Cooling problems with undertrays?

[M_a_t_t]

For anyone interested Julian Edgar has made a (maybe a few?) video(s) on reducing pressure under the hood to increase effectiveness of the radiator/intercooler. He has more in his book too.

https://www.youtube.com/watch?v=o60jC6EIgPU&t=1s

[Nautilus]

Thank you

I've read carefully Mr. Julian Edgar's experiments on Autospeed ever since 2011.

[JulianEdgar]

Originally Posted by Nautilus

Thank you

I've read carefully Mr. Julian Edgar's experiments on Autospeed ever since 2011.

That's great - thank you - but my aero stuff has progressed a very long way since the material I wrote on AutoSpeed.

For example, it's quite possible to get measurable downforce on a road car at normal speeds, and there's no need to guess aero pressures now - just measure them.

[Nautilus]

Thank you, Sir

The experiments on Autospeed gave me the right info to solve a very annoying problem, which comes directly from VW design: the engine bay of the 1.8T is murderously hot.

The cooling system keeps the block in a bearable range of temperature and the oil cool enough to be touched by finger. But everything around the engine block is slowly cooked. Wire insulation hardens, hoses crack, factory intercooler barely cools. Small plastic fittings in the cooling system crack over time - they need to be changed at each timing belt change. And this happens even on stock engines, leave alone mods to raise horsepower in the 250 hp range.

Solutions:

Cooling system: full undertray from a 1.9TDI engine, from front splitter lip to the front side of the front axle subframe. With aluminum additions to the rear of the subframe. The decrease in static pressure below the sump and axle draws / pumps the air outside the engine bay, towards the exhaust tunnel.

Intercooler: 2 small undertrays at the sides of the engine bay, which protect the lower part of the intercooler from dust, dirt and hot pavement. Slits in the plastic fender liner, so the rotation of the wheel acts like a fan to suck air from behind the intercooler. All slits in the front bumper which do not lead to a radiator carefully closed, so all air is funneled in the intercooler or cooling radiator.


Intercooler itself had been replaced many years ago with a much bigger, almost cubic model from Boost Factory. They do no longer make them, but there are similar coolers on the market from Tyrolsport, Eurojet and other manufacturers.

Battery box: small duct from battery box to air filter box, so the engine draws plenty of air through the battery box.

Ducts and hoses: insulated coolant recovery tank and coolant hoses. Like a Prius, the coolant tank keeps coolant warm for many hours, and the pump draws directly from it upon engine start, so the block warms up in minutes, even in winter.

Also insulated fuel hoses and fuel rail, so the heat does not "boil" or "bubble" the fuel.

The improvements are visible in the fact radiator fans operate much less in moderate traffic.

Fact: fans operating alternately in high and low speed add about 0.1-0.2 liters/hour to fuel draw. With no fans, indicated idle fuel draw is 0.6-0.7 l/h, with fans in full speed 0.8-0.9 l/h

[JulianEdgar]

Originally Posted by Nautilus

Thank you, Sir

The experiments on Autospeed gave me the right info to solve a very annoying problem, which comes directly from VW design: the engine bay of the 1.8T is murderously hot.

The cooling system keeps the block in a bearable range of temperature and the oil cool enough to be touched by finger. But everything around the engine block is slowly cooked. Wire insulation hardens, hoses crack, factory intercooler barely cools. Small plastic fittings in the cooling system crack over time - they need to be changed at each timing belt change. And this happens even on stock engines, leave alone mods to raise horsepower in the 250 hp range.

Solutions:

Cooling system: full undertray from a 1.9TDI engine, from front splitter lip to the front side of the front axle subframe. With aluminum additions to the rear of the subframe. The decrease in static pressure below the sump and axle draws / pumps the air outside the engine bay, towards the exhaust tunnel.

Intercooler: 2 small undertrays at the sides of the engine bay, which protect the lower part of the intercooler from dust, dirt and hot pavement. Slits in the plastic fender liner, so the rotation of the wheel acts like a fan to suck air from behind the intercooler. All slits in the front bumper which do not lead to a radiator carefully closed, so all air is funneled in the intercooler or cooling radiator.


Intercooler itself had been replaced many years ago with a much bigger, almost cubic model from Boost Factory. They do no longer make them, but there are similar coolers on the market from Tyrolsport, Eurojet and other manufacturers.

Battery box: small duct from battery box to air filter box, so the engine draws plenty of air through the battery box.

Ducts and hoses: insulated coolant recovery tank and coolant hoses. Like a Prius, the coolant tank keeps coolant warm for many hours, and the pump draws directly from it upon engine start, so the block warms up in minutes, even in winter.

Also insulated fuel hoses and fuel rail, so the heat does not "boil" or "bubble" the fuel.

The improvements are visible in the fact radiator fans operate much less in moderate traffic.

Fact: fans operating alternately in high and low speed add about 0.1-0.2 liters/hour to fuel draw. With no fans, indicated idle fuel draw is 0.6-0.7 l/h, with fans in full speed 0.8-0.9 l/h

An interesting mix of modifications!

[Nautilus]

Citroen had a similar dilemma back in the 1950s, when they designed the DS: due to the pointy nose needed to get good aerodynamics, the scoop for the radiator could not be made big enough.

So they placed the narrow scoop at the extreme front end, practically in the steel bumper, where the air pressure at speed is highest. The boat-prow shape of the nose dictated a slanted steel undertray from the bumper to the front wheels (which funneled the air below front axle... good ole static pressure again).

