GrauGeist
Generalfeldmarschall zur Luftschiff Abteilung
The Merlin coincidently had the same CID as the Packard V-1650 (Liberty L-12).
Best tank engines of WWII
- Thread starter z42
- Start date
Ad: This forum contains affiliate links to products on Amazon and eBay. More information in Terms and rules
Just been catching up on this very interesting thread.
Have a look at this link about the genesis of the big Ford V8 - which explains why it's piston and cylinder dimensions were the same as the RR Merlin
Sounds like Henry Ford being Henry Ford!
The Ford's, the Merlins and Maurice Olley
Origins of Ford Aero and Tank Engines, the American Merlin, and the writings of Maurice Olley. Before the entry of the United States in the Second World War,…thunderboats.ning.com
Some quotes from the article in italic:
This opinion is proved that to be specious by their British Factory which, starting in 1941, made over 30, 000 Merlin's without one reject! Like Packard, the British Ford factory redrew the RR drawings and tightened the tolerances so the engines could be mass produced.
Not the case - RR at Derby were mass-producing Merlins before the British-owned and British-built factory, that Ford of England was running, went online. For the best of the parts count (ie. discounting the US specifics, like the prop spline, S/C drive and carb), these were interchangeable since RR used tight tollerances.
The Ford GG aircraft V12 showed great potential, producing over 1800 Hp on its initial dyno test!
Ford V12 went to the dyno tests as a V2 (vee two) test rig. The full V12 was never there to be tested.
A. Through the efforts of their designer, Charles Sorrenson, aka "Cast Iron Charlie", Ford was the first carmaker to mass produce "enbloc" V engines, namely the Ford "Flathead" V8's and the Lincoln V12's. The aluminum 1650 C.I. enbloc V12 Aero engine was well within their capability. It was far more rigid than the multi piece Merlin or the GM Allison blocks and could have been pushed far harder and been capable of more horse power.
Merlin 1-piece block was not as good as the Merlin 2-piece block, so no.
The Merlin contract would have given Ford the entry into the market, which is why it was pursued.
Ford was already tasked with the production of R-2800 engines, IOW they were already in the market.
ThomasP
Chief Master Sergeant
Hey Chris MacD,
Thanks for posting the above stuff. Very interesting, particularly the notes from Maurice Olley.
However,
First, the statement "This opinion is proved that to be specious by their British Factory [Ford of England] which, starting in 1941, made over 30, 000 Merlin's without one reject!" is at best silly hubris. There is no historical evidence to justify this statement as it is used to imply superior workmanship.
Second, I feel obliged to point out that Jim's statement "Like Packard, the British Ford factory redrew the RR drawings and tightened the tolerances so the engines could be mass produced." is very incorrect. I do not know if his reference comes from the myth that the tolerances on the RR blueprints were not tight enough for volume production (and/or to have interchangeable parts) and he is just repeating the myth, or if it is from a misunderstanding of tolerancing and how the blueprints were made in England and the US at the time.
1. Ford of England did not redraw the blueprints like Packard did. For production of the Merlin, Ford of England used the standard 1st angle projection blueprints (see llnk below) just like RR at Derby. In fact they used copies of the same prints RR used. The reason for this is simple - 99%+ of the workforce at Ford of England were UK citizens, were educated in British colleges and trade schools, had gained most of their experience in UK workshops, and as such were very familiar with 1st angle projection drawings, while 99% were very not familiar with 3rd angle projection.
Ford of England may have redrawn some RR prints at some point (under the supervision of RR) for replacement of worn copies, or in order to declutter and update prints with current Change or Revision notes. They certainly did make their own feature specific prints (simplified blueprints covering only a smaller number of features and used on the shop floor in order to make it easier for the machinists to separate out the details and dimensions required for the particular operations they are performing - usually called production drawings or production blueprints) - but the blueprints used at Ford of England would have remained 1st angle projection, and the end tolerance requirements for function and fit would have remained the same as on the blueprints they received from RR.
