Alternative to Final Stage/Fan Controller For Engine Cooling.
#31
Pin 12 will give you 10 % power on its own and was originally intended for stability but is rarely needed, pin 13 is actually an I/O line, it will read the potentiometer setting, but will also adjust the temperature, effectively turn the controller off when grounded, and over ride the temperature on when 12V is applied. Pins 12 and 14 combined will give you near 100% when combined. The two pages not included with the controller are included with the instructions from the site.
If the wires into the original output stage provide 12V to turn the fan on, it should include all of the functions needed including the torque converter temp. That would avoid the extra wire run also. If they ground to turn the fan on, one of the temperature over ride pins, with a series 10 k ohm resistor would also work.
It will probably be fine here, but you generally want to avoid pins 85 and 86 on a relay, particularly when its triggered with a switch. The Tyco app note claims a worst case 1,000V spike without any suppression. It will be less from arcing in the switch but still pretty high. The compressor coil will usually have a suppression diode. As mentioned, the impedance, which is about 10k ohms on those pins sets the current. Pin 87 would be better.
If the wires into the original output stage provide 12V to turn the fan on, it should include all of the functions needed including the torque converter temp. That would avoid the extra wire run also. If they ground to turn the fan on, one of the temperature over ride pins, with a series 10 k ohm resistor would also work.
It will probably be fine here, but you generally want to avoid pins 85 and 86 on a relay, particularly when its triggered with a switch. The Tyco app note claims a worst case 1,000V spike without any suppression. It will be less from arcing in the switch but still pretty high. The compressor coil will usually have a suppression diode. As mentioned, the impedance, which is about 10k ohms on those pins sets the current. Pin 87 would be better.
#32
Electron Wrangler
Lifetime Rennlist
Member
Lifetime Rennlist
Member
It will probably be fine here, but you generally want to avoid pins 85 and 86 on a relay, particularly when its triggered with a switch. The Tyco app note claims a worst case 1,000V spike without any suppression. It will be less from arcing in the switch but still pretty high. The compressor coil will usually have a suppression diode. As mentioned, the impedance, which is about 10k ohms on those pins sets the current. Pin 87 would be better.
Alan
#34
Electron Wrangler
Lifetime Rennlist
Member
Lifetime Rennlist
Member
As noted they are all individually switched to ground for activation at the controller - the unit has internal pull-ups on all these.
The fan final stage has 2 individual (FET) fan output drivers - they are designed to always do the same thing: both off, both half-speed or both full-speed. There is a feedback loop from the final stage to the controller, if either fan fails the other will run at full speed whenever the fans need to run at all.
The final stage just gets a PWM signal for the fan speed from the controller, there are no digital signals to the final stage
The intake switch normally causes 1/2 speed operation, the TC temp and refrigerant pressure switches cause full-speed operation. Having all do full speed is acceptable I think.
It would be nice to disable the fans during starting - do you provide for that? We have an accessory bus (X-bus) that is active in accessory/ignition but not during starting. The clutch activation is based on this - so that will help for most modes.
Alan
The fan final stage has 2 individual (FET) fan output drivers - they are designed to always do the same thing: both off, both half-speed or both full-speed. There is a feedback loop from the final stage to the controller, if either fan fails the other will run at full speed whenever the fans need to run at all.
The final stage just gets a PWM signal for the fan speed from the controller, there are no digital signals to the final stage
The intake switch normally causes 1/2 speed operation, the TC temp and refrigerant pressure switches cause full-speed operation. Having all do full speed is acceptable I think.
It would be nice to disable the fans during starting - do you provide for that? We have an accessory bus (X-bus) that is active in accessory/ignition but not during starting. The clutch activation is based on this - so that will help for most modes.
Alan
#35
Nordschleife Master
Thread Starter
You wizards talking about using this as the feed to the relay for the three sensor fan activation ? Yes I hope, otherwise your are losing me.
