On the Meredith Effect

Ad: This forum contains affiliate links to products on Amazon and eBay. More information in Terms and rules

The P-40 did not use the Meridith effect in its design ? It's a high drag airplane , so was wondering if it's air scoop had anything to do with that .
 
The P-40 did not use the Meridith effect in its design ? It's a high drag airplane , so was wondering if it's air scoop had anything to do with that .

It was probably pretty average for single engined fighters of WWII, i.e., zero lift drag coefficient of about 0.022. The Mustang was much better, at about 0.017, about 30% better than the average.

Cooling system design for fighters was hard, and nobody did it better than the P-51 team.
 
Was there any Merideth effect in the P-40 design ? I doubt it highly as the P-40 inlet was considered high drag. Thought I'd ask though. Perhaps someone may know if it was at least attempted. I do know that the prototype had a mounting similar to the P-51 before it was moved forward.
 
In fact the designer of the P-40 said that the air intake of the D and later models was too large, causing air to spill out the front of the scoop and upset the airflow down the fuselage and that was why the later P-40's needed a tail extension, because they did not fix the engine cooling air intake problem.
 
Was there any Merideth effect in the P-40 design ? I doubt it highly as the P-40 inlet was considered high drag. Thought I'd ask though. Perhaps someone may know if it was at least attempted. I do know that the prototype had a mounting similar to the P-51 before it was moved forward.
I wish people would read the various GB patents held by TP de Paravicini. He covered the negative drag in various patents including air cooled radial engines.

Of course final designs needed to go through a few iterations to get optimum performance.

The first paper was in 1932. He was at Farnborough before Rolls Royce hired him.

Then we would see he covered almost all variations. I taking from leading edge of wing was a chosen method but wing engineers did not want to redesign wings. So it mostly ended up as a streamlined shape device under the fuselage.
 
Was there any Merideth effect in the P-40 design ? I doubt it highly as the P-40 inlet was considered high drag. Thought I'd ask though. Perhaps someone may know if it was at least attempted. I do know that the prototype had a mounting similar to the P-51 before it was moved forward.
The Meredith effect produces thrust from the heat of the engine. The idea is to reduce the overall cooling drag, I suppose it is possible to have a lot of "Meredith effect" but still have massive cooling drag because of the inlet and other "stuff". In principle, by description, there is little difference between the arrangement on a Hurricane and a P-51, in practice the difference is huge.
 
You can patent the idea or the formulas and attach drawings to illustrate the idea. It doesn't mean that something built to those drawings is actually going to work.

I am not trying to take anything away from TP de Paravicini. His drawings may have worked. However there was a lack of actual experimentation/ test back up at the time.

I believe the British only got a 24 ft wind tunnel into operation in 1935/36? You might be able to squeeze a bipane into it with the wing tips cut off. I have no idea how fast the airflow was.
Remember it was the guys running this wind tunnel that thought (or provided the formulas to the airframe makers) that the Beaufighter I was good for 370mph and the early Typhoon should have gone 460mph.
Lets just say that practical experience was years behind the theory.
We also have to forget the idea that any and all radiators placed on the belly of an airplane were somehow using the Meredith effect just because the Mustang had a radiator on/in the belly and did use the meridieth effect.

The XP-40 with the belly radiator had trouble breaking 300mph so obviously the under nose radiator set-up was better. Still doesn't mean it used the Meredith effect, or used it to any real benefit.

With the P-40 you had 4 or more different nose radiator setups. the XP-40 went through more than one(?), the P-40/P-40B/P-40C used another. the P-40D/E and later Allisons used the 3rd and the P-40F & L used 4th.
This is complicated by the fact that the P-40 used some different Allisons with different power ratings and used the Merlin which had different radiator/oil cooler requirements than the Allisons. These different engines affect the cooling airflow requirements and thus the size of the radiator/oil cooler inlet and/or outlet. Once you get passed the XP-40 the long nose planes used a 1040hp engine (if it wasn't being abused) while the D & E got a 1150hp engine. I have no idea how much larger the radiators got if they got any bigger.

P-40coolers.jpg

Upper picture ins the long nose P-40, lower ones are the radiators/oil cooler in the later Allison models. There is little or no expansion of the airflow before it hits the radiators.

rear view of the radiators
P-40shutters.jpg
Photo credited to Scott Murphy.

It is a museum plane and while the outlet is adjustable the actual airflow in the "duct" is short and has some sharp corners to get around. Not a lot of expansion space.

when they changed to the D & E you got this
p-40-nose.jpg

Please note the centerline of the prop moved up 6 inches and the prop moved back. I don't know if they lowered the bottom of the radiator cowl any but the inlet opening was sure raised at the top.

The P-40F used a Merlin rated at 1300hp and used a bigger radiator and two oil coolers, It did drop the radiator inlet opening.

