Can someone explain the fan states?

[davepl]

I'm trying to re-enable discrete control of 2 fans in the eRod LS3 ecm. I've done the hardware side of splitting out the fan control wires (which ship hooked up to the same relay). I know there are a couple of other people doing the same so I imagine any responses will help them as well.

I do not currently understand the "Fan State Transition % vs Current State".

Working backwards, I start with my ECT. I look that up in "Desired % vs ECT", which is 0/100 depending if you're under/over 206F.

I have seven "States", and my confusion starts because I don't know what a state is, why there are seven of them, when a transition occurs.

After that I think I understand the "Output Control vs State".

So, could someone give us a little brain dump on how the states work? I did search for "fan state" but didn't find much - but if it's been done before, by all means just point me at the thread.

Thanks!
Dave

PS: Perhaps the default non-useful values in my ECM table are confusing me - if there weren't all 100, I could see sort of how it'd work. So I'm guessing these values need to change to "define" the states? Well, I guess you can see I'm over my head and need to source a little assistance :-)

[FunkedOut]

This table here lets you specify what value of desired fan % triggers the listed state change.
There's another table that lists how each stage behaves.

Set stage 0 to have both fans off.
Set stage 1 to have fan 1 on and fan 2 off.
Set stages 2-7 to have both fans on.

Then you can set the 0>1 trigger to 30% and the 1>2 trigger to 60%.
Then set the 2>1 trigger to 50% and the 1>0 trigger to 20%.
Leave the rest at 100%.

Now on your desired vs ECT table, set the ECT you want to kick one fan on to 30%.
Set the ECT you want the second fan to come on at to 60%.
That defines how it will work as the ECT rises.

As the ECT is dropping, pick a temp you want the second fan to turn off (something colder than when it turned on) and set that to 50%.
Pick the temp for cutting off the last fan, set it to 20%.

Fill in everything between the 50% and the 60% with 55%.
Fill in everything between the 20% and the 30% with 25%.
Fill in everything above the 60% with 99%.
Fill in everything below the 20% with 1%.

Clear as mud?

[davepl]

Almost clear... but what are those percentages a percentage OF? Maximum cooling capacity? Or are they just a convention that maps a desired % to a fan state?

I do need to make sure that it's not actually doing PWM on the outputs, because I'm using relays, and you don't want those cycling 128 time a second... if they even could.

I'll follow your info and set up my tables and maybe post them for review... that way I have to understand it well enough to do the work, and I can get it checked by someone who already knows what they're doing!

[gmtech16450yz]

http://www.hptuners.com/forum/showth...ight=fan+logic

[davepl]

Thanks. Based on the foregoing, here's what I think I know.

I set the fans to discrete, since I'm relay based on not solid-state (PWM would just buzz my relays). I set it to 2 fans.

I then decided I have 3 states:

[State 0] - No fans
[State 1] - Fans on low
[State 2] - Fans on high (my relays take care of the high/low speed part based on the two trigger wires)

I then defined those states in the "Output Control vs State":

[State 0] - 0 0
[State 1] - 0 1
[State 2] - 1 1

My tables are below; my biggest concern is the "Fan Desired" transition from one state to another, I'm not sure I grok that entirely. I read it as:

"Fans go from State 0 to State 1 at 50% demand" which happens at 203F.
"Fans go from State 1 to State 2 at 100% demand" which happens at 210F.

"Fans go from State 2 to State 1 at 50% demand"
"Fans go from State 1 to State 0 at 0% demand"

I'm not sure about this last part, since it's not clear to me how you transition from one state to another.

Does this look right, or am I "off by one" or similar on the transitions? Or am I totally lost? Or am I close?

[FunkedOut]

I never tried it with the same values for different state changes.
Also, I think you're going to have the values in the state change table 49 and 99 so that th values in the ECT vs desired are larger than and smaller than for the corresponding state changes; not certain though.

I think you got the idea now, just minor details.
Try it and see.

[gmtech16450yz]

http://www.hptuners.com/forum/showth...rol-discovery!