Cooling fan circuit re-design

[Farrar]

I don't have software actually designed for producing schematics, but I think this will give the general idea of what I plan to do with my cooling fan circuit.

4467

FYI, Fuse #7 is indicated on my schematic because I have removed many things from that circuit and as such there was a wire already available. If one were to implement this, one could also run a new "hot" wire and insert an inline fuse. Originally, my goal was simply to provide protection for each fan, but after reading about Welmoed's constant-power modification, I decided to implement that as well.

Farrar

[DMCMW Dave]

If you do it like this you have the breakers in the relay trigger circuit rather than the load source for the fans. The breakers need to be on the side of the relay to the far left in your picture.

You will find that without a diodes in front of the compressor, any time the otterstat turns on the fans the compressor will enagage too. And probably burn out the otterstat since it's not made for that much current.



I don't have software actually designed for producing schematics, but I think this will give the general idea of what I plan to do with my cooling fan circuit.

4467

FYI, Fuse #7 is indicated on my schematic because I have removed many things from that circuit and as such there was a wire already available. If one were to implement this, one could also run a new "hot" wire and insert an inline fuse. Originally, my goal was simply to provide protection for each fan, but after reading about Welmoed's constant-power modification, I decided to implement that as well.

Farrar

[Farrar]

I've left the factory diodes in place; they're simply not black on the diagram because the drawing software I use does not allow for it.

I am not sure what you mean about the circuit breakers; my goal is to have one for each fan. Did I put them in the wrong order like the factory did? I will look again.

Thanks,
Farrar

[Ron]

He means put the breakers in the yellow wire(s) feeding the relay.

As is, assuming this is a Bosh style relay, it or the feed wire (yellow) will melt down as soon as you connect the battery. (87a is connected to 30 when the relay is in the normal state(unenergized), so, you have 12V unfused > 30 > 87a > ground. Sparky wouldn't like it.) Don't use 87a here, otherwise OK.

You should always label the relay pinouts (or draw the guts).
30 - Feed
85 - Relay Coil Ground
86 - Relay Coil Feed (Trigger; low power)
87 - Power Output (high power) - Normally Open Contact
87a-Power Output (high power) - Normally Closed Contact

[Farrar]

You mean like this?

4486

Farrar

[DMCMW Dave]

Yes, but (as in the first version) your indicator light circuit has effectively tied the outputs of the two relays (and so the fans) together.

[Ron]

...and you still have 87a connected to ground, a dead short.

[Farrar]

Yes, but (as in the first version) your indicator light circuit has effectively tied the outputs of the two relays (and so the fans) together.

I remember reading on the DML years ago that there was a way of getting the light to come on when the fans were running. How was this accomplished?


...and you still have 87a connected to ground, a dead short.

That's what it looks like they did in the original circuit, so I just kept it that way:

http://dmctalk.org/attachment.php?attachmentid=106&d=1306264555

Farrar

[Ron]

I remember reading on the DML years ago that there was a way of getting the light to come on when the fans were running. How was this accomplished?
For "Fan Running", run each 87 through a diode>>bulb>>ground. (Actually "Fan has power")
For "Fan FAIL", run each 87a through a diode>>bulb>>ground (actually "Circuit Fail), and, switch the main feed, it should not be always hot anyway.



That's what it looks like they did in the original circuit, so I just kept it that way:
That's a different type of relay...notice it has two outputs (two 87s so to speak).

[Bitsyncmaster]

For "Fan Running", run each 87 through a diode>>bulb>>ground. (Actually "Fan has power")
For "Fan FAIL", run each 87a through a diode>>bulb>>ground (actually "Circuit Fail), and, switch the main feed, it should not be always hot anyway.

That's a different type of relay...notice it has two outputs (two 87s so to speak).

"Change over relay" has 5 pins. "The common changes over to the other output.

Ron is correct, you will short circuit the +12 volts. Pin 30 is the common pin which will need to go to the fans if you want to ground pin 87a.

[Farrar]

If I understand correctly, then I should do something like this?

4505

I really appreciate the help, y'all. I'll buy you a coffee at DCS '12 :)

Farrar

[content22207]

You can't go to DCS -- you don't have a fuel pump (at least not one that is actually mounted in the car).

Bill Robertson
#5939

[Farrar]

You can't go to DCS -- you don't have a fuel pump (at least not one that is actually mounted in the car).

I have more than one car.

Farrar

[content22207]

If you don't install that pump before next June, I am going to drive down there and put the damn thing in myself. It's smaller than a Coke can, for crying out loud. Surely there is *SOMEWHERE* in a DeLorean to mount the thing.

Bill Robertson
#5939

[Farrar]

If you don't install that pump before next June, I am going to drive down there and put the damn thing in myself. It's smaller than a Coke can, for crying out loud. Surely there is *SOMEWHERE* in a DeLorean to mount the thing.

I only have a few hours a week with the car at this point, but my schedule should improve within a few months.

