• NAXJA is having its 18th annual March Membership Drive!!!
    Everyone who joins or renews during March will be entered into a drawing!
    More Information - Join/Renew
  • Welcome to the new NAXJA Forum! If your password does not work, please use "Forgot your password?" link on the log-in page. Please feel free to reach out to [email protected] if we can provide any assistance.

Actual Measurements of Airflow for Triple Electric Fan Setup

FrankGRUN

NAXJA Forum User
Location
Reno, NV
I'm in the process of installing a new all-aluminum bar and plate radiator (BC1193) from Champion Cooling in my 1996 Cherokee. It turns out that in 2013, I purchased the Triple Threat Extreme fan set with integrated shrouds from FF Dynamics. I didn't install them at the time because of a number of negative reviews about how electric fans could not match the mechanical/electric factory design for airflow. I put them in attic storage, and decided to look at them again for this new radiator install.

The three shrouds all closely ganged together cover the the radiator core rather exactly. I was not impressed with the nylon push-through tie-downs supplied to mount the shrouds to the radiator, so I made up an aluminum angle custom mount to fix the shrouds as tightly as possible to one another and sealed the shrouds to the core surface with high-temperature silicone edge gasket material. The modified mount holds the fan shrouds tightly to the core body.

I setup a test rig, first mounting the fans firmly to an 18 inch long flat surface, then took a series of air flow rate measurements. I then added the radiator, bolted it to the fan assembly, and repeated the measurements. Then, I added an air conditioner condenser and did then again. In each case, I measured the applied voltage and the current draw for each airflow measurement. I measured the fan rotational speed for each condition with CenTech Photo Tachometer. I measured the airflow with a handheld Mastech MS6252A Digital Anemometer.

The results are:

The peak current surge when the fans are triggered to start is 22.8 amps at 13.96vDC. After this surge, the fans settled to the results cited.

Open air (no radiator) fan. V=13.98vDC, A=13.68 amps, RPM=1890, Airflow ranged from 1830 to 1905 ft./min. Airflow measured both behind and in front of the fans.

With BC1193 installed in front of fans (not in Cherokee).
V=13.75vDC, A=11.60A, RPM=2210, Airflow ranged from 1350 to 1410 ft./min measured at the rear of the fan. Interestingly the Airflow measured in the front of the radiator (at the radiator core surface) was a very uniform 970 ft./min.

With the AC condenser mounted in front of the radiator,
V=13.57vDC, A=11.92A, RPM=1820, Airflow ranged from 1250 to to 1310 ft./min.

The fan blade diameter is 9.25 inches for an area of 67.2 sq. in or 0.47 sq. ft. This says, just taking the cross-sectional area of the fan that under open air conditions, the individual fans flowed 878 CFM. Against the radiator, they flowed 649 CFM and with the AC condenser added, they flowed 601 CFM. These numbers are per fan and indicate that the 3-fan package is pulling 2634, 1947 and 1803 CFM.

Interestingly, the very uniform air flow over the whole core takes into account the contribution of the three shrouds. The core area corresponds to 2.387 sq. ft. and therefore the total air mover by the three fans amounts to 2315 CFM.

As a final note, all three fans are drawing a total of 11.60 amps with fan installed on radiator.For automotive fans mounted against radiators we expect of the order of 100 CFM per amp. By this guideline, the flow with radiator installed should be around 1200CFM, so the fans are moving more air per amp than expected by nearly a factor of two.

These measurements are crude using inexpensive instruments and under far from laboratory conditions, but they clearly show the differences in airflow from free air to radiator mounted. Also these measurements were take at an elevation 6,125 ft. above sea level.
 
Is your XJ currently operable? Would be interesting to put the anemometer behind the factory mechanical fan and electric fan.

When I did it on my XJ with a new NAPA 272318 mechanical fan clutch, I got 850 ft/min on a cold engine and 660 ft/min on a warm engine. The estimated flow rate based on swept area is very low - only 775 cfm for a cold engine and 580 cfm for a warm engine. Something is probably wrong with my measurement because people claim that the factory mechanical fan flows 3000 cfm.

A new TYC electric fan with curved blades produced an impressive air speed of 2250 ft/min which results to a flow rate of ~1150 cfm when measuring the fan swept area and assuming uniform air speed. There's not much room behind the fans, so the anemometer was pretty close to the fans, probably about 2 inches away. Was positioned roughly mid-distance between the base and tip of the blades. This was all with a single row radiator and AC condenser installed. About 1000 ft elevation. For comparison, the straight blade Dorman that was on my XJ when I bought it produced an air speed of 1175 ft/min and an estimated flow rate of 781 cfm (greater swept area than the TYC curved blade fan).
 
Good point! I had thought about it but thought my fingers were in jeopardy! The Cherokee is fully operational as yet till I do the radiator exchange. Hope to start Saturday. I'll try to rig up something to get in there. I can certainly measure the airflow through the radiator/condenser. Tomorrow. BTW in any of my many different measurements of airflow, I have never found values even approaching accepted values for mechanical or electric fans...
 
