Ki-61 Hien (Tony) Radiator / Meredith Effect?

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

Bronc

Banned
57
0
Oct 6, 2009
Did the design of the radiator on the Ki-61 Hien (Tony) take advantage of the
Meredith Effect and thus provide additional propulsion? The ventral placement
on the Ki-61 is similar to the P-51 and there is even an adjustable exhuast door.

Bronc
 
Most liquid cooled aircraft engines are going to have an ajustable door/flap on the radiator to control airflow through the radiator for different flight conditions. THe Meredith effect was put out in a paper in 1935/6 (along with exhaoust gs propulsion) so the Japanese might have been aware of it. It seems it was quite a trick to actually get "propulsion" from it verses a reduction in cooling drag.

Say your "standard" radiator installation caused 50lbs of drag and you tried for the Meredith effect in a new installation and hoped for 20lbs of positive thrust. Say when tested the new installation provided no positive thrust but showed a total drag of only 15 lbs (50lbs drag -35lbs thrust=15lbs drag) are you using the Meredith effect?
 
I've heard that the Mustang's radiator design added thrust to the effect of an extra 300 hp, but I'm very unsure about that to say the least.
 
Did the design of the radiator on the Ki-61 Hien (Tony) take advantage of the
Meredith Effect and thus provide additional propulsion? The ventral placement
on the Ki-61 is similar to the P-51 and there is even an adjustable exhuast door.

Bronc

Welcome Bronc

Listen, each time you have P1/P2 and T1/T2 pressure and temperature increase in a airflow tube you always obtain the Meredith effect whatever you want it, or not.

It was first discovered, used and patented for the statoreactors (and pulsoreactors) by the frenchman René Leduc. The laws are far from being linear and even equationnable as Maredith thaught himself and it takes Karman-Nikuradze curves and other abaccus to 1) establish flow regimes, then 2) choose the right thermodynamic equation systems in order to etablish real thrust values.

In my opinion both T and P increase are to small inside the P-51 radiators duct to provide valuable thrust. Even if it was optimized for that.
 
Last edited:
...each time you have P1/P2 and T1/T2 pressure and temperature increase in a airflow tube you always obtain the Meredith effect whatever you want it, or not.

In my opinion both T and P increase are too small inside the P-51 radiator duct to provide valuable thrust...
That wasn't really the question though, was it? The originator asked if Japanese engineers made any attempt to exploit Meredith Effect.

Nice to hear your opinion. Can you explain how a small, sleek aircraft like the Spitfire IX required 200hp more to fly at the same 400mph airspeed as the P-51? And why the P-51 went on to be in the region of 25mph faster? What do you think wasn't being significantly overcome in the Spitfire, that was in the P-51?

Out of interest, what are the T and P figures for the P-51 radiator duct?
 
That wasn't really the question though, was it? The originator asked if Japanese engineers made any attempt to exploit Meredith Effect.

Ok, i don't know if they take advantage or exploited it. But for sure they had advantage from Meredith effect, even if they were unaware about that principle.




Nice to hear your opinion. ... And why the P-51 went on to be in the region of 25mph faster? What do you think wasn't being significantly overcome in the Spitfire, that was in the P-51?

Well 703-657 km/h = 54 km/h so 33mph.

First i would say that Yak-7l (l for laminar) was 617-573 = 44 km/h faster than the serial one. Even considering the lighter Yak-1 from that times it makes at least 617-590 = 27 km/h increase. This for 17,15 m² wings.
For 22 m² wings like Mustang or Spitfire it (laminarity) should give from 35 to 55 km/h in speed increase.

Secund ,it should not be ignored the hudge difference in manufacturing quality from the P-51 to the Spitfire. It's not a secret that the elliptic wing had often hollows and humps. So for the rest of wet aeras due to rugosity and friction drag, it certainly reduced the Spit max speed from other 15 to 20 km/h at least from the Mustang.

Third, the Mustang had higher wing load, but it plays not a great rule in the max speed value. But yes , at the same WL as the SpitIX it (theP-51D) should be a little faster due to induced drag reduction, and it can be explained by some cooling drag reduction this time.

Fourth, there is a favourable radiator position the Mustang airframe, respecting someting like the "aeras low" rules. (No brutal increase, no brutal mean section decrease that means no brutal decomprssion, no brutal recompression). I don't know for the Mustang but the MiG-3 gained about 25 mph from the MiG-1 removing aft it's radiator bath.




Out of interest, what are the T and P figures for the P-51 radiator duct?

