Fan Test (long post with many pics)

[Photoman]

Dan called a week or so ago and the part of the conversation was about the stock cruiser fan and the SC fan. It made me curious about how much air the fans actually put out etc. I thought I would throw together a little experiment to test the output. Well, it turned out to be quite a project that degenerated into a test of wills and a week of work later I have something to do some rough calculations.
I loaded a piece of ¾” plate on to a little trailer I built as a base for the electric motor and fan. I got the pulleys from an old drill press and then had to machine up adapters for them and the fan. Looking around I spied one of the spare tire spindles I had purchased from Slee Off Road. I had to do a little modding as the spindles were never intended to spin at any RPM’s. Bet you never thought your spare tire spindle would do 1725 RPM’s eh Christo? Next, I built a wood frame work that has the same size opening as the (97) cruiser front. It was made so I can move it back and forth over the fan to check for different airflow at different fan depths and centering. I used the stock shroud with foam weather strip insulation to seal it.
This post is premature because I won’t have a laser tach till Tuesday or Wednesday to check what RPM’s the measurements are taken at. I thought I would post it now to see if anyone has any suggestions or things they would like tried before I rip it apart or the rain gets to the particle board.
In the one pic there is a round tube I am building to confine the air somewhat for sort of direct test as opposed to the air blowing freely.
I did a quick run with the stock fan and the SC fan; both installed on a new blue fan clutch. I do have my old fan clutch and may try it.
Random thoughts and observations: The stock opening on a 97 has the hood latch support sticking down and I don’t know how that affects the opening size. Same at the bottom. For the test, there is no rad, cooler, or AC stuff to affect the flow. There is very little airflow near the center of either fan. The max airflow seemed to be about 4 inches in from the outside of the blades. The airflow is erratic, not steady. The Spal electric fans seem to direct the airflow toward the center. When I tested the electric 16” fan in the tube the highest air speed was in the upper right, lower right, and center. The tube is approx. 19 ½ inches in dia. and 3 feet long. I took 9 readings around the tube, averaged them, converted them to CFM, and came up with 2393 CFM for the electric. The manufactures rating is 2070 CFM for this fan. Truck was running with jumper cables to electric fans.
The dog came over and whizzed on my tig rod tube holding up the fan for the test. Pretty hazardous work this testing business. Guess I should be glad a cow did not come over.
The blue fan clutch would wind up and throw max. air about 10 seconds after turning the motor on. Then it would drop back and put out about half the amount of air. I went all over the fan clutch with the IR thermometer and the temps ran from 81.8 to 88.3F. Not exactly what I would call getting hot. I let the fans run for about 10 minutes and the fan never kicked back up. Without the tach, I will assume this was at about 1725 RPM’s, as I was running off 1 to 1 pulleys with a 1725 electric motor. FWIW, when I start the motor up and the fan clutch seems to kick back in, the whole setup shakes a good bit as I never balanced anything. It’s a little shaky having my hands (and face) 7” from the blades while this is happening. I did notice my wife thumbing through the life insurance policy. Hmmm. If I run the motor at 50 percent on the variac the fan clutch still seems to engage, but in a much longer time and more gradually. Really need the tach for anything meaningful and I also think I need to run all the fans through the tube. I don’t want to cut the aluminum yet before I run any final tests with the electric fans.
Instruments: Kestrel 4000 to measure airflow in FTM (feet per minute). Variac. IR temp gun.

Pics are Front, Back with tools, and side.


Bill

[Photoman]

Other pics are, test in progress, and electric fan test with tube.

[turbocruiser]

Bill, wow, I repeat wow, you are a gifted genius man. I'm amazed and awed!!!

[reffug]

Originally Posted by turbocruiser

Bill, wow, I repeat wow, you are a gifted genius man. I'm amazed and awed!!!

DITTO.

[IdahoDoug]

Bill,

You can be my neighbor anytime. Please.