Late model P-40s used engines rated for 1200hp and were officially rated for WER power levels.

How much the later/higher powered engines also impacted the need for a longer tail I don't know.
 
On the P-38 the radiators for the early models, located back in booms, had a simple scoop intake. The J, L, M models had radiator scoops that were further out in the airstream, and featured a lip that separated the boundary layer air from the faster moving outer air. Even though the engines on those models were more powerful, in reality the radiators were the same size as the earlier models. Moving the radiators out where the faster air could get to them gave better cooling.

On the P-51 Mustang they put a lip on the bottom airscoop as well. And the P-80 had splitter plates on the jet engine intakes for the same reason, to separate the boundary air.
 
You can patent the idea or the formulas and attach drawings to illustrate the idea. It doesn't mean that something built to those drawings is actually going to work.

I am not trying to take anything away from TP de Paravicini. His drawings may have worked. However there was a lack of actual experimentation/ test back up at the time.

I believe the British only got a 24 ft wind tunnel into operation in 1935/36? You might be able to squeeze a bipane into it with the wing tips cut off. I have no idea how fast the airflow was.
Remember it was the guys running this wind tunnel that thought (or provided the formulas to the airframe makers) that the Beaufighter I was good for 370mph and the early Typhoon should have gone 460mph.
Lets just say that practical experience was years behind the theory.
We also have to forget the idea that any and all radiators placed on the belly of an airplane were somehow using the Meredith effect just because the Mustang had a radiator on/in the belly and did use the meridieth effect.

The XP-40 with the belly radiator had trouble breaking 300mph so obviously the under nose radiator set-up was better. Still doesn't mean it used the Meredith effect, or used it to any real benefit.

With the P-40 you had 4 or more different nose radiator setups. the XP-40 went through more than one(?), the P-40/P-40B/P-40C used another. the P-40D/E and later Allisons used the 3rd and the P-40F & L used 4th.
This is complicated by the fact that the P-40 used some different Allisons with different power ratings and used the Merlin which had different radiator/oil cooler requirements than the Allisons. These different engines affect the cooling airflow requirements and thus the size of the radiator/oil cooler inlet and/or outlet. Once you get passed the XP-40 the long nose planes used a 1040hp engine (if it wasn't being abused) while the D & E got a 1150hp engine. I have no idea how much larger the radiators got if they got any bigger.

View attachment 538400
Upper picture ins the long nose P-40, lower ones are the radiators/oil cooler in the later Allison models. There is little or no expansion of the airflow before it hits the radiators.

rear view of the radiators
View attachment 538401Photo credited to Scott Murphy.

It is a museum plane and while the outlet is adjustable the actual airflow in the "duct" is short and has some sharp corners to get around. Not a lot of expansion space.

when they changed to the D & E you got this
View attachment 538402
Please note the centerline of the prop moved up 6 inches and the prop moved back. I don't know if they lowered the bottom of the radiator cowl any but the inlet opening was sure raised at the top.

The P-40F used a Merlin rated at 1300hp and used a bigger radiator and two oil coolers, It did drop the radiator inlet opening.

Late model P-40s used engines rated for 1200hp and were officially rated for WER power levels.

How much the later/higher powered engines also impacted the need for a longer tail I don't know.

Bottlehead replies. There are numerous patents that TP de Paravicini alone or with his boss at Rolls Royce Edwin Ellor filed that covered various methods of
cooling by means of utilizing waste heat. TP de Paravicini was at Farnborough research prior to being recruited to Rolls Royce.

It appears the Americans latched on to Meridith. His patent is preceded.
Rolls-Royce Crecy engine which was to be installed in Spitfire
.
 
Bottlehead replies. There are numerous patents that TP de Paravicini alone or with his boss at Rolls Royce Edwin Ellor filed that covered various methods of
cooling by means of utilizing waste heat. TP de Paravicini was at Farnborough research prior to being recruited to Rolls Royce.

It appears the Americans latched on to Meridith. His patent is preceded.
Rolls-Royce Crecy engine which was to be installed in Spitfire
.

Technical Data Digest


The net has lost a lot of text. So continuing.

Here is a list of GB patents by Ellor and de Paravicini. GB463303, GB471371, GB472334, GB472555, GB472820, GB473340, GB473341, GB484300 Then one of his patents on exhaust ejectors GB471177.

All of these relate to cooling of engines and the various methods.
I don't have access to papers publish by him in the Aeronautical Society.
 