To get more accurate CFM estimates shouldn't you guys be accounting for the blockage due to the fan hub area?
i.e., fan swept area = pi((blade radius)^2 - (hub radius)^2)
 
To get more accurate CFM estimates shouldn't you guys be accounting for the blockage due to the fan hub area?
i.e., fan swept area = pi((blade radius)^2 - (hub radius)^2)

That's what I've done for all my CFM calculations. Its not hard to believe that this will give a reasonable value for the electric fan, but the mechanical fan blade has very non-uniform shape that very likely affects air flow at function of radial position.

On a side note, I realize why my mechanical fan CFM estimates are so low - they were done at idle. Will definitely be much higher cruising along at 2000 engine rpm, although 3000 cfm seems unlikely except perhaps at 3000+ engine rpm.
 
Just setup the anemometer for some measurements at the condenser face of the operating 1996 Cherokee 4.0L with the factory dual mechanical clutch fan and electric AC fan system.

The air speed measurements:

Idle (cold) mechanical fan side - 950 ft/min

Idle (engine at 210F) mechanical fan side - 625 ft/min

2000 rpm (engine at 210F) mech fan side - 1397 ft/min

3000 rpm (engine at 210F) mech fan side - 1995 ft/min

AC fan side, eFan off, idle (cold or hot) - 110 ft/min

AC fan side, eFan on, any condition - 765 ft/min

The air speed is quite uniform over the condenser surface for each of the two zones. The mech fan numbers tank right at the shroud boundary. I leave it to the reading observer to calculate an effective area to generate CFM numbers!
 
Given that the air speed numbers drop off so sharply at the mechanical shroud boundary, I think it is reasonable to choose the mechanical fan area value for the air flow calculation to be the dimension of the core bounded by the shroud. On my Jeep, the there is no frontal air path other than the radiator (I have sealed the rest of the front to the sides of the radiator). By my numbers this area is 17 x 11 inches or 1.30 sq ft.. This gives mechanical fan numbers of 812 CFM at hot idle, 1816 CFM at 2000 rpm hot and 2593 CFM at 3000 rpm. Using the same approach for the AC fan, (14 x 11 inches), I get an airflow number of 909 CFM.

So, with AC Fan on (by AC of fan override switch) the available numbers become 1721 CFM at idle, 2725 at 2000 and 3502 at 3000 rpm. This is the measured airflow over the whole core with the vehicle at rest and hood sealed.

Just a reminder, the total airflow for the Triple shrouded efan system is 2315 CFM. So, the triple eFan system is competitive with the mechanical fan past 3,ooo rpm and withbthe factory dual fan system up to 2000 rpm.

BTW, for full disclosure I'm not running the stock fan clutch, rather I'm using the often recommended Grand Cherokee 4.0L one that is identical to that used on the 2.5L Cherokee.
 
I was surprised how uniform and how repeatable the air speed numbers are, both for the radiator with triple fans mounted in the garage on my test bench and for the Jeep in the driveway. I not only carefully moved the anemometer around the core surface, I also did the measurements flush with the condenser core, back 1, 2, 3 and 6 inches. Same numbers +/- 2%!
 
As a FYI note, the air speed measured is, among other factors, a direct function of the pressure differential across the radiator. The fan generates a reasonably localized low pressure zone and air flows in to the pressure gradient. These measurements were taken at 6,125 ft. elevation, therefore, relative to sea level, the air pressure is lower, the air density is lower and therefore the corresponding values measured would have to be adjusted higher to correspond to sea level data. As indicated above the air density is a key factor in the velocity response to a pressure differential and the density is a function of temperature and relative humidity. For these measurements, the air temperature was 50F (had the garage door open) and the RH was around 30%. Correcting for elevation would add about 12% to these values and correcting to a nominal temperature closer to 80F would reduce these values by about 3%. For a sea level view at a reasonable temperature, these values would become;

Idle w AC fan - 1875 CFM
2000 rpm w AC fan - 2970 CFM
3000 rpm w AC fan - 3817 CFM

Triple Fan w shrouds - 2523 CFM

YMMV!
 
Never saw this thread when it was new. Epic data. Curious how the larger fan options (Taurus/Volvo 2 speed fan for example) stack up. Also curious what the airflow looks like at higher temperatures, as I don't think there's a lockup mechanism, so the hotter it gets, the more the goo in the clutch slips. Could be wrong, but it's interesting that the CFM went down as the engine warmed up.

Thanks for doing this. Makes a good case for using well maintained stock cooling fans.

Depends on the use case. For towing or heavily laden with steep grades, yeah, mechanical fan rules. For rock crawling and other low-speed stuff, I've noticed the mechanical fan lacks in airflow at low engine speeds for elevation at higher temperatures like a lot of us in the Southwest experience during the summer.

Not flow data like above, but anecdotally:

XJ in question is a 97 4.0 Sport, usual armor, spare tire carrier, trail tools, AW4 with 35s and 4.56s. ZJ clutch, 185*F thermostat, stock radiator/pump/housing, hood vents. Call it 4500lbs with driver/passenger.

