Cooling Fan Tech

[realsquash]

Location and ambient temps have a big effect of radiator and fan needs. What works for rock bouncers and technical crawling with light weight buggies doesn’t work for continuous throttle KOH racing on a hot year. Radiator location also has big effect, catching air flow at speed vs a radiator down low with fan only air flow.

Believe it or not, ambient air temp over the 32-100F range has little effect on a radiator's efficiency. Like maybe 5% over that range. Getting natural air flow through the rad is useful for sure, but not because of ambient air temp.

 

[realsquash]

My new cooling package could either twin 12" 500W (each) fans side by side that overlap the tanks a little or a single monster 19" 850W unit that doesn't overlap.

Both options will be Spal brushless units.

The single fan route is substantially cheaper but I like the redundancy of 2 fans.

I'm using an 850w brushless, but from a BMW, and making my own shroud from scratch. It is on a "square" radiator (22x19 but thicccccc) so I had no choice. I only mention this because the BMW fans are dirt cheap and in every junk yard in the land. Look up a 2015 550i for example. The ATS/Camaro fans are much less common in the junkyards and cost more when you find them. You can unbolt the fan motor from the BMW shroud and build your own. It's still about a 19" OD fan blade.

 

[HYDRODYNAMIC]

Believe it or not, ambient air temp over the 32-100F range has little effect on a radiator's efficiency. Like maybe 5% over that range. Getting natural air flow through the rad is useful for sure, but not because of ambient air temp.

The DeltaT equation is based upon incoming air temp as well as incoming fluid temp. That is how one determines radiator capacity.
I am not following you on how ambient air temps are insignificant.

 

[Bebop]

I'm using an 850w brushless, but from a BMW, and making my own shroud from scratch. It is on a "square" radiator (22x19 but thicccccc) so I had no choice. I only mention this because the BMW fans are dirt cheap and in every junk yard in the land. Look up a 2015 550i for example. The ATS/Camaro fans are much less common in the junkyards and cost more when you find them. You can unbolt the fan motor from the BMW shroud and build your own. It's still about a 19" OD fan blade.

Nice ! They are indeed dirt cheap.

Now the question is the following : How do you PWM control them ?
I doesn't appear that they are SPAL, the motor looks way different.

 

[realsquash]

Nice ! They are indeed dirt cheap.

Now the question is the following : How do you PWM control them ?
I doesn't appear that they are SPAL, the motor looks way different.

They work exactly the same - "switched ground" PWM, vary duty cycle from 5 to 95% (might be 5-92, I don't remember) and a frequency of 128hz works fine. A GM ECM runs it without issue. Or you can rig up an Arduino type thing with a temp sensor on its own to run the PWM part by itself.

 

[Bebop]

They work exactly the same - "switched ground" PWM, vary duty cycle from 5 to 95% (might be 5-92, I don't remember) and a frequency of 128hz works fine. A GM ECM runs it without issue. Or you can rig up an Arduino type thing with a temp sensor on its own to run the PWM part by itself.

I have a Holley ECU that has a programmable PWM output so that's not a huge deal. Glad to see that the frequencies are known. That's another option to take into account, cool and thanks for sharing !

 

[realsquash]

The DeltaT equation is based upon incoming air temp as well as incoming fluid temp. That is how one determines radiator capacity.
I am not following you on how ambient air temps are insignificant.

Check out SAE 2000-01-0579. They designed a model, then tested and compared it to a real car and adjusted, and then came up with some conclusions. They aren't testing edge cases where a radiator is undersized or doesn't have enough airflow. In addition to airflow and air temp they tested coolant flow and it had a larger effect but didn't go into more testing on that. If your incoming air is 200F that's a different story, obviously.

 

[realsquash]

I have a Holley ECU that has a programmable PWM output so that's not a huge deal. Glad to see that the frequencies are known. That's another option to take into account, cool and thanks for sharing !

Most automotive PWM stuff is designed to take the least amount of CPU power needed to calculate whatever they are doing. They usually work over a frequency range of like 100-250hz and need to account for inconsistent CPU cycles during the PWM counting. None of these devices need to be dead locked on a particular frequency. They are almost all switched ground because it faults out more "nicely" when wires get pinched and it's not particular about voltage. Water pumps, fans, fuel pumps, etc. Most of them work the same way.

 

[bdkw1]

I'm not sure why they say that, but since there's no justification,

They may be referring to the shitty fans that get stuck directly to the core with the plastic tube saws and no shroud.

If your using a shroud and the fans are at least an inch off the core, overlap is fine as long as the shroud is sealed pretty well.

