Does anybody know or have experience with purge valve solenoids? I have one on my 90 w/ MSPNP and currently use the purge valve solenoid from the charcoal canister to kick up my idling when A/C is turned..I know I know A/C is for wimps but I'd rather cruise in comfort than sweat my *** off. I also have the BMW TPS and its been dyno tuned profesionally and have no issues with stalling when AC is on..however I can feel the solenoid doesn't react quite fast enough to catch the compressor. As you can see in my picture I have it plugged into the TB and as I said works pretty good, but I'm looking for one that leaks air quicker and with more volume...anybody know a good one to use? TIA

well, its a 20-year old AC system, don't expect it to be flawless. I had my AC **** running flawlessly with no bog on my NON-TPS car with an old *** 1.6.

I'm not so sure it's an issue with the "quickness" of the solenoid. The OEM one is plenty quick. It turns on and off dozens of times per second during normal use.

The purge solenoid probably doesn't move enough air to keep up with the load increase of the A/C compressor, it wasn't really built for that.

I keep forgetting that there isn't a fast idle input (like from A/C signal) to MS, what a silly thing to not include.

I have plans to do something like that... it would be pretty easy, I think.

There are plenty of other air solenoids available, some of which are intended specifically to be used as fast-idle valves. I only chose the purge valve initially because it's something that everybody already has.

Can't beat the convenience of it, I definitely agree. Have you considered using two solenoids in parallel or something similar, I expect 2 of the purge solenoids may have a better effect than just one.

I just wrote some code that adds a fast idle input. You can connect the AC or blower inputs to it. When the input goes low, it changes the idle target to the fast idle speed. It works on the stim, but the battery on my laptop died before I got a chance to test it on the car.

Do you think I'd be better off raising the idle speed, or increasing the minimum dc?

I don't have AC, and I've never actually tried to use the heater in my car (with the MS). My car idles fine. I'm just doing this for you guys.

__________________
92 Miata STS #7 using AFM (no MAP sensor) and custom code in OEM case (custom adapter board). Interfaced with atmega328 for emissions and CEL control.

I'd think that both would be ideal.

Raising MinDC will give you an immediate burst of idle accel, whereas relying upon idle speed alone will still give it a chance to stall while the idle routine slowly ramps up DC to seek the new target.

Meanwhile, raising target idle will keep is smooth once it's back into stable closed-loop, without you having to raise minDC so high that it never climbs above it.

Oh, and major props on the software mod. I've never ventured into editing the code. Which release did you use as the base?

__________________

I've been running a slightly modified version of hr_10g on my car for a while, so that's what I used. It would be simple to port it to 11c2 or whatever everybody runs here.

I made another change that seemed to resolve the idle droop issue on my car (stalls when coming to a stop). I have both TPS inputs going to the cpu (TPS 1N and 2L). The cpu artificially sets the tps values:

0v = no throttle
2.5v = partial throttle
5v = full throttle

I need to do more testing though.

I'll go ahead and play around with different minimum DC steps, and see how it handles heater + headlights, then I'll post a firmware for you guys to try.

If you have any other ideas, let me know!

PS What inputs do you recommend? My car is not setup like everybody elses, so I'm not sure what inputs you guys have leftover.

__________________
92 Miata STS #7 using AFM (no MAP sensor) and custom code in OEM case (custom adapter board). Interfaced with atmega328 for emissions and CEL control.

I'm currently using 2 solenoids in parallel, it provides just enough air flow but the rpm raise slowly with these solenoids while the A/C will drop the rpm immediately.

I then get a delay circuit and put it in between the solenoid and the A/C compresser. So I switch on the A/C, then the solenoid get powered immediately, the delay circuit will provide power to the A/C compresser after 2 second.

So the rpm will first raise up to 1200rpm slowly and then drop to 950rpm.

However sometimes in very hot condition, the air provided by 2 solenoids is still not enough to maintain a stable idle. Since the resistance made by the compressor is somehow increased with temperature.

The code has been tested, and I moved it over to a "clean" version of hr_10g (attached). At this point, all it does it wait for JS4 to go low, and it changes the idle speed to whatever you have set for your Fast Idle Speed RPM.

I tested turning on headlights & blower at the same time, and it was able to prevent a stall every time. The idleDC increases quickly and catches it. The downside is that when you turn the A/C off, the idleDC drops quickly, and temporarily increases your idle speed (with my tune, at least).

I haven't messed around with changing the minimum DC yet. It would be a bit more complicated, but I will do it next week. I'm not sure it's necessary, since the idleDC increases quickly on it's own due to the increased idle setpoint... but we'll see.

For testing, I wired it to a switch. You'll want to wire it to the A/C input (1Q) or the blower input (1S). Both of those go low when switched on, and are 12v while off. So you'll need to use a voltage divider to keep it below 5v.

I'm sure there is a "smoother" way to do this. In fact... it might work better in open loop mode. But I like to start with the simple things first

Joe - Feel free to split this thread up if you want. This really doesn't have anything to do with the purge solenoid, although we both have the same final goal in mind.

Attached Files

msns-extra-hr_10g_m.zip (25.5 KB, 78 views)

__________________
92 Miata STS #7 using AFM (no MAP sensor) and custom code in OEM case (custom adapter board). Interfaced with atmega328 for emissions and CEL control.

That is what i have now and it certainly is better than one. Just to be clear I have no stalling issues even w one solenoid, the tuner did an excellent job with idling, however I still think a larger single faster one is better. There is still idle dip but its kept at a drop of only 150 rpm and settles nicely to 1K w a/c..my normal idle is 900 a/c off.

These solenoids were designed for a different purpose, so Joe if you know of this particular solenoid you speak of it would be greatly appreciated. thanks

Now aren't you the devious one? I was hoping to take a look at your implementation and what do I find? An .s19 but no .asm file.

Am I understanding correctly that some of you are looking for a more suitable solenoid? I know older Toyotas used A/C idle up solenoids that they sometimes refer to as a VSV (vacuum switching valve). Don't know if they'd work any better than our purge solenoids, but at least they were originally used for our intended purpose. I don't think they still make the same old part for my 25-y.o. 4runner but some of the Toyota truck guys report successful replacement with Toyota part number 88690-32280, which is originally for a Camry. You can buy that one from a dealer (Jay Marks out of Houston is a good outfit) or just try pulling one from a bone yard, should be plenty of them in random Toyotas.

__________________

Hahah noooo . I just didn't want to waste my attachment quota. I've been working on open-source projects for years, and I can tell ya that someone actually caring to see the code is rare!

Attached Files

hr_10g_m-src.zip (121.1 KB, 76 views)

Has anybody tried the modified firmware with A/C? I don't have A/C, so no way of knowing whether it helps or not.

Yes I have that solenoid as well as one from a Mitsubishi V6 and still the purge valve from the Miata leaks air most.
I'm now thinking maybe looking into Chevy's, Ford's etc..

I just wished DIY or somebody could build in a dedicated circuit for the MSPNP to address A/C.

Ya don't need a circuit, you need modified code... and I'm working on it!

Looking forward to it!