holy driver

Thanks Matt.

holy driver

I measured the resistance of the purge valve solenoid and it's about 26 Ohms. Which means it'll draw about 0.5A. That's more than I was hoping. Does anybody know how much the A/C relay draws? This might be getting to close to 1A for comfort.

AbeFM

iTrader: (3)

When I wanted to know how much stuff drew, I just flipped the voltmeter over to the little "A" and measured it. :-P

I think that's about the average, though, my fans were ~500 mA to run both.

holy driver

Heh, if only it were that easy. I'd have to break the circuit to measure it properly, and I'd rather not do that. I don't have an inductive loop current meter.

I was also more asking out of laziness because the A/C relay (or what I think is the A/C relay) is a little hard to reach.

AbeFM

iTrader: (3)

I was thinking you could do it at the OEM connector...

holy driver

You mean the connections at the ECU? If you've disconnected the ECU how are you powering the circuit to measure the current? It seems rather cumbersome.

In any case, there is *a* way to measure or look up the current draw and/or resistance of the switch, it's just more work, and now we've kind of gotten ourselves out into the weeds.

I still think the "right" solution to this is to use a spare input on the MS and jump it to 1Q. Then have the idle code bump up the idleDC when 1Q is grounded. Though I guess that's also more work than we're discussing here.

AbeFM

iTrader: (3)

yes - having the MS do ANYTHING is always way more trouble than it's worth. The only thing harder is finding out what it actually does from the people who wrote it. :-)

Joe Perez

iTrader: (8)

You want to be rally safe? Dike the purge valve out from the factory wiring. Splice the + side (Wht/Red) into the same wire feeding the A/C condenser fan (Blk/Yel), and connect the - side (Yel/Red) to ground. That will isolate it from the MS completely, and cause it to open when the compressor is on.

AbeFM

iTrader: (3)

Joe, I might not do that, if you have your fans wired together, it'll turn on the air leak whenever the fans are on. If you don't have them wired that way, you've got some 'splaining to do

holy driver

I don't have them that way because I read something somewhere that the one fan relay isn't rated for 2 fans or something.

edit: although presumably you could do this inside the MS so the signal that triggers the regular fan also triggers the A/C fan so each runs off its own relay.

edit2: meh, I'll probably just buy a couple of relays and just do this properly since I've been whining about over-current throughout this entire thread.

Joe Perez

iTrader: (8)

Fine. Hook it up to the wire the drives the mag clutch.

SideDraft

Joe I like your your idea, simple and easy and it seperates it from MS.

Have you actually done this? Will the purge solenoid leak enough air to compensate for a/c idle?

AbeFM

iTrader: (3)

Skip Edit 2, edit 1 is the right thinking. I just soldered a piece of wire between the two fan drives on the OEM harness.

For years I had these wired together (on the signal side of the relays), when I went MS, I jumpered the same thing on the harness (much easier). Works great, you stress nothing, as long as the drive transistor is rated at least 500 mA. You can get one at radio shack if need be.


I imagine that makes more sense. I think I'd have to go and actually read to figure out why you can't hang something off the harness, though. Couldn't you just protect the PnP with a diode or some such?

Joe Perez

iTrader: (8)

Clearly someone didn't read the very first paragraph in the very first post in this thread:Of course I haven't actually done this. I don't need to, and I think you guys are all pussies for having A/C. I did quickly test the theory (and you can too) by just pulling the hose off the solenoid. It causes a momentary surge in idle which is compensated for by the IAC valve after a couple of seconds. But my A/C compressor is in a landfill somewhere, so I can't tell you if it's enough to work for that.



Yeah, you could cut the factory wire that goes from the MS to the A/C pushbutton (by way of the safety switch) and install a diode inline with that, then tie the solenoid in on the non-MS side of the diode. I'm not sure whether the additional VF across the diode will cause the MS to not turn on the A/C, but it's worth a shot.

holy driver

Alright folks, I did this and no sparks yet. The valve appears to open and close as advertised. However, I think we need to use a bigger valve. The vacuum leak does help somewhat and the recovery is quicker than before, but there's still some droop and shudder. *shrug* It's progress though.

So what should I use to filter the valve? Stockings? :P

AbeFM

iTrader: (3)

Well, if you have a spare out on the MS, try what we were kicking around earlier, drive the AC from a relay, have the MS turn it off when your RPM dip. You could also have that same control turn on the air leak.

But Alberto's car, which only ran AC for a couple days till the wiring gave up (suspected for unrelated reasons), the AC didn't bother it. Somehow, he has this amazing idle. Totally uncontrolled (no IAC valve at all) and the AC generally doesn't hurt it.

holy driver

This doesn't make sense, because the dip occurs right when the compressor kicks on because the idle code doesn't respond fast enough. If you turn the A/C off whenever the idle dips then the A/C will basically never come on or it'll start oscillating on and off.

AbeFM

iTrader: (3)

If you set the switch point at, say, 600 rpm:

1) RPM droops
2) IAC starts to react
3) RPM hits 600 rpm
4) AC shuts off
5) IAC starts to effect motor
6) RPM increases over (theoretical) 750 RPM AC enable
7) additional AC load easily handled by now-open IAC valve
8) We all have a party in HolyDriver's new cool car.

I'm sure there are values for which it would oscilate, but it would at least keep the car from stalling, and with carefully tuned values might at most turn on-off twice. And the same relay could be used to turn off AC turning slow takes offs from a stop, or full throttle romps where you want to have all the power at the wheels. Sure would beat turning off the AC by hand every time you see some porsche you want to stomp.

holy driver

Hmm, I see what you're saying, I don't really have stalling problems though, just the annoying shudder and droop. Also it would seem that my earlier suggestion of preemptively bumping up the idleDC as soon as the compressor comes on would work better, no? (although i guess we're also talking about different situations since you're using an MS2 right?)

Anyway, I think I'm less concerned about this whole thing now. It'll get me by while I try to come up with some super fancy control function that depends on the day of the week and the phases of the moon.

AbeFM

iTrader: (3)

Oh yeah =- that would be ideal. Here was my idea from a while ago, but you'd have to rewrite the idle code to accept a DC offset, variable gains, etc

Basically, you'd take an ADC (analog in) and have all your devices (your big honking stereo, AC compressor, power steering pump, basically anything that 1) puts load on the motor, and 2) has an output to tell the motor it's on) and run those to a series of resistors, which would put a analog voltage on the ECU in.

Then, for higher voltages on that pin, you would increase the idle accordingly. The OEM already does it, that's why the outputs are there.

I would imagine you could do something similar with battery voltage, but don't know if it's worth the trouble.

And a simpler version would be to tie all those outputs together, and use a digital input pin on the ECU, having a two state "high idle/regular idle" set up.