'165 switching both fans from the ECM?

[Peter Gargano]

Kevin Crain wrote:
> 
> And to conclude this project...after spending a few hours in the corner w/
> CSH in place...I accepted the conclusion that the '165 ECM can only manage
> one fan (relay)

If you're not into writing code, then see the last two paragraphs.

If you're prepared to write some code (as opposed to modifying existing 
tables), the '165 can be made to a lot of things that were not originally 
planned. The '165, (and variants that have both quad drivers) have 9 driver 
outputs:

  A1              from active high driver U13 (fuel pump driver)
  A3, D1, A7, C2  from active low quad driver U3
  A4, A5, A2, C1  from active low quad driver U6 

C1 is nominally the FAN output. Any other unused output(s) could be 
assigned to a second fan.

David C. Allen wrote:

> you only have the one A/D Ch 4 (input Pin C10) that is doing the resistance
> to temperature conversion in the ECU.

Note also that the FMD (U12) supplies the Hi/Lo resistor switching to 
provide better CTS resolution.

> So it looks like you would have to write the routine that looked at  A/D Ch
> 4 and then controls output ECM Pin C11

(?) C11 is nominally a MAP input (for speed density systems). The existing
routine to read the CTS (A/D 4) already provides a temperature value.

> You could also do a 2 speed single fan with low speed/(on off) controlled
> via output ECM Pin C1, and high fan controlled by ECM Pin C11

I'm not sure what you're saying here - C11 is an input.

> Maybe a better solution would be to Pulse Width Modulate output ECM Pin C1
> and drive the fan via a control circuit instead of using a relay. Now you
> have a continuously variable cooling fan that runs at a speed based on
> engine temp or A/C pressure.

This should be possible, but I'm not convinced it would achieve anything
too dramatic that the existing system (comprising thermostat, coolant, 
etc.) doesn't already handle.

But yes, a spare quad driver could be used to control a fan relay. And just
to complicate it further, a spare A/D channel (out of the 10 coming from 
input pins) could be used to dynamically set the mid-point hysteresis (trip)
temperature.

Bottom line is, unless you've got the time to write and test this stuff,
a simple external circuit connected to C1 and either a switch or a 
temperature sensor and some electronics, is probably going to be the easiest
way...

Kevin Crain wrote:
> 
> I found the following mfg/part# info on switches that close at 212degF and
> open at ~190degF:
> [snip]

Okay, forget the temp sensor and electronics - just use a preset 
temperature switch! But I'd suggest that a sensor connected to an LM555 
connected to a relay, is the cheapest and most versatile solution, as the 
trip temperature can be controlled by a variable resistor hooked to the 
555. You're not stuck with an (expensive) fixed temperature switch.

Peter
----------------------------------------------------------------------------
To unsubscribe from gmecm, send "unsubscribe gmecm" (without the quotes)
in the body of a message (not the subject) to majordomo at lists.diy-efi.org