Wiring an Electric Cooling Fan - Holley EFI

[andrewb70]

A lot of good information can be found here: http://www.lateral-g.net/forums/show...614#post539614
The info is mostly about using a stock ECU to run fans PWM, but some of the info is applicable to using the Holley.

I'm working on a 1967 Cougar with a LS swap, and will be using the Dominator ECU.
I have a C6 Corvette fan & fan controller (solid state relay) that will be connected to the Dominator.
By all accounts this should work very well. Andrew

[racindego]

I'm still not thrilled with the PWM setup as I have it. I can't get an answer on frequency anywhere after several calls to JEGS and calls to every electric fan manufacturer I can find on the internet. I have tried frequencies ranging from 60 Hz to 1000 Hz. All work and the system seems to work fine with any frequency, but the fan makes a very audible noise that changes drastically with frequency, as you could imagine. This noise is louder than an electric fan on full blast, no matter what the frequency. So it defeats the purpose of the PWM to some extent. I'm also concerned about the life of the fans if I don't get this right. So I called Holley, maybe a mistake. I thought, hey, maybe they can at least give me a ballpark. Wrong. I was told that what I'm doing is NOT what the system was designed for and that if I wanted it to work correctly, I need to use the Holley PWM output to power a PWM controller, and use a brushless fan designed to work with that controller. I think similar to a Derale. If you aren't aware, this is probably $500 worth of crap I thought I didn't need. I was also told that the Holley PWM output on the Holley systems is designed for use with nitrous or meth solenoids - basically intermittent use. And although many use it for fans and fuel pumps, that is not what it was designed for. I'm probably going back to regular outputs and standard relays and stop wasting my time with this PWM. I feel like I'm fighting the system for no real benefit and a net loss in reliability.

[andrewb70]

I'm still not thrilled with the PWM setup as I have it. I can't get an answer on frequency anywhere after several calls to JEGS and calls to every electric fan manufacturer I can find on the internet. I have tried frequencies ranging from 60 Hz to 1000 Hz. All work and the system seems to work fine with any frequency, but the fan makes a very audible noise that changes drastically with frequency, as you could imagine. This noise is louder than an electric fan on full blast, no matter what the frequency. So it defeats the purpose of the PWM to some extent. I'm also concerned about the life of the fans if I don't get this right. So I called Holley, maybe a mistake. I thought, hey, maybe they can at least give me a ballpark. Wrong. I was told that what I'm doing is NOT what the system was designed for and that if I wanted it to work correctly, I need to use the Holley PWM output to power a PWM controller, and use a brushless fan designed to work with that controller. I think similar to a Derale. If you aren't aware, this is probably $500 worth of crap I thought I didn't need. I was also told that the Holley PWM output on the Holley systems is designed for use with nitrous or meth solenoids - basically intermittent use. And although many use it for fans and fuel pumps, that is not what it was designed for. I'm probably going back to regular outputs and standard relays and stop wasting my time with this PWM. I feel like I'm fighting the system for no real benefit and a net loss in reliability.

What fan are you using? What fan controller are you using? The C6 fan controller is about $100 and it's looking for a 128 Hz frequency. Andrew

[racindego]

The fans are JEGS part number 52101. I was told by JEGS that they are PWM compatible. I'm not using a fan controller, I'm using the Holley ECU to control the fans via a solid state relay. The relay is rated for between 1 Hz & 1000 Hz.

[andrewb70]

The fans are JEGS part number 52101. I was told by JEGS that they are PWM compatible. I'm not using a fan controller, I'm using the Holley ECU to control the fans via a solid state relay. The relay is rated for between 1 Hz & 1000 Hz.

If you read through the link that I posted, many guys are using the C6 Corvette fan controller. (This is basically a solid state relay and it's looking for 128 Hz at 90% duty cycle.)
http://www.amazon.com/ACDelco-15-808...words=10377609
People have used them with both stock and aftermarket fans. Andrew

[racindego]

I read through that link. If that relay is the same as a solid state relay, it won't answer my question or help at all. You also imply that the frequency would change vs. duty cycle. This is not how the Holley software works. The frequency is fixed at all duty cycles. I'm not sure where the disconnect is, but unless I'm completely misunderstanding, I don't need that module when using the Holley software to control the fans. And I have verified as much, my system works fine. I just don't know what to set the frequency at in the software, and no one can tell me. Combine this with the facts that I get the distinct impression that I'm the only one doing it this way (even though I followed the instructions in this thread) and also that Holley recommended against it, I think I will just forget about PWM. Unless someone who has actually done this can offer some information on a setup that actually works using the Holley software.

[carbuff]

Racindego,
Before I try and type this up, let me put a disclaimer in that I have not implemented the PWM setup myself yet. I'm thinking about it though, and I've read most all that I can find on the subject. And if my attempted explanation is overly simplified, I apologize in advance.

I think Jegs may have misled you a bit when they claimed the fan is "PWM compatible". The fan itself isn't the piece of the puzzle that matters. A fan is just an electric motor that spins the blade. When you apply a voltage to the motor, it spins. If you apply a low voltage, it will spin slower. And if you apply a higher voltage, it will spin faster. That's exactly what happens if you have ever used any kind of fan with different speeds. The switch causes the supply voltage to be reduced to the motor, which in turn spins the blade at a slower speed, and the max setting of the switch would supply the full voltage to the motor, giving the highest speed.

The fans for our cars work the same way. With a typical relay, you are either supplying full voltage or none at all, depending on if the relay is on. If the relay is connected to the battery or alternator, then that's the voltage going to the fan. As you probably know, when the car is running, and the alternator is charging, the systems will typically see about 14.4 volts or so. When you stop the engine (really, stop the alternator from charging), that will often drop to the battery voltage, which is about 12 volts. You've probably heard your fans pitch change when they are running and the car is not running, so they are being supplied from the battery. That 2.4 volt drop causes the fans to run slower (a little less than 20% slower).

