All2kool

OK, a few months ago I posted about how to keep the AC Compressor from coming on with the Defroster. Several people pointed out how the AC Compressor acts as a Dehumidifier in conjunction with the Defroster. My AC Compressor is a little noisy and I didn't want it coming on with the Defroster. Solution? I unplugged the AC Compressor. My Defroster works just fine without it (I have always used the Defroster as the Heater as I don't like hot air blowing in my face or at my feet anyway). By disconnecting the AC Compressor I have created a Manual Fan Switch - by just turning the HVAC Control to Max AC. My car never overheats but it is comforting to know I can turn on the Fan whenever I choose too.

tequilaboy

In the 86 bins that I've see, the fan on and off temps are the same whether A/C is on or off. (107.75 C on and 104.75 C off)

The A/C high pressure signal is also needed to make use of the A/C on fan settings. This would not be the case with the compressor disconnected unless the switch was shorted.

When I've played with different A/C on vs. A/C off fan settings, I've only seen a difference when the compressor is running and A/C pressure is high.

I don't understand how simply disconnecting the compressor and turning on the A/C via the switch would have any impact on the fans, unless the programming was also modified to support this.

All2kool

OK, I'm a bonehead. The fans does NOT come on as I described, my theory was wrong. Please refrain from further post's on this thread.

GIMMESOME

I tried this on my 86,and my fans did not come on,are they supose to?
Is there something wrong? I thought you could only force the fans on with the aldl jumper,or a manual switch.Is mine broken?

GIMMESOME

OK thanks,I was starting to worry there was something wrong with mine.

SunCr

Though the '86 and '93 are different, neither system works on temperature.

For the '86, a 5 volt reference is delivered to a normally closed switch mounted on the high pressure line. It's designed to open at 220 psi and when it does, the voltage sensed by the ECM rises to the reference and the ECM grounds the main fan relay. Design keeps the switch open until the pressure falls to about 190 psi. It then recloses and the ECM cuts ground to the Relay. Granted, pressure is temperature, so if you want to look at it that way, 190 to 220 psi equals a temperature of about 130 to about 150 degrees (F) at the condensor outlet. That's nowhere near the temperature posted about the programming (which isn't how it works anyway - 0 volts sensed by the ECM - don't ground the Rely - 5 volts ground it - simple enough?). You can get the main fan running by simply disconnecting the harness at the switch, but if you do this, it won't turn off. The compressor doesn't have to be running as the ECM doesn't control compressor engagement nor does the programming even consider the a/c request signal (it's used for idle control only). For programming purposes, it would be extremely complicated to change how it works The better (if not the only) way to effect a change is to obtain a switch with a different calibration (I don't think it exists, but if you've swapped over to an R134 charge, it might be worth the effort to try and find one as that charge does better with a little less pressure and you achieve that by turning on the fan at a much lower pressure).

The '93 is a bit more sophisticated as the ECM controls compressor engagement, but I won't venture into it.

tequilaboy

Fan control always seems to be a confusing topic.

It is further complicated by speed.

Below 11 mph, the A/C on Fan on/off settings are always used.

Below 35 mph with A/C pressure high, a constant duty cycle is employed, typically 100%.

Otherwise, above 11 mph, separate A/C off Fan on/off settings are available.

There is also a coolant temperature dependent duty cycle available, although it is also typically set to 100% across the board.

Here are some of the fan related parameters within the ecm bin file (taken from ABTR as an example):

Fan Off Coolant Temp Threshold (A/C On) 104.75 Deg C
Fan On Coolant Temp Threshold (A/C On) 107.75 Deg C
Fan Off Coolant Temp Threshold (A/C Off) 104.75 Deg C
Fan On Coolant Temp Threshold (A/C Off) 107.75 Deg C

Fan PWM Duty Cycle Vs. Coolant Temp.

Deg. C % Duty Cycle
80 99.6
92 99.6
104 99.6
116 99.6
128 99.6
140 99.6
152 99.6

Other relevent parameters include the 11 mph and 35 mph speed thresholds and the constant duty cycle already mentioned.

I don't know if these relays can actually handle a low duty cycle without overheating and sticking as I haven't tried it.

SunCr

If what you're describing (a/c only) is available through '89, I've never seen it - it's simply an on/off signal from the switch and only the main fan is ECM driven. There's no duty cycle (other than what's built into the switch). Nor is there any high or low speed for fan control regardless of the signal - a/c switch or CTS. Speed - assuming a functioning system, will keep the pressures low enough so that switch contacts remain closed (and if it doesn't keep your fingers crossed that it pops the high pressure switch to shut it down). '90 and above are a different story as the ECM took over compressor control (along with the fans) and the '96 adds another wrinkle with dual speed fans. The design of any a/c system is to maintain operating pressures within a desired range and if it gets too low or too high - cut it off (and up until '89 that's accomplished with switches that are connected to the compressor circuitry). Before '90 it wasn't terribly sophisticated as those switches communicated absoluely nothing to the ECM. After that, an all in one pressure sensor relays a great deal more for compressor and fan control and the low pressure switch is monitored for system charge.