2. Packard actually did redraw (over a period of about 6-8 months) the entire package of prints sent from England, converting them from 1st angle projection to the 3rd angle projection (again, see llnk below). They had to since 99% of the available workforce in the US were used to 3rd angle projection and were not familiar with 1st angle projection (for the same reasons as the UK workforce was familiar with 1st angle projectionand not with 3rd angle projection). They did not, however, initially make any significant changes in the dimensions or tolerances - this is because they were competent. They would not have done so as they had no experience with the Merlin to give them any way of knowing what needed changing, and doing so could have invited disaster when it came time to produce, assemble, and run the engines. Note that the blueprints were redrawn before any Merlin engines were manufactured by Packard (I think). I cannot say whether Packard eventually tightened up any tolerances or not, or for what specific reasons, but I can say that there would have been no initial need to do so in order to reliably manufacture an engine with interchangeable parts. That is not how dimensioning and tolerancing works relative to ease of manufacturing. The tolerances on part 'A' simply having to be "tight enough" to fit the mating par 'B' whose tolerances have to be "tight enough" to reliably allow assembly, and the tolerances on both part 'A' and part 'B' would have to allow fit to any mating part 'C', etc, etc. The tolerances as specified by RR would already have been "tight enough" or RR could not have effectively made any working automobile or aircraft engines prior to the Merlin. Mechanical dimensioning and tolerancing is something that any contemporary US or UK journeyman level draftsman or mechanical engineer would have understood.
By the time Ford and Packard were approached, RR had been making reliable aircraft engines like the Kestrel for 14 years (which would have used the same basic dimensioning and tolerancing methods as used on the Merlin). The functional tolerances may have had to be tightened for the Merlin in order to make reliable engines (due to the significantly higher performance requirements), and the fit tolerances tightened to match the functional requirements, and this was OKed and done by RR as the engine evolved, but that is all. I do not think there can be any rational argument that any tightening of tolerances by Ford or Packard was required in order to speed up production or improve ability to produce interchangeable parts - at least not any that RR was not already using in practice.
Also - although it seems counter-intuitive - in actual production of machine(ed) parts, if you want to speed up production and/or make parts more interchangeable, you need to loosen tolerances on at least 1 of the mating parts, and possibly both - not tighten them.
3. As has been pointed out by tomo pauk above and elsewhere in this forum, RR had already produced several thousand highly functional Merlins with interchangeable parts by the time Packard was ready to begin production. There is no practical way they could have done so without a through understanding of the needed tolerances.
Link to a good basic explanation of 1st and 3rd angle projection as related to blueprints:
"Understanding First and Third Angle Projections in Blueprints — CMM QUARTERLY"
A good explanation of tolerancing is significantly more complicated, since the dimensions often include functional tolerance requirements as well as fit requirements. However, if you leave out the functional tolerances, a good basic explanation is included in any good textbook on GD&T (Geometric Dimensioning and Tolerancing). While GD&T had not been concentrated in one formal methodology at the time, the principals embodied in GD&T methods applied during WWII just as they do today - only the language/symbology changed to make a more unified, standardized, and easily teachable international system. The link below is to a good basic explanation of the principles of GD&T:
"Introduction to GD&T — CMM QUARTERLY"
bleh
Thanks for posting the above stuff. Very interesting, particularly the notes from Maurice Olley.
However,
First, the statement "This opinion is proved that to be specious by their British Factory [Ford of England] which, starting in 1941, made over 30, 000 Merlin's without one reject!" is at best silly hubris. There is no historical evidence to justify this statement as it is used to imply superior workmanship.
Second, I feel obliged to point out that Jim's statement "Like Packard, the British Ford factory redrew the RR drawings and tightened the tolerances so the engines could be mass produced." is very incorrect. I do not know if his reference comes from the myth that the tolerances on the RR blueprints were not tight enough for volume production (and/or to have interchangeable parts) and he is just repeating the myth, or if it is from a misunderstanding of tolerancing and how the blueprints were made in England and the US at the time.
