The Great A/C FAQ
#61
![Default](https://www.miataturbo.net/images/icons/icon1.gif)
Wait, the MSPNP people would be pissed about showing the inside of the box? I didn't realize that anything they did was trade secret-y... I thought I was paying for the nice, neat/clean assembly (which is pretty nice btw).
There's continuity to ground on the PNP box for pin 1Q.
The thing I realized is that maybe I should've used pin 1J instead since we *know* the compressor relay current must travel through there. The PNP doesn't just jump 1Q to 1J, the two are actually connected via diodes/transistors I think. I was going to try to reverse engineer the circuit to see what's going on. I think it's limited to the daughterboard and doesn't actually touch the MS. I assume I blew one of the transistors by overheating it with the extra current (it didn't fail immediately, but after some time of working correctly). Awesomely, everything else in the MS still seems to work.
There's continuity to ground on the PNP box for pin 1Q.
The thing I realized is that maybe I should've used pin 1J instead since we *know* the compressor relay current must travel through there. The PNP doesn't just jump 1Q to 1J, the two are actually connected via diodes/transistors I think. I was going to try to reverse engineer the circuit to see what's going on. I think it's limited to the daughterboard and doesn't actually touch the MS. I assume I blew one of the transistors by overheating it with the extra current (it didn't fail immediately, but after some time of working correctly). Awesomely, everything else in the MS still seems to work.
#62
Boost Pope
![](/mt2011/images_td/ranks/supporter1.gif)
![](/mt2011/images_td/ranks/smod.png)
Thread Starter
iTrader: (8)
Join Date: Sep 2005
Location: Chicago. (The less-murder part.)
Posts: 33,339
Total Cats: 6,793
![Default](https://www.miataturbo.net/images/icons/icon1.gif)
This was a MSPNP?
Those directions were for a regular DIY MS.
By connecting the solenoid to 1Q, the current through the solenoid did not pass through the MSPNP, it went through the switch to ground like it was supposed to, and that switch is perfectly capable of dealing with the tiny amount of current we're talking about here.
However, whenever the A/C was off, placing the solenoid here caused +12V to be applied to the input of the PNP daughterboard which is only expecting a closure to ground! I haven't seen the exact circuit that DIY is using, but I'm pretty sure that is what has been fried. You probably overvoltaged whatever switching mechanism they're using.
On the plus side, it should be relatively easy to fix. The A/C portion of the circuit lives entirely on the daughterboard- it does not interact with the MS main board at all.
And yes, you're correct. Putting it on 1J would have been fine, as on the MSPNP that pin is rated to see 12v.
![Robertfmelbert](https://www.miataturbo.net/images/smilies/robertfmelbert.gif)
By connecting the solenoid to 1Q, the current through the solenoid did not pass through the MSPNP, it went through the switch to ground like it was supposed to, and that switch is perfectly capable of dealing with the tiny amount of current we're talking about here.
However, whenever the A/C was off, placing the solenoid here caused +12V to be applied to the input of the PNP daughterboard which is only expecting a closure to ground! I haven't seen the exact circuit that DIY is using, but I'm pretty sure that is what has been fried. You probably overvoltaged whatever switching mechanism they're using.
On the plus side, it should be relatively easy to fix. The A/C portion of the circuit lives entirely on the daughterboard- it does not interact with the MS main board at all.
And yes, you're correct. Putting it on 1J would have been fine, as on the MSPNP that pin is rated to see 12v.
#64
Boost Pope
![](/mt2011/images_td/ranks/supporter1.gif)
![](/mt2011/images_td/ranks/smod.png)
Thread Starter
iTrader: (8)
Join Date: Sep 2005
Location: Chicago. (The less-murder part.)
Posts: 33,339
Total Cats: 6,793
![Default](https://www.miataturbo.net/images/icons/icon1.gif)
Erg...
Yeah, most of the serious complaints seem to be coming out of the MS2 camp in this regard. I can certainly see it happening with MS1, particularly when initially letting off the gas if you haven't taken the time to properly adjust the minimum(dc) value.
Ok, let's see a pic (top and bottom) of the relevant section of the daughterboard and we'll try to point you towards fixing it.
