Automotive - Radiator and condenser dual fan / two speed controller

[Buzz pop smoke-oops]

Hello Electro-Tech-Online enthusiasts,

I'm new to the forum and new to circuit design.

I have a need for a 12vdc dual fan / two speed control circuit (two fans, each have two speeds)

Here's my theory of operation:

* Controller receives power from ignition key switch so that power is only with key in position 2 (engine running)
* To buffer load on the alternator diodes, no two fans will be energized simultaneously. Time delay prevents two fans from starting at the same time
* Inputs for fan1 will be a) temperature switch in radiator when closed calls for low speed (ground closes contact) b) signal from Engine Control Module engine temperature output will call for high speed
* Inputs for fan2 will be c) clutch engagement (12vdc to refrigerant compressor clutch from vehicles' automatic climate control (ACC) calls for low speed d) pressure switch @ refrigerant receiver/drier (ground closes contact) calls for high speed
* Power to controller is negated at engine cranking (no power to fans while starting the engine) (12vdc from starter bendix)

Here are my thoughts of components choices for control of operation:

I think filtering 12vdc via voltage regulator LM7805 will clean up the dirty / noisy line input. Obviously this results in operating the circuit in 5vdc

A hex inverter MC14069UB will handle the inputs. Where two inputs come in simultaneously, an XOR gate CD74AC86 will prevent two speeds of either fan from being energized (based on logic truth input1=1, Input2=1, output=0)

Where power to both fans is called for, regardless of fan speed, a 555 timer will space out energizing one fan before the other.

Outputs will energize fan speeds through emitter follower switches via NPNBC109 utilizing the flyback diode to prevent relay chatter. The emitter followers will ground coils in standard automotive 5 pole relays to close contacts which will energize each fan speed. These relays will receive power from the vehicles line bus.

All outputs will be fused. Inputs of 12vdc will be fused prior to the controller.

I've chosen components mostly based on what I have on hand. I may be off base in some of these component combinations but they seem to all have the compatibility, versatility and robustness to handle the work sustainably.

However, I have questions - which is why I'm posting on the forum:

* Is it reasonable and/or necessary to regulate line voltage? Is my choice or regulator sufficient?
* Have I added undue complication by regulating to 5vdc for my working voltage?
* Am I thinking in terms of what is electrically possible by attempting to ground relay coils with 5vdc through each transistor?

 

[Danwvw]

I think the Auto Parts Stores carry those "Engine Cooling Fan Modules" Depends on the year of the car. Cars in the 80-90's just used a couple of relays and big heat sinked in line resistors to do it. What Year and model of car? I suppose you could design and build one Probably the 5 volt reg might play a part somewhere in the circuit however you would not use it on the Power to the fans. That could be handled by Power Mosfets like a FDP025N06 driven to desired fan speed by oscillating 7555's that your logic could switch in as needed.

 

[Buzz pop smoke-oops]

Thanks, Danwvw,
I've thought about asking parts stores, I may still do that.
I've wanted to take a shot at circuit design and when this came to me I though this could be my chance...
The bulk of my electronics knowledge and experience comes from industrial fabrication and machine maintenance / repair. So I can stumble my way through a rudimentary design, but this one's got a bit more substance to it.
The application is an aftermarket dual fan assembly in a fabricated shroud, made by VDO. It was installed on a clients' car, a modified 1987 Jaguar Series III XJ6 what's had a General Motors LT1 V8 engine installed by my clients' neighborhood hack / con artist (come to find out) before the job was pulled away from that joker and brought to me.
I agree and understand 5vdc is not suitable for powering the fan motor windings, I'm needing clarification on grounding each relays coil through my emitter follower switch circuit to close each set of contacts.
I will clean up my drawing and post a pic here, maybe you and others will have a look and see if what I'm trying to do makes since.
I'm glad I posted here today, I'm looking forward to some good conversation and chances to broaden my understanding so I can make more things work.
Kind Regards,
Austin James

 

[Danwvw]

The BC109 Transistor could drive a 70 ohm relay but maybe can't handle just any kind directly, nor could other CMOS Logic components. When I need a high gain transistor to drive something drawing several amps, I go to the 2N3055(npn) or the 2N2955(pnp) (Total Over Kill Though) also power N channel FET's like the IRF-540 and FDP025N06 and many others will ground a relay or the fan the FDP025N06 (because it's only 0.0000025 ohms when on) could easily handle the fans current for that matter.

