[Greenflight]

Man, this thread has really become a wealth of info... Matt, those links are great. I'm definitely planning on putting a belly pan (at least up front) on my truck.

 

[mattW]

I am also curious to ask people who already have working EVs how hot does your motor and controller get? Can you touch them after a period of driving? how much airflow do you think is necessary to keep them from overheating? If decent airflow was required it would definately affect how far you could go with fitting a belly pan because if air is to flow through the engine bay there needs to be somewhere for it to go, whether its air vents on the hood or from underneath the car (most cars vent under the car). So a belly pan would prevent air from flowing through the motor compartment unless you made room for vents in the pan. You can read about airflow through the engine bay here, the site is to do with getting the most out of your turbo but it has implications for electric car design. Basically we can assume that if 85% of an ICE is lost as heat compared to 15% in an electric car then there is probably 5-6 times more airflow in our cars than is required to keep the components cool. For the most aerodynamic car we want to minimise the flow to the minimum possible level without the motor or components overheating. Adding a small belly pan, only on the front section of an already aerodynamic prius in the site i linked in a previous post increased fuel economy by by 10% at a cost of just AUD$66. I hoping this thread will help people realise that they can increase their range and top speed considerably without the cost and weight of more or bigger batteries.

 

[Greenflight]

I happen to have a little experience with cooling... My motor burned out at about 500 miles due to overheating.

After that experience, I decided I didn't want it to happen again, so I ducted an electric radiator fan to the motor endbell. With the fan operating whenever the ignition is on, the motor gets warm to the touch, but not quite "hot," to use technical terms.

During the summer, when it's hot and humid, the motor still gets a little hotter than I'd like, but not nearly as hot as before the fan.

The controller stays cool as a cucumber.

I'd say your estimate of about 5-6 times more airflow than needed is about right for a good quality motor. To be safe, I'd avoid blocking more than 2/3 or 3/4 of available airflow.

I think the ultimate solution is the belly pan under the front only, with small guides by the rear wheels, and maybe a few mods on the back end. But that's just my uneducated guess.

 

[John]

One way to minimise you cooling air requirement would be to duct it from your cold air intake at the front of the car to where it is needed. I would say the majority would be needed by the motor. Those with direct drive would even benefit from fan assist for this particularly when travelling slowly up hills. I have a liquid cooled controller so it will be easy to duct air to the tiny little radiator. Those with air cooled controllers could build a shroud to duct air over the heat sink where it is needed.

I'm thinking about making circular ports at the stagnation point on the front bumper passing into a conical diffuser designed for maximum pressure recovery dumping into an air box to slow the flow and allow dust and water to fall out and drain away. The ducting would connect that air box to the motor and controller.

Even with a belly pan under the entire engine compartment there would still be openings at the transmission tunnel and drive shaft and steering rack ports without needing to create special vents through the pan. I suppose I could duct the hot air from the motor out of the compartment to prevent it elevating the temperature. Then I would only need to supply a tiny amount of ventilation air to the compartment to remove a small amount of heat and some hydrogen from the batteries.

 

[Mr. Sharkey]

In my case, I have ducted air from the grille of the car with blower assist.

The controller in my car handles only 10 amperes of current, so heat dissipation is negligible there.

Cooling for the motor is collected from an opening in the grille, which is accomplished via a custom-made aluminum box with a removable tray that holds a 4x6" sheet of blue polyester filter media to collect dirt and bugs. The output of this filter box feeds a 4", 12 volt DC squirrel cage blower, which in turn feeds the motor shroud from the brush end. Spent cooling air is exhausted from the shaft end of the motor.

The blower is connected to a PWM speed controller, and runs at about 1/4 speed anytime the ignition switch is on. When motor temperature gets up to about 75 degrees C, a temperature sensor signals the PWM controller to feed 100% duty cycle to the blower motor, bringing it up to full speed.

The combination of positive air flow from the motion of the car and the blower running at low speed seems to be more than enough to keep the motor at a low temperature on city streets. The blower usually only runs 100% at highway speeds, and usually throttles back to 25% by the time I have exited the freeway and driven home to park (based on my previously living in the city).