Yeah, most of the serious complaints seem to be coming out of the MS2 camp in this regard. I can certainly see it happening with MS1, particularly when initially letting off the gas if you haven't taken the time to properly adjust the minimum(dc) value.
Ok, let's see a pic (top and bottom) of the relevant section of the daughterboard and we'll try to point you towards fixing it.
#65
![Default](https://www.miataturbo.net/images/icons/icon1.gif)
I'll have to get it tonight. I'm acutally finding the idle remarkable difficult to tune, especially when dealing with the IAT heatsoak etc, so I figured I'd work around this stuff.
I read what you were saying about minimum DC previously and that's less of my problem. The problem I have is just the car stumbling when the compressor comes on. It usually recovers, but the risk of stall is annoying.
I read what you were saying about minimum DC previously and that's less of my problem. The problem I have is just the car stumbling when the compressor comes on. It usually recovers, but the risk of stall is annoying.
#66
![Default](https://www.miataturbo.net/images/icons/icon1.gif)
FIY, the MS-II guys have, totally out of the blue I guess, decided to rewrite their code. After several people have independently written their own.
It sure would be nice if they'd SAY they were going to, or thinking about it, to waste less time. They still haven't said what they did, or how, or why. So I guess another 12 hour marathon of reading and trying to decode their code will be called for, to see if it's worth switching away from what I have.
It sure would be nice if they'd SAY they were going to, or thinking about it, to waste less time. They still haven't said what they did, or how, or why. So I guess another 12 hour marathon of reading and trying to decode their code will be called for, to see if it's worth switching away from what I have.
#67
![Default](https://www.miataturbo.net/images/icons/icon1.gif)
Thanks Joe. Haha, I figured I was taking a risk with the MSPNP, with the famous last words of "What could possibly go wrong?" ...
Thanks for clearing that up. BTW, DIY claims that the idle code, "properly tuned," should be able to handle the A/C compressor kicking on without issue.
![](https://www.miataturbo.net/images/smilies/smile.gif)
I look forward to seeing what works for you in the end.
#69
![Default](https://www.miataturbo.net/images/icons/icon1.gif)
Alright, I took some photos of the daughterboard and spent some time testing continuity.
The image with all the lines is what I *think* the circuit currently looks like. The red lines are measured continuity that I think might be wrong (i.e. shorts to ground that maybe shouldn't be there).
I tested D1 and D2 and they have no continuity in reverse-bias and read ~.55-.56 on the diode test in forward bias, which seems correct as far as I can tell. The thing is, if the continuity from the node between R13 and R7 to GND shouldn't actually be there, then what could possibly have failed to cause that? It doesn't totally make sense. The other disturbing thing is that it seems like parts of this circuit *do* talk to the MS.
The image with all the lines is what I *think* the circuit currently looks like. The red lines are measured continuity that I think might be wrong (i.e. shorts to ground that maybe shouldn't be there).
I tested D1 and D2 and they have no continuity in reverse-bias and read ~.55-.56 on the diode test in forward bias, which seems correct as far as I can tell. The thing is, if the continuity from the node between R13 and R7 to GND shouldn't actually be there, then what could possibly have failed to cause that? It doesn't totally make sense. The other disturbing thing is that it seems like parts of this circuit *do* talk to the MS.
#70
![Default](https://www.miataturbo.net/images/icons/icon1.gif)
Joe - or anyone with something intelligent to say:
There's a whole bunch of valves on the '99. How did you pick the valve you did for the additional air leak?
I was trying to put something together, and saw a 'purge control valve', a 'canister leak valve' and a something-or-other boost sensor. Since all mine are in storage, I couldn't sort through them. I assume the purge one is the one going to the motor, so it's what I used, but I'm a little unsure if it's the best choice. I'd like to get the largest capacity one, if I could. And one which won't fry if it's run at 100% d.c.
Holy: Interestingly, when I looked at your picture, it seemed to me some of the traces were either wrong, or harder to follow than you expected. I can see some traces with my own eyes that don't agree with how I interpreted your drawing. My guess is it's the interpretation that's wrong, but maybe a straight, top down view would be better. Also, there's a lot of glare in the fotos. Try not using a flash and instead steadying the camera on something.
Also, who spells it fotos?
