Electric vs. belt driven cooling fan
 

I just got done replacing the belt driven cooling fan on my GMC Sonoma with an electric fan, for the purpose of trying to save fuel. But after doing this I figured out a way to measure (guesstimate actually) the amount of fuel savings I might expect.

The electric fan draws 6 amps at 14 volts (84 watts). If one horsepower is 745 watts then the electric fan uses 0.11 horsepower. My electric fan seems to blow about as much air as the original mechanical fan. This suggests that the original fan was probably using about the same amount of power, which would be about 1% of the total cruising speed power used by the vehicle if cruising power used is around 11 HP.

Note, almost no energy is consumed by my electric fan because it is almost never running. In other words, this pickup rarely needs a fan at all under the conditions which I use it. So this is saying that I have removed the 0.11 horsepower robbing mechanical fan, with almost zero added energy deficit of the new fan.

I realize my method of measuring the energy expense of the original belt driven fan is a bit crude but I could not think of any better way to calculate it.

My question is, what are the opinions of others on how much fuel might be saved using an electric fan vs. mechanical, and how did you come up with your figures?

The only things to consider are:

* Your truck probably came with a fluid clutch fan. While it was still pulling energy, I doubt it was "0.11HP" all the time.

* Your electric fan gets its energy via the alternator. The 0.11HP it consumes is magnified by whatever your electric system efficiency is (i.e. 0.11HP may actually be 0.15HP).

You will definately see better mileage (no idea how much, though). I just wonder how long the fan will take to start making you money. My guess is that you won't turn a profit for a few months, but it should help in the long term.

I'm just anal, but I find it valuable to calculate overall costs. We've all heard of the guy driving 15 miles for gas 5c cheaper. :p

- LostCause

I did not even realize that there were any vehicles made in the last 15 years that did not have an electric fan. However, here is a link that says 16 to 41 hp used for big trucks.
A comparison of viscous drives and on/off fan drives for diesel engine cooling | Diesel Progress North American Edition | Find Articles at BNET
I think what you did is worth while. If you had the opportunity, you could measure manifold vacuum at a particular speed on a particular stretch of road with the old vs new cooling fan. Then, with the electric fan in place, go back to that road, and speed up till your manifold vacuum is the same as it was with the non-electric fan.

Now you know (more or less, not exactly) how much faster you can go and use the same amount of gas. If you do a roll down test of the truck you can calculate .cd, and from that calculate the power it takes to move it at different speeds. From that you get the power the old fan took, and with a guess at engine efficiency, you can back out the value of MPG.

But in my case, the electric fan is almost using no power as I mentioned. This is because it almost never comes on. In fact, I drove around with no fan at all for three days before I got the electric fan installed and the temp gage never rose above its normal range.

I drive about 10,000 miles/year and I calculated, using my 0.11 HP figure, that it will increase my mileage by 0.33 mpg and thus take 6.6 years to pay for itself @ $4/gal and the 33 mpg that I get. The fan w/T-stat cost $80.

I guess I would not have done this mod had I done this calculating before buying the new fan, but that's water under the bridge now.

Anyway, thanks for the input.

Thanks for the link.

Gee, if a truck fan consumes a minimum of 16 HP then my little pickup fan may consume a whole 1 HP perhaps instead of the 0.11 HP I was calculating. But, I wonder if my clutch actually ever engaged. The reason is, as I mentioned earlier, I have driven around a few days without any fan at all and had no temp problem at all. My new electric fan only comes on near the end of the 1/2 mile hill I must climb in 1st gear to get to my house. Otherwise the normal driving airspeed provides adequate airflow.

I drive a school bus and I can feel the loss of power when its fan clutch engages. It is not real pronounced but it is definite.

Measuring manifold vacuum as you suggest would be a good idea but it is too much work to remove and reinstall fans so I guess I will pass.

One thing that I thought of that might be beneficial is that the electric fan may allow the engine to warm up faster when first started since it does not blow any cold air over the motor, since it is off until needed. The mechanical fan blew air constantly, even though it has a fluid clutch, and that must have had some cooling affect on a motor that is trying to get warmed up.

Thanks for the input.

That's a good point (warm-up times of fluid clutch vs. electric).

