The Lotus Cars Community banner
641 - 660 of 698 Posts

·
Registered
Joined
·
9,786 Posts
Discussion Starter · #641 ·
Andy, I was looking at your data on the airflow velocity in the intercooler and tried to look at it another way. I tried to calibrate the vertical axis in the same units (mph) as the vehicle speed.
With your corrections (between average IC airspeed versus sensor), I end up with the flow through the IC approximately 5% of the vehicle speed.

This is way below what a front mounted intercooler would provide. Have I got this right?
Hmmm... I'm not sure. If you look at the Y-axis, it's in feet/minute. Your red (5%) line intersects the 80mph line at about 450 fpm, which is (450 * 60 / 5280) = 5.11 mph. 5.11 / 80 = 6.4%... a bit above 5%.

I see what you're getting at though... yes, the flow through the intercooler is below the potential flow that an unobstructed, properly ducted front mounted intercooler would receive. Keep in mind that even a front mounted cooler would not see a flow equal to vehicle speed... it creates quite a bit of drag and would slow the airflow down just by itself... not to mention other obstructions in front or behind in the path of the cooling air. The airflow you see in my graphs is *after* the intercooler.
 

·
Registered
Joined
·
3 Posts
Hmmm... I'm not sure. If you look at the Y-axis, it's in feet/minute. Your red (5%) line intersects the 80mph line at about 450 fpm, which is (450 * 60 / 5280) = 5.11 mph. 5.11 / 80 = 6.4%... a bit above 5%.

I see what you're getting at though... yes, the flow through the intercooler is below the potential flow that an unobstructed, properly ducted front mounted intercooler would receive. Keep in mind that even a front mounted cooler would not see a flow equal to vehicle speed... it creates quite a bit of drag and would slow the airflow down just by itself... not to mention other obstructions in front or behind in the path of the cooling air. The airflow you see in my graphs is *after* the intercooler.
I was not sure if your plot had already applied the 284/374 (average/sensor location) factor that you mentioned later. I you had already corrected it, you are quite right.
I have had a couple of Evo's and suspect that the airflow would be about 50% of the incident velocity. That is still 10X better than 5! The Lotus intercooler is much deeper, I suspect to compensate for the lower velocities.
 

·
Dog Pilot
Joined
·
75 Posts
Physics. -poke-

Given,
Zeroth law of thermo: if temp A = temp B and temp B = temp C, then temp A = temp C
where
temp A = Before IC
temp B = IC temp
temp C = After IC
things in a linked system will eventually reach thermal equilibrium, so [IC temp] reaches [before IC] -> [before IC] = [IC temp] -> [before IC] = [after IC].

Proof:
Specific Heat equation:
Q=cm*dT
where
Q = unit of heat(energy in transfer). [constant in our case]
c = object in particular's heat capacity(aluminum, air). [constant in our case]
m = mass. [constant in our case]
dt = final temp - initial temp. [variable]

in the case of 0 airflow at equilibrium, aka 0 cooling.
final temp IS initial temp,
then dt = final temp - initial temp = 0,
so Q = cm*0,
therefore Q = 0, aka no heat transfer, ergo heat soak.

---
To take a step back to the original discussion:
Given Q = cm*dT and therefore Q/cm = dT, you may think doubling mass (2m) results in halved change in temp as in Q/(2cm) = dt/2
However, because all things reach equilibrium eventually, it just means it'll just take twice as long.

To put it simply:
If you place 1 pound of aluminum block and a 2 pound block on a heat plate(outputs energy at a constant rate), they'll both will reach the same temperature(equilibrium) as the heat plate. The difference is the 2 pound block will take twice as long. If you get a 100lbs block of aluminum on that heat plate, it'll also reach that thermal equilibrium. It'll just take that much longer.

In our case:
Say, if you hit 'heat soak' at lap 1 with 10lbs IC, you will have heat soaked at lap 2 with 20lbs IC.

To be fair, RLS IC could have better cooling efficiency due to larger surface area and/or bar&plate design when subjected through same amount of air.