To use the IRF-540 or any other N Channel Power FET Transistor as a fan relay driver to ground the Coil:
The high side of the coil will go to + 12 volts,
The ground side of the coil goes to "D" (Drain) of the IRF-540,
The gate "G" of the IRF-540 will make the IRF-540 conduct between "S" (Source) and "D" when it is High (ie 5 volts) and stop conducting when it's low (ie less than 0.5 volts),
Connect "S" to ground.

There are so many approaches a design could take to deal with the control of the fans including they could be handled by a ECU of some type. But really with the State of the Art in Power Mosfets being so good and their compatibility with CMOS components, relays are not needed to run and control the fans.

 

[shortbus=]

The gate "G" of the IRF-540 will make the IRF-540 conduct between "S" (Source) and "D" when it is High (ie 5 volts) and stop conducting when it's low (ie less than 0.5 volts),

Are you sure about that? I didn't know the IRF540 was a logic level mosfet, won't using 5V on the gate make a lot of heat in it? The data sheet says it's a non-logic level. https://www.vishay.com/docs/91021/irf540.pdf

 

[Danwvw]

It's probably not considered a logic level FET, that is correct. However I have tested it and some others like trench HD FET's etc and 5 volts is plenty to get enough saturation that they won't heat up. For example a fet at 60 milli ohms (Source to Drain) will only generate 35amps X 0.000060 .06 ohms X 35 = 2.1 Volts X 35 is 73.5 0.00021 watts at 35 amps.
A Power Mosfet Gate has a bit of capacitance but had very high impedance to DC. If you feed one a pulse width modulated 0-5 volt square wave you can control the fan motor seed anywhere you want it. It's important that the gate wave form have a quick rise and drop time to avoid any heat in the FET's operation and if it's done right they don't need a heat sink. If you had a analog temperature sensor on the head of the engine you could have variable control of the fan speed then just override the circuit to drive both fans high speed when the AC turns on.

 

[Diver300]

That power calculation isn't correct.
60 milli Ohms is 0.06 Ohms.
At 35 A, that will be 2.1 V across the MOSFET, and the power will be 73.5 W of heat generated.

 

[Danwvw]

Yep I did volts got to multiply X 35 amps again for watts! I have been thinking mOhms was Power of -6, I should check this. But the FDP025N06 could handle it.
Ok Your right so using the FDP025N06 with just 2.6m ohms would be 2.6/1000 = .0026 ohms at 35 amps which would be .091 volts and that X 35 amps would be 3.2 Watts. So yeah, Probably need a heat sink. I have an application using Power FETs on at about 10 amps that mount to my voltage regulator housing that gets 100' C pretty easy from the engine alone but it worked well with all of the Power FETs I tried however it's not over 50% duty.

 

[Danwvw]

Man there are a lot of these things already invented! Here a nice one: PWM Module from Speedway.

 

[shortbus=]

It's probably not considered a logic level FET

I'm strictly a self taught, mostly from web forums, amature at electronics, but I've had many people both here and AAC tell me to get closer to 10V G-S on these type mosfets. The 5V just seem to close to the max threshold voltage (4V) to be a good idea. Wasn't trying to be a smart a$$, just trying to keep up with how things are changing in their use.

 

[Danwvw]

This is very true, one thing is when 12 volts is available, Use it on a pull-up resistor to the Gate the cmos open collector ic's etc allow it.