1. Ford of England did not redraw the blueprints like Packard did. For production of the Merlin, Ford of England used the standard 1st angle projection blueprints (see llnk below) just like RR at Derby. In fact they used copies of the same prints RR used. The reason for this is simple - 99%+ of the workforce at Ford of England were UK citizens, were educated in British colleges and trade schools, had gained most of their experience in UK workshops, and as such were very familiar with 1st angle projection drawings, while 99% were very not familiar with 3rd angle projection.
Ford of England may have redrawn some RR prints at some point (under the supervision of RR) for replacement of worn copies, or in order to declutter and update prints with current Change or Revision notes. They certainly did make their own feature specific prints (simplified blueprints covering only a smaller number of features and used on the shop floor in order to make it easier for the machinists to separate out the details and dimensions required for the particular operations they are performing - usually called production drawings or production blueprints) - but the blueprints used at Ford of England would have remained 1st angle projection, and the end tolerance requirements for function and fit would have remained the same as on the blueprints they received from RR.
2. Packard actually did redraw (over a period of about 6-8 months) the entire package of prints sent from England, converting them from 1st angle projection to the 3rd angle projection (again, see llnk below). They had to since 99% of the available workforce in the US were used to 3rd angle projection and were not familiar with 1st angle projection (for the same reasons as the UK workforce was familiar with 1st angle projectionand not with 3rd angle projection). They did not, however, initially make any significant changes in the dimensions or tolerances - this is because they were competent. They would not have done so as they had no experience with the Merlin to give them any way of knowing what needed changing, and doing so could have invited disaster when it came time to produce, assemble, and run the engines. Note that the blueprints were redrawn before any Merlin engines were manufactured by Packard (I think). I cannot say whether Packard eventually tightened up any tolerances or not, or for what specific reasons, but I can say that there would have been no initial need to do so in order to reliably manufacture an engine with interchangeable parts. That is not how dimensioning and tolerancing works relative to ease of manufacturing. The tolerances on part 'A' simply having to be "tight enough" to fit the mating par 'B' whose tolerances have to be "tight enough" to reliably allow assembly, and the tolerances on both part 'A' and part 'B' would have to allow fit to any mating part 'C', etc, etc. The tolerances as specified by RR would already have been "tight enough" or RR could not have effectively made any working automobile or aircraft engines prior to the Merlin. Mechanical dimensioning and tolerancing is something that any contemporary US or UK journeyman level draftsman or mechanical engineer would have understood.
By the time Ford and Packard were approached, RR had been making reliable aircraft engines like the Kestrel for 14 years (which would have used the same basic dimensioning and tolerancing methods as used on the Merlin). The functional tolerances may have had to be tightened for the Merlin in order to make reliable engines (due to the significantly higher performance requirements), and the fit tolerances tightened to match the functional requirements, and this was OKed and done by RR as the engine evolved, but that is all. I do not think there can be any rational argument that any tightening of tolerances by Ford or Packard was required in order to speed up production or improve ability to produce interchangeable parts - at least not any that RR was not already using in practice.
Also - although it seems counter-intuitive - in actual production of machine(ed) parts, if you want to speed up production and/or make parts more interchangeable, you need to loosen tolerances on at least 1 of the mating parts, and possibly both - not tighten them.
3. As has been pointed out by tomo pauk above and elsewhere in this forum, RR had already produced several thousand highly functional Merlins with interchangeable parts by the time Packard was ready to begin production. There is no practical way they could have done so without a through understanding of the needed tolerances.