IC is a device that depends on forced convection. The way to make it work better is by forcing more air through rather than adding mass. Besides, who'd want to add mass to the highest point away from the CoG?
On the other hand, A2W cooling would be more dependent on 'thermal mass', especially because water is has HUGE thermal capacity at ~4200J/kg*K. It takes a huge amount of energy to raise a kilo of water 1degreeC. It will stay at initial(ambient?) temperature for longer before it goes up in temp. Aluminum is comparatively tiny at ~900J/kg*K it'll go up in temp at the slightest chance.
Interesting thing is once the water in A2W heat soaks(eventually, preferably not before the race end), it'll just be as useless as heat soaked aluminum IC.

The point of my wall of text is not to bash RLS. I love their product. I'm just discussing science here.

---
West oz loti, there's no 'camp'. It's true whether or not one believes in it.
Prove me otherwise, then I'll stand corrected.

Thanks for throwing the math out there. I was too tired to try to make a cogent post with math in it. :)

The thermal carrying capacity of the stock IC or even the RLS is very, very small in the grand scheme of things. It will reach equilibrium almost immediately on a dyno run. If there's no airflow, there's no cooling and you're heatsoaked.

Another point I'd like to make with dyno runs is that the fins on the RLS and on the stock IC are pretty dense, you need a lot of static pressure to get a good amount of air through them. Just blowing a fan in front of the car or even straight in front of the intake(s) is not going to yield a lot of airflow. If you really want to test the effects of an IC on a dyno, you need to put a high static pressure fan or fans on the IC itself, meaning you have to have a specific blade pitch on the fans designed to increase the SP and not the CFM. You can have all the CFM in the world, but if there's no pressure behind it, very little will make it through the IC itself, instead creating backpressure on the fan's airflow and robbing you of cooling.
 

·
shay2nak
Joined
·
25,205 Posts
And did your dyno runs have air flowing over the IC? If so, then it doesn't have anything to do with what we are discussing.

I also find it very difficult to believe you could "feel" a 3% increase in torque. 3% is easily within the margin of error for most calculations not done on a dyno, butt dyno's not withstanding. I know some fellas that can swear they feel the torque increase with their Type-R stickers, too... but we all know how that goes. But maybe your butt-dyno is more sensetive than most people, so I'll give you the benefit of the doubt on that one.

At the end of the day thermodynamics are thermodynamics. You can't change them. If you want to burn money on stuff, I'm totally fine with that. I completely support your right to throw money at problems left and right. Spend $15,000 on 2lbs of weight reduction and 1hp increase. Lord knows I've wasted money on many-a-car chasing rainbows and the need to believe that you didn't just spend $3000 for little to no improvement is strong. I've been there. I understand. This is like trying to argue with an audiophile about how wooden knobs and $10,000 power conditioners make the sound "richer" and "fuller."

I just have a problem when others entice people to waste their time and money on something that is simply not possible through all known laws of physics and thermodynamics. Lets stick to the known sciences and not woo-woo magical beans and psychic airflow.

I've said what I have to say here on this subject and I don't want this to turn into a flame war, so I stand by my statements and believe they stand on their own merits at this point. The rest is up to the reader.
No, but we had a small fan blowing onto the intercooler each time I went to the dyno. Engine lid propped and a fan sitting in front of the trunk blowing at the back of the IC. You may not believe it, but I could feel the difference. I can also feel the weight difference in the car when I fill it up with gas. Not a full tank, but 5 or 6 gallons each time. 10whp in this car is almost like 20whp in a regular car because of its light weight.
 

·
Registered
Joined
·
297 Posts
I have linked to real life track testing

Others have presented the maths......

What more can be said! Do whatever floats your boat. I personally am going to work on maximising the airflow. The data and maths supports this path.

I will post my project when I start it, but everyone is free to do whatever they want to their car.

Just don't tell me a stationary car is less heatsoaked just by changing an intercooler. None will work without airflow (and lots of it!), that is the facts!

And by the way, that is how it is designed to work!

The watts of heat energy are significant when at full throttle, it is like trying to cool an electric radiator with a pedestal fan, it will have little if no effect!
 