DougM

[Rusty Phillips]

im curious about one thing - doesnt the fan clutch need heat to lockup?

because its not locked up, I believe this would lead to the fan speeding up and slowing down, and thus giving you some variation in your cfms

if you really want to test the various fan's cfms & flow patterns (variation from hub to blade tip) wouldnt a better test be to ditch the clutch and direct drive them at 1725 rpm?

another thing to consider is that one blade style may spin faster than the other with an unlocked clutch

[cruiserdan]

<<<"The dog came over and whizzed on my tig rod tube holding up the fan for the test. Pretty hazardous work this testing business. Guess I should be glad a cow did not come over.">>>

<<<" I did notice my wife thumbing through the life insurance policy.">>>

LOL Bill! I just spit coffee all over my keyboard........

I am quite fond of the headlamps too. They add sooo much....


Um, do you know Red Green by chance?





D-

[turbocruiser]

Originally Posted by cruiserdan

LOL Bill! I just spit coffee all over my keyboard........

I am quite fond of the headlamps too. They add sooo much....

D-

Don't overlook the properly placed Slee Offroad sticker!

Bill, I really am awed at all of this.

For grins and giggles I'd love to see the same tests with the ringed fan from the supercharger kits and the runners.

Also, any way we can incorporate the dog piss into an external liquid evaporative cooler to spray on the rad???

[cruiserdan]

Originally Posted by turbocruiser

Also, any way we can incorporate the dog piss into an external liquid evaporative cooler to spray on the rad???

You can Beta that one yourself..........

[turbocruiser]

Originally Posted by cruiserdan

You can Beta that one yourself..........

Yea, but everyone knows turbocharged vehicles do not overheat like supercharged ones, so loan me yours and i'll get goin on the beta tests, well actually, the dogs will get goin!!!

[cruiserdan]

[Photoman]

Thanks folks! Your kind words are what give me the incentive to keep going on this thing and the cruiser.

Well, as soon as I hit the send button for the thread, we had a powerful storm here and the electric went out for 19 hours. Just back now.

Rusty - Part of the testing that I wanted to do was to see if, when, and at what RPM the fan locked up. Right now I don't understand it. As you could see in the figures the temp of the fan clutch never was above 89 degrees F and yet the air output would double for maybe a minute, then the fan would uncouple and although the electric motor would continue to spin at its rated RPM, the fan would slow down. In some ways it makes sense. Supposedly, the fan should only be needed at lower RPM's and speed. You are also correct about the fan seeming to vary constantly. That is why I have to wait to get the tach on the thing where I can actually see what the RPM's are doing. In addition, I still have my old fan clutch that I want to put on just to see how it varies from the blue one. After that I can weld it into a locker and do what you said about a direct drive test. What I would like to do along the same lines is vary the RPM's to see how the air output changes with the RPM's. I would like to run the thing up to 2500-2700 to see what it does at highway cruising speed. I got the pulley; just don't know if I got the stones.

Dan - You know farmers. You got to have a bit of humor. The worst part is I screwed the headlights on, only to realize they were on the wrong sides. Figured someone would jump me on it, so I switched them. As far as the dog goes (so to speak), that was one advantage of working a lifetime in the sewers (like Norton). It wasn't the first time to be whizzed on. Break a sewer main while trying to make a tap. For those that don't know, sewer mains don't have shutoffs.

Turbocruiser - The ringed fan that I am testing is the one from the SC kit. The other is the stock fan. Is that what you meant? If someone sends me another fan fairly soon that would bolt up, I would be glad to give it a spin.

Bill

[cruiserdan]

The operation you describe is normal. The fan spins up briefly with a cold start and then slows down. In order to get it to kick in you will need to figure out a way to apply heat to the bimetal spring in the middle. Since we are trying to determine air movement of the different blades at the same RPM I suggest that you hard-mount the blades so that they turn at the known motor RPM. I would also excercise caution about the speed you run them at as I do not think the fan clutch ever actually locks solid. I think your motor RPM of 1750 would be a pertinant test speed. I can send you a 65020 blade for comparisson purposes.