CentralCoaster

You can have the main fan run all the time by disconnecting the high pressure switch on the refrigerant line near cylinder #6

Damien89

When i turn the air con on my 89 the fan at the radiator isn't coming on. What could be the problem? I already bought those two switches that centralcoaster mentioned but i haven't installed them yet because somebody told me that one of them could be the culprit.
Any ideas?

tequilaboy

What is the coolant temperature when you turn the A/C on? Or is the main fan never coming on?

The 105 C fan on temp is 221 F unless modified.

Damien89

The main fan works normally ecxept that when the ac is switched on the main fan doesn't come on like it used to do.

tequilaboy

Thinking a little more on the fan subject, I thought of a method to turn the fans off at speed which does not occur by default.

Since the A/C on fan on/off settings are used below 11 mph:

If the A/C off fan on and off temps are raised above the A/C on fan on temp, this should have the effect of turning the fan off above 11 mph when the is A/C off.

I think this would be desireable at least to reduce the alternator load at speed, and free up some fractional HP that the fan motor consumes ( a few hundred watts anyway). It may make sense to raise the 11 mph threshold to ensure adequate cooling airflow at low speeds.

I'll test the idea when I get a chance and report back.

CentralCoaster

If I was starting from scratch, and being as efficient as possible,


I'd have these settings...



Temp1 small fan
Temp2 large fan
Temp3 both fans
Speed over X mph = all fans off.

AC_on small fan

SunCr

TequilaBoy - you try that and let us know - I might recommend that you make sure you've got some gages plastered to the windshield just to make sure the pressures are within performance specs. For the rest of us with something built below '90, the ECM isn't grounding the main relay based on operating temperature (a/c operation). If you want to change it from the present design, you'll need a different switch or you're going to have to re-engineer the ECM. You can drive around all day with the fan running by simply disconnecting the a/c fan switch harness and it won't turn off regardless of speed or engine temp. It's made that way because pressures can become extreme at cruise - especially when it's plugged up. Before it blows up, it should shutdown, but along the way, the radiator is going to get a bit toasty. The guys who designed the thing were told to keep it as safe as possible and keep the warranty claims to the a/c system - not the whole friggin' engine.

If the main fan doesn't come on with a/c, the pressure is either too low to close the switch contacts - it needs to be 220 psi - or the switch is bad. Since the ECM grounds the main relay when it sees voltage on this circuit, you test the ECM by disconnecting the harness at the switch. If the fan doesn't come on, the wire is shorted to ground or the ECM is bad. Assuming it comes on, the system didn't reach 220 psi because 1. It's cold out - below 50 degrees or 2. there's no gas in it. Check it out further with a Manifold Gage Set or go to an a/c shop where they know how it works - amazingly, a lot do - they actually make you take a test to work on these things.

tequilaboy

We've already established that A/C pressure high signal and speed less than 35 mph sets a 100% duty cycle for the main fan instead of the coolant temp based duty cycle table (which is also 100% but could be reduced if desired).

It is not completely clear that the A/C on fan on/off coolant temp settings are used as an "or" condition or as an "and" condition for main fan activation, but from what has been posted I can buy that the high pressure signal is a reasonable condition for activating the main fan as an "or".

I have no reason to doubt that it works as described, however that is only part of the total fan control and it is irrelevant in this context.

The changes that I plan on testing (to disable the fan at speed) with the A/C off, will only have an effect when the A/C is off, so there is no risk of overpressurization to the A/C system and no serious re-design required.

The A/C on fan operation will be unchanged except for lower than factory temp settings, roughy 200 F on and 180 F off that I currently run.

The A/C off fan settings will simply be raised to the factory setting to enable a window between 180 F and 226 F where the fan will be disabled when A/C is off and the vehicle speed is above 11 mph.