·
6.94915254 lbs/hp
Joined
·
2,101 Posts
Thanks for throwing the math out there. I was too tired to try to make a cogent post with math in it. :)

The thermal carrying capacity of the stock IC or even the RLS is very, very small in the grand scheme of things. It will reach equilibrium almost immediately on a dyno run. If there's no airflow, there's no cooling and you're heatsoaked.

Another point I'd like to make with dyno runs is that the fins on the RLS and on the stock IC are pretty dense, you need a lot of static pressure to get a good amount of air through them. Just blowing a fan in front of the car or even straight in front of the intake(s) is not going to yield a lot of airflow. If you really want to test the effects of an IC on a dyno, you need to put a high static pressure fan or fans on the IC itself, meaning you have to have a specific blade pitch on the fans designed to increase the SP and not the CFM. You can have all the CFM in the world, but if there's no pressure behind it, very little will make it through the IC itself, instead creating backpressure on the fan's airflow and robbing you of cooling.
I will agree that thermal capacitance is really overrated since it means you still need to cool it down again...

However I am still trying to understand why you are fixated on zero air flow conditions. I mean yes, the stock IC and RLS IC will perform identically if there is no air flow, but since in the real world there is air flow the larger cooler (RLS) will make more efficient use of the air flow and outperform the smaller cooler (stock) every time, as long as there are the same air flows. It is pretty simple.

With that said, I think we can all agree that increasing air flow is your biggest bang for the buck right?!

Also, if anyone wants a definitive answer, I have access to a wind tunnel that can replicate wind speeds up to 120 mph, solar radiation, ambient temps from -20F to 120F, and it has a 4 axle dyno. Will cost a pretty penny though!
 

·
Registered
Joined
·
4,893 Posts
Pitot tubes are an awesome tool that may come in handy here. Use one to determine actual wind speed (not vehicle speed) and one for airflow. Pitot tubes are immune to temperature changes whereas those fan-style wind speed meters read low in hot air (air had a lower density when hot).

By using a pitot tube for wind speed, it may not change your results but will definitely remove some noise that comes from gusts and breeze.

Also, that dip in the data at 45mph is probably related to the boundary layer. With cars, you get turbulent flow on even a perfectly smooth plate very easily because that boundary layer builds with speed and length until it turns turbulent. You are probably seeing some of that effect since the scoop is on the surface of the roof. Just an educated guess though

Correction: they still depend on knowing density. It's bee. Too long since I looked at the equations and thought density was canceled out.

Sent from AutoGuide.com Free App
 

·
Dog Pilot
Joined
·
75 Posts
I will agree that thermal capacitance is really overrated since it means you still need to cool it down again...

However I am still trying to understand why you are fixated on zero air flow conditions. I mean yes, the stock IC and RLS IC will perform identically if there is no air flow, but since in the real world there is air flow the larger cooler (RLS) will make more efficient use of the air flow and outperform the smaller cooler (stock) every time, as long as there are the same air flows. It is pretty simple.
I'm not fixated on zero airflow. I was responding to oldmansan:

My point is that sitting on a dyno with a fan pointing at the front of the car (effectively no air reaching the top half scoop or side scoops) three back to back runs were within .55 rwhp of each other. There wasn't a hint of heat soak, even without airflow. I think this speaks favorably of the RLS IC (and heatshield).
I was pointing out that without airflow, the RLS IC was doing absolutely nothing for his testing and trying to use that data to describe why the RLS IC is better than the stock IC is flawed. Without airflow, the IC is effectively heatsoaked, which is the diametric opposite of the claim and that's what I pointed out.

With that said, I think we can all agree that increasing air flow is your biggest bang for the buck right?!

Also, if anyone wants a definitive answer, I have access to a wind tunnel that can replicate wind speeds up to 120 mph, solar radiation, ambient temps from -20F to 120F, and it has a 4 axle dyno. Will cost a pretty penny though!
How much does the wind tunnel cost to operate? What kind of pennies are we talking about here?
 

·
Registered
Joined
·
2,364 Posts
Heatsoak is not a term that has any strict engineering context. In fact after hearing it for so many years I can't even think of a sensible interpretation.

