wasn't a question of whether or not Rolls Royce was mass producing Merlin engines. They obviously were doing this before Packard got into the game.
It was more about the design of the engine that required a team to fit the pieces as I understand was the process used by Rolls Royce (and one which required skilled labor) as opposed to much less skilled labor required in a production line arrangement more typical of automobiles and whether that was a conscious choice for an expectation of very limited production. Parts interchangeability is not really a great concern when the total production numbers are very low.

When it became obvious that production numbers would be much higher, did tolerances change to guarantee parts interchangeability which apparently was not a feature of the earlier engines? By various accounts, this was one of the issues that Packard tried to address before they began production.

- Ivan.

 

[pbehn]

Hello PBehn,

You answered a question that I wasn't asking. It wasn't a question of whether or not Rolls Royce was mass producing Merlin engines. They obviously were doing this before Packard got into the game.
It was more about the design of the engine that required a team to fit the pieces as I understand was the process used by Rolls Royce (and one which required skilled labor) as opposed to much less skilled labor required in a production line arrangement more typical of automobiles and whether that was a conscious choice for an expectation of very limited production. Parts interchangeability is not really a great concern when the total production numbers are very low.

When it became obvious that production numbers would be much higher, did tolerances change to guarantee parts interchangeability which apparently was not a feature of the earlier engines? By various accounts, this was one of the issues that Packard tried to address before they began production.

- Ivan.

As per the previous thread, until Packard arrived and pushed us out of our mud huts the British first machined a piston and then machined a cylinder to fit it, this was done many times until we got 12 that were almost the same, later they cast an engine block to fit the cylinders and this engine block was measured in order to produce an aeroplane that fitted it. This irregular method of quality control resulted in fighters and bombers with extra ordinary properties, some would only climb and dive while others only turned to port or to starboard some wouldn't take off and others wouldn't land. Happily within a squadron the British bumpkins usually had a plane that could do something like that what was required which kept the Hun at bay from 1939 to 1943 when PACKARD arrived to save the day. If you think I am treating you with complete contempt, read your own post you patronising goon.

 

[Ivan1GFP]

Hello PBehn,

You are assigning a motive that was never was there. You are thinking that this is a "National Pride" issue and the truth is pretty far from that. I really have no stake in what happened in either Great Britain OR the United States 80 or so years ago.
There is no question as far as I am concerned that the US Army made a pretty serious mistake in choosing no to invest in a better supercharger for the Allison engine before the war. They made a prediction that didn't match what would be required when the war began.
EVERYONE makes mistakes when trying to predict the future. To deny that is foolish.

The question I asked was because of YOUR statement about the very limited production numbers of prior RR engines and lack of contract for the Merlin when it was first designed. My question about tolerances and parts interchangeability was really one of whether that was a choice because of expected production, or something else such as perhaps technology limitations, etc. or whether it simply was not true.

The reason I "Liked" your post was that I had a pretty good idea that your response would end up where it went even though it wasn't the question I was asking. (Shortround6 actually answered the basic question in the other thread WITHOUT bringing up mud huts.)
It sounds like the issue actually existed but that it was Ford in UK that did the most to tighten the specifications and not Packard in US.
The second question of WHY the work was even necessary still hasn't been answered, but maybe we will get there eventually.

You are also making an assumption that my comment about production lines versus hand assembly is an assessment of quality. It isn't.
Tighter fit between parts also isn't necessarily a quality improvement either, but that kind of discussion is getting way off topic.

- Ivan.

 

[pbehn]

The reason I "Liked" your post was that I had a pretty good idea that your response would end up where it went even though it wasn't the question I was asking. (Shortround6 actually answered the basic question in the other thread WITHOUT bringing up mud huts.)
It sounds like the issue actually existed but that it was Ford in UK that did the most to tighten the specifications and not Packard in US.
The second question of WHY the work was even necessary still hasn't been answered, but maybe we will get there eventually.
.

The issue is one obvious in production engineering. How you set up a production facility depends on how many units you need to make per day/week/month. The Merlin was no ones favourite engine until 1940. It was not scheduled to be fitted to any bomber and the Spitfire and Hurricane were due to be replaced by Vulture and Sabre powered planes. Slowly the Halifax was switched to use the Merlin at the design stage, the RR Vulture engine was shelved so the Manchester became the Merlin powered Lancaster. From a peace time requirement that probably wouldn't have seen more than 5,000 made (as per the Kestrel) the requirement shifted to 10s of thousands. It was Rolls Royce who did the tightening of the specification because it was their engine, the best people to speak to on the subject of mass producing engines are those who do it like car engine makers, but it is not a "given" that what Ford did on a car engine could be done on an aero engine because they are bigger. Also what Packard considered mass production of engines would not be today, for todays car manufacturers of car engines.

To produce a prototype engine frequently takes a long time because you don't yet have the machines to produce it, so the machining is done by skilled people who can produce all the pieces working from drawings with few specialised machine tools or may even be sub contracted to people who have the machines to do it. When you go for mass production you can specify much tighter tolerances, if you have machines to meet those tolerances or can accept the rejection rate that the tighter tolerances may cause. However to do that you need lots of machines and semi skilled or skilled operators. At the start of its life the Merlin was no more guaranteed to be produced than the Peregrine. Now imagine how much Packard would lose if they set up a factory for the Peregrine which was shelved and only 380 units produced. Packard protected themselves from such a loss by not accepting an order for less than 5,000. However the facilities built by or for Packard were not for 5,000 engines in total but for around 20,000 per year at peak which shows how quickly things changed around 1940-41. From being nothing special and little future it became almost the only game in town, specified in almost every UK fighter and bomber as the main choice or as a contingency should a factory get demolished. This apart from Merlins fitted to US aircraft which exceeded the number originally ordered by the British.

 

[fastmongrel]

As per the previous thread, until Packard arrived and pushed us out of our mud huts the British first machined a piston and then machined a cylinder to fit it, this was done many times until we got 12 that were almost the same, later they cast an engine block to fit the cylinders and this engine block was measured in order to produce an aeroplane that fitted it. This irregular method of quality control resulted in fighters and bombers with extra ordinary properties, some would only climb and dive while others only turned to port or to starboard some wouldn't take off and others wouldn't land. Happily within a squadron the British bumpkins usually had a plane that could do something like that what was required which kept the Hun at bay from 1939 to 1943 when PACKARD arrived to save the day. If you think I am treating you with complete contempt, read your own post you patronising goon.

I thought everyone knew that RAF aeroplanes and engines were carved from a single piece of Oak by a little old man who had a workshop in a forest glade.