What-if upgrading considerations.

[Shortround6]

1. Changing/adding supercharger gears.
Superchargers had some definite limits. One was impeller tip speed. If the tips of the impeller exceed the speed of sound (at least by very much) shock waves form and disrupt the air flow through the supercharger severely limiting it's airflow. Due to the temperature and pressure inside a supercharger the speed of sound is higher inside the supercharger than in the open air outside but tip speeds around 1250ft/sec seem to be a good guide line. If the supercharger is already operating at or near this limit upping the gear ratio will do no good.
2. Dry engine weights vs "powerplant" weights. Dry engine weights not only do not include oil and coolant they do not include starters, generators, exhaust systems, gun synchronizers, vacuum pumps and other odd bits and pieces the engine actually needs to function in the aircraft. Bigger engines need bigger propellers and cowling. As a rough rule of thumb about 2 pounds per horsepower seems to come out well for powerplant weights, a bit lighter for single stage supercharged engines and a bit higher for turbo engines. Navy R-2800s come out at about 3900lbs not including fuel system.
3. Extra drag of some upgrades, Boulton Paul Defiant went from the Merlin III to the Merlin XX and from 1030 hp to around 1200hp at altitude and picked up 9mph in speed. The Merlin XX needed a bigger radiator and the extra drag of the radiator killed most of the extra performance. Maybe with more time and experimentation a lower drag radiator setup could have been worked out. ALL of the extra power of an upgrade or swap is NOT available for extra performance, some of it is used up in extra drag.
4. Other "improvements" needed. The wing or wing group of the Hawk 75 went from 842lbs of the commercial model to about 1000lbs on the early P-40s (The Hawk may have needed a little beefing up any way, reports of wing skin buckling on P-36s) and then to 1120-1130lbs on the P-40E and up. Different gun bays or the wing had to be beefed up to keep the 12 "G" load requirement as the plane went from 5692lbs to 6800-7500lbs (P-40-P40C) to 8300-8500lbs (P-40E and up) ??? or a bit of both??? landing gear weights also changed.

 

[davebender]

Wouldn't 12 Gs kill the pilot?

 

[Shortround6]

Wouldn't 12 Gs kill the pilot?

It would if they actually expected the plane to pull 12 "G"s and if the plane held 12 "G"s for more than a moment in time.

They actually wanted an 8 "G" limit with a 50% safety margin. Stress analysis was a fairly new field and in the 1930s nobody really know what the limits were, throw in metal fatigue, dynamic loading (instead of static loading with sandbags) and production tolerances and the high limit doesn't look so strange.

 

[tyrodtom]

In 1954 a John Stapp withstood 46 Gs for a brief part of a second, 25Gs for 1.1 second on a USAF rocket sled.
12-16 Gs can be withstood without fatalities for almost a minute, though the person wouldn't be able to see at the lower limit, or conscious at the higher limit. Of course if you're flying the aircraft, either could be fatal.

There have been some Formula 1 crashes that have recorded close to 100g peaks, and the driver survived.

 

[tomo pauk]

As a rough rule of thumb about 2 pounds per horsepower seems to come out well for powerplant weights, a bit lighter for single stage supercharged engines and a bit higher for turbo engines.

Were the both liquid- and air-coled paying same weight penalty for HP delivered?

 

[Njaco]

In 1954 a John Stapp withstood 46 Gs for a brief part of a second, 25Gs for 1.1 second on a USAF rocket sled.
12-16 Gs can be withstood without fatalities for almost a minute, though the person wouldn't be able to see at the lower limit, or conscious at the higher limit. Of course if you're flying the aircraft, either could be fatal.

There have been some Formula 1 crashes that have recorded close to 100g peaks, and the driver survived.

D@amn!! Talk about a thrill ride.

 

[parsifal]

You would have thought that with that load limit, you would suffr multiple broken bones.....

 

[Shortround6]

Were the both liquid- and air-coled paying same weight penalty for HP delivered?

It comes out close. The air-cooled doesn't have the radiator of course but then most air-cooled engines didn't quite get the same HP per pound that the liquid cooled engines did (dry weight). This, of course, bounces all over the place as different models and boost limits are used. A few pound per HP figures.
Wright R-1820 at 1200hp-----1.10lbs per HP
P&W R1830 at 1200HP -------1.22lbs per HP
Allison V-1710C at 1090HP----1.23lbs per HP
Allison V-1710F at 1150hp----1.14lbs per HP
Allison V-1710F at 1325hp----0.99lbs per HP
Packard V-1650-1 at 1300hp-1.16lbs per HP
Merlin XX at at 1480HP-------0.98lbs per HP
Wright R-2600 at 1700HP-----1.16lbs per HP

Please note the weight penalty paid by P&W by going to a two row design for the radial. They do get a smaller frontal area though
Air-cooled engines almost never got a WER rating without ADI.
Some figures for engine installations. numbers are pounds per horsepower.