Well i can't tell nothing, without Mustang radiators "porosity" (pressure loss) value and overall section. Probably no more than 373 K/293 K ~1.3 and T1/T2 value and about 1.2 - 1.5 in P1/P2 due to speed decrease inside the radiator elements. And i m even not sure from that last, since there might be no increase, but at opposite some pressure drop inside the radiator walls/tunnels.

In fact the real and indsputable drag economy in the big Mustang's radiator is mainly due to Bernouilli law (speed decrease from outside, in divergent, radiator washed at lowered speed, convergent duct) than Meredith effect. In my opinion...

qv being constant, a 10% increase in cross-aera (q) makes 10% decrease in speed so (10%)²= 21% in drag.

In final words, there is enough good reasons to explain Mustang performance even without evocating a questionnable Meredith effect.
 
Last edited:
Ok, i don't know if they take advantage or exploited it. But for sure they had advantage from Meredith effect, even if they were unaware about that principle.

I agree this, and further speculate that obtaining Meredith effect thrust was not calculated in the radiator cowl design. The original design concept was pruchased from Curtiss and then significantly improved relative to drag reduction first on the XP-51, then further refined on the P-51B, and last on the XP-51J/P-51H.

Secund ,it should not be ignored the hudge difference in manufacturing quality from the P-51 to the Spitfire. It's not a secret that the elliptic wing had often hollows and humps. So for the rest of wet aeras due to rugosity and friction drag, it certainly reduced the Spit max speed from other 15 to 20 km/h at least from the Mustang.

I would tend to disagree this point but could be convinced with facts.

First for this to be a common condition it would have to be proven that the 'hollows and humps' were in the leading 25% chord region where normal attached flow could be disrupted to earlier boundary layer separation. Second, existance of such humps and flows in the regions past 25% chord would perhaps contribute to additional but nevertheless very small increments in parasite drag. Last, the regions between the leading edge and 1/4 chord point is a shorter distance with implied greater stiffness and regularity in the airfoil skin over that region. The last is very speculative but one would really have to screw up the tooling to get significant variations from loft contours here


Third, the Mustang had higher wing load, but it plays not a great rule in the max speed value. But yes , at the same WL as the SpitIX it (theP-51D) should be a little faster due to induced drag reduction, and it can be explained by some cooling drag reduction this time.

This is a tricky assumption. The Spit wing had greater theoretical efficiency factor due to elliptical wing (small) and slightly greater aspect ratio. Induced drag should be slightly less for the Spit at the same speed. For the same speed the Lift loading (WL) will be smaller than the 51 and becaues the Spit Max CL slope is higher the Spit should be flying at a slightly lower angle of attack - with primary difference really reduced to the both the drag of the Wing and parasite drag of the Fuselage/Radiator system. The Mustang was clearly superior for both of these designs relative to parasite drag.

The parasite drag difference is dominant at top speeds.


Fourth, there is a favourable radiator position the Mustang airframe, respecting someting like the "aeras low" rules. (No brutal increase, no brutal mean section decrease that means no brutal decomprssion, no brutal recompression). I don't know for the Mustang but the MiG-3 gained about 25 mph from the MiG-1 removing aft it's radiator bath.

The cross sectional fuselage area at the radiator cowl location slightly increases from the cross section just forward. In my opinion the intake design and the position relative to the boundary layer was the number one factor in the drag reduction as a % of parasite drag. Resulting sepration past this point would have far less effect than a cowl mounter radiator - but equally if flow separation was a significant factor for both the P-51 and Spifire downstream of the radiator - the Spit would win because there is nothing to create parasite drag behind its wing mounted radiator cowls - so I tend to discount this as a major factor.

Well i can't tell nothing, without Mustang radiators "porosity" (pressure loss) value and overall section. Probably no more than 373 K/293 K ~1.3 and T1/T2 value and about 1.2 - 1.5 in P1/P2 due to speed decrease inside the radiator elements. And i m even not sure from that last, since there might be no increase, but at opposite some pressure drop inside the radiator walls/tunnels.

Neither can I

In fact the real and indsputable drag economy in the big Mustang's radiator is mainly due to Bernouilli law (speed decrease from outside, in divergent, radiator washed at lowered speed, convergent duct) than Meredith effect. In my opinion...

qv being constant, a 10% increase in cross-aera (q) makes 10% decrease in speed so (10%)²= 21% in drag.

In final words, there is enough good reasons to explain Mustang performance even without evocating a questionnable Meredith effect.