Gregte, are you tracking your fuel consumption? Just curious. (Feel free to enter your truck in the garage: Fuel Economy, Hypermiling, EcoModding News and Forum - EcoModder.com)

Well maybe 8 hp, but I would guess more than 1.
I thought the fluid clutches spun the fan all the time, but sometimes they spun it more and sometimes less? On the schoolbus (ha I used to drive one when I was in high school) does the fan have a clutch like an AC compressor does? Where it is locked up hard OR completely free?
The other way you could test the fan HP is to mount that old fan on an electric motor that you have set up as a dyno (you let the motor pivot on an arm, and measure the force the arm pushes with using a scale. This force * the length of th arm is torque. Then multiply by rpm to get hp) Then run the motor at diffrent speeds and you will get the curve of fan power VS rpm.

The bus I drive (Navistar T444E) has a viscous fluid clutch just like my GMC Sonoma had. It is always spinning but faster when engaged.

Now this sounds like a good idea! I may even consider doing it if I don't run into too much trouble mounting the fan onto an electric motor.

The fan on my GMC just looks terribly inefficient at moving air. My attached picture doesn't make it look quite so bad as it does in person.

That's exactly how some eddy-current dynamometers work. Just make sure that you include the distance from the center of the armature axis to the outside of the motor can when you measure your arm length. It's also important that this arm is the ONLY way you have to keeping the motor can from rotating. Most motors have a pinch clamp bolt where the bracket attaches to the motor can or housing, so you would need to freely loosen that up or mount it on bearings so that all the counter-rotational force of the housing goes through the load cell on the arm.

I was thinking in terms of mounting the motor on a plywood square, which would have a narrow triangular support under the middle, under the center of the armature as you say. Then the support arm would be adjusted so it is level. With a narrow or sharp (knife edge) support the accuracy could be quite good.

Wow. I agree.There was a time when fans had 5 or 6 blades..

I finally measured the load that my original belt driven fan puts on the motor of my 1994 2.2L GMC Sonoma. I mounted the fan (viscous fluid clutch type) onto my 1 horse electric table saw motor. I set the motor on a steel bench in such a manner that it was free to roll and used a spring scale to measure force to measure ft. lbs. of torque and thus calculate HP. The motor turned at 3450 RPM which equates to the GMC's crankshaft turning at 2500 RPM due to the pulley sizes involved. The speed of the unengaged fan does not matter as I am only concerned with the torque load on the motor itself.

It turns out that the fan load measured 0.19 horsepower. I then heated the bi-metal coil on the clutch to engage the fan. At this point my setup was too unstable to use my spring scale for making a measurement so I used my best judgement for feeling the pounds of twist on the motor and came up with 5 pounds. This calculated to be 0.996 HP.

It should be noted that the clutch virtually never engages on this fan in my normal driving, so therefore I am not including it's power load in my calculation below.

Assuming that the fan requires 0.19 HP to cruise down the highway at 55 (no tach so I don't know if that's 2500 RPM but seems it must be close) and assuming that it takes perhaps 10 HP to cruise 55 (just a WAG) and also assuming 34 MPG at this speed (quite easily achievable) I wonder if this means that the fan would cost about 2 percent of my fuel to turn it since 0.19 HP is about 2% of 10 HP.

If the above assumptions are true then that fan was costing me about 0.68 MPG.

I drive about 10,000 miles/year. At 34 MPG and $4 gas it must cost me $1176 per year.

Removing the belt driven fan would then be saving me $23.53 per year. I paid $85 or so for the fan w/thermostat so it will take 3.6 years to brake even with $4 gasoline.

Also, my new electric fan uses almost zero power since it only comes on for a very short time, only on very hot days, as I climb my 1/2 mile hill in 1st gear coming home. Therefore I consider its energy use too insignificant to matter. It only uses about 85 watts when it is turning and that is brief.

With the information I have now I would not have installed the electric fan. It was fun however so all is not in vain. It makes me a bit suspicious of some claims that I have read of people saving 3-5 MPG by replacing a mechanical clutch fan with an electric fan.

Or, maybe all of my measurements are screwed!!

