Why were V engines 12 cylinders?

[Deleted member 68059]

..the turbocharged Allison out performed the 20 series single stage merlin !!!

Actually on 2nd June 1940 Wright Field estimated that the Merlin XX + P-38 was 20mph faster than the turbo allison at 25,000 ft*

* US National Archives II, College Park, NARA-6547493 – "Rolls-Royce Buzzard, Eagle, Merlin, Vulture" 1929-1940. {see "Table x" on page dated June 2, 1940.}

 

[Shortround6]

Plus the Hurricanes (1940-43) and later Lancasters and Mosquitos produced in Canada.

Packard built over 26,000 single stage Merlins, it is doubtful that the US took the whole 3000 it was supposed to get form the first contract. The need for more than 23,000 Merlins to meet Canadian production requirements seems a bit slim.
On the other hand, Ford built around 24,000 R-2800 radials, anybody trying to claim that Ford should have built Merlins should try to figure out how to power planes with twin R-2800s with a pair of single stage Merlins.

 

[gjs238]

Packard built over 26,000 single stage Merlins, it is doubtful that the US took the whole 3000 it was supposed to get form the first contract. The need for more than 23,000 Merlins to meet Canadian production requirements seems a bit slim.
On the other hand, Ford built around 24,000 R-2800 radials, anybody trying to claim that Ford should have built Merlins should try to figure out how to power planes with twin R-2800s with a pair of single stage Merlins.

What were the timeframes for Ford Merlins vs R-2800's?

 

[Shortround6]

Ford was talking to RR and the American government in June of 1940, Ford got millions to build a new factory in Sept 1940 and broke ground the same day. Packard signed the official contracts in the Sept of 1940. Ford had built 262 engines by the end of 1941 and Packard had built 45 engines by the end of 1941. Ford sent their engineers to P & W East Harford and basically duplicated the layout of the production floor and the flow of parts and assembly. Packard got two (yes two) British engineers to help them.

I really can't see Ford doing much better than Packard in getting the Merlin into production. Packard was looking at the engine blue prints and sample engine in July about 2 months before the official contracts were signed so Ford's supposed head start would only be a few weeks (4-5?)

 

[xylstra]

RR Merlin and Griffon, DB 605, Allison V-1710, Mikulin AM-35/38/42, Klimov M-105, Hispano-Suiza 12Z/Y.... all V-12 liquid cooled engines.

Why were all V configuration aero engines of the 1930s onwards 12 cylinders? Why no V-8, V-10 or V-16? Of the latter there was a postwar prototype.

Above 12 cylinders crankshaft torsional vibration and flexing is a big, BIG problem. Manifolding can also be a problem.

 

[kiwi2wheels]

A quibble: fork and blade rods don't require offset between the cylinder banks; plain rods would. Each type has its pros and cons. One of the cons of plain rods is they force an offset between banks. Fork-and-blade rods require a bearing surface between the big ends. Articulated rods, if the geometry isn't done right, will result in different strokes for the two cylinder banks. I suspect they're also the heaviest of the three types.

Comparing the width of fork and blade rods ( not to mention the added complexity ) to plain rods, wouldn't the f & b crank be longer slightly overall, because of the need for longer rod journals ? And how were the differences in rod weights compensated for ( e.g. assembled alternating them in the banks ) or isn't it an issue ?

 

[swampyankee]

Comparing the width of fork and blade rods ( not to mention the added complexity ) to plain rods, wouldn't the f & b crank be longer slightly overall, because of the need for longer rod journals ? And how were the differences in rod weights compensated for ( e.g. assembled alternating them in the banks ) or isn't it an issue ?

It would be fairly straightforward to make sure the two rods in a fork and blade pair had the same weight and have close to the same mass distribution, probably easier than for articulated rods, as used by many German engines (also, unless one is very careful, many engines with articulated rods will have differing strokes in the two cylinder banks.) I am not an expert in bearings, but I think the distance between the webs is going to be about the same for fork and blade, articulated, and plain rods.

 

[wuzak]

Comparing the width of fork and blade rods ( not to mention the added complexity ) to plain rods, wouldn't the f & b crank be longer slightly overall, because of the need for longer rod journals ? And how were the differences in rod weights compensated for ( e.g. assembled alternating them in the banks ) or isn't it an issue ?

I would think that the outside width of the fork rod would be very similar to two "plain rods", if not the same, simply because the area of the two bearings combined would need to be the same as the bearing for the blade rod.

"Plain rods" would have to sit side by side on the crank, which means that one bank would be offset from the other. Car vee engines use this method because it is cheaper.

 

[elbmc1969]

RR Merlin and Griffon, DB 605, Allison V-1710, Mikulin AM-35/38/42, Klimov M-105, Hispano-Suiza 12Z/Y.... all V-12 liquid cooled engines.