The designer of the P-40, Berliner, claimed that the longer tail on the P-40 was required due to the front of the radiator being too large, which caused excess air to spill out of the front and flow back along the fuselage. He considered the fuselage length extension to be a crude way to handle the problem, when redesigning the air scoop would have been better. He left Curtiss over such issues.
P-40-Guadalcanal.jpg
 
I believe the British only got a 24 ft wind tunnel into operation in 1935/36? You might be able to squeeze a bipane into it with the wing tips cut off. I have no idea how fast the airflow was.
Remember it was the guys running this wind tunnel that thought (or provided the formulas to the airframe makers) that the Beaufighter I was good for 370mph and the early Typhoon should have gone 460mph.
Lets just say that practical experience was years behind the theory.
We also have to forget the idea that any and all radiators placed on the belly of an airplane were somehow using the Meredith effect just because the Mustang had a radiator on/in the belly and did use the meridieth effect.

Bottlehead reply,
Meridith is with a capital M please,
You are being very disingenuous about patents. and theory,

Tom de Paravicini had done all the theory as he was the best mathematician at Rolls Royce at the time and also became the youngest to become a member of the Royal Aeronautical Society at the time,

Rolls Royce cross licensed many patents with the Germans prior to WW2 and that is how the ME109 got the same cooling arrangement as well as being offered the Merlin engine design.

His theory was born out in practice by the increased speeds achieved in the Spitfire coupled with his exhaust ejectors overall added about 50 mph.
He based his calculations on a base of 250 mph.
He called it "Negative Drag" not effect, Thermo-dynamics was his forte, Later wrote papers on gas turbines etc.

Further more a scale model can be tested in a wind tunnel. as did designs of Concord in the 1960s.

There is a patent of a wind tunnel design he did at Bristol Aircraft (Armstrong) in 1942 Patent GB536189 to test engines in a real high speed air stream.

I suggest you read the CV of Trescilian a bright engineer.
Stewart S. Tresilian - a C.V.
 
Supercharged or turbocharged Meredith cooling system has much more favorable characteristics.
I-250(MIG-13) and SU-5 motorjet aircrafts had a compressed air cooling system with afterburner. Afterburner was only used for a maximum speed 825 km/h. During cruise at 700-750 km/h higher nozzle pressure ratio was improving Meredith effect, compensating cooling system air resistance and air compressor power. Even greater effect was estimated by using compound exhaust gas turbine to power the air compressor and additionally heating the cooling air with exhaust gases.
The other example was 4000 hp motorjet engine R-130 with the weight of 1400 kg and takeoff thrust 2000 kg, and fuel consumption 600 kg per hour at Mach 0.7 (1500 kg thrust).
 

Attachments

  • MIG-13 motorjet aircraft.jpg
    MIG-13 motorjet aircraft.jpg
    304.5 KB · Views: 166
  • R-130 4000hp motorjet.jpg
    R-130 4000hp motorjet.jpg
    56 KB · Views: 135
  • su-5 motorjet.gif
    su-5 motorjet.gif
    207 KB · Views: 209
  • WW2 motorjet aircrafts.jpg
    WW2 motorjet aircrafts.jpg
    99.5 KB · Views: 111
The aeronautics community, at least until the mid-1930s, was both fairly small and quite international. For a wonderful example of this, see the 1935 Volta Conference. With the rise of pathological nationalism, e.g., in Germany, this ended.
 
It would require an efficient inlet, to slow the air, a minimal air pressure drop in the radiator, and an efficient nozzle. Having the highest possible temperature in the radiator would also help.
I get an efficient inlet, an efficient nozzle, the highest reasonable temperature differential between radiator air: What I don't get is a minimal pressure change at the radiator. I'd figure you'd want the heat to rapidly heat and cause a maximum pressure change.

to add to that the exit nozzle/area, needs to be variable in area in small steps, A fixed exit or two position exit is only going to be efficient an certain airspeeds and air densities.
Yeah, you'd need to adjust the nozzle to deal with the airflow changes through the duct.

Looking at the P-40F

28626081280_562b6e2c62_b.jpg


While the basic shape seems good, it seems that the basic problem is that there's no splitter to remove turbulent airflow from the duct. It was highly important that the airflow into the duct be smooth. I remember that in a book I read on the Spitfire (and it was a problem in the Spitfire).

The designer of the P-40, Berliner, claimed that the longer tail on the P-40 was required due to the front of the radiator being too large, which caused excess air to spill out of the front and flow back along the fuselage. He considered the fuselage length extension to be a crude way to handle the problem, when redesigning the air scoop would have been better. He left Curtiss over such issues.
That's why he left Curtiss? Over the inability to design the radiators how he wanted?
 
Looking at the P-40F

View attachment 573550

While the basic shape seems good, it seems that the basic problem is that there's no splitter to remove turbulent airflow from the duct. It was highly important that the airflow into the duct be smooth. I remember that in a book I read on the Spitfire (and it was a problem in the Spitfire).

How comes that basic shape seems good?
 

Users who are viewing this thread

Back