Well if you want share your raw data we can check all your math at least :)
Did you take a pic of your setup?
Do you still have the measurement of the lever arm etc?

The lever arm used was a bracket on the motor housing that measured 3.5 inches from motor shaft center. I measured 1 pound of force pulling perpendicularly from the radius between shaft and the point on the bracket when I attached a fish scale. This tells me the force in ft. lbs. was 0.2916.

I used 3450 RPM X 0.2916 / 5252 to get 0.19 HP.

No pics, but just imagine a basically cylindrical motor laying on a steel table top. It is free to roll. However, when it tries to roll it cannot because I have a fish scale attached to a bracket on the motor. I see to it that the scale is pulling perpendicular to a radius which extends from the motor shaft to the scale's attaching point. This radius is 3.5 inches long. Therefore, with the motor and fan running, I divide the fish scales reading (1 pound) by 3.428 (3.428 is 12 inches/ 3.5 inch radius) to get the actual foot pounds of torque which is 0.2917 ft. lbs.

I assume the motor speed is 3450 because it's plate says it is. The actual RPM of the fan (it's clutch is not engaged, but it still spins) is irrelevant. Only the torque load seen by the motor and the motor's RPM are needed.

Anyway, I gotta get to work now lest I have even bigger problems ;-)
Later

In the attached sketch
Is this the setup? Assume that the things close to being in line or at 90 degrees to each other are exactly so. The fan has been omitted for clarity.
Is dimension DIST-AA and DIST-BB 3.5 inches, or close enough?

Yes, that is it exactly and the AA and BB are indeed 3.5 inches.

The motor sits on the table on its very smooth and round surface.

To solve a statics problem, the sum of all the forces AND torques must be zero.
So there must be a force to offset or balance the force on the spring scale. This force is the red arrow marked REACTION FORCE on the diagram. This force must be equal in magnitude to the spring scale force but opposite in direction.

If we choose the center of the motor shaft as the pivot point for the torque calculations it makes things easy, because the lever arm length is the same for both the spring scale and the reaction force. Adding the two values gives a result that is twice what was measured on the spring scale. Alternately we could have chosen the point at which the motor touches the table as the zero point, in this case the spring scale force would be as measured but the lever arm would have doubled in length. Both solution methods give the same result. So either way, the torque on the fan/motor is twice what was initially calculated.

Statics problems are rich in opportunities to make errors. Hopefully I have not made one here. ;)

here is a idea for electric fans, I had a old chevy vega that i put a big b lock in , well needless to say a mechanical fan would not work for space reasons.. so i put in a electric fan and wired it directly to the coil wire so it was always on when i was running the motor as my radiator was rather small for the motor so i wanted to stay on top of the heat issues..

any ways when id shut the car off in a slight wind blowing and the car faced into the wind it would stay running as the fan was being spun by the slight wind and powering the ignition cloil.. the coolest part is that when not windy and faced into the wind, it was very easy to time the fan spinning down, so I could shut the car off, get out and walk a few yards away and snap my fingers just as it shut off.. people thought it was funny , like the Fonz on the jukebox..

However you easly could make that electric fan generate power when its not used, with a simple electric circuit.. to run it throw a relay to feed it power , when not running the other side of the relay feeds back intot he electric system and youd just needa (is it a diod?) to make a "one way valve" that if there is more than 14 volts comming back from the fan it charges the system.. that fan is always spinning when running down the freeway , recapturing some of it seems like a decent idea.

I have no #'s for my supra fan swap (mechanical - electric- mechanical) but i don't think that the fuel differance was much. However the hot high rpm performance was very noticeably worse.
However even 4 e fans (1 16" 1 12" 2 6") with 50 amps peak draw were woefully in-adequate cooling under track conditions.

Thanks for that information! I would not have figured that out myself but after reading what you wrote I do understand what you are saying and it does make sense.

I had originally concluded that it would take 3.6 years of fuel savings to pay for my electric fan. Your insight reduces this to 1.8 years. And this is all at $4 gas. If gas doubles in the meantime (hey, it might) my mod becomes even more worth the doing. In all honesty, I think with me it is more a matter of foolish pride than of the actual fuel savings although saving money, no matter how little, is always good.

This is very interesting!