#36
The inputs would have to have more function than ground and 12V in order to get 50% output, it would either have to be pwm, frequency modulation, or a variable dc voltage, not enough inputs for a binary word (assuming two inputs). I'm referring to the inputs directly into the controller
#37
Found a rough schematic,I'm referring to the pwm and sense inputs to the final stage. Are either one of those 12V under the conditions when any of the three are true?: the AC is on or the trans temp or the intake temp is high
#40
I think it's already good, I think its just the details in finding the easiest way to do it and getting the closest to the part of the original function that you want to keep. If the driver stage remains functional without being connected to the main amplifier, that would be the cleanest and give the closest to the original function. If not, a P-type FET could replace the relay now used for the three functions and would take up very little space and, I think already mentioned, one of the original signal wires to the main amplifier could be used in lieu of routing a wire from the relay
#41
Electron Wrangler
Lifetime Rennlist
Member
Lifetime Rennlist
Member
The inputs would have to have more function than ground and 12V in order to get 50% output, it would either have to be pwm, frequency modulation, or a variable dc voltage, not enough inputs for a binary word (assuming two inputs). I'm referring to the inputs directly into the controller
The feed to the final stage is PWM from the controller - all the smarts are in the controller. The final stage is just the drivers and the SC/OC feedback.
I don't see any value in trying to use the stock controller in any way since only by interpreting its PWM output can you determine what its trying to do.
Alan
#42
That's pretty much what I'm saying, I would guess the original driver already intelligently processes those signals, along with several others, like freon pressure. The PWM logic is most likely very simple, probably 0 V for each fan off, 12V for full on, and some percent duty cycle for 50%. There would most likely be little or no extra circuitry, just the effort to either measure or research the outputs, and then understand them. The only possible issue would be if the information on line is correct in that it also has feedback from the output stage (most likely fan current) and if not having that feedback will cause the driver to not operate correctly.
Whether or not it would be worth it depends on whether or not there is any real processing, other than temperature in the driver.If there is no processing, it wouldn't add much value.
Whether or not it would be worth it depends on whether or not there is any real processing, other than temperature in the driver.If there is no processing, it wouldn't add much value.
Last edited by brian baskin; 08-19-2014 at 10:55 PM.
#43
Nordschleife Master
Thread Starter
Brian - the controller has an AC active input and inputs from the intake temp switch, torque converter temp switch, high refrigerant pressure switch as we have disussed. It also has the coolant (radiator) temp sensor (we didn't discuss this before because your system fully replaces this) - this temp sensor is analog and controls the fan speed transitions - all the other inputs are simply switched.
The feed to the final stage is PWM from the controller - all the smarts are in the controller. The final stage is just the drivers and the SC/OC feedback.
I don't see any value in trying to use the stock controller in any way since only by interpreting its PWM output can you determine what its trying to do.
Alan
The feed to the final stage is PWM from the controller - all the smarts are in the controller. The final stage is just the drivers and the SC/OC feedback.
I don't see any value in trying to use the stock controller in any way since only by interpreting its PWM output can you determine what its trying to do.
Alan
Many a discussion about an alternative to the final stage failures. The dc unit looks good for it as a stand alone temp controller. Variable speed fan function relative to radiator temp. One switched function to 50% on AC activation. A bonafide tie in for 928 atf, intake, refrig temps which are simply switched just like the original though not to full speed fans. I think an offset to the lack of full speed when these sensors switch is the fact that the fans are always running at least at the lowest speed - this give some degree of cooling to atf cooler - also how often to these three trigger ? - not often. I do note that 50% dc fans when I press the AC button seems like more fan than I got with the old unit when the AC was turned on.
Now back to the forest.
#44
Drifting
A bonafide tie in for 928 atf, intake, refrig temps which are simply switched just like the original though not to full speed fans. I think an offset to the lack of full speed when these sensors switch is the fact that the fans are always running at least at the lowest speed - this give some degree of cooling to atf cooler - also how often to these three trigger ? - not often.
#45
Nordschleife Master
Thread Starter
Actually, as the 3 928 temp sensors are simply switched (no variable speed) on the original and the dc unit, I might just tap the sensors in to the 'standard' AC switch in addition to the optional input I have them at now. When triggered in extreme conditions, will throw the fans even higher then the way I have them currently arranged.