Yes. The wing selection, the radiator cowl, the exceptionally clean fuselage design including windscreen and canopy (malcolm hood for P-51B), and manufacturing quality/surface regularity - all played a role

I am with you that a.) true analytical calculation of thrust must take into account true porosity of the radiator in the equations and I personally have never seen mass flow rate calculations or Temperature/Pressure values in any drag profile for the Mustang. I am willing to believe Meridith effect or at least suspend disbelief pending those data.

Oh I forgot. It is probable that the Ki 61 radiator design with respect to Boundary Layer control was as rigorous as the P-51. I have no opinion on Meredith effect increment to thrust for all the reasons I am agnostic on the Mustang.

As to the primary factor, I remain on the side of analysis that says the parasite drag reduction was a combination of geometry and boundary layer control more than any other factor - until proven otherwise.
 
Last edited:
and from the great man himself

An Engineer's Perspective on the Mustang
by J Leland Atwood
from Flight Journal June 1999

North American Aviation's (NAA) Mustang fighter is generally credited with a 20 to 30mph speed advantage over most of its WWII contemporaries. This speed advantage also permitted a considerable increase in range that required more fuel, but not enough to significantly reduce speed. Records show that some 275 U.S. aces were 'made' in P-51s. The reasons for the Mustang's significant performance capability have never been clearly explained and I hope to clarify why its aerodynamic features enabled this capability.

To begin: in 1940, the British Purchasing Commission, which I dealt with, had a member - H.C.B. Thomas from Farnborough whom I found to be familiar with the Meredith Report. This report outlined a feature that could enhance the performance of any internal-combustion engine at high speeds by using a radiator form of heat dissipation. A low-velocity airflow through the radiator was one element of this, and it was apparent to me that the larger the radiator, the lower the speed of the air flowing through it; this approached one of the Meredith Report's objectives.

I therefore offered Mr Thomas sketches and other descriptions of a Mustang design that had the main radiator in the rear of the fuselage. The alternatives were wing radiators such as those used on the Spitfire and the Bf 109, and under-engine radiators such the P-40's; both positions limited radiator size and the length and size of the ducting that could be used to handle and control the cooling air.

In addition to the radiator's rearward position, after the design contract had been awarded and at the recommendation of NAA's aerodynamics group, it was decided to use a new airfoil of a class generally designated as 'laminar flow'. This was being developed at NACA (later NASA) at Langley Field, Virginia. A 1939 report by Eastman Jacobs and others at Langley contained the results of the tests of some small laminar-flow airfoils. The drag on these small models was quite low, and there was some hope that laminar flow could be achieved much farther back on an airfoil than had been predicted by previous investigators. The publishers of the report, however, warned that they had not been able to obtain laminar flow on wings of anywhere near the size of those required for actual aircraft and that their tests were to be taken only as the results from laminar-flow models of not more than six inches in width.

In spite of this warning however, both Ed Horkey (leading aerodynamicist at North American) and Bell Aircraft's chief engineer, Robert Woods, decided to try laminar-flow profiles on the P-51 and the P-63 respectively. These airfoils were incorporated on the Mustang and the Kingcobra airplane with the hope that laminar flow could be extended well back on their wings. Extensive efforts were made to polish and protect the P-63 wing's leading edge profile, but the results were equivocal. Those who advocated the laminar flow wing felt that the Mustang's outstanding performance resulted from laminar flow over most of the wing. Kingcobra designers felt they were getting a similar effect, although that aircraft's performance did not justify this conclusion.

With respect to the Mustang, many tests - including some in recent years - have shown that extensive laminar flow was not developed on the Mustang wing and that the drag of the wing was probably no less than that of conventional wings of the same thickness and taper ratio. On the other hand, the Mustang's cooling drag was much lower. This was the result of using a ducted radiator with a large area and a slow-speed airflow through it (Pr and P2); closing up the exit and creating a backpressure restored the momentum of the cooling of air (momentum lost in radiator transit). This was possible because of the radiator's cooling capability which, to be adequate in a full-power climb, was much more than that required at high speed and high dynamic pressure. According to calculations given in a supporting paper, the drag created by momentum loss in passing through the radiator can be reduced from some 400 pounds to close to 30 to 40 pounds because of the offsetting momentum of the jet thrust from the radiator exit (V2).

Since these two effects ie the wing drag and the radiator momentum recovery have never been disentangled in the literature, a technical reason for the Mustang's performance has never been clearly identified.

NACA had taken the lead in airfoil development and had worked out a large series of airfoils that were used generally throughout the industry. For instance, the Spitfire wing was of the NACA 2200 series - 13 percent thick at the root and 6 percent thick at the tip. This is the same airfoil series as is used on the DC-2 and the original North American BT-9 and AT-6 trainers. To improve the stall characteristics, I later changed the NACA 2200 series on the AT-6 trainer to the 4412 at the tip. It is quite probable that the Spitfire's wing, being only 6 percent thick at the tip, had a lower drag than the Mustang's wing as actually incorporated.