There's a whole bunch of valves on the '99. How did you pick the valve you did for the additional air leak?
I was trying to put something together, and saw a 'purge control valve', a 'canister leak valve' and a something-or-other boost sensor. Since all mine are in storage, I couldn't sort through them. I assume the purge one is the one going to the motor, so it's what I used, but I'm a little unsure if it's the best choice. I'd like to get the largest capacity one, if I could. And one which won't fry if it's run at 100% d.c.
Holy: Interestingly, when I looked at your picture, it seemed to me some of the traces were either wrong, or harder to follow than you expected. I can see some traces with my own eyes that don't agree with how I interpreted your drawing. My guess is it's the interpretation that's wrong, but maybe a straight, top down view would be better. Also, there's a lot of glare in the fotos. Try not using a flash and instead steadying the camera on something.
Also, who spells it fotos?
#71
![Default](https://www.miataturbo.net/images/icons/icon1.gif)
The photos were shot without the flash and with a tripod, I think I didn't get the focus right though. The lines I drew do not correspond to traces directly, but correspond to where I found continuity. I just drew them wherever I wanted. Looking at the visible traces alone is not that helpful as its a multilayer PCB and there are a lot of traces that just aren't visible.
Anyway, the stuff I drew on there is a bit of a red herring. I talked to DIY about my problem and it turns out that the only thing of importance that 1Q connects to is the quad op-amp chip in the lower right corner. I think I fried one of the op amps inside it so I'm going to try replacing that guy. The circuit for 1Q/1J is actually pretty simple. It's just 1Q going into the op amp and the output of the op amp drives transistor Q2 whose output connects 1J to ground.
Regarding the valve: I was under the impression that Joe just picked a valve that was already connected to the intake manifold and was unused.
Anyway, the stuff I drew on there is a bit of a red herring. I talked to DIY about my problem and it turns out that the only thing of importance that 1Q connects to is the quad op-amp chip in the lower right corner. I think I fried one of the op amps inside it so I'm going to try replacing that guy. The circuit for 1Q/1J is actually pretty simple. It's just 1Q going into the op amp and the output of the op amp drives transistor Q2 whose output connects 1J to ground.
Regarding the valve: I was under the impression that Joe just picked a valve that was already connected to the intake manifold and was unused.
Originally Posted by Joe Perez
Since your MS went in, your charcoal canister purge solenoid has been sitting there lonely and forgotten. Why don't we put it to use? (The purge solenoid is on the cold-side shock tower right next to the igniter.)
#73
![Default](https://www.miataturbo.net/images/icons/icon1.gif)
Hooray, my A/C works correctly again. I just swapped out the op-amp chip on the daughterboard. For those of you who were curious about how the MSPNP circuit works, it's basically like this (attached).
By putting +12v directly to 1Q with no resistor I caused the op-amp to fail with its (-) input (1Q) shorted to GND (hence A/C on all the time). I think this meant I was actually bypassing all the safety switches and nothing would've shut my A/C off.
By putting +12v directly to 1Q with no resistor I caused the op-amp to fail with its (-) input (1Q) shorted to GND (hence A/C on all the time). I think this meant I was actually bypassing all the safety switches and nothing would've shut my A/C off.
#74
Boost Pope
![](/mt2011/images_td/ranks/supporter1.gif)
![](/mt2011/images_td/ranks/smod.png)
Thread Starter
iTrader: (8)
Join Date: Sep 2005
Location: Chicago. (The less-murder part.)
Posts: 33,339
Total Cats: 6,793
![Default](https://www.miataturbo.net/images/icons/icon1.gif)
Well, that's great news. I think we can try it again, this time with the solenoid paralleled with the A/C relay, on 1J. That pin will deal with 12 volts just fine.
Abe- the reason I chose that particular solenoid is that this hack was originally written for a 1.6 car. In '90-'93, they don't have a lot of surplus air solenoids under the hood. Aside from the IAC valve, they have exactly one- the canister purge valve.
Abe- the reason I chose that particular solenoid is that this hack was originally written for a 1.6 car. In '90-'93, they don't have a lot of surplus air solenoids under the hood. Aside from the IAC valve, they have exactly one- the canister purge valve.