In fact, since I can barely, or not even, hear my new electric fan, I wired a small light bulb across its power terminals and mounted the light bulb on my dash so I could monitor when the fan is running and thus be able to adjust its thermostat for best efficiency. Well, having done this, the first thing I noted is that the fan spins with the incoming air from vehicle motion, enough so that it lights the bulb, however dimly.

Actually, I am thinking that it is not desirable to have the fan spin when not being powered because that is putting undue wear on its bearings and brushes. I could configure a relay to put a dead short across the fan's power terminals when it is not being powered. This would greatly reduce its ability to freely spin, thus saving bearing/brushes wear.

Thanks for the honest opinion. Most of what I have read from guys who have put on e fans has said "it gained me 3-5 mpg!" or similar unlikely sounding numbers.

My guess was that the fuel savings would be minor. That is why I decided to try to actually measure what the old belt driven clutch fan was actually using after getting the good suggestion/technic from ttoyoda. There is no accurate way for me to measure the before and after FE gain with a mod like this unless the gain is very large or unless you want to do an extensive A B A type of test. Again, my only need to know the FE gain is for pride/ego, since I had already spent the money. Thanks again.

I would call it science. :thumbup: You are now smarter, you have done a test, taken data, and when you hear someone spout off with inaccurate info, you will be able to correct them. I'm smarter too, I had to think about how to do the calculation, so thanks. :)
I feel that 1.8 years is a pretty good payback time. I do home energy mods that have much longer payback.

I like this calc. Seems reasonable to me.
And we have to recognize - car makers and modders alike have spent more time and money on other 'FE' mods that yield less improvement.

I too put an e-fan in my car in summer '05. I also put in a light (green LED) to let me know when it's running. Fan is too quiet to hear under nearly all conditions.

I drive more miles. Currently about 14K /year (car pooling saves miles!) but since '01 it's averaged about 40K /year. So my payback was shorter.

A non-turbo Volvo 240 is shall we say, not over powered, at 115 hp and 3050 lb. So I'm glad to have taken that load off the engine and also get some FE benefit. OK, you can't have the FE and power benefits both at the same time - but I get either at any given time, controlled by my right foot.

Some older vehicles don't have a clutch on a belt driven fan. They would have better efficiency with the electric fan because the mechanical fan was always under load.

I replaced my belt driven fan in my '95 Chevy S10 4-cylinder with an electric fan from my Geo Metro.

Fan works fine, but like you said, it seems like a fan is never needed anyways. I think a big part of this is that the truck has the same size radiator whether you have the 4 or 6 cylinder.

I have ofter heard that radiators are designed for heat of the bigger engine size when it is towing.

If that's true, than a little S10/Sonoma should never need a radiator fan at all, other than maybe stuck in traffic on a hot day, with your engine idling.

What did you use as a switch to operate your fan turning on and off? Mine has been rigged up to run manually. If you have a specific thermal switch, could you please list the model number or where you got it from?

Thanks,

-Ben

bennelson I'm with you here. I put an e-fan on my b2300 mazda (ford ranger) and the fan never turned on except when i turned on the a/c or was sitting in traffic a long long time, also when playing in the dirt. Back then I did it for performance, not economy. As long as a vehicle is traveling at least 20mph, the airflow is enough to cool the radiator.

some ecus do have a fan control pin, even if the vehicle has a belt fan.
I used a performance 5.0 mustang fan i had left over when i sold it. It had a thermoswitch, an additional a/c turn on relay AND a manual override.

You can swipe the switches out of cars that came with e-fans from the junkyard.

how do you do that please? does it have a reostat?

also i need to get an electric fan for my rig. which vehicles use large high CFM
fans that i could pull from the JY??

there are 4 choices . all have pros and cons.

the electric need not run much , in fact only at stop lights .signs.

but the clutch fan is disengages. and uses virtually no power. Slips.

a directly bolted fan will indeed suck tons of power an fuel from any motor.
1950s all over again.?

then there is a flex fan that uses almost no energy and costs little. (tops)

the electric fan needs an inline radiator hose coupling with ECT to work reliably. (the ones that clamp to the hose are no good)
i tried others and all fail.

i installed my first in 1970 so , it's not new . (OTC kit)


the electric fan efficiency is low and so is the alternator, they compound
to huge heat losses total.

direct drive is the bench mark for 1:1 efficiency, IF ever there was.
same blades (like a submarine screw)
same playing field.

