The Stovebolt Site Forums The Shop Area The HiPo Shop Electric Fan Controller

I could use some advice concerning Electric Fan Controllers. I understand completely how they work, that's not it, it's more what would be advantageous to include in a proper controller...

A) Currently, to use an Electric Fan you need to find a place within the water stream to install the sensor. Very problematic since we only have one location at the top of the manifold for our Mechanical Water Temp Gauge. There are ways, but none of them are very attractive. Here is some testing that resulted in learning how to use a more modern approach:

http://devestechnet.com/Home/WarningSystem#WaterTemp

I solved this problem by using a more modern Automotive Grade Thermistor Circuit. The sensor is simply bolted to the Thermostat Housing and is adjustable from room temp to +257 degrees so fully within range and very sensitive (+/- 2 degrees same as the water stream type). With that solved, it makes electric fan's much easier to use, but...

B) Most electric fan controllers are just a relay, a sensor and some wiring. Are there more features that could be nice to have? For example, adjustable speed control? How about an On/Off switch? If you were on the market for the very best Fan Controller, what would you include that is not mainstream already?

According to the engineers at Spal, the fan can spike the amperage as high as 40 amps on startup and since some applications use two fans, I would incorporate two relays for that purpose. This is, as usual, research for another How-To on Mechanical Fan Alternatives. Everything will be documented so you can make it yourself. Thoughts?

If I were using a dual fan setup, I would want them to come on in stages, one first, then the other if needed. If they need to both be on at the same time, at least have one delayed a second or two, that would lessen the current spike on startup. I do like a manual switch as well, just in case all else fails and you need it on!

So I understand.. the manual switch should work...

1) OFF meaning... let the thermal switch do its job.
2) ON meaning.. turn it on NOW and override the thermal switch
right? So this way, there is no chance the fan will miss a cycle and somehow lose its ability to cool the engine.

So, ON, if left on by mistake would just run the fan (whenever the Ignition switch is ON) forever. But if OFF, it would run the fan according to the Thermistors setting. I like this. There is no chance of ruining the engine due to no fan at all that way.

I can see using a dual thermistor circuit in a perfect world, where you have one go off at the standard 185 degrees, and the other starting at 190 or so. Ok, that is also a new idea. Remember, if I do two Thermistor circuits, it would be two thermistors in the same lug cavity, set 'differently'. Meaning YOU set the temperature of one, and YOU set the temperature of the other via a small potentiometer on the circuit board for each. So this is configurable by you on both counts. Yes?

I can also add a 10 second delay circuit, but that is not really exactly what you are going for. You are going for optimal engine cooling. Keepem Coming Folks! Great Post Seattle!

Usually you have them come on at different temperatures. So say one comes on at 170, the other at 210.

Deve,

Is this what you what to replicate?

This unit even takes a signal from your AC unit to run the fan/s when the AC is on.

I have this controller in my truck and really like it.

Mike B

Mike.. Link? and yes. The idea is to make the Worlds Most Awesome Fan Controller BAR NONE. Then, make the plans to make your own free of charge to everyone.

oops, sorry here's the link!

https://www.ronfrancis.com/prodinfo.asp?number=AR-99

Mike B

Mike, There are certain companies that are just THE standard by which everyone else wishes they could emulate. Steele Rubber for Rubber Products for our vintage trucks, and Ron Francis Wiring for the Electrical system. I am an admirer of Ron Francis.

Yes. That is exactly what I am going for. But, as you know from the Critical Systems Warning Circuit Thread here that you are aware of, I have been working on this problem now for quite a long time, and completely SEPARATE from Ron's research, I came up with the exact same conclusion.

An Automotive Grade NTC Thermistor Circuit, and the one that I created on my own here:

http://devestechnet.com/Images/Projects/WarningSystem/WaterTempSchematicFinalc.jpg

which was augmented by a friend to include something Ron's solution didn't incorporate, a Supervisor Circuit is in operation and in quality everything Ron did and then some, really proves WE CAN!

That basic circuit is better because with Ron's solution, if a wire was broken or the circuit was broken in some way, the only time you would know is if you noticed the fan did not come on when it was supposed to. My circuit monitors the health of the circuit and ensures all wires and components are working correctly. When they are not, an Amber LED on the box lights to tell us a wire is broken (the most likely) at the Sensors lug housing. Of course, it is NOT necessary in some people's mind because you SHOULD know there is something wrong if the Fan does not come on as expected. But, it's the little things.

So, without being critical of Ron's work, which I am always amazed at, and they know it, how can we make that just a little better? (We already have.)

Mike, what do you think of the idea of an 'Interrupt' Switch as stated above? (off being normal op, On being an override so the occupant can turn on the fan by hand?)

Fan speed control? The AC override is something we would also need to add. Even AutoWires web site shows how to add that feature with their old Thermal Sensor technology.

I am not annoyed by it, but I am not emulating Ron's System, rather doing something completely separate and without any consideration for what is out there already. It's nice to see what's out there, but it should in no way change our minds as to what we can do to make THE ultimate Fan Controller. I learned about the Automotive Grade Thermistor direct from Vishay, the manufacturer's Engineering team when researching the CIS circuit. I then went about making my very own Lug Mounting for Vishays Thermistor. I have room to put about four of them inside one Lug Housing, so we can give each fan their own sensor very easily. YOU were instrumental in guiding the thought process, then months of playing with other peoples designs led me to an even better idea. Make it from scratch!