D-

[Photoman]

Send blade – receive info.

Oh, you are entitled to yet another star for your forehead for spotting the 1FZ. Sharp eyes.

Bill

[Photoman]

I locked the fan clutch up to test at one speed as Rusty had suggested. The pulley speed was reduced to approx. 431 RPM to see how it would work out. I held the meter about 7” from the back of the pulley blades. The airflow still varied. With the stock fan it read 936 to 1030 FPM (not CFM). The conversion to CFM is multiply the FPM by the area of the opening. I don’t feel comfortable using the 20” shroud diameter (r = 10) opening till I get the tach and use the metal tube I made. With the SC blade the numbers were 767 to 825 FPM. Interestingly, at the edge of both fans, there was no airflow. By that I mean I held the meter ½” outside the edge of the spinning blades and parallel to them. Very close.
So at this slow speed, and crude measurement, it can be seen that the stock fan puts out more airflow than the SC fan.

Pic of fan clutch lockup and taking a measurement.

Bill

[Tools R Us]

Originally Posted by cruiserdan

Um, do you know Red Green by chance?

D-

Red Green is my hero!

In the playing we have done with fan clutches a heat gun pointed at the thermo coil makes the clutch engage.

The big thing missing in your test is the pressure difference caused by the radiator and condenser. An air stalled fan works differently than one working with all the air it can get. Instead of adding a radiator and condenser maybe another sheet of wood with a pattern of holes to simulate the restriction?

[concretejungle]

Originally Posted by Tools R Us

The big thing missing in your test is the pressure difference caused by the radiator and condenser. An air stalled fan works differently than one working with all the air it can get. Instead of adding a radiator and condenser maybe another sheet of wood with a pattern of holes to simulate the restriction?

very interesting point. I can just feel myself getting smarter.................



.................i think i'm getting a head ache.

[Photoman]

Tools R Us

Thanks for your suggestion. From my original post –

For the test, there is no rad, cooler, or AC stuff to affect the flow.

I was concerned that the lack of the rad etc. would affect the accuracy of the results. Realizing the conditions are not optimized for each fan, the conditions are the same for all fans, so it should basically be apples to apples. My intention was just to see what fan put out the most air and hopefully provide the best cooling for the cruiser. What do you think?

Bill

[Rusty Phillips]

ok, im curious about a couple of things, but first things first

I think that as an airfoil (fan blade or a airplane prop) moves thru the air the higher pressure air on the underneath side spills over the wingtip and creates a vortice (mini tornado) with the low pressure air on the upper side

this should be minimized with a fan carefully fitted to a duct, right?

but blah blah blah...... the fact is that your taking measurements on the down wind side of the fan - you are catching air that is spinning whirling in every direction at once.....

I suspect that if you take your cfms ahead of the spinning fan you might get cleaner less turbulent air and more repeatable numbers with less variation.

lemme think some more

[Rusty Phillips]

and well, I thought of something that was absolutely brilliant and Im sure could have saved you a ton of time and etc, but I forgot what it was.......


whatever..... couldnt have been too great of an idea as I have completely forgotten it




I am wondering just what the different speeds of the fan clutches are

the original (eaton?) one was a two speed - does that mean freewheel and engaged? (and is it fully locked 100% of input speed or some lower ratio?)

and the blue (aisin?) is a three speed - does this mean freewheel, low speed (what is the %? 50?), high speed (what is this %- 75% or 100%?) of input?

just what are the numbers, and at what temps do the various steps occur?



PS - I too wish you were my neighbor..... nice setup.

[Tools R Us]

Originally Posted by Photoman

I was concerned that the lack of the rad etc. would affect the accuracy of the results. Realizing the conditions are not optimized for each fan, the conditions are the same for all fans, so it should basically be apples to apples. My intention was just to see what fan put out the most air and hopefully provide the best cooling for the cruiser. What do you think?