If you mean to say the material condition is carrying alot of heat, the term "hot" or "higher temperature" is pretty ubiquitous here

Perhaps the context is the IC is nearing some portion of steady state full load conditions

or perhaps it implies delta temperatures across the IC are below a certain threashold

Its not a term that makes sence to discuss in the context of thermal science without a rigorous definition.

I certainly agree the thermal capacitance is a negligible factor here.


I guess It has been too long since I have been @ school because for some reason I was thinking a joule was defined as the work required to raise 1 g of water 1 deg C @ STP. I guess that number is closer to 4 J. I guess it would be too much to ask for to coincide with 1 newton of force acting for 1 meter. Perhaps this is te first time the metric system has let me down!
 

·
6.94915254 lbs/hp
Joined
·
2,101 Posts
Heatsoak is not a term that has any strict engineering context. In fact after hearing it for so many years I can't even think of a sensible interpretation.

If you mean to say the material condition is carrying alot of heat, the term "hot" or "higher temperature" is pretty ubiquitous here

Perhaps the context is the IC is nearing some portion of steady state full load conditions

or perhaps it implies delta temperatures across the IC are below a certain threashold

Its not a term that makes sence to discuss in the context of thermal science without a rigorous definition.

I certainly agree the thermal capacitance is a negligible factor here.
+1 ...Well said!
 

·
Registered
Joined
·
2,200 Posts
Ummmm last I recall it was around $7,000/day, not including set-up costs... :crazyeyes

Needless to say you understand why I haven't just done this myself!
Yeah, that's a little pricey. If it were under $1000 I might consider it. For science!
Uh.... Kickstarter/Gofundme? :crazyeyes
I'd rather spend that money on some other useful studies, though...

Heatsoak is not a term that has any strict engineering context. In fact after hearing it for so many years I can't even think of a sensible interpretation.

If you mean to say the material condition is carrying alot of heat, the term "hot" or "higher temperature" is pretty ubiquitous here

Perhaps the context is the IC is nearing some portion of steady state full load conditions

or perhaps it implies delta temperatures across the IC are below a certain threashold

Its not a term that makes sence to discuss in the context of thermal science without a rigorous definition.

I certainly agree the thermal capacitance is a negligible factor here.


I guess It has been too long since I have been @ school because for some reason I was thinking a joule was defined as the work required to raise 1 g of water 1 deg C @ STP. I guess that number is closer to 4 J. I guess it would be too much to ask for to coincide with 1 newton of force acting for 1 meter. Perhaps this is te first time the metric system has let me down!
Agreed. :cool:
 

·
Dog Pilot
Joined
·
75 Posts
Heatsoak is not a term that has any strict engineering context. In fact after hearing it for so many years I can't even think of a sensible interpretation.

If you mean to say the material condition is carrying alot of heat, the term "hot" or "higher temperature" is pretty ubiquitous here

Perhaps the context is the IC is nearing some portion of steady state full load conditions

or perhaps it implies delta temperatures across the IC are below a certain threashold

Its not a term that makes sence to discuss in the context of thermal science without a rigorous definition.

I certainly agree the thermal capacitance is a negligible factor here.


I guess It has been too long since I have been @ school because for some reason I was thinking a joule was defined as the work required to raise 1 g of water 1 deg C @ STP. I guess that number is closer to 4 J. I guess it would be too much to ask for to coincide with 1 newton of force acting for 1 meter. Perhaps this is te first time the metric system has let me down!

In this context, "heatsoak" refers to a state the IC is in where the delta between ambient and the IC temperature is so small as to yield little or no meaningful heat transfer wherein the intake temp and the output temp are virtually the same. Or even a worse condition where the output temp is higher than the intake.

Although that is a good question, where did the term heatsoak even come from and why is it used?
 