In recent years, considerable effort has gone into attempts to extend the laminar flow on wings of commercial aircraft, and the actual airfoil profiles around today have some of the characteristics of the 1939 attempts to make laminar-flow wings. But in general, attempts to extend the laminar flow have been unsuccessful without boundary-layer control devices such as perforated wing surfaces and powerful pumps. The wings on today's commercial aircraft use airfoils that are more wedge-shaped, but they do not attempt to develop extended laminar flow. These airfoils do, however, have one characteristic that is very important to modern airplane operators, and that is the increased aircraft speed at which portions of the local airflow go supersonic.. This is usually referred to as the wing's 'critical Mach number'.

The early aircraft wings had rather bulbous shapes (first developed by Joukowsky) and more of a teardrop appearance. These were capable of a good lift coefficient at relatively low speed, but the sharp curvature of the upper portion of the leading edge caused the air to speed up excessively, and parts of it would approach sonic speed at a relatively low airplane velocity. This characteristic was not inhibiting in earlier aircraft because of their generally low speeds.

On today's commercial aircraft, the wedge-shaped airfoil in use has increased the aircraft speeds at which supersonic air velocity is developed over the airfoil's top. This increase in airplane speeds from Mach .75 to about Mach .85 is of great significance to managers and operators of commercial airlines involved in scheduling necessities and traffic control throughout the world. The critical-Mach-number increase to that point enabled aircraft to cruise at around Mach .85 without incurring large drag or fuel-flow penalties. On the other hand, there is no evidence of a greatly increased laminar flow at lower speeds on these aircraft.

The point of all this is that nearly all WWII fighters operated at Mach numbers of .65 or less. The primary advantage presented by the so-called laminar-flow wing was therefore not in drag reduction but in high-speed dives, where temporary airspeed shock waves were created on the wing's upper surfaces and a loss of control and lift occurred as the critical Mach number was exceeded. This was a phenomenon we called 'compressibility', and it became the subject of a huge amount of research. The Mustang pilot, with his laminar flow wing, had a higher critical Mach number, so he could point the nose down and know he could out-dive virtually any airplane and recover relatively easily. The P-47 and P-38 however, with their older, fatter wings would hit compressibility and have to use their dive flaps to recover safely. So, besides being an overall clean design, the legendary Mustang's speed and range rest as much on carefully designed radiator airflow as on anything else. As is often the case in aircraft design, it was the seemingly small details that counted.
 
Last edited:
Colin - good find on the Atwood article. It would be very hard to argue his persepective.

On the other hand Lednicer's VSAERO model showed very good correlation between the finite element model results for drag contribution and the various part and full scale wind tunnel tests on the Mustang. This is the same consultant (and modelling technique) employed by the big bore racing teams as they continuously strive to reduce drag. Lednicer's study and results were accomplished after Atwood passed away in 1999.

What is without debate today is that the hoped for laminar flow did not stay attached all the way to the predicted (and hoped for) 40% chord point where the pressure gradient should have increased enough to create BL separation. Separation in fact did occur closer to 25-30% chord similarly to non laminar flow wings of similar T/C.

The Laminar flow wing airfoil did produce some positive results as predicted. 1.) the Bondary Layer build up was more gradual resulting in less CD0 for the wing, 2.) the transonic shock wave did in fact arise where predicted - namely at point of peak velocity - at the approximate 40% chord line. Where the BL separates and what the profile looks like is huge in wing parasite drag and the Mustang laminar flow wing in fact yielded positive results in this area.

The former resulted in less drag than a similar, conventional, airfoil with same thickness and taper.

The latter had the effect of negating any pitch forward due to increase of Moment Coefficient at critical mach - found in all conventional wings with peak velocity distribution around the 1/4 chord point. Both the 51 and the 47 had nearly identical critical mach numbers but the Mustang did not attempt to pitch down.
 
On the comparative build quality of Spits I was watching a Discovery doco interviewing an engineer who was restoring both a Spit and a Messer, and also a RAF pilot who flies Spits in the Battle of Britain airshow who gave a pilot evaluation of the cockpits (this episode is available on You Tube in five parts called something like Facts and Figures of the Spitfire).