In the meantime, I have noticed another error in my schematic. Attached is a corrected version.

Farrar

[Ron]

In the meantime, I have noticed another error in my schematic. Attached is a corrected version.
That looks correct. ...you realize that the light can come on even if only one fan is powered don't you.

[Farrar]

That looks correct. ...you realize that the light can come on even if only one fan is powered don't you.

Thank you! That's the best news I've read all day. :)

Yes, I know about the light -- I don't feel like adding another light somewhere, so I will simply use the "fan fail" indicator which is already there.

Farrar

[Ron]

Just wondering...Why don't you like a fan fail light? ...it seems like it's easier to notice a light that comes on than one that went out?

[Farrar]

Just wondering...Why don't you like a fan fail light?

Mine has never worked. A previous owner of my car removed the socket and all.

I think it's a great idea, but I have no idea how to implement such a thing.

Farrar

[Ron]

I think it's a great idea, but I have no idea how to implement such a thing.

For "Fan FAIL", run each 87a through a diode>>bulb>>ground (actually "Circuit Fail), and, switch the main feed, it should not be always hot anyway. .

EDIT: Sorry, that won't work...the light would lie if the ac and otterstat were both off.
You could do it similar to the original way...I don't know what was wrong unless it was that it, as all of the above, doesn't tell you if the fan is running, just that power is being sent to its input wire.

[DMCMW Dave]

Thank you! That's the best news I've read all day. :)

Yes, I know about the light -- I don't feel like adding another light somewhere, so I will simply use the "fan fail" indicator which is already there.

Farrar

I recall doing this to mine maybe 10 years ago and I built a one-transistor circuit so that both fans had to be powered to make the light go on. If either failed, the light would go off. Any of these circuits of course only tell you that power is being sent to the fans, they don't tell you if the fans are not actually turning.

[Farrar]

The only thing I can think of for a real "fan fail" indicator would be installing a normally-open reed relay between each fan and ground. However, I have never seen a reed relay rated at a high enough amperage to work for a radiator fan.

(Obviously, Dave M has found a solution to this problem, but only for stock circuit configuration.)

Edit: Come to think of it, aren't reed switches very small? Perhaps that is what the original fan fail module is: a reed relay with some resistors to drop the current across the switches. I'm only guessing, though -- I have never taken an electronics course or even read a book on the subject. (... a deficiency soon rectified by a trip to the library.)

Farrar

[Bitsyncmaster]

I recall doing this to mine maybe 10 years ago and I built a one-transistor circuit so that both fans had to be powered to make the light go on. If either failed, the light would go off. Any of these circuits of course only tell you that power is being sent to the fans, they don't tell you if the fans are not actually turning.

Or an open (bad connection) would not be detected.

[Bitsyncmaster]

The only thing I can think of for a real "fan fail" indicator would be installing a normally-open reed relay between each fan and ground. However, I have never seen a reed relay rated at a high enough amperage to work for a radiator fan.

(Obviously, Dave M has found a solution to this problem, but only for stock circuit configuration.)

Edit: Come to think of it, aren't reed switches very small? Perhaps that is what the original fan fail module is: a reed relay with some resistors to drop the current across the switches. I'm only guessing, though -- I have never taken an electronics course or even read a book on the subject. (... a deficiency soon rectified by a trip to the library.)

Farrar

I've never seen an original Fan Fail unit but I would guess it had a relay with two coils. Then if each coil (each fan current) was the same they would cancel and not turn the relay contacts on.

My new unit measures each fan current and software determines if both fans are working properly. So it would detect, an open fuse, an open connection or a high current fan (jammed). A short circuit would blow a fuse.

[Ron]

This is one way to tell you that AT LEAST one fan input wire is being fed (light off = good):
4512

This is one way to tell that BOTH fan input wires are being fed (light on = good):
4513

For homework- Combine the different ways of using the diodes-relays to do the latter but light on = BAD...or, I'll show you if you don't want to play :wink:

[Farrar]

For homework- Combine the different ways of using the diodes-relays to do the latter but light on = BAD...or, I'll show you if you don't want to play :wink:

Oh, boy homework -- and classes just started this week! :lol:

I will take a stab at it. Probably wrong, but what else is there to do on a hot Friday afternoon? ;)

4517

Farrar

[Ron]

Oh, boy homework -- and classes just started this week! :lol:

I will take a stab at it. Probably wrong, but what else is there to do on a hot Friday afternoon? ;)
You got!!!

You can do it this way too (save a couple of diodes).

Dave probably has enough jumk laying around to do it a lot easier w/o relays...

4576

[robvanderveer]

I've been following this thread for a while now. Forgive me for being stupid, but aren't you actually making a lamp to indicate the proper workings of a relay? Wouldn't that be similar to designing a bulb inside a light switch to see that the room is lit?

What we really need is a way to detect that the fans are actually spinning when either the otterstat or the A/C demands it. Or better still: that the fans are NOT spinning even though they are powered.