engine.....................................1.000-1.300
Propeller and controls.................0.300-0.350
Engine mounts..........................0.045-0.080
Cowling...................................0.063-0.100
Exhaust system........................0.027-0.087
Carb. air scoop.........................0.015-0.025
Oil system (empty)....................0.040-0.060
Fuel system (except tanks).........0.020-0.040
Starter....................................0.020-0.045
Generator................................0.025-0.035
Misc.......................................0.015-0.030
radiators and coolant.................0.200-0.280
Increase over single stage supercharger
Mechanical drive with inter-cooler..0.150-0.250
Turbo with ducts, intercoolers etc..0.300-0.400

Liquid cooled engines tend to be at the lower end of the Lbs per Hp ratio to begin with (when the Allison was giving 1090hp the R-1820 was usually rated for 1100hp and the R-1830 was 1050-1100hp), they usually needed smaller cowls and perhaps smaller oil systems/coolers. Most V-12s used simpler exhaust systems than most radials.
This chart is from "Aircraft Power Plants" by Fraas, 1943 and is for engines between 1000-2000hp.
Exceptions can always be found.

 

[tomo pauk]

Thanks for the informations

 

[merlin]

Shortround6:
You mentioned the Defiant, considering the P.94 (single seat version) with Merlin XX had a max speed of 360 mph - how much less would it have been with the Merlin III?
The Boulton-Paul P.88a had an est. max speed of 337 mph with a 1,500 hp Hercules, how much less with the earlier engine - mk 1 = 1,290 hp, mk 2 = 1,375 hp., and more with the mk 4 at 1,650 hp??

 

[Shortround6]

Some of those are estimates. Another consideration is that most of those HP figures are take off and not power at altitude.

 

[drgondog]

The sled was doing about 600mph on tracks and slowed by running the tracks through a water trough. The entire exercise was in part to understand the stresses of punching out at very high speeds.

 

[Shortround6]

The power required for a particular aircraft's speed increase goes up with cube of the speed. Doubling the speed requires 8 times the power.

MK I Spitfires with 880hp did about 280mph at sea level. MK XIV Spitfires with 1850hp did about 360mph at sea level, a 28.6% increase for double the power. This actually works out closer than I thought it would. Take the cube root of 880hp. Multiply by the 28.6% increase and the cube the result. it comes out at 1861hp with a bit of rounding off. Considering some of the modifications to the Spitfire this seems fairly close.

A 50% increase in power should give a 11.45% increase in speed.

 

[tyrodtom]

It's not likely you could get a 50% increase in power, without more weight. With no other changes aerodynamically the aircraft would have to produce more lift during take offs and landings to support the increased weight, so both are at higher speeds. Plus more fuel is used to feed the hungrier engine, so more weight, or less range.

But everybody always wants more power, so all fighters usually got more powerful, bigger, heavier.

How much did the landing speed go up from the Spitfire I to the Spitfire XVIII ? I see the top speed at altitude increased by about 100 mph .

 

[Shortround6]

How much did the landing speed go up from the Spitfire I to the Spitfire XVIII ? I see the top speed at altitude increased by about 100 mph .

From the Spitfire performance site it appears that the stalling speed with flaps and gear up went from 78mph ASI to 87 mph ASI and with gear and flaps down it went from 68mph ASI to 75mph ASI.

 

[merlin]

So for a quick 'rule of thumb' calculation for every % increase in HP divide by three, and multiply the previous speed by the answer.
But how does it work with decrease in the expected power available!?

 

[Shortround6]

You have to take the power of the existing engine, find the cube root of it. Then figure increase in speed, multiply the percentage increase times the cube root and then cube the answer. as in 1000hp has cube root of 10. times a 10% increase (110% of original speed). 10 X 1.1=11 or for a 20% increase in speed 10 X 1.2=12. the cube of 11 is 1331 and the cube of 12 is 1728.

For less power try going the other way. 10 X 0.9 for a 10% reduction in speed. 9 cubed gives 729hp needed for a speed 10% lower than 1000hp will give.

For us math challenged people this web site does all the hard work.

Cube And Cube Root Calculator

 

[drgondog]

The 12G is an engineering reference for the 'point' the a/c is expected to have structural/catastrophic failure and doesn't reaaly concern whether the pilot can take the load because it is considered a 'moot' question.

 

[drgondog]

Recognize that as speed increases, thrust of the Propeller/engine system decreases (T=propefficiency*BHp/Velcity) and drag increases with square of velocity - so there are asymptotic diminishing returns on V=f(Hp) for a fluid with real friction (like air).

 

[Shortround6]

If I understand the drag thing right, while the drag increases with the square of the speed that is the "force" of the drag at any given moment. As in going from 300 mph to 350mph will increase the drag "force" at a given moment by about 36% but since you are also going about 16% further in the same period of time (second, minute or hour) the work being done goes up with the cube of the speed.

Does that sound right?