Cooling system tests, week #2

raygreenwood · Post by **raygreenwood** » Sun Feb 08, 2004 9:51 am

A couple of pieces of information are missing. (1) what is your advance rate when running at constant speeds over 55 mph? (2) With that power pulley, you may be actually increasing load on the engine at a rate faster than the fan creates cooling air velocity/volume. Fan speed is not linear with output volume, due the fact that compression and cavitation may be happening between the blades as the fan "overspeeds" past its design volume. Cooling air volume may actually decrease when the fan reaches acertain rpm. This is very common on centrifugal fans. This is why HVAC charts are careful to list rpm and volumetric limits on each fan style and blade pitch. (3) Where is your air intake for the fan. The airflow over vehicles is not the same at every speed as it compresses over the body and under the body. You get turbulence in vearous rear areas. This may actually be causing a disturbance that is limiting the entrance of air into the fan. Just some thoughts. Ray

Bruce2 · Post by **Bruce2** » Sun Feb 08, 2004 1:06 pm

raygreenwood wrote: (1) what is your advance rate when running at constant speeds

???? If I'm running at constant speed, there is no rate of advance.

raygreenwood wrote:(2) With that power pulley, you may be actually increasing load on the engine at a rate faster than the fan creates cooling air velocity/volume.

Its not a power pulley, its larger than stock by about 12%. Air flow during brief acceleration spurts to 7000 rpm is irrelevant. Its low rpm or steady state cruising that an engine operates in during 99.999% of its life. A 3-4 second blast to high rpm does nothing to the temps.

raygreenwood wrote:Fan speed is not linear with output volume,

Absolutely true.
Go to this site: http://www.offroadvw.net/tech/ and click on the Cooling Fan link. This is an example of BAD information on the net. The guy shows a graph of CFM vs. RPM, and claims it to be perfectly linear to 10,000 rpm fan speed. BULL$H!T No fan is perfectly linear, that's why their output graphs are called "fan curves".

raygreenwood wrote:due the fact that compression and cavitation may be happening between the blades as the fan "overspeeds" past its design volume.

Cavitation does not exist in an air moving fan. Cavitation is the phenomenon of a liquid (usually water) evaporating due to the low pressure on the intake side of a fan blade (propeller), then that vapour violently condensing at the surface of the blade when pressure increases enough.

You may be confusing the transition between laminar air flow and turbulent air flow with cavitation. It is well known that when air flow is turbulent, you will get less CFM than if you had laminar air flow. At the transition you will see a sharp change in the output CFM.

raygreenwood wrote:(3) Where is your air intake for the fan. The airflow over vehicles is not the same at every speed as it compresses over the body and under the body. You get turbulence in vearous rear areas. This may actually be causing a disturbance that is limiting the entrance of air into the fan. Just some thoughts. Ray

Fan intake is in the stock location on a stock late Beetle.

We've all heard what VW recommended in the past. "Its top speed is its cruising speed". Since a stock late Beetle could go 85-90mph, it is reasonable to assume that the cooling fan is within its design range at that RPM. Since I don't cruise at that speed, I'm confident that the cooling fan in my car is NOT being overdriven outside its design range, even though I have a larger lower pulley.

mharney · Post by **mharney** » Sun Feb 08, 2004 1:20 pm

Bruce2 wrote:You may be confusing the transition between laminar air flow and turbulent air flow with cavitation. It is well known that when air flow is turbulent, you will get less CFM than if you had laminar air flow. At the transition you will see a sharp change in the output CFM.

If you are talking purely dynamics of a fluid through a simple cross section, no. If you mean at the fan, yes. In laminar flow you have a gradient of fluid flow, that is stationary at the walls of its passage, while the maximum is the point that is the centroid of the cross sectional area, pretty much. This provides better heat conduction. The turbulent flow increases resistance at the contact surfaces, which requires significant pressure changes to cause an increase in flow. The point at which it starts to transition to turbulent flow is not actually much different from what the volumetric flow rate was when it started to transition.