YMMV, heheeheheh

ps:
it does take load of pump bearings.
it is quieter then when clutch is hot and fully engaged.
and if clutch fan seizes, then this is a go.
I do like the beautiful simplicity of a flex fan.

You are confusing viscous clutch with an electric or mechanical on/off fan. 2 totally different technologies. A viscous clutch fan uses some power even when it is 'disengaged'. But it's never really free of friction especially above idle.

Nissan still used a clutch fan on the Z cars, and probably on some of their trucks (not sure about the trucks).

One thing to consider on the clutch fan is when it is not needed the airflow over the fan could actually reduce the load on the engine to drive the water pump amd the fan, to the point where there might not be any losses at higher speeds (speculation).

I have seen cars driven with broken water pump belts, where the clutch fan was actually providing enough coolant circulation to keep the engine cool. One time on my Z car the water pump belt broke and I limped home using pulse and engine off glide to get 6 miles back to my shop to replace the belt. When I got there the coolant temperature was actually lower than normal.

The clutch fans are controlled by a valve actuated by the small bimetallic spring on the front of the fan, which changes the valve restriction inside the fluid coupling depending on the temperature of the air flowing through the radiator.

I used to demonstrate to customers when their (clutch fans) were worn out by wrapping my hand in a shop rag and stopping the fan blades from spinning and actually spinning them backwards while the engine was running (definitely not recommended).

When they were worn out the customer complaint was overheating at idle.

On a lot of cars, like my 95 Riveria the radiator is angled at about 45 degrees to the vertical and you almost never heard the electric fans come on. I would attribute this to the fact that the angle of the radiator would allow air flow to increase when the vehicle was not moving and reduce the necessity for any fan operation.

I never heard of the electric fan creating electrical current from the air moving across it and turning the blades, making it a generator. Thats neat. I wonder if it actually happens on a regular installation, but I doubt it.

regards
Mech

no rain on any parade.
 

there sure are lots of speculation and dreaming here, that is for sure.
a brush motor and a generator are one in the same, cept the brush timing
is set to optimize one or the other , on some air craft, they are both in one.
a starter and a generator , same unit ,with special dual brush arrangement.\
(or similar trick)

your fan clutch test method SHOULD never be repeated , a bad idea if ever there was. Why do that?. wow.!

The FSM of each car explains how to do it with out taking any risk.
this one does well.
Saturday Mechanic - Fan Clutch Replacement - Popular Mechanics

with hand move fan , engine off ,cold, the fan must fell like a knife in natural room temp. butter. if it feels seized to the pump , it is.

NOW start the motor , cold.
and then strobe light (a tool) the fan hub and then the pump pulley.
note the huge diffr in rpm at idle and 2000 rpm. a the 2 points. That is the slip.
next ,follow the above link for the hot test.

most ppl that report huge MPG , gains.
are:
1: dont know how to do it. ( maybe 10000 posts here like that)
2: or the fan hub was bad, why on earth compare something good to something new and good.

google #1 and find out how to do it accurately, it's NOT easy.

averaging , is one way.
same driver, same road, same weather, same every thing.
in science this is called holding the constants and changing only one variable.
control the variables or its all just for fun and is useless.


one more link
Fan Clutch Testing: How To Test A Cooling Fan Clutch | Suite101.com


ok some fan clutches have a spring trap. (opens and closes at a specific heat thru rad) [ bi-metal spring is only one type ]
other fan clutches have a thick Silicone grease inside that gets hard when hot (feels like butter at room temp)
buy a strobe light- tachometer and strobe your fan and learn what it really does. It slips like crazy when cold and then trips on , on hot days , at a red light.

YMMV. really. testing MPG below 2-3mpg is no simple matter.
in fact this is the natural background error, in the best of conditions.
reporting gains inside this range, requires publishing your study and proving all variables were controlled.
as each car has differences and conditions of repair.


if you don't understand it? , take it apart and behold, see what it really is.