Link to a good basic explanation of 1st and 3rd angle projection as related to blueprints:
"Understanding First and Third Angle Projections in Blueprints — CMM QUARTERLY"
A good explanation of tolerancing is significantly more complicated, since the dimensions often include functional tolerance requirements as well as fit requirements. However, if you leave out the functional tolerances, a good basic explanation is included in any good textbook on GD&T (Geometric Dimensioning and Tolerancing). While GD&T had not been concentrated in one formal methodology at the time, the principals embodied in GD&T methods applied during WWII just as they do today - only the language/symbology changed to make a more unified, standardized, and easily teachable international system. The link below is to a good basic explanation of the principles of GD&T:
"Introduction to GD&T — CMM QUARTERLY"
bleh
Last edited:
Elvis
1st Sergeant
What I question is why would anyone have to second guess any engine designed and built by R-R?
They built some very good engines, all by themselves. Look at the Silver Ghost cars. Those engines are STILL running and they're well over 100 years old.
They built some very good engines, all by themselves. Look at the Silver Ghost cars. Those engines are STILL running and they're well over 100 years old.
It makes for a good story. Invent the 'villain', and then invent the - local - 'superhero' that can do stuff better.
To hell with the facts, since we have newspapers (or clicks nowadays) to sell.
Shortround6
Lieutenant General
It was quite possible to build high quality engines without the tight tolerances of mass-produced cars where everything had to be interchangeable.
And even after using tighter tolerances some parts still needed to be hand selected.
But please forget that "file to fit" nonsense. They had measuring tools to measure things like the bore size and they selected the appropriately sized piston to fit the bore of the cylinder to give the clearance desired. Problem was getting 12 pistons that were close in weight to each other. That problem did not go away even if they tightened the tolerance on the diameter. It got less but did not go away.
A lot depends on manufacturing machinery and rate of production too. If, for instance on a part that is 6in in diameter you have a tolerance of +- .010 in and you tighten it to +- .008 you are getting rid of the extremes, which is good. But what does it take to keep up production at the tighter tolerance/s. It is just workmen who are not taking their time or is trying to tale to much material off in each cut and the tool post/support is flexing under pressure? is the lathe itself running true? Trying to build pression machinery on old/worn machines is a real problem.
My Grandfather made parts for bombsights in WW II, gyroscopes, on old machinery. It could be done but it took a lot of small cuts and a lot of measuring. When you build new factories with new machinery perhaps you can use tighter tolerances at the same time. Trying to build Merlins on machines that RR Eagle engines in WW I might be asking a bit much?
And even after using tighter tolerances some parts still needed to be hand selected.
But please forget that "file to fit" nonsense. They had measuring tools to measure things like the bore size and they selected the appropriately sized piston to fit the bore of the cylinder to give the clearance desired. Problem was getting 12 pistons that were close in weight to each other. That problem did not go away even if they tightened the tolerance on the diameter. It got less but did not go away.
A lot depends on manufacturing machinery and rate of production too. If, for instance on a part that is 6in in diameter you have a tolerance of +- .010 in and you tighten it to +- .008 you are getting rid of the extremes, which is good. But what does it take to keep up production at the tighter tolerance/s. It is just workmen who are not taking their time or is trying to tale to much material off in each cut and the tool post/support is flexing under pressure? is the lathe itself running true? Trying to build pression machinery on old/worn machines is a real problem.
My Grandfather made parts for bombsights in WW II, gyroscopes, on old machinery. It could be done but it took a lot of small cuts and a lot of measuring. When you build new factories with new machinery perhaps you can use tighter tolerances at the same time. Trying to build Merlins on machines that RR Eagle engines in WW I might be asking a bit much?
GrauGeist
Generalfeldmarschall zur Luftschiff Abteilung
It's interesting how that "file to fit" myth still persists.
"Somewhere out in the shire, old bespectacled elves in a vine covered cottage, sit around Merlin engines, tapping and filing away by lamplight"
Meanwhile, in the real world, the engines produced by Rolls-Royce and Packard had about a 70% (+/-) interchangability rate.
"Somewhere out in the shire, old bespectacled elves in a vine covered cottage, sit around Merlin engines, tapping and filing away by lamplight"
Meanwhile, in the real world, the engines produced by Rolls-Royce and Packard had about a 70% (+/-) interchangability rate.