Bill

Fans are optimized to a certain flow, being able to put out big flow numbers without restriction won't do any good for cooling the truck, so the data is useless? The fans job is to develop a negative pressure/vacuum in the shroud to pull air through the radiator, measuring the output of the fan may not tell much, because it's not taking into account air that's recirculating/leaking around the edges of the fan and through the clutch? Measuring the flow pulled from the front side and the vacuum pressure in the shroud would get better data?

The easiest way I see to do it is to put a piece of wood in front that is about two inch larger than the opening in all directions, attach it with about half an inch spacers so the air is pulled in around the edges and put a/some hole/s to use the airflow meter from the front. you may need to adjust the spacers/holes to approximate the airflow to what it is on the truck. Use the locked clutch so the rpm can be controlled.

If you do this I predict that the air flow will become more complicated, you will see backwards flow through the fan clutch and maybe around the edge of the ring shroud fan.

Fans eat air, the more you give them the more power it takes to drive them. The easiest way to test this is with a shop vac, put an amp meter on the power cord and a vacuum gauge on the input, run the vac without a hose, record the vacuum reading and amps, then restrict the flow, the airflow will fall, the vacuum will rise and the amps will fall, if you close the intake, the flow will be zero, the motor RPMs will race, the vacuum will be high and the amps will be the lowest! If you pulled the fan out of it's housing, giving unlimited air, the airflow reading would be the highest, almost no vacuum and the amp reading would be the highest.

This is why the shroud and "seals" are so important, controlling the air so the fan only sees what air is being used to cool the system, any extra air only costs power and overloads the clutch at the expense of cooling. The seal at the shroud, radiator to core support, hood to core support and the tin pan under the radiator all prevent air from recirculating.

The blade tip to shroud interaction is critical, on an nonringed (standard 80) fan the tips of the blades cause a radial airflow that blows against the parallel part of the shroud causing a high pressure turbulent seal between the fan and shroud, the most efficient design would have the parallel "stovepipe" portion of the shroud as long as the depth of the blades. If the blade is set too deep that flow will blow into the shroud and recirculate, if too shallow the flow will blow outside of the shroud, but will leak air back into the shroud. Both setups will produce big airflow numbers behind the fan and require lots of power, won't pull much air through the radiator!

I hope my feeble ramblings make some kind of sense!

[Tools R Us]

Originally Posted by Rusty Phillips

and the blue (aisin?) is a three speed - does this mean freewheel, low speed (what is the %? 50?), high speed (what is this %- 75% or 100%?) of input?

just what are the numbers, and at what temps do the various steps occur?



PS - I too wish you were my neighbor..... nice setup.

The blue hub Aisin appears to be linearly variable, the hotter it gets the stiffer it gets, no steps and is adjustable.

[SamB]

Awesome dude, loving your work

I'd be interested to know how the flow drops as you add things infront of the fan i.e, Rad, aircon, trans cooler, intercooler...

[Photoman]

A little more info. I tried testing the airflow on the outside (front of cruiser) as suggested. The airflow was still turbulent and was approx. ¼ that of the inside (motor side) about 7 inches from the fan. A couple of other interesting things. With the SC ringed fan, there was a negative airflow (air flowing to the front of the cruiser) between the ring and the shroud. The air flow at this same gap stayed positive (back to the motor) with the stock fan. Also, on the motor side of the stock fan at the fan edge there was zero airflow. The pics show these more graphically. Weird stuff.

SamB - I will stick the rad and an intercooler in front to test when I get the tach to check RPM's.
Been through Wellington to Picton IIRC. Spectacular country. Spent five weeks there photographing.

Bill

[alia176]

Originally Posted by Tools R Us

Fans eat air, the more you give them the more power it takes to drive them. The easiest way to test this is with a shop vac, put an amp meter on the power cord and a vacuum gauge on the input, run the vac without a hose, record the vacuum reading and amps, then restrict the flow, the airflow will fall, the vacuum will rise and the amps will fall, if you close the intake, the flow will be zero, the