Turbulence at the fan is a whole other subject with entirely different rules, and the turbulence acts as a barrier to air flow by way of the resistance caused by the chaotic eddy currents.

Baja Wes · Post by **Baja Wes** » Sun Feb 08, 2004 3:05 pm

Bruce2 wrote:Absolutely true.
Go to this site: http://www.offroadvw.net/tech/ and click on the Cooling Fan link. This is an example of BAD information on the net. The guy shows a graph of CFM vs. RPM, and claims it to be perfectly linear to 10,000 rpm fan speed. BULL$H!T No fan is perfectly linear, that's why their output graphs are called "fan curves".

Well I am the guy that wrote that article, it was part of an engineering thesis, it was given a very high destinction from the head of fluids at the university, and the graph is perfectly accurate.

If you knew anything, you would know your fan laws. There are 3 basic ones;

? fan air delivery varies directly as the speed
cfm2 = (rpm2/rpm1) x cfm1

? fan pressure varies as the square of the speed
SP2 = (rpm2/rpm1)² x SP1

? required fan horsepower varies as the cube of the speed
hp2 = (rpm2/rpm1)³ x hp1

Notice what the first one says, directly meaning a linear relationship. Go on, do a search on the net, you will see what I say if true. You might want to get an education before you go saying well known engineering laws are crap.

Fan curves are called fan curves because most engineers are concerned with the pressure vs flow graph of a curve, which is a squared relationship and is curved. See the second fan law.

Baja Wes · Post by **Baja Wes** » Sun Feb 08, 2004 3:18 pm

raygreenwood wrote: Fan speed is not linear with output volume, due the fact that compression and cavitation may be happening between the blades as the fan "overspeeds" past its design volume. Cooling air volume may actually decrease when the fan reaches acertain rpm.

Every part of that statement is not true. Again see the fan laws. Also cavitation is only a liquid phenomena, it is boiling of a liquid due to a localised pressure drop below the vapour pressure of the liquid, you cannot boil a gas.

You will see a fan choke when it approaches sonic air velocities, but trust me the VW fan will explode long before that is ever going to happen.

Farmer out of town · Post by **Farmer out of town** » Sun Feb 08, 2004 3:37 pm

Now it gets interesting. - Perhaps we should start a new thread with this ? - Administrator ?

Wes I´m NOT saying you are wrong. And I know of Mackerle´s law.

My question is :
How did you measure the airflow, in the T1 shroud or in some setup where the flow is easier to measure ?
And at what backpressure - None, linear, progressive ?

If what you state is correct, (if I understand what you mean) a better flow producing fan in the T1 housing (or a higher reving fan) "should" be the way to more efficient cooling. Only it´s not working. Tried that already. The T1 fan housing simply could´nt handle more air. The fan just stalled and made a whine almost like a turbine. We found out because during one of the tests, one of the freshairplumbings fell out. and the fan just boosted air out trough the hole and cooled the engine better. So we tried with both holes free and evened out the cooling on both sides.
After that we gave up and went back to stock type fans.
Torben

Baja Wes · Post by **Baja Wes** » Sun Feb 08, 2004 4:21 pm

ok, we tested the vw engine complete, a running motor, at running temp, with a complete shroud.

We measured the air flow by putting an inlet on the fan shroud. We used a 150mm PVC pipe, about ~2000mm long. We moulded a bellmouth to the entry of the pipe to reduce entry losses and reduce the effect of the inlet pipe on the fan's air flow. We then mounted a pitot static tube in the center of the pipe, where the air flow was nice and smooth. We moved the pitot static tube across the duct cross-section at predetermined points and therefore calculated a velocity profile across the duct cross-section. We then used the area under the velocity curve to give us the air flow rate. We followed the guideance of an australian standard (AS2936-1987 also known as the SAA fan code) on the entire testing procedure, a recognised test method for fans. There is an equivalent american standard on how to do it too.