[Farrar]

Ron -

Okay! So let me make sure I understand WHY that works:

When the fan signal is being sent from the compressor or otterstat, the relay which turns ON the light has current. However, when the fans ARE on, the coil in that relay is energized, breaking the circuit and turning the light OFF. Correct?

Rob -

I'm the stupid one; remember, I'm new at this! :) Originally, I was planning on having the light come on when the fans were on and simply putting black tape over the word "fail." However, as you can see to implement a "fan fail" indicator as per stock function, even though I have both fans on separate relays and circuit breakers. In either case, you only have proof that the fans are being "told to come on," not proof that they are actually running. It would take someone much better educated than me to figure out how to have a light come on only when the fans have seized.

Farrar

[Ron]

Farrar,
You got it down dude!
To actually confirm it is spinning you could use a lot of things (e.g. led & detector, or counter, air flow detector, or ......

=====


I've been following this thread for a while now. Forgive me for being stupid, but aren't you actually making a lamp to indicate the proper workings of a relay? Wouldn't that be similar to designing a bulb inside a light switch to see that the room is lit?

What we really need is a way to detect that the fans are actually spinning when either the otterstat or the A/C demands it. Or better still: that the fans are NOT spinning even though they are powered.
LOL, yes.
...well, similar, but not the same. Here, the light switch is not in the same room...but yeah I totally see what you are saying.
If you think about it, knowing the fans are/not spinning is also similar.
We only need to know if the engine is getting too warm. If one likes to rely on lights (v/s gauges) or likes a backup (or extra kick when violating DUI laws or Sweet Suzy is being distractive, etc.) then I'd do it like this (and omit the buzzer):
http://dmctalk.org/showthread.php?1247-Dual-Fan-Relay-Mod-w-opt-buzzer

[Farrar]

Here is what confuses me:

According to the wiring diagram in the Workshop Manual, the otter stat's feed is live at all times. (Through Fuse #5.)

According to the wiring diagram posted here on DMCTalk, the otter stat's feed is switched. (From Main Relay.)

It looks like the labels of "Feed Live at All Times" and "Feed Main Relay Live When Ignition Is On" got switched somewhere along the way.

Which one is correct -- Workshop Manual or DMCTalk?

Farrar

[Bitsyncmaster]

Here is what confuses me:

According to the wiring diagram in the Workshop Manual, the otter stat's feed is live at all times. (Through Fuse #5.)

According to the wiring diagram posted here on DMCTalk, the otter stat's feed is switched. (From Main Relay.)

It looks like the labels of "Feed Live at All Times" and "Feed Main Relay Live When Ignition Is On" got switched somewhere along the way.

Which one is correct -- Workshop Manual or DMCTalk?

Farrar

The fan relay is live all the time. The otterstat is switched power. If the otterstat was always live, when you shut the engine off the fans would continue to run until the otterstat cooled down.

[Farrar]

The fan relay is live all the time. The otterstat is switched power. If the otterstat was always live, when you shut the engine off the fans would continue to run until the otterstat cooled down.

OK, that's what I thought. Thanks, Dave!

Public Service Announcement: Houston's re-printed Workshop Manual reproduces the original printing error.

Farrar

[content22207]

What we really need is a way to detect that the fans are actually spinning when either the otterstat or the A/C demands it. Or better still: that the fans are NOT spinning even though they are powered.

If you have OEM fans, you know when they're spinning -- they shake the whole car.

Bill Robertson
#5939

[Farrar]

Farrar,
You got it down dude!

Whew! Big sigh of relief! Now I need a beer.

This was a lot harder than adding a relay/off-delay to the power windows! :lol:

Farrar


If you have OEM fans, you know when they're spinning -- they shake the whole car.

I plan on replacing mine with the lower-power-consumption ones I bought a little while ago; they're not nearly as loud.

Farrar

[Ron]

If you have OEM fans, you know when they're spinning -- they shake the whole car.

Bill Robertson
#5939That's why I asked him in the other thread 'how much the replacement fans, and acoustic boards cost'.

[Farrar]

That's why I asked him in the other thread 'how much the replacement fans, and acoustic boards cost'.

My mistake -- I thought you were asking the original poster in that thread, not me.

The fans cost me $35 each plus shipping. What's an "acoustic board"?

Farrar

[Ron]

My mistake -- I thought you were asking the original poster in that thread, not me.

The fans cost me $35 each plus shipping. What's an "acoustic board"?

FarrarYou know, the boards (panels) made of sound absorbent material people cover the cabin-side of the firewall, floor, etc with. ...like Dynamat but more rigid.

[DMCMW Dave]

If you have OEM fans, you know when they're spinning -- they shake the whole car.

Bill Robertson
#5939

Only if one of the blades is missing.

[Farrar]

You know, the boards (panels) made of sound absorbent material people cover the cabin-side of the firewall, floor, etc with. ...like Dynamat but more rigid.

I haven't invested in any such material yet.

Farrar

Revised schematic:

4684

Farrar