SaparotRob
Unter Gemeine Geschwader Murmeltier XIII
I thought they were knitted from steel mesh by groups of delightfully dotty aunts at their book clubs.
Reluctant Poster
Tech Sergeant
- 1,776
- Dec 6, 2006
Rolls Royce didn't make over 87,000 Merlins - more than Ford and Packard combined - using left over WWI machinery. Crewe and Glasgow were build from scratch using new machine tools and Derby was updated with the latest machine tools.
There is no evidence to suggest that Rolls Royce was incapable of making pistons to a standard diameter. As to weighing pistons that was not unusual in the automobile industry. I posted some time ago a video of Ford weighing pistons for their flathead V-8s.
Post in thread 'could the Allison engine have done what the Rolls Royce Merlin did?'
could the Allison engine have done what the Rolls Royce Merlin did?
yosimitesam
Senior Airman
The debate at the time of the Sherman being perferable to the Pershing was all about SHIPPING the things to Europe. Fewer M26's could be shipped and UNLOADED in all the places they needed to be. Not all docks used by the Allies in NW Europe had their own unloading cranes. In fact most did not. (The bigger ports such as Antwerp had some dock cranes but most smaller ports did not.) The crane loading capacity on the Liberty ships was the bottleneck, especially in unloading. From "ww2ships.com", "Liberty Cargo Ships":
So only one hold could generally be used for M26 whereas probably two holds for the M4. Not to mention the greater dimensions of the M26 meaning fewer in a given space. Since the vast majority of war shipping to Europe was done with Liberty ships, the number of M26's that could be transported AND UNLOADED would have been substantially less. This simple fact overrode everthing else in the debate about sending heavier tanks to NW Europe in the latter part of the war.
Elvis
1st Sergeant
Boy I know that one for a fact!
I worked for a DDA dealership for a short time when I was in my 20's.
Our milling table, for shaving cylinder heads, wasn't straight, although you couldn't tell the shop owner that.
We'd shim the heads with Blower Shims and make test cuts to make sure it was cutting true. Also, you didn't want to bolt the head to the table too tight, because you could use that to "pull" the head straight with a little extra snugging of one bolt or the other.
When you're getting used to working with equipment like that, it can be a REAL head scratcher at times!
I walked away from that job thinking that's a hell I wouldn't want to put my worst enemy through.
SHEESH!
Elvis
1st Sergeant
By the time the Pershing was operational and being shipped to Europe, wasn't the main cargo ship the larger and faster Victory ship?
Maybe a better question is, did it have more powerful cranes on board?
The victory ship was slightly larger than the liberty but had three distinct upgrades.
First was speed. The liberty type was vulnerable to U-boats as it was not all that quick, trundling along around 11 knots.
The victory ships could cruise at 4 knots more making them more survivable.
Second was the hull which was around 15 feet longer and 6 or so feet wider than the liberty, giving larger area in the holds
but not a massive amount. the framing was different as well allowing more flex which had caused the loss/damage of some liberty
vessels (too rigid and could actually break up).
Third and very important was the change from steam driven equipment (cranes etc) to electric. This allowed the victory ships to
unload themselves. They were far less dependant on dock facilities so unloading was more efficient.
First was speed. The liberty type was vulnerable to U-boats as it was not all that quick, trundling along around 11 knots.
The victory ships could cruise at 4 knots more making them more survivable.
Second was the hull which was around 15 feet longer and 6 or so feet wider than the liberty, giving larger area in the holds
but not a massive amount. the framing was different as well allowing more flex which had caused the loss/damage of some liberty
vessels (too rigid and could actually break up).
Third and very important was the change from steam driven equipment (cranes etc) to electric. This allowed the victory ships to
unload themselves. They were far less dependant on dock facilities so unloading was more efficient.
The Liberty remained the main US war built cargo ship to the end of the war. The last of about 2,700 was delivered in Oct 1945.