We also measured the actually fan shaft rpm at each point using a spindle tachometer capable of 20,000rpm. The advantage of doing this is we could tell when the VW fan belt started slipping. Let me tell you the VW fan belt starts slipping early, and silently (starts to whine when it is slipping real bad). This is why people get confused and think the fan stalls. It isn't the fan stalling, it is the fan belt that isn't capable of transmitting enough power to turn the fan any faster. This is due to the power required being proportional to the cube of the fan speed.

How did you complete your test? It seems to me that you may have made the mistake of measuring engine speed. That way the little VW fan belt was not capable of transmitting any more power to the bigger fan than it was to the VW fan, so resulting in no additional air flow. It would also explain the whine. If this was not the case I am interested to discuss what else could have happened with you in a civilised manner

Shall be something new for a car forum

MASSIVE TYPE IV · Post by **MASSIVE TYPE IV** » Sun Feb 08, 2004 8:32 pm

The engine was running at 3500 crank RPM.. I did not notice that the belt did slip with high drive ratios (356 pulley up top)

The testing I was doing was to test systems just off the shelf. Thats how it will benefit most enthusiast that are not engineers..

The Porsche 911 systems slip the belt MUCH more easily than the VW systems do- FYI (we saw it at as low as 2500RPM crank speed)

Baja Wes · Post by **Baja Wes** » Sun Feb 08, 2004 9:33 pm

Bruce2 wrote:No, you read it backwards. When I went to the taller tire it caused the rpm to drop, and instead of lower oil temps like you predicted above, the oil temps went up significantly.

I decided to read the rest of this post and found this comment. Bruce, I would be looking at your jetting. If you noticed this trend than you probably have a lean spot if your rev range. Before you were cruising above it, now you are cruising in it. Get a dyno check of your motor done, and get them to plot the air fuel ratio against rpm. It's amazing how much difference in temp the jetting will make.

Jake, I am not trying to take anything away from your test, I am just defending my own. I tested the VW fan the same way as most industrial fans are tested, so it's funny that people will still argue.

I did basic tests on the fan flow curves. You have taken it further and started to look at how it effects engine temperature. I don't know how you did it, so I can't comment on your methods obviously. But I will say that the inside of the shroud is a complex piece of gear.

A certain shaped deflector/foil inside the shroud may work good with a certain air velocity, but when you go above that you may have seperation of the air stream from the foil and low pressure air pockets form where a cooling stream of air once was, causing cooling issues. I think you may have encountered this problem in your tests, and you can then see why the VW factory carefully chose the pulley ratio's they did.

keep up the good work.

James aka fastfugitive · Mon Feb 09, 2004 1:39 am

Interesting stuff, guys - keep it coming!

Oh, going back to page 2, I use 97 Octane Fuel, run at max 32deg BDTC...

I also had tall tyres at the time - 30" Wranglers...

Could go some way to explain the high cruising temps - with low revs and high load...
Although the car only weighs 700Kg approx with me in it.

However, this year I have gone race only and it now sits on 25" rubber and is somewhat closer to the road

MASSIVE TYPE IV · Post by **MASSIVE TYPE IV** » Mon Feb 09, 2004 5:01 am

I was not testing the "Fan" as much as other shrouds! We kept the same fan till the end of the test when I exploded it on purpose!

Plastermaster · Post by **Plastermaster** » Mon Feb 09, 2004 6:34 am

has anyone found a difference in shrouds with heater ducts vs without? Jake I know your results are waiting for the mag article...Anyone else?

Ron

MASSIVE TYPE IV · Post by **MASSIVE TYPE IV** » Mon Feb 09, 2004 6:49 am

you can count on a 30 degree cooler engine when the ducts are blocked off.... That goes for pretty much any shroud we tested.

Plastermaster · Post by **Plastermaster** » Mon Feb 09, 2004 6:53 am

Wow! In the winter the engine can probably do better to have them open and in the summer when you don't need the heat, close them for better cooling.

Thanks
Ron

MASSIVE TYPE IV · Post by **MASSIVE TYPE IV** » Mon Feb 09, 2004 7:10 am

Thats all according to if you have heater boxes are not. If you have heater boxes they MUST stay hooked up or the heater boxes fry the heads! Nothing to cool them off otherwise.