The same concept applies when you are using PWM. But the piece you may be missing is something which converts the PWM signal from the Holley box into a variable voltage.

Did I lose you? Let me try to explain the control side now.

PWM, for anyone who doesn't know, stands for Pulse Width Modulation. What it means in practice, is that a signal is turned on and then turned off for a percentage of time over a fixed period of time. That time is determined by the frequency setting. The Corvette controller that Andrew referenced runs at 128 Hz. A Hertz is equivalent to 1 divided by the frequency, so 128 Hz equals 1/128 of a second, or 7.8 mS. The Holley allows you to program a range of frequencies (I haven't looked to see the max), but all that means is how precise the control will be. In this case, the resolution is 7.8 mS on or off, which is relatively coarse, but for this application is plenty.

Back to the explanation of PWM. PWM works by causing a signal to be turned on & off for a percentage of time. This can be accomplished in different ways. The simplest would be to turn it on for the specified percentage, then turn it off for the remainder of the time, then repeat. In the case of a 50% duty cycle, that means turn it on for 50% of the time then off for 50% of the time. For a 25% duty cycle, you would turn it on for 25% of the time, then turn it off for 75% of the time, then repeat. How does that relate to the frequency though? Well, it's my assumption that inside of the Holley box they have a free running oscillator that runs at whatever frequency is specified. Then they use a counter and multiplier to determine the on & off times.

If I specify 128 Hz and 25% duty cycle, the counter will start counting from 0 and counter up to 31. That's a total of 32 time slots, or 128 * 0.25. Then when the counter transitions to 32, it will turn off the output and leave it off until it counts to 127. The counter will reset back to 0 at that point and turn the output back on. The box could alternatively turn the output on for 1 cycle at time 0, then turn it off for 3 cycles during 1, 2, and 3, then turn it on for 4, and repeat. But this is more complicated and only works nicely for certain duty cycles.

The other important thing to note about this output is that it's switching from 0 volts to full voltage (12 or 14.4, depending on what the Holley outputs) whenever it switches. That means that whatever device is receiving this output is seeing the full voltage swing each transition.

Now we get to the missing piece... A fan is a 'dumb' device, as I described earlier. It will only spin a motor at a certain speed based on the voltage it receives. To have it run continuously at a steady rate which is lower than full blast, you need some device to convert the full-range on & off output of the Holley box into a constant voltage that ranges between 0 and full (12 or 14.4) voltages. That is what the Corvette Controller box does. If, in my above example, I set the duty cycle to 25% at a specific condition, and I send that information to the Corvette Controller via PWM, that controller will generate a voltage which is 25% of the full-range supply voltage, or 3.0 or 3.6 volts in this case.

Since that controller box is not part of the Holley ECU, that is why you have to know its frequency. You need to send a pattern at a rate that it will recognize and thus understand what you are telling it. If you tell it the wrong frequency, it won't understand the correct duty cycle you are sending it. Said another way, the Holley and the Controller will be out of sync.

Now, my guess is that you have the Holley PWM output running a relay that is switching at high speed, but that is switching the full voltage. So your fan is seeing the voltage turn on & off very quickly. That is not good, and I expect you will eventually wear out your fan by doing this. It's just like switching a light switch on & off as fast as you can: you will kill the bulb eventually. And that's also why I think you are experiencing erratic behavior. The fan is just cutting on & off really quickly and not actually achieving a steady state speed.

Phew...let me try and summarize the full picture with fewer words now.

The ECU will output a signal that turns on & off at a frequency that you specify in the PWM Configuration panel.

The percentage of time which that signal will be on or off will be dictated by the Duty Cycle you enter into the Configuration panel table.

You need an external device to convert the high frequency switching signal from the Holley into a steady state voltage, the value of which is determined by the percentage on/off time from the Holley output.

The output of that external device is the voltage you connect to the fan. It will range anywhere between 0 and full-voltage. The varying voltage will cause the fan to spin at different speeds: higher voltage -> faster speed.

Phew...with all of that said. I do not know of any other device besides the Corvette Controller Module which performs this function. I'm sure there is one out there somewhere, but I haven't found it yet. There are plenty of devices that can do the conversion in the electronics world, but those operate at much lower current levels than the fans in a car require. That's the other piece of the puzzle: having a module capable of supplying 20 - 60 amps, depending on your fan configuration.

Hopefully my long-winded explanation made some sense and helps clear up your issue a bit.

[racindego]

Carbuff, thanks for the response. I understand everything you said. I do know that the system works the way I have it, it's just loud (a symptom that many encounter when using PWM on computer cooling fans, I have learned - usually fixed with a capacitor). Also, I'm not sure what the Corvette controller does, as one member on here has said that it is just a solid state relay. One of you must not be correct. One thing I can say for certain - finding someone who has done this using the Holley software is next to impossible, as far as I can tell. And Holley tech seems to know less about it than us here in this thread. And I'm not bragging, since I'm pretty sure my setup is wrong, even though it works for the moment. LOL.

[Danny Cabral]

After reading Carbuff's post, it seems I may have misled Racindego. (If so, I apologize.) I thought some Holley EFI users have done it with just a solid state relay (without the controller). I have no desire to PWM control my electric fans, but now I realize a PWM electronic fan controller is more than just a large solid state relay. It converts the PWM signal (from the ECU) into variable voltage.

[racindego]

No need to apologize to me. We are all after the same thing here. I appreciate all the time you put in on this site.