SunCr

Believe me - the signal, an open at the pressure switch is the only signal for a/c main fan operation (below '90). If you want to eliminate the compressor and ancillary controls from the accessory drive train, find a totalled Prius and hijack it's a/c system, though it might tax the stock alternator so a 16 volt alternator in place of the compressor, might be needed or could be a better way to make it work. From what was originally posted, this isn't/wasn't a non-a/c topic. If it was, I wouldn't have responded.

tequilaboy

From a simplified logical perspective,
the fan control within the ecm for 86-89 ($32 and $32B) looks like this:

if < 35 mph and A/C high pressure (switch open)
Fan duty cycle = 100%

else if > 35 mph and A/C high pressure
Fan duty cyle = value from coolant temp based table (also 100%)

else if < 11 mph
{
Compare coolant temperature to A/C on fan on/off threshold
If coolant temp > threshold
Fan duty cycle = value from coolant temp based table (also 100%)
}

else if A/C off (and > 11 mph)
{
Compare coolant temp to A/C off fan on/off threshold
If coolant temp > threshold
Fan duty cycle = value from coolant temp based table (also 100%)
}

Note: this description is not complete, as there is also a 15 second minimum on time for the fan that is not considered here. Fan off details are also not included in the example above.

To summarize:

For speeds below 11 mph, or with A/C on without A/C high pressure (switch closed), the fan is controlled by the A/C on fan on/off temperature thresholds and duty cycle table.

For speeds greater than 11 mph and A/C off, the fan is instead controlled by the A/C off fan on/off temperature thresholds and duty cycle table.

The change in threshold usage that occurs at 11 mph for the A/C off state is what results in most of the confusion regarding the useage of these thresholds when tuning the fan on/off temperatures for both states. For this reason, it is common for both A/C on and A/C off thresholds to have the same settings.

Updated: Here's what the real eprom disassembly looks like from $32B

;*************************************** ******
; >>> Cooling Fan Tables & Params <<<
;
; COOL CAL VAL'S = = (deg c + 40) * (256/192)
;*************************************** ******
DF82: LDF82 LDAA L0065 ; Filtered MPH
DF84: BRSET L0037,#$40,LDF94
; ... else
DF88: CMPA LC2F6 ; 35 MPH, Fan D.C.=LC2F9 if
MPH LT THRESH & A/C Press Hi
DF8B: BCC LDF94 ; BR IF MPH GT THRESH
; ... else
DF8D: LDAA LC2F9 ; 99.6% FAN DC
DF90: BEQ LDFC0 ; BR IF Z
; ... else
DF92: BRA LDFB9
;
DF94: LDF94 LDX #$C2FC ; A/C OFF FAN PAIR (ON/OFF)
DF97: BRCLR L0037,#$80,LDFA0 ; BR IF b7,
; ... else
DF9B: CMPA LC2F7 ; 11 MPH, If LT THRESH MPH
USE AC ON CALS
DF9E: BHI LDFA3 ; BR IF Vss GT THRESH
; ... else
DFA0: LDFA0 LDX #$C2FA ; A/C ON FAN PAIR (ON/OFF)
DFA3: LDFA3 BRCLR L00F4,#$FF,LDFA8 ; BR IF NOT $FF
; ... else
DFA7: INX ; INCR INDEX TO FAN
OFF VALUE
DFA8: LDFA8 LDAA L005B ; COOLANT TEMP
DFAA: CMPA 0,X ; GET FAN ON/OFF THRESH
Page 153
ABTC_HAC.SRC
DFAC: BCS LDFC0 ; BR IF COOL LT THRESH
; ... else
;---------------------------------------------
; Fan Duty Cycle
; D.C. vs Coolant Temp
; Tbl Val = %D.C. * (256/100)
;---------------------------------------------
DFAE: LDAB #$A0
DFB0: LDX #$C2FE
DFB3: JSR LF337
DFB6: TSTA
DFB7: BEQ LDFC0
DFB9: LDFB9 LDAB LC2F8 ; 15 Sec, Min Fan On Time,
(sec * 5)
DFBC: STAB L00F5
DFBE: BRA LDFCD
DFC0: LDFC0 BRCLR L00F5,#$FF,LDFC9
DFC4: DEC L00F5
DFC7: BRA LDFCF
DFC9: LDFC9 CLRA
DFCA: BCLR L0034,#$08
DFCD: LDFCD STAA L00F4
DFCF: LDFCF RTS

SunCr

As far as I know - doesn't exist. Who gave you this or this your interpertation of the code? Switch open - 5 volt rise at the ECM, main fan is commanded. Switch closed - 0 volts at the ECM, main fan off. There isn't any so called duty cycle. Drive around and you'll see that (though it's really tough on the hood so it's better on the dyno).

tequilaboy

I can burn a chip with reduced duty cycles and I can send you a copy of the scan data logs if you wish to see them.

I was planning on doing this anyway just for the hell of it.

With factory settings, both the fixed duty cycle and duty cycle table are set to 100%, but these are calibratable values and can be reduced. The duty cycle is also present the the ALDL datastream so it can be observed and recorded.

I'm a little concerned with burning up the contacts in the relay, but I think it will be ok for a short duration test. I can probably get my hands on some solid state relays if I choose to go this route on a more permanent basis.