Why were all V configuration aero engines of the 1930s onwards 12 cylinders? Why no V-8, V-10 or V-16? Of the latter there was a postwar prototype.

Cooling. With the same proportions, the surface area increases with the square of the dimensions, but the volume increases with the cube. Since cooling is primarily around the "barrel" of each cylinder (primarily!), cooling area increases much more slowly than the volume of the cylinder. If you want to increase power by 50% per cylinder (to have the same effect as increasing the number of cylinders by 50%), you're generating 50% more heat, but the surface of each cylinder increases by much less than 50%.

 

[overbeck]

Cooling. With the same proportions, the surface area increases with the square of the dimensions, but the volume increases with the cube. Since cooling is primarily around the "barrel" of each cylinder (primarily!), cooling area increases much more slowly than the volume of the cylinder.

It's the head and barrel that we want to keep cool. We'd prefer not to remove any heat at all from the volume within the cylinder, if that were possible.

 

[Shortround6]

It's the head and barrel that we want to keep cool. We'd prefer not to remove any heat at all from the volume within the cylinder, if that were possible.

A text book answer on efficiency.
Melting holes in the tops of pistons is a pretty quick way of ruining efficiency
three ways of keeping the piston crown cool,
1, cool intake charge coming through the intake valve, cool being relative, 200-300 degree intake air temp (after supercharger) is a lot cooler than the piston head and cylinder.
2 Splashing or spraying oil on the bottom of the piston (which were often made with fins to assist in dissipating heat
3. Heat traveling by conduction from the piston crown to the piston walls and then through to the cylinder walls.

The cooler you can make the cylinder the closer you can come to the detonation limits of the fuel (higher boost which means more fuel burned for more power).

fuel efficiency (power per pound of fuel burned) and max power (maximum amount of fuel burned in a given time period) are often at odds.

 

[pbehn]

A text book answer on efficiency.
Melting holes in the tops of pistons is a pretty quick way of ruining efficiency
three ways of keeping the piston crown cool,
1, cool intake charge coming through the intake valve, cool being relative, 200-300 degree intake air temp (after supercharger) is a lot cooler than the piston head and cylinder.
2 Splashing or spraying oil on the bottom of the piston (which were often made with fins to assist in dissipating heat
3. Heat traveling by conduction from the piston crown to the piston walls and then through to the cylinder walls.

The cooler you can make the cylinder the closer you can come to the detonation limits of the fuel (higher boost which means more fuel burned for more power).

fuel efficiency (power per pound of fuel burned) and max power (maximum amount of fuel burned in a given time period) are often at odds.

Another idea is to keep the heat away from the piston load bearing pieces. There were rumours that Honda used an insulating ceramic crown on the pistons of its racing engines, so the heat stayed in the combustion chamber.

 

[elbmc1969]

It's the head and barrel that we want to keep cool. We'd prefer not to remove any heat at all from the volume within the cylinder, if that were possible.

True, but that leads to the idea that an infinitely large cylinder is ideal for cooling, since you'd have the minimum cylinder surface area for unit of volume. However, cooling that would be a nightmare.

 

[overbeck]

A text book answer on efficiency.
Melting holes in the tops of pistons is a pretty quick way of ruining efficiency
three ways of keeping the piston crown cool,

I'm aware of piston cooling jets, etc. I think what you're getting at is the center of the piston crown being further from the walls in large bore cylinders makes cooling the piston more difficult. Do you have figures on how much heat is conducted away through the piston rings? Oil jets should remove far more heat than that.

But this is distinct from cooling the gasses within the cylinder. Are you suggesting that cooling the gasses by conducting heat into the head/walls is significant for lowering the piston temperature? These surfaces are probably too far away and the time too brief to have much effect. The gasses cool much more quickly by expansion.

True, but that leads to the idea that an infinitely large cylinder is ideal for cooling, since you'd have the minimum cylinder surface area for unit of volume.

It is, but there are many reasons not to use one very large cylinder.

EDIT: the head, walls and piston crown may be thicker in a large capacity cylinder for structural reasons, which makes cooling more difficult. Again, while this is related to swept capacity, it is distinct from cooling the volume within the cylinder, which is undesirable.

 

[Shortround6]

Cooling the cylinder was a major problem and/or limit to the power produced per cylinder of air cooled engines, It may have been for liquid cooled but I have seen no articles or graphs comparing different liquid cooled cylinders.
I have no figures on heat conducted away through the rings but transfer of heat from the piston to the cylinder walls was fairly important, whether through conduction (through the oil film?) of radiation.
P & W had made two different 9 cylinder Hornet engines, the R-1690 which was fairly common and the R-1860 Hornet B which was built in small numbers and according to one account, helped steer P & W to the "more but small cylinders" path.
More importantly are the graphs that show for the Wright Cyclone the power increases over the years (at least up until 1940) and accompanying graph showing the increases in the area of the cooling fins per cylinder. No increase in power happened without an increase in cooling fin area. P & W was constantly trying to improve cooling fin area on it's engines. Wright finally resorted to cutting groves into the cylinder barrels and rolling sheet metal into the groves when they couldn't cast/forge/machine the needed fins.