 

[Greenflight]

OK Sharkey, fill me in here. HOW does your car's controller only handle 10A?

John, I'm thinking if you place the ports at the stagnation point you won't get much flow for the drag produced. If you don't need much air, wouldn't it be logical to use flush scoops in attached (heck, maybe even laminar, if you have a smooth car) flow locations?

 

[John]

Placing the ports at the stagnation point will cause the stagnation point to move somewhere else on the surface. My theory is that by putting the ports there (basically at the apex of the vehicles shape) the flow will have travelled the minimum distance across the surface of the vehicle possible before entering the port causing minimum boundary layer growth for maximum diffuser efficiency and pressure recovery.

 

[mattW]

If you place a scoop on an area with attached flow you create an area of turbulance behind it, look at the area behind the hood scoop of this WRX undergoing DIY aero testing. I think your idea sounds good john, will you be using the existing grill or creating your own ports? Also it would be good to get a photo of the front end of your car so we can see what you mean. Sharkey do you have a photo of under the hood that would show the ducting or are there batteries in the way?

 

[Mr. Sharkey]

OK Sharkey, fill me in here. HOW does your car's controller only handle 10A?

Simple: Shunt Wound Motor. The controller's only load is the field coils in the traction motor. A single TO-3 high current/high voltage transistor handles the current. It's mounted on a small aluminum plate, enclosed within the larger controller enclosure. No heat sinks, cooling fans or atmospheric radiation necessary.

The armature in the motor of my car is connected directly to the battery pack through a high current contactor. The fields receive 100% PWM duty cycle from the controller when the motor is idling (and during full regen), and the field is weakened (lower PWM cycle) for acceleration.

Matt:

I was looking for a decent image showing my filter box and ducting, but don't think I have one, perhaps I can go out and snap one tomorrow when it's not storming. Only one battery under the hood, a 26 Ah gel cell for 12 volt accessories. Lots of room to view and move under there.

 

[Greenflight]

I see... so, what are the downsides? Sounds really cool...

I was thinking more along the lines of a flush scoop- one that draws the air down from the surface without projecting into the boundary layer. This type of scoop is designed to cause the least possible disturbance.

It seems like the stagnation point would split in two around the port (just guessing here ). Since the greatest loss of boundary layer energy generally occurs in the area on either side of the stagnation point, it seems as though you would end up with an excessively thick boundary layer at the front of the car, causing flow detachment sooner at the rear.

Anyway, just my $0.02...

 

[TexomaEV]

I've added a belly pan under the the frontal area of the car (blackened Coroplast), mainly to protect the electric motor, and components from splashing water/debris. I suppose it could help with Aerodynamics as well.

 

[Greenflight]

Well, I used up almost a whole roll of duct tape and a couple sheets of pink (I know, I know...) insulation foam today, and I now have something that could possibly be described as a belly pan. It goes all the way back to the front axle in the middle, and it also covers the area in front of the wheels.

One of the pics also shows a lexan grill cover I put on a few days ago.

The results? Well, they don't seem to make a whole lot of difference. If I had to guess, I'd say they reduce drag a little. I can't really say for sure.

When I go to test stuff like this, I usually go about a mile one way on the flat stretch of road in front of my house, and then return, shifting into third and allowing it to reach top speed.

I noticed maybe a 1 mph increase, but it's hard to say...

Well, it can't hurt anything, right?

I'm going to do a little more work on the area in front of the wheels. I think I can get some improvement with a little tweaking.

 

[John]

What you’re talking about is an NACA submerged scoop. They don't suffer from the loss of efficiency you would expect from a regular scoop due to it ingesting the boundary layer. Something about a vortex generated by the flow over the leading edge of the NACA scoop ejecting the boundary layer and causing it to ingest clean air.

Boundary layer growth might be greatest at the stagnation point but the boundary layer would be thinnest at that point due to that being the first point of contact of the airflow. Placing an intake there wouldn't cause an increase in boundary layer growth. First air flow contact would simply move out to the edges of the intake.