Air entered the engine bay through a narrow scoop, then expanded in a funnel sealed with rubber-fabric to the front of the radiator, so all air was forced through the core, and exited the engine bay through wheel wells, aided by the rotation of the wheels and inboard brake rotors.

Side note: the engine was fully behind the front axle, almost buried under the dashboard. This helps weight distribution a lot and remained for all aerodynamic Citroens, through CX, GS, XM, even the modern-day C6.

[JulianEdgar]

Yes I found a good contemporary ad on the Citroen and did a video on it -

[Nautilus]

If the Bugatti 100P didn't fly in the original 1939 configuration, there is another plane which made good use of similar drag reduction for aerodynamics and cooling: the P-51 Mustang.

The Mustang wing was a high lift
configuration, as well as low drag. . . the Mustang in squadron service
was not laminar to the same extent as the wind tunnel development
models. Not one day in the past 34 years (the book was written in 74)
has it performed in that manner for any or all of the reasons just
given.

So if it wasn't the laminar flow wing that gave it it's high speed and
extensive range, what was it?

The most prominent speed secret was the dramatic reduction of cooling
drag. Placing the airscoop on the belly just in front of the rear edge
of the wing removed it as far as was practicable from the turbulence of
the prop and placed it in a high pressure zone which augmented air
inflow. Tests in the wind tunnel with the initial flush mounted scoop
were disappointing. There was so much turbulence that cooling was
inadequate and some doubted that the belly scoop would work. The
breakthrough was to space the scoop away from the surface of the belly
out of the turbulent boundary layer of the fuselage. Further testing
showed that spacing it further out would increase cooling but at a cost
to overall drag. Various wind tunnel tests established the spacing at
the current distance which represents the best compromise between
spacing out from the turbulent flow of the fuselage, drag and airflow.

With the flow into the scoop now smooth and relatively nonturbulent,
the duct leading to the radiator/oil cooler/intercooler was carefully
shaped to slow the air down (the duct shape moves from narrow to wide,
in other words a plenum chamber) enough from the high external speeds
to speeds through the heat exchangers that allowed the flow to extract
maximum heat from the coolant. As the air passed through the radiators
and became heated, it expanded. The duct shape aft of the radiator
forced this heated and expanded air into a narrow passage which gave it
considerable thrust as it exited the exhaust port. The exhaust port
incorporated a movable hinged door that opened automatically depending
on engine temperature to augment the airflow. The thrust realised from
this "jet" of heated air was first postulated by a British
aerodynamicist in 1935. The realization of thrust from suitably
shaped air coolant passages is named after him and called the "Meredith
Effect". Some have said that at certain altitudes and at a particular
power setting the Meredith effect was strong enough to actually
overcome all cooling drag; this is not regarded as being accurate by
most aerodynamicists. It greatly contributed to overall efficiency of
the cooling system but never equaled or overcame cooling drag.

[JulianEdgar]

The Mustang aircraft is interesting (and much quoted) but the low drag outcomes are of limited benefits to cars.

But Porsche actually did achieve effectively zero cooling drag on one of their road cars:

[Nautilus]

1. Working by hand, with no wind tunnel or other aero instrument, it's pretty hard to reproduce the Porsche design.

2. However, the cooling improvement by leaving the rotation of the forward wheel to suck out the air from the intercooler bay has been tried and proven by thousands of people ever since 1999, when the Golf Mk4 platform with side intercooler became widespread.

3. It only works since a design quirk placed the intercooler very close to the wheel and separated from the engine bay with a plastic wall. Some insulative material on the wall and behind headlight in the upper engine bay, and the IC bay is thermally separated from engine bay entirely.

4. The car lacked some aero features which had become ubiquitous after 2004. Being designed during 1998-1999 crisis, when crude oil dropped to 10 USD/barrel, there was little incentive to cling to the smallest efficiency improvement. Official Cd had been 0.32

5. There is fitted from factory a partial grill block just behind the S badge, as visible here. Side upper grills may be covered on the inside in the same manner, but there is little benefit into doing so. Airflow for cooling is driven by the pressure differential across the radiator, more than by straight flow into the radiator bay.

BEFORE / AFTER

No deflectors before front wheels / Boat-prow deflectors before front wheels;
Half undertray below engine / Full undertray from front lip to the rear end of the front axle
Parachute-like shape of the rear bumper on the inside and rear axle exposed / Full undertray from rear axle well to the rear edge of the bumper;
Nothing behind rear wheels / Boat tails behind rear wheels
Big and ribbed muffler / Flat aerodynamic muffler
Lower side of the intercooler open towards ground / Side undertrays left and right beside engine undertray
Exhaust tunnel made a big bend before fuel tank / Flat aluminum plates around the exhaust bend (but not covering it)
Sheet steel front control arms / Control arms plated over, like fish flippers parallel to ground
Useless gaps in the front bumper, leading nowhere / All gaps which can't steer the airflow into intake or some radiators closed oveer
Gaps in the bodywork / All gaps in the bodywork closed with rubber gaskets

Audi has found out by experiment that smoothening the underbelly as much as possible may give a delta Cd around 0.024.

If deflectors before front wheels really give a delta Cd of 0.01, this drops the overall Cd from 0.32 in the 0.28 range, which is actually very good for a daily-driven car.