In comparing the Spit construction to the Messer (and American warbirds of the era) the engineer noted the Spit was less "clean and neat" with its skinning and general construction, less refined and what he termed "more old fashioned in building style." Joins were not as smooth, it had drag inducing triple rolled edges and folds, didn't have comparable quality flush riveting, the best part of it he said was the engine. By comparison the Messer (like American warbirds) had very refined build quality and he was actually surprised by it being higher than he expected, it reminded him of German Formula One racers of the era and he said it was actually probably even over-engineered for its requirements, but overall a very high build quality.
I presume you'd probably find similar differences with a Mustang.
 
Vanair, I saw that show and here's a few comments.

Every time I hear one of these so-called "experts" comment on the quality of construction, it makes me wonder. Most of the major sub assemblies that made up WW2 fighters were built in a production jig, and things like rivet lines, and gaps between skins are dictated by that tooling and engineering requirements and there were limits constantly inspected as the production tooling did wear out. Expense and the need to keep the production line rolling usually dictated where and when that tooling was to be refurbished, repaired or replaced.

As far as the quality of the flush riveting - I wondered what he was making his assumption on??? A typical AN 426 flush rivet is installed in either a dimpled or countersunk hole and usually carries a tolerance if the head isn't exactly flush with the skin its driven into. This is usually due to the countersunk hole or dimple not being properly prepared, to not enough pressure being applied from the assembler shooting the rivet. Later in the war a took called a micro shaver was developed that allowed a small portion of the rivet head to be "shaved down" so better flushness was achieved.

To make a fair comparison one would have to look at the model built, the production lot and factory it came from and probably the year it was built. I'd bet dollars to donuts you saw variances in the so called production quality based on this as it was evident when you compared German aircraft built early in the war when compared to late WW2 production.
 
I find it hard to believe that 109s produced from early 1944 onwards were that well constructed, being turned out like hot rolls. There is the minutes of a meeting on 1.98sts boost for the 109K that questions going to that boost when the airframes were of such bad construction.

The Me262 had to have body putty applied to the joints because panels met terribly.
 
I find it hard to believe that 109s produced from early 1944 onwards were that well constructed, being turned out like hot rolls. There is the minutes of a meeting on 1.98sts boost for the 109K that questions going to that boost when the airframes were of such bad construction.

The Me262 had to have body putty applied to the joints because panels met terribly.

Butt joints degradation depends entirely on the location (i.e upstream or downstream) of boundary layer separation. I had an opportunity to look the first MiG 21s that arrived in the US after the 1967 Arab/Israeli unpleasantness. Their mfg quality was atrocious but if the gaps were downsteam od boundary layer build up - it just doesn't mattter
 
Butt joints degradation depends entirely on the location (i.e upstream or downstream) of boundary layer separation. I had an opportunity to look the first MiG 21s that arrived in the US after the 1967 Arab/Israeli unpleasantness. Their mfg quality was atrocious but if the gaps were downsteam od boundary layer build up - it just doesn't mattter

Typical on many east bloc aircraft.
 
Well yes the strict comparison between a Battle of Britain era Spit and a 1942 109G-2 is not exactly fair but the point remains that I still find it highly likely the overall finish quality of the Mustang was higher than that of the Spit during wartime production, in this light.

As another example look at how much extra speed was gained in the TsAGI tests of refinished post production Yak-1 during 1942, as much as 20km/h gained simply by refinishing the Yak at an airfield, things like pulling off the panels and reattaching them properly :p
Just to say the quality of the manufacturing finish can have a tremendous effect on performance. This is also something the Spitfire was noted for, a refinished test example frequently performed much better than one taken directly off the production line but this was not really the case with US fighters like the Mustang, essentially because I guess without bombers over your home territories and extreme demands on your industry, plus the infamous US mass-industrial capacity allowed for an environment where finish quality on aircraft produced was extremely consistent and very high.

That's the point I was trying to make, when comparing Mustang performance with the Merlin 60/70 and Spitfire performance with it. The Spit does have a really nice wing, and a high critical Mach, surely a professionally and lovingly refinished test model taken from production would start to close the speed gap quite a bit from the average "listed max level speed ratings" of both a/c.
A test 'Stang is still pretty likely to get around the 700km/h mark spot on I think. But would a refinished, test MkXVI say for contemporary production and identical engine, come close to 680km/h and that wouldn't be such a big difference really would it? I think it a realistic possibility.
 
Wasn't the leading edge of the Spitfire wing also part of the structural strength of the wing?
I think it was refered to as a "D" spar? with thiicker than normal skining compared to the rest of the wing?

Not that it couldn't b dented but perhaps a bit more resistant to causual damage?
 

Users who are viewing this thread

Back