·
Registered
Joined
·
1,012 Posts
I'd say it is more likely that the IC was effectively immediately heatsoaked and the power numbers you are seeing are with a heatsoaked IC. That would account for your consistent pulls across the three without any air. The IC doesn't do any good if there is no air flowing across it... Without air it is by default then heatsoaked or the functional equivalent thereof. The usefulness of an IC only comes in to play when air is flowing across the fins, otherwise it might as well just be an inert piece of really thick piping. You'd see the same numbers if you just had a straight pipe in place of the IC, assuming there is indeed no airflow through it.
That is completely incorrect. The IC doesn't need to have air flowing across it to disperse the heat. It obviously helps tremendously when there is airflow, but it's not necessary. If the air temperature around the IC is lower than the air going through it, there will be heat transfer through the fins. You yourself pointed out in a subsequent post that the purpose of the fins is to increase the surface area for the heat transfer to occur. Unless the ambient air around the IC fins is hot enough to be essentially equivalent in temperature to the air going through it, heat transfer will occur. That heat transfer will be signficantly better with an IC than with a straight pipe, regardless of airflow, because of the enhanced surface area.
 

·
Registered
Joined
·
2,200 Posts
That is completely incorrect. The IC doesn't need to have air flowing across it to disperse the heat. It obviously helps tremendously when there is airflow, but it's not necessary. If the air temperature around the IC is lower than the air going through it, there will be heat transfer through the fins. You yourself pointed out in a subsequent post that the purpose of the fins is to increase the surface area for the heat transfer to occur. Unless the ambient air around the IC fins is hot enough to be essentially equivalent in temperature to the air going through it, heat transfer will occur. That heat transfer will be signficantly better with an IC than with a straight pipe, regardless of airflow, because of the enhanced surface area.
No one's going to argue that one. There will always be some heat transfer to the ambient air since ambient air will nearly always be at lower temperature than the output temperature of the compressor side.

Few assumptions had to be made in order to move the discussion forward. Such assumptions are useful, for example, in the case of whether mass has any significant effect on the final velocity of free falling objects. Assuming no air resistance,the equation ends up being like vf=sqrt(2gh) so you can see mass as no effect on objects in free fall. It's unlikely that such condition will exist, but it is a useful tool for understanding the physics behind it.

In the case of the thermal mass vs airflow discussion, cooling from passing convection is assumed to have negligible effect(in comparison to forced convection) especially since the car is assumed to be nearly always in motion getting airflow through the IC.
Plus, doubling mass doesn't double surface area when density is same due to square-cube law, therefore it is also unlikely that doubling mass also doubles passive cooling due to increased surface area.
 

·
Registered
Joined
·
3,317 Posts
Back to the original starting topic of testing (but still relevant to the current conversation of flow vs larger IC) One thing I would suggest is a more controlled road test. No wind tunnel needed as long as you control your conditions. One good test to do is a run of say, 5 standing max throttle starts to, say, 60 mph and log the IAT. As long as you do the standing starts within a set period of time, as short as possible of course, and make sure your intercooler goes back to equilibrium temp at idle before you start another test run, you should have a pretty repeatable test. A back to back test of different hardware configurations will give you a good idea how effective each change will do.
1. stock config
2. Stock with 4" tube shroud or 3 chamber shroud
3. Upgraded IC with stock airflow
4. Upgraded IC with 4" tube or 3 chamber shroud
5. fans.. etc.
 

·
Registered
Joined
·
297 Posts
Back to the original starting topic of testing (but still relevant to the current conversation of flow vs larger IC) One thing I would suggest is a more controlled road test. No wind tunnel needed as long as you control your conditions. One good test to do is a run of say, 5 standing max throttle starts to, say, 60 mph and log the IAT. As long as you do the standing starts within a set period of time, as short as possible of course, and make sure your intercooler goes back to equilibrium temp at idle before you start another test run, you should have a pretty repeatable test. A back to back test of different hardware configurations will give you a good idea how effective each change will do.
1. stock config
2. Stock with 4" tube shroud or 3 chamber shroud
3. Upgraded IC with stock airflow
4. Upgraded IC with 4" tube or 3 chamber shroud
5. fans.. etc.
Basically this convo is going around in circles.
I suggested something like this earlier with the manufacturers/distribitors supplying parts.

A similar test was done on exige.com, interesting reading.

Got my car now so will be looking at my mod in the near future.

:)
 
641 - 660 of 698 Posts
This is an older thread, you may not receive a response, and could be reviving an old thread. Please consider creating a new thread.
Top