But, I still like a flex fan. with the nice curved blades.

I hope my directions and ideas are well received.
All my life , I've fought to kill measurement errors.

otherwise, it's all for nothing.

I would sure think flex fans are more lossy than electric.

Why?

Flex = always turning

Electric = very seldom on.

OK Maybe in Death Valley the flex could be better.

i never said one is better than the other, just pointing out facts few take in to consideration

• Electric conversion losses are the worst. (double conversion)
• noise
• heat.
• cost
• reliability (complexity is almost never the best solution )
  
  

the flex fan uses little energy ,
but does waste some in very cold climates, because at idle it is not needed
in prudhoe bay.

your choice will always be personal. Floating your own boat....?
I just want to give choices.

many ppl have bad clutches and don't even know it.
or dont know how to test it , or have wrong assumptions of how it really works. ( it just slips most the time)


Best post , for me?
saving motor with broken fan belt. fan made pump spin just enough to save the motor. Chock one up ,for a real fan.
never happened to me i do real PM's on cars. (since 65) Gatorbelts ! Rule.
retired yesterday .

Aside from the pickups and classics, who even has a fan clutch? :confused:

do the math on the Alternator in series with the losses on the electric brush motor.
they are both very wasteful, and in series they are pathetic.
start with a drop of gasoline can work that to the spinning blade (you will not err)

if your car is in Arizona and you drive in bumper to bumper traffic. for hours.
it is ,in fact ,wasting way more energy, then a simple direct drive flex fan.

so consider where you live too. and driving predominance's.

but thinking one is better than another , well that depends.
for MPG , then one that works best is the one that works the least.
and for many , its working full time all summer long.

all engineering is a compromise.

cheers

Of course the electric system has more built in efficiencies...

UNTIL you consider the duty cycle.

The belt was one month old and broke at the splice, brand new Nissan part.

Seems a little arrogant to ASSUME someone else does not do PMs.

Maybe your boat has a few leaks you have not discovered.

Spinning a fan blade backwards was an effective demonstration to a customer, very convincing to those who had been jerked around by other unscrupulous mechanics, many who recommended all kinds of PM that had nothing to do with solving their problems.

Kind of an admission fee to their little world of "I am the only really good mechanic on the planet" syndrome.

Maybe a different customer philosophy, but PM can be done by any amateur, while real diagnosis and repair of the cause of the problem is the test of 30 years without a summons to appear to explain your procedures.

No assumptions in this post, eh?

regards
Mech

Again all depends on the run time / run speed / conditions you operate in.

If you are a Las Vegas cab driver in the summer the Engine drive will be more efficient.
If your cooling fans run so rarely that you can't tell (until it hits 95F) when they fail (ME) the the electric will be much more efficient.

I'm not sure about that, ever notice how many cooling fins are on a belt drive fan clutch? Even with all of those fins and high airflow it still gets pretty darn hot. I'm not sure about this one, but if I had to guess it would be similar. Now if you could do an on/ off clutch we'd have a solution.

Depends on the fan both ways. I've had belt driven fans that were stupid loud, and I've seen electric fans that were stupid loud.
Not sure what you are talking about here.
Cost is going to depend on the application, but since most cars have E-Fans to begin with, I'd say that the E-fan would win. and if you were designing a system from scratch then E-fan would almost certainly win.
[*]reliability (complexity is almost never the best solution )

The flex fan is a pig at low RPM's and drains *less* horspower at high RPM's in an attempt to equalualize airflow/ horsepower loss. If your application needs constant airflow then it is a great option, if you drive your car either in stop and go traffic, have short trips (always on waste the most energy at start up) or on the interstate (high airflow regardless) it isn't such a good option.
Agreed, just use good Data

If they can run their fan slowly/ not at all then they have no reason to have an always on fan. They should either stick with the broken clutch, or run an E-fan with thermostat.

If their car doesn't run hot in their application, then it is of no concern to them.

What?
One could argue that several ways.
1) If you had reduced the harmonics of the fan you wouldn't have thrown the belt.
2) You should have watched the belt more closely.

What are PM's on cars?

I've run GoodYear Gatobacks before, and they do work very well, particularly with superchargers.