Beginning of an Automatic Climate Control System, all juiced up on Fusion

**h3rk** · 11-12-2007, 08:49 PM

Ok here's where I am at with automatic AC controls with the brain. I have a pretty good idea of how I'm going to take control of the blower with the brain, All of the dampers are pretty much no brainers (I have vacuum actuators for all of my mode doors, but my temperature door will be a servo-amp type like many other vehicles).

I haven't gone too far into where I'm going to put the sensors yet, as this is a conversion from a truck with manual AC controls, to some super-sweet automatic AC controls; so I'm gonna need some new sensors.

From the factory, my vehicle had a knob on the control unit that ported vacuum to each actuator (through some monsterous maze of ports inside). I found a mainfold that contains little 12V solenoids

Here's how it incorporates into the existing vacuum circuit. This is going to be an easy swap in for me. And it will work for anyone with a number of vacuum solenoids they need to operate with the brain.

The only door left to drive needs a servo-driver, it's much like the ones I've found in the skimmies for tons of other cars so, this may help you if your car only has electric servos as well.

There is more going on in the block than they're showing, for instance; the two leads going to the motor are not in and out, they are 12V and control, on this servo, control needs 2.5V to hold position, 0V closes, and 5V opens.
The return path (not really shown) is the top terminal. 5V reference is needed for the third terminal down, and the 2nd changes (position pot) as the servo moves, in the OEM automatic setup 0-5V is turned into 0-255 counts and used to compare with the desired position (I'll get into controls and alogrithms soon, just not right away)

I think I'd be safe using an analog out if greenman100 can make the values instantaneous like he wants to, there's virtually no draw on the control terminal. I have been playing around with 2 relays and 2 resistors as an alternative way of coming up with 0-2.5-5 volts as well.

that's pretty much it for the mechanics of temp control...Oh here's the servo I'm using.

Now...Blower control...ugg.
I'm not going to go into how much I hate most of my alternatives. I want variable speed, and I'm not settling for 4 speeds or ten relays, Honda uses power transistors, hell even the version of my truck with auto control uses PWM, the big problem with that is it uses pwm to drive the motor AND a PWM control signal drives the driver, and I don't think we're quite up to PWM outputs yet so... I think I'm going with the power transistor style setup used by Honda, Jag, Audi, etc. What I've found appealing about those is they seem fit for using a low-power, analog control signal, and I get a feedback signal as well.

Here's some rough stuff that I've thrown together so far on this method.

That's pretty much all of the outputs and a couple inputs. I hope some of the more engineering savvy will help guide me with my circuits, keeping me from making too many rookie mistakes (the tech that I am).

I will make an initial placement decision on the temperature, humidity, and sunload sensors soon and post them as well.

A note on my sunload sensor, I need to see if the one already in my truck for my headlights is the one they use for the automatic AC system as well.

As for the temperature sensors, I'm sure that I am going to be poking at least one little hole in my airbox, ones going in the grill area, and one is going in the people-tank.

More to come soon...

**Uraijit** · 11-13-2007, 09:03 AM

Looks interesting. But that particular transistor will probably be difficult to find. Generally blowers are replaced as a complete unit, and so electronic parts are not sold individually.

Can you just get a radio-shack transistor or the proper size? I don't really know enough about electronics to know what to look for, so I can't really help you there.

Your other option would be to get one off a blower at a salvage yard.

**thekl0wn** · 11-13-2007, 09:55 AM

Awesome project! I'll be watching this one closely, as I plan on doing the same thing later on!

**inh** · 11-13-2007, 11:28 AM

Plan on doing the same thing in my CRX

Lucky for me, all my switches for vent selection are electronically controlled, so a small relay bank is all I need.. Temp mixture is cable controlled though. I think fan speed might, but in my head that doesnt make sense, and all my diagrams are on my now dead laptop hard drive.. I do know there are two cables for the climate control though..

For the cables, I plan on hooking them up to one or two servos, and driving the servos either with a serial port servo driver, or using the brain, if they ever get that servo output thing done =]

**h3rk** · 11-13-2007, 01:05 PM

PM me with your year/make/model and I can help you out with your documentation. ANd answer that 2 cable question.