Forged machined fins on the cylinder head, sheet metal fins on the cylinder barrel. Such fins were used on the 1300hp and up R1820s, the 1900hp R-2600s and the R-3350s used on the B-29s and other WW II aircraft, the very early R-3350s used normal fins.

The Cylinder head, walls and piston are never going to be cooler than the incoming mixture once the engine is running. This is one of the problems. Local hot spots can ignite the mixture before the spark plugs do. supercharged aircraft engines often used excess fuel as a coolant. The change from a specific fuel consumption in the .40s (lb per HP/hr) when cruising to into the .60s if not the low .70s at full power is well beyond the change in stoichiometric mixture of going from lean to rich.
BTW the V-8 Renault air cooled engine in WW I also used a very rich mixture to assist cooling ( heads were said to glow in dim light)

 

[pbehn]

Cooling the cylinder was a major problem and/or limit to the power produced per cylinder of air cooled engines, It may have been for liquid cooled but I have seen no articles or graphs comparing different liquid cooled cylinders.
I have no figures on heat conducted away through the rings but transfer of heat from the piston to the cylinder walls was fairly important, whether through conduction (through the oil film?) of radiation.
P & W had made two different 9 cylinder Hornet engines, the R-1690 which was fairly common and the R-1860 Hornet B which was built in small numbers and according to one account, helped steer P & W to the "more but small cylinders" path.
More importantly are the graphs that show for the Wright Cyclone the power increases over the years (at least up until 1940) and accompanying graph showing the increases in the area of the cooling fins per cylinder. No increase in power happened without an increase in cooling fin area. P & W was constantly trying to improve cooling fin area on it's engines. Wright finally resorted to cutting groves into the cylinder barrels and rolling sheet metal into the groves when they couldn't cast/forge/machine the needed fins.
View attachment 592081
Forged machined fins on the cylinder head, sheet metal fins on the cylinder barrel. Such fins were used on the 1300hp and up R1820s, the 1900hp R-2600s and the R-3350s used on the B-29s and other WW II aircraft, the very early R-3350s used normal fins.

The Cylinder head, walls and piston are never going to be cooler than the incoming mixture once the engine is running. This is one of the problems. Local hot spots can ignite the mixture before the spark plugs do. supercharged aircraft engines often used excess fuel as a coolant. The change from a specific fuel consumption in the .40s (lb per HP/hr) when cruising to into the .60s if not the low .70s at full power is well beyond the change in stoichiometric mixture of going from lean to rich.
BTW the V-8 Renault air cooled engine in WW I also used a very rich mixture to assist cooling ( heads were said to glow in dim light)

Was it that around around 1940 they came up against another law of physics. Increasing surface area only dissipates more heat up to a point, making thinner fins closer together reaches a limit of thermal conductivity and then it is the air flow temperature and density that makes the limit?

 

[Peter Gunn]

Personally I think Packard did a magnificent job producing the Merlin considering how little guidance they got from Rolls Royce. Ford was just Ford in 1940, good at mass production with an adequate engineering staff, Packard on the other hand...

Yeah yeah, Ford built a bunch of R-2800's, I'll assume the quality of those R-2800's was up to Pratt and Whitney standards or the the Air Corps wouldn't have used them.

 

[Deleted member 68059]

Personally I think Packard did a magnificent job producing the Merlin considering how little guidance they got from Rolls Royce.

Rolls-Royce sent over their Chief experimental engineer, chief production-quality engineer and chief designer to stay in America on-site with Packard.

Barrington was dead by 1943 from pure stress, Ellor returned in 1944 and was dead in six years (very prematurely) , Reid (production-quality) had to retire that same year too.

They literally sacrificed three of their most valued engineers in the middle of a war. I`m not sure you CAN get more support than that?

 

[Peter Gunn]

Rolls-Royce sent over their Chief experimental engineer, chief production-quality engineer and chief designer to stay in America on-site with Packard.

Barrington was dead by 1943 from pure stress, Ellor returned in 1944 and was dead in six years (very prematurely) , Reid (production-quality) had to retire that same year too.

They literally sacrificed three of their most valued engineers in the middle of a war. I`m not sure you CAN get more support than that?

Point taken.

Just my dislike of Ford and attending products showing.

 

[Shortround6]

Ford was able to send a team to P&W Hartford and copy the factory layout and parts flow. Not to mention drawings of parts/tooling didn't have to be redone.
Ford did NOT show P&W how it was done, at least until somewhat later during the 2nd/3rd expansion when they came up with new ways of making cylinders.