8 x Z-EC2-S-C2 Liberties were built as dedicated tank carriers being rearranged internally with 4 holds with larger hatches instead of 5. But the booms were a mix of 5 ton (for the aftermost hatch) and 15 & 30 ton (for the 3 remaining hatches).
The first Victory didn't complete until Feb 1944 and only 416 were built in the period to just beyond WW2 (plus another 117 as attack transports for the USN). Again however most of the booms had a 5 ton capacity with there being a single 30 & 50 ton capacity boom aft and also forward in some ships.
Given the general cargoes carried, and the methods then used for handling those cargoes (manually by large contingents of stevedores in each port) 5 tons was usually plenty. Heavy loads could be handled by shoreside cranes or floating cranes when necessary.
Production of the M26 Pershing didn't begin until Nov 1944 at the Fisher Plant and March 1945 at the Chrysler run Detroit Tank Arsenal. The first shipment didn't arrive in Europe until Jan 1945, with only 310 arriving by 'VE Day, only 200 or so being issued to combat units. In the Pacific, the first shipment of 12 left the USA on 31st May, being offloaded at Okinawa on 4th Aug.
So Pershing did not represent a significant cargo problem in WW2.
Elvis
1st Sergeant
Thanks guys.
- Thread starter
- #356
I don't understand this. Liberty ship has boilers lit to produce steam for various auxiliary equipment, be they directly steam driven or then electrical where power comes from a steam driven generator. Similarly the Victory ship has boilers lit to run the generators which gives electricity for electrically powered equipment. How does this make the Victory ship able to unload itself and the Liberty ship not?
Using shore based electricity is AFAIU a more recent phenomenon to reduce emissions from ships in port, and even to this day it's not that common.
For another anecdotal data point, my father worked on tankers in the late 1960's / early 70'ies, and apparently at least back then much equipment like cargo pumps was steam driven in order to avoid sparks from electrical equipment that could cause fires. Not sure what current praxis is.
Last edited:
Sorry. Didn't explain that very well. The electric winches, cranes etc were more efficient as torque was available all the way through any movement.
The part I missed was the Liberty ships had 5 ton boom cranes with upgrades to 10 tons being made I think in 1944. The Victory ships had 14 5 ton
cranes but also had one 30 ton and one 50 ton crane as well. These allowed the self unloading as any item carried could be self lifted. The Victory
ships had the masts built into the hull which was what allowed the heavier cranes to be used.
Last edited:
yosimitesam
Senior Airman
No. There were 2710 Liberty ships built and only 534 Victory ships, with most launched in 1944-1945. The first Victory ship was not launched until January 26, 1944 and its shakedown finished on April 1, 1944. Most were lauched during the summer and fall of 1944. As for the cranes, I'm unable to find the "standard" crane capapcities for them. I suspect they (or at least some) were fitted with heavier cranes, e.g. 50 tons or more, but I can't verify this.
Edit: They were faster (16-17 kn, at least) which made them almost immune to Atlantic U-boats. None were sunk by U-boats and only 3 were sunk in the Pacific by kamikazes.
Edit:
According to "Ships for Victory", Fredric C. Lane, ISBN 0-8018-6752-5: Cargo handling gear was similar ... [to Liberty ships] ... consisting of 14 5-ton booms, one 30-ton, and one 50-ton boom. So unless special modifications were done to some, they had the boom capacity about the same as a Liberty Ship.
Took me a while to find that book in my library.
(I accidentally replied to my own post previously instead of editing this one. I may need a nap.)
Last edited:
yosimitesam
Senior Airman
They were faster (16-17 kn, at least) which made them almost immune to Atlantic U-boats. None were sunk by U-boats and only 3 were sunk in the Pacific by kamikazes.
swampyankee
Chief Master Sergeant
- 4,306
- Jun 25, 2013
Another issue was the a significant portion of the US Army's bridges and river crossing equipment was often unable to manage the M26's weight and size. Nicholas Moran, aka "The Chieftain" has discussed some of this. The issues with engineering equipment make sense, as building field bridges with too much extra capacity is wasteful.