I am thinking we could easily make each fan have its own sensor, fully adjustable, complete with the supervision circuit, and an Interrupt Switch with AC compatibility.

I am thinking of mounting the box near the Fans. Two 40 amp Relays would be inside the box with the rest of the circuitry. I am a fan of redundancy, so I am designing this circuit to use two 20 amp relays that are soldered direct to the PCB, THEN use those two relays to fire the two 40 amp ones. The 20 amp ones will be very solid and unlikely to ever die, but the 40 amp ones would be mounted in the box separate so they are easily replaced.

Sorry for the novel, but this is EXCITING!!

Deve,

In my case I wanted less clutter in the engine bay, so I mounted ALL of my electric controllers/relays under the dash on the firewall (Fuse Panel, Headlight Controller, Fan Controller and Power Antenna Relay).

As for two sensors...I only want/need one because I only have one fan. Are you think of putting two sensors in one lug to help eliminate clutter?

The bypass/override switch woud be a good idea as lomg as it's in the 40 amp circuit so the controller could be completely turned off in the event of a failure.

Mike B

Yes. Two sensors in the same lug side by side will give me consistent temperatures since they are both in the same location.

The really cool thing about thermistors is they can 'represent' a heat value. So if the outside of the thermostat housing is really only 160 degrees when the actual water temperature is 180, we can adjust the thermal characteristics to compensate for the difference. I have tested this carefully and it works flawlessly. Then when you mentioned Ron Francis (awhile back) I felt gratified that someone else was doing the same thing.

I would just include both thermistor circuits in the package since they are so cheap.

http://www.mouser.com/ProductDetail/Vishay/NTCLE203E3104HB0/?qs=%2fha2pyFaduhaIbtOc0OaTdeNfzbCZzQ0F1R5X49TLWtdxj2enfIeQA%3d%3d

http://www.vishay.com/docs/29118/ntcle203.pdf

Yes, the switch would have to be in the higher amp circuit alright. I always include a fuse and SPST mini-toggle on the box that is separate from that as well so you can kill power to the circuit.

I wonder how many people would put this under the dash as opposed to near the fan.

Mike, how many wires does the Ron Francis solution have going to the Thermistor Sensor? Where did you put it? Mine has one wire to connect (per sensor) and it goes on the Thermostat Housing screw.

The Ron Francis thermister has a mini coax cable to carry the signal. I mounted mine on the t-stat housing to manifold bolt.

So you are thinking about putting two thermisters in one lug with a multi wire harness to keep it clean?

I built my truck to look stock so there are no extra switches or holes cut in my dash or hanging under it. I would have mounted both the kill and override switches out of sight.

Mike B

I had to do so much of this same research for the Critical Information Circuit that its almost all done for me. This is what the lug that I designed looks like:

http://devestechnet.com/Images/Projects/WarningSystem/WarningSystem0041lg.jpg

http://devestechnet.com/Images/Projects/WarningSystem/ThermistorLugHousing.jpg

The Thermistors are held in by bending the tabs over. This puts them up against the lug metal, then high temp silicon covers everything. This has been tested over and over and the results are every bit as good as an immersed sensor.

What does the lug look like that connects to yours?

The reason I asked about one or two wires is because I figured out how to use one wire to the thermistor with the other one soldered to the lug itself. Less wires to break is always a good thing. Then since there will be two thermistors, I will use shrink tubing for the first foot or so from the lug, then small pieces of shrink tubing every 6 inches to the box.

The box would have the circuit kill switch mounted to it since it would be pretty rare you would ever use it, if at all. The override switch can go anywhere you choose. The box would need to be mounted so you can get a screwdriver inside the holes for the small circuit board potentiometers to adjust the temperature of each fan ON condition. I will preset them so FAN1 comes on at 185 and FAN2 comes on at 190 or whatever you guys think makes sense. For one fan, just ignore the FAN2 part.

I hear that Mike. Keeping things stock looking is the hard part with some of this stuff, but its all possible with a little creativity.

The lug on my truck is an aluminum barrel lug with the thermister epoxied in and then heat shrink was added to hold the coax cable to the lug. I wonder if they used the coax shield to protect the signal wire from outside electrical interference?

Mike B

Maybe they thought they needed to, but since this is a ohm value solution using 12v and voltage dividers, there is no frequency or PWM issues, so I don't see a need. But it probably looks pretty cool. Coax isn't a bad idea since the shielding is used for ground. My variation doesn't require a return ground wire so it would be wasted. If coax isn't used for fuel sending units, it shouldn't be needed here.

The Thermistor is a variable resistor. The control variance is temperature. With the NTC type, Ohm value goes down from the stated value. A 100K Thermistor's ohm value is 100K at room temp and as little as 50K at high temp. It's a really elegant solution for those having trouble trying to figure out how to get a conventional water temp sensor in the water jacket. It also lasts longer because they are not snapping open and closed all the time.

Thanks for the info Mike.. if anyone has any other ideas on what to include in this project... Did we decide that controlling the speed of the fan is probably not a desireable thing?