I live in the mountains in East County San Diego and have a 3 mile 10% grade near my house that I have to climb most of the time I drive my Jeep (which if I'm coming from town, is preceded by an average 7.2% grade over 3.4 miles).

With the above setup, on a 95*F day, coolant temps with the AC on and maintaining 55mph speed limit (sometimes involving lots of throttle in 2nd gear), coolant temps would hover juuuust below the 220*F (ish) tick mark on the gauge.

Switching out the ZJ clutch/stock mechanical fan for a Volvo 2 speed fan (low speed comes on above 160*F, high speed at 215*F) on the same grade/temps, it would hover right at the 220*F mark. This means (anecdotally) the stock mechanical fan outperforms even the mighty Taurus/Volvo 2 speed setup. At speed.

Rock crawling and sub 25mph on forest roads/trails at 80-90*F at 6-7000' elevation, on the other hand, the thermostat never fully opens and it camps out at about 190*F, where before I had to constantly monitor the temps and eventually turn off the AC or stop for a bit to cool the motor. AC at low speed is considerably colder, and I no longer get constant misfire codes when offroad due to boiling fuel, nor do I burn my hand opening the hood at anything less than breakneck speed. :D

To me, switching to the Taurus/Volvo fan took a lot of anxiety away for my Jeep's primary use: off-roading. If you're towing a boat, mechanical fan is where it's at.
 
In addition, forcing the Jeep to maintain higher RPM by downshifting seems to yield little in cooling "gains" since the 4.0 seems to just make heat instead of horsepower at high RPM. :D
 
Good point on the value of a e-fan for low speed, relatively low rpm driving. I think there may be several ways to skin this cat though. I was out on double track this weekend on some extended climbs, probably only 4-5% grade at 2000 ft and 85F, and I was able to do ok with a mildly modified factory setup which consisted of a Cold Case radiator, high flow water pump, high flow water neck, 180F tstat, and a manual switch on the efan. I could hold at 195F (which seems to be the tstat control point) no prob with the efan running.

However, AW4 temps were higher than I wanted. I was seeing 215F on the outlet and 197F on the inlet. I need to get a transmission fluid cooler.

With that said, I do contemplate the switch to all efans. I've not heard about the Taurus/Volvo efan setup. Any links?
 
Thanks for the real world numbers. They do match what info I have come across. I do have the Dirt Bound plate for the e-fans. Then the SPAL 10" fans, and the electrical stuff from K-Suspension. Its not installed.
Some of the issue I see is that the SPAL fans are rated at 800 cfm at 12 amps. The Aux Fan for the XJ is 1100 cfm (curves blades) at 24 amps or is it 48 amps. So, a down grade of fans. Two stock fans would put out as much as the 3 fans.
I did look into and have a SPAL 10" fan that is rated for 1100 cfm at 24 amps. A bit thick though. I did buy one. Some issues with fitting it. Also, I just purchased a RHD Aux fan, used. I figure a stock fan LHD and RHD, and something in the middle. Needs to clear the water pump snout. Then a Nick-n-Time controller.
https://nickintimedesign.com/
I did purchase two of them. Might be a while though. Still, a different view or path to think about. And another possibility is to mount the third fan in front as a pusher.
 
Thanks for the real world numbers. They do match what info I have come across. I do have the Dirt Bound plate for the e-fans. Then the SPAL 10" fans, and the electrical stuff from K-Suspension. Its not installed.
Some of the issue I see is that the SPAL fans are rated at 800 cfm at 12 amps. The Aux Fan for the XJ is 1100 cfm (curves blades) at 24 amps or is it 48 amps. So, a down grade of fans. Two stock fans would put out as much as the 3 fans.
I did look into and have a SPAL 10" fan that is rated for 1100 cfm at 24 amps. A bit thick though. I did buy one. Some issues with fitting it. Also, I just purchased a RHD Aux fan, used. I figure a stock fan LHD and RHD, and something in the middle. Needs to clear the water pump snout. Then a Nick-n-Time controller.
https://nickintimedesign.com/
I did purchase two of them. Might be a while though. Still, a different view or path to think about. And another possibility is to mount the third fan in front as a pusher.

"Blowfish", you never post a pic of anything you talk about but you've own or have bought everything in the world you must have the largest garage in the world!
 
Good point on the value of a e-fan for low speed, relatively low rpm driving. I think there may be several ways to skin this cat though. I was out on double track this weekend on some extended climbs, probably only 4-5% grade at 2000 ft and 85F, and I was able to do ok with a mildly modified factory setup which consisted of a Cold Case radiator, high flow water pump, high flow water neck, 180F tstat, and a manual switch on the efan. I could hold at 195F (which seems to be the tstat control point) no prob with the efan running.

However, AW4 temps were higher than I wanted. I was seeing 215F on the outlet and 197F on the inlet. I need to get a transmission fluid cooler.

With that said, I do contemplate the switch to all efans. I've not heard about the Taurus/Volvo efan setup. Any links?

I need to make a thread on it. Came up with a decent belt routing that magically uses the stock belt length, removes the fan clutch, and has good wrap on all pulleys.
 
Back
Top