 

[DesertCJ]

Geeze, I'm still living the simple life with a Taurus fan in a custom shroud running off a 75 amp relay and a thermostat switch. I suppose one day, I will have to figure out all the new fangled pwm, brushless fans and join the modern era...

 

[HYDRODYNAMIC]

Check out SAE 2000-01-0579. They designed a model, then tested and compared it to a real car and adjusted, and then came up with some conclusions. They aren't testing edge cases where a radiator is undersized or doesn't have enough airflow. In addition to airflow and air temp they tested coolant flow and it had a larger effect but didn't go into more testing on that. If your incoming air is 200F that's a different story, obviously.

My understanding of the model from the description is saying that there is not much change when the ambient air temp and the coolant inlet temp rise together. In other words the ETD entering temperature difference is the same as long as both coolant temp and ambient temps rise together.
That is very different than saying ambient air temps do not effect the heat transfer while not raising the coolant temp to keep the ETD the same.

Attached is the graph of oil coolers with 100ETD which is 100F air and 200F oil with a DeltaT of 100F which would be the same if the air was 50F and the oil was 150F.
With engine coolant and oil coolers the coolant temp is set and can not raise just because the ambient air temp rises. So the ETD will not be constant.
That SAE study just goes to prove that the ETD equation is possible because the relationship between ambient air and entering coolant are linked in almost a linear rate regardless of temp range.

 

[WaterH]

Although two fans sounds redundant, if one dies then you have a big hole in the shroud that the other fan will pull air through. Both fans need to run together to get max cooling. For a low temp one fan on one off then it’s not a big deal of bypass is occurring. But both on is required for high temp cooling.

If you want the redundancy of two fans and max performance with one fan down, you need a rubber divider between them. That way the one remaining fan pulls all its air through the cooler.

 

[HYDRODYNAMIC]

If you want the redundancy of two fans and max performance with one fan down, you need a rubber divider between them. That way the one remaining fan pulls all its air through the cooler.

Agreed.
I ran into an issue on an industrial application where when one of the fans was off the other fan would start pulling air through the dead fan and spin it backwards. When the second brushless fan went to turn on it was spinning backwards just fast enough that the motor rotor could not turn slowly enough to find home to turn on and ramp up. When brushless fans turn on they wiggle back and forth and find home and then start ramping up. So if you are running brushless fans independently they need to be isolated from each other if there is a chance they can spin the others backwards. Ideally brushless fans would both be variable controlled at the same time and have no need to be turned on and off independently but if you are running a manual override or simple temp switch without variable speed someone could run into this issue.

 

[rockdog57]

Geeze, I'm still living the simple life with a Taurus fan in a custom shroud running off a 75 amp relay and a thermostat switch. I suppose one day, I will have to figure out all the new fangled pwm, brushless fans and join the modern era...

Me too. Been working for years

 

[HYDRODYNAMIC]

Bebop mentioned a new candidate for brushless Spal fans in the 12" size range. Chevy Volt. Dual 300 watt fans but it also comes in a really nice shroud that looks easy to fit onto the rectanglish radiators that most are running. Maybe 16"-18" x 26"-28" shroud based upon pictures.
It appears that the cooler and fans are not for the small engine but rather for the Drive Motor Inverter Cooler. Which makes sense since dual 300 watt Spals can easily cool a V8. I assume that the electric system and batteries can pull V8 heat loads in a short time period and this is the need for such large cooling. Who would have thought that hybrids would be so inefficient.

 

[gt1guy]

They may be referring to the shitty fans that get stuck directly to the core with the plastic tube saws and no shroud.

If your using a shroud and the fans are at least an inch off the core, overlap is fine as long as the shroud is sealed pretty well.

Here's a picture of a custom C&R Racing radiator that's 26" wide and running 2 of the new 12" Spal fans. Fans overlap the side tanks and the shroud has foam sealing it to the tanks. Shroud is also pretty deep.

Just showing that a top tier radiator company has no problem overlapping fans.

 

[realsquash]

Bebop mentioned a new candidate for brushless Spal fans in the 12" size range. Chevy Volt. Dual 300 watt fans but it also comes in a really nice shroud that looks easy to fit onto the rectanglish radiators that most are running. Maybe 16"-18" x 26"-28" shroud based upon pictures.
It appears that the cooler and fans are not for the small engine but rather for the Drive Motor Inverter Cooler. Which makes sense since dual 300 watt Spals can easily cool a V8. I assume that the electric system and batteries can pull V8 heat loads in a short time period and this is the need for such large cooling. Who would have thought that hybrids would be so inefficient.

On the Volt those two fans cool everything - engine, electronics, A/C, etc.