**Enforcer** · 11-13-2007, 01:42 PM

subliminal

**inh** · 11-13-2007, 01:54 PM

Already subsonic =p

**h3rk** · 11-13-2007, 11:14 PM

subliminal...ha, just a typo.

Well, I've listed and categorized my I/O:

Digital Outputs:

Mode 1 - (12V) direct connection to small vacuum solenoid: Panel/Foot Mode master damper Open signal (works as a pair with Mode 2)

Mode 2 - (12V) direct connection to small vacuum solenoid: Panel/Foot Mode master damper Closed signal (works as a pair with Mode 1)

Mode 3 - (12V) direct connection to small vacuum solenoid: Heater Mode damper signal (works as a pair with Mode 4)

Mode 4 - (12V) direct connection to small vacuum solenoid: Defrost Mode damper signal (works as a pair with Mode 3)

Mode 5 - (12V) direct connection to small vacuum solenoid: Recirculation Mode damper signal

AC Comp Req - (12V) passes a '1' "request compressor on" signal through refrig. high pressure sw. to the powertrain control module (may not need relay)

Fan Enable Relay - (12V) relay turns on to connect fan's high side to fused, IGN 12V

Fan Max Speed Relay - (12V) Picks up to parallel over fans low side driver circuit, for max fan speed.

Analog Inputs:

Outside Temp - (0-12V) or (0-5V?) new sensor in grill.

Inside Temp - (0-12V) or (0-5V?) I may go with an asiprated sensor here, new sensor.

Outlet Temp (upper) - (0-12V) or (0-5V?) New sensor, upper center panel vent.

Outlet Temp (lower) - (0-12V) or (0-5V?) New sensor, lower heater vent.

Sunload Sensor -(0-12V) New sensor, my book says there's a seperate one for HVAC.

Fan Speed Feedback - (0-12V) voltage sensed across driving transistor 0V=max speed.

Temp Door Feedback - (0-5V) There seems to be a problem with this, I'll show the circuit below, after I'm done with this, I was hoping to treat the feedback circuit as a normal 3 wire sensor.

Analog Out (crosses fingers):

Temp Door Drive - (0V,2.5V,5V) 0-close,2.5-open, 5-open, because analog out will take 2 outputs it's probably more cost effective to drive 2 relays with 2 resistors to get these voltages, but I kept it here for now.

Fan Speed Control - input to power transistor or darlington circuit like I've shown above.

Ok, thats about it as far as the brain is concerned, the rest of the circuit needs to be finished as well, setting up the supply voltages and commons and such.

There are a few concerns that I haven't worked out yet. One is driving the vacuum solenoid block. There are 6 wire connections 5 discrete inputs, and one common return. I assume the easiest way will be to test the solenoid to see what each draws, make sure that they're ok for direct connection, apply +12V to common and let each output switch the return path.

Can I tie the + side of several outputs together with no problems?

Now on the temperature door actuator with position indication, (230$ at the dealer, 70$ at RockAuto.com):
The block that says "Solid State", I assume has something like a bridge driving circuit for a bidirectional motor like in microchip's AN893. If not that, then something similar, I included a simple circuit from that below.

The problem is, the top connection below is used for both motor return and low reference in position indication. I am really not sure how to approach this, as I was planning to tie all 3 of those wires to an analog in.

If one of those 3 pins is a 'low' on the analog array, can it take the servo return current? Can I use only 2 pins for analog, and tie to a more stout ground connection with the low and still have the 'analog in' feeback work?

**nserpa9248** · 11-13-2007, 11:28 PM

this may be a dumb question, but can the fusion brain do all of its programmed functions when not connected to a (running) computer? (say sending an input, which would make an output do a certain thing?)

That way, if the computer was off, the fusion brain could have power and still allow you to have HVAC control?

If it had that ability, it would be a key feature to the brain, and would prob end up getting one myself, since that would make converting my manual (cable) HVAC system, to a digital system (using actuators / servos), a much more realistic idea

**h3rk** · 11-13-2007, 11:33 PM

Yeah..., right now, the only way that is going to happen is with a 3rd controller in parallel, or some fancy PC startup/shutdown stuff.

I think there is hope for that, with the brain in the far off future. But for now, afaik, there's no way to dump any of the functions to the chip on the brain.

I'd keep my hopes up though; with a little patience, who knows.

That must have been my first or second question, way back when this first started.