 

[mattW]

Greenflight you could also try to measure your power output at the same speed before and after. Has your overall top speed improved? If your speed is limited by the revs of the motor it probably wont make a difference but if you were struggling to get enough torque to get to higher speed then you should have a higher top speed with the changes you made. You could also try getting up to a specific speed (the higher the better) and then let the vehicle coast to measure the deceleration. This experiment and the maths needed are described in detail here. You definitely need to be doing testing to make sure what you have done will make a practical improvement to your efficiency. Aerodynamics is a hard thing to predict and sometimes is very counter intuitive. I would say that if you can't find a measurable improvement after adding something then its probably not worth the extra weight.

 

[Greenflight]

Yup, NACA submerged scoop. Words weren't coming to me. I guess the concept of intakes at the stagnation point just sounds weird. Probably works fine. I'll have to look into it some more.

I run out of torque long before I run out of revs. Top speed in 3rd is pretty predictable, around 45. After I lowered the front, it went up to about 47, which is about where it is now. Tomorrow, if I get the chance, I'll add some deflectors in front of the wheels and see what happens. I'm going to try those wedge-shape deflectors. If I can get them to funnel some air towards the center of the car (where the pan extends furthest back), the pan should help more. We'll see. I'm also going to try gap sealing the top edge of the bumper to the grill, with a large fillet. Should help flow around that area a bit.

I'll let you know what I find out.

All this duct tape is creeping me out... Great for trying out mods, though.

 

[Greenflight]

I finally got the chance to throw together some deflectors for the front wheels. They seem to improve performance slightly, but it's hard to tell because it's pretty windy out. Calm conditions would be much better.

From what I can tell thus far, these deflectors are probably worth their weight. I think it would be worth making a nice fiberglass (maybe even carbon fiber? ) belly pan with molded-in deflectors. Another project for when I get bored.

If I were to lay up a belly pan, it would be easy to incorporate a new bumper cover as well, possibly picking up some more speed there.

I'm going to take it for a spin again this evening when the winds have (hopefully) calmed down and I'll post some more results, along with pics.

 

[John]

I've been wondering about the effect of the incidence of the belly pan to the road. The belly pan slopes down from the front towards the road. This would cause the air flowing under the car to accelerate towards the back of the pan (to a rearward velocity relative to the road) which would in turn cause the pressure to drop below atmospheric pressure sucking the pan (and the car it's attached to) towards the road. The down force generated might have a negative effect on the overall aerodynamic drag. If the belly pan wasn't ridged enough the effect would exaggerate it self as the pan flexed towards the road. The high velocity flow is then dumped when it reaches the back of the front axle into a large open bottom cavity formed by the rest of the engine compartment where it mixes with dead air. There is then no chance of diffusing the flow to pick up some forward pressure to mitigate the drag.

I think Kevlar would be a better choice for a pan than carbon fibre. It has a similar strength to weight and is much tougher it is just not as ridged.

 

[Greenflight]

True- except the decrease in pressure would probably be compensated by upwash at the rear of the belly pan, plus the increase in pressure that would likely be caused by the sudden slowing of the air. Hard to say, though.

It is true that a lower angle of incidence would probably improve performance significantly. In fact I was just thinking about that. If I were to build a custom bumper cover I'd compensate by lowering the front of the bumper low enough to flatten out the pan. Plus it would reduce overall airflow underneath the car, which would probably also have a pronounced effect.

Kevlar might be a better choice... I just don't know much about it. I'll probably just end up using fiberglass anyway.

I drove the car a little more and concluded that there was a slight increase in performance from the deflector. A better design would probably have a much more noticeable improvement.

I like these ideas! Good, cheap ways to improve performance... Gotta love it.

 

[John]

I was thinking that effectively dumping the high velocity flow into a large cavity containing dead air would prevent the low pressure area from going further back than the back of the belly pan and would reduce any up wash effect. Pressure recovery in this area without a diffuser would be quite poor don't you think.

I would prefer the notion of not creating any negative lift. My car also has sloping front pan. I'm guessing it was built that way for a clearance angle for steep driveways and the like.

 

[John]

Your lower front bumper (spoiler) doesn't appear to have a very pronounced lip. I imagine it would spill quite a lot of air into the underside flow especially considering how flat over all the front of the vehicle is with a lack of tapering off at the corners.