 

[DesertCJ]

Bebop mentioned a new candidate for brushless Spal fans in the 12" size range. Chevy Volt. Dual 300 watt fans but it also comes in a really nice shroud that looks easy to fit onto the rectanglish radiators that most are running. Maybe 16"-18" x 26"-28" shroud based upon pictures.
It appears that the cooler and fans are not for the small engine but rather for the Drive Motor Inverter Cooler. Which makes sense since dual 300 watt Spals can easily cool a V8. I assume that the electric system and batteries can pull V8 heat loads in a short time period and this is the need for such large cooling. Who would have thought that hybrids would be so inefficient.

What's the deal with the Camaro fans? They're half the price of Volt fans on Rockauto...

Edit: 2015 Camaro fans...2016 up are crazy expensive

 

[realsquash]

What's the deal with the Camaro fans? They're half the price of Volt fans on Rockauto...

Edit: 2015 Camaro fans...2016 up are crazy expensive

2015 aren't brushless. And the newest Camaro shares the platform with the Caddy ATS, and those are all brushless. There are actually a ton of donor cars for these kinds of fans - Merc, BMW, Chevy of many kinds, etc. There are WAY more BMW fans on the market than anything else as far as I can tell. 400, 600, and 850 watts.

 

[Bebop]

The Volt fans are cheap as hell on eBay and the stock shroud would be easy to cut and repurpose.
The ones I’m looking at are from the 2011-2015 since in 2016 they went to a big single 300W fan.

Thats a reworked set of Volt fans.

 

[Grendel]

I like the BMW fans, thanks for that.

 

[WaterH]

. If you dont use a sensor, one signal wire is grounded (-) and one is powered (+) and it doesn't need anything else to run.

I just ordered a brushless fan for testing. From what the guy said, your above statement is not correct. He says the ground would have to be "pulse modulated". I have to admit, you sound like you know more about it than him.

 

[realsquash]

I just ordered a brushless fan for testing. From what the guy said, your above statement is not correct. He says the ground would have to be "pulse modulated". I have to admit, you sound like you know more about it than him.

What fan did you get?

 

[WaterH]

What fan did you get?

I got the 10" fan. I'm toying with putting two of them in the Delorean.

 

[realsquash]

I got the 10" fan. I'm toying with putting two of them in the Delorean.

Chevy Volt fan?

 

[WaterH]

Chevy Volt fan?

I don't know if it's the same. I should have checked because $$. But there aren't any volts in junkyards around me anyways.

 

[HYDRODYNAMIC]

I just ordered a brushless fan for testing. From what the guy said, your above statement is not correct. He says the ground would have to be "pulse modulated". I have to admit, you sound like you know more about it than him.

The GEN 2 motors which are the current ones can run analog, PWM, and to ground to turn on. The GEN 1 had to have a signal and is inverse of the GEN2. I believe the same goes for GM motors being controlled similar to GEN1 motors.
I believe the 10” brushless is a newer release and should only be GEN2.

 

[holdmypocket]

Of course I've run into this thread the day after I finished installing a www.autocoolguy.com PWM controller with a MK VIII/Dorman 620-118 fan on my Ford 400 with three-core Champion radiator. I sprung for this controller because I kept reading about relays failing/being unreliable, wanted soft-start, only wanted it to run as fast as it needed, and the fail-safe is nice. Reading this article sold me: Dave's Volvo Page - 4-Speed Mark VIII Cooling Fan Harness Project and Diagrams

With all that said, I've only driven the truck around town in 85-90 degree weather, rocksteady at 180 where I set it. After a minute with the fail-safe switched on (forces it to full speed) idling in my driveway it drops from 180 down to 160. Seems pretty sweet; not sure. Would I see an improvement going to a brushless fan?

Anyway pics of mounting using angle-iron..(going to bend one up out of sheet metal that covers all of the radiator when I have more time and patience)

 

[Bebop]

The GEN 2 motors which are the current ones can run analog, PWM, and to ground to turn on. The GEN 1 had to have a signal and is inverse of the GEN2. I believe the same goes for GM motors being controlled similar to GEN1 motors.
I believe the 10” brushless is a newer release and should only be GEN2.

I agree with you

 

[realsquash]

For what it's worth, I have a Volvo non-brushless fan on a project and I'm PWM-ing it with the controller from a Ford Fusion using an E67 GM computer. It doesn't have the advantages of brushless motors, but it's a very good setup. I also control the A/C compressor with the E67 so that varies the fan constantly to try to keep the A/C high side pressure down (cool better). This was $15 at the pick n pull with a handful of wiring, but the connectors are available aftermarket. Most of the time the fan never runs fast enough to even hear it unless it's very hot and the A/C is on.