When you click on links to various merchants on this site and make a purchase, this can result in this site earning a commission. Affiliate programs and affiliations include, but are not limited to, the eBay Partner Network.
Always been my theory. I had wired them up to some old megasquirts back in the day and hopes people would try opening the solenoid to increase fuel pressure to combat the warm fuel issue.
If my math is right all that would do is boost fuel pressure approx 10psi at idle. Assuming 43psi base, that's an extra 30% fuel from the table at idle that tapers to nothing extra at full load. Might work on stock injectors, highly doubt it works worth a damn if you have larger injectors and deadtime is a larger percentage of actual pulsewidth. Still pretty useless for a wandering heatsoak idle lean-out.
It's a jank fix, akin to the 1.6 idle valve wart. Glad they ditched it on the NB for obvious reasons. If you were going to add features to the megasquirt to use the valve for anything other than open time after starts (pretty sure seconds is a option, getting actual seconds from crank-to-run, though?) it seems like you'd need custom firmware and good ******* luck with that. At that point may as well just ask for a ASE taper by RPM and forget about the valve...
The only one who could realistically do that is too busy dicking around with bullshit nobody asked for and specifically said he isn't adding anything else till he gets testers for that. Seeing as how nobody asked for these features, I expect development is going to be stalled for another 3+ years.
In other words, don't get your hopes up for anything new from megasquirt firmware.
I'm considering figuring out rusefi, because megasquirt is pretty much dead in the water any more.
The only one who could realistically do that is too busy dicking around with bullshit nobody asked for and specifically said he isn't adding anything else till he gets testers for that. Seeing as how nobody asked for these features, I expect development is going to be stalled for another 3+ years.
In other words, don't get your hopes up for anything new from megasquirt firmware.
I'm considering figuring out rusefi, because megasquirt is pretty much dead in the water any more.
I ran RusEFI in my NB1. Awesome ECU.
I’m not sure if it’s stalled however. Their software dev is just one guy, but he knows his stuff.
I have an unused Frankenso sitting around. AFAIK, it’s circuitry is good.
NVM - RusEFI has made a Microsquirt competitor. 220$ assembled on eBay. Damn.
Speeduino is good. But the software is limited, featurewise. The creator isn’t too fond of adding in features.
I had a moment and remembered you could do fuel temp comp in MS3. The flex fuel sensor is too slow to be useful for anything other than logging, but you can use a generic sensor input instead. If a thermistor on the fuel injector does actually coincide with leaning out with heat, then you could theoretically correct it in advance instead of relying on EGO or ASE. Everything you need is already in the firmware, assuming it works like it's supposed to.
I never got around to sticking on on my injectors, still need to do that...
If my math is right all that would do is boost fuel pressure approx 10psi at idle. Assuming 43psi base, that's an extra 30% fuel from the table at idle that tapers to nothing extra at full load. Might work on stock injectors, highly doubt it works worth a damn if you have larger injectors and deadtime is a larger percentage of actual pulsewidth. Still pretty useless for a wandering heatsoak idle lean-out.
It's a jank fix, akin to the 1.6 idle valve wart. Glad they ditched it on the NB for obvious reasons. If you were going to add features to the megasquirt to use the valve for anything other than open time after starts (pretty sure seconds is a option, getting actual seconds from crank-to-run, though?) it seems like you'd need custom firmware and good ******* luck with that. At that point may as well just ask for a ASE taper by RPM and forget about the valve...
Honestly, I never really had the issue, cause I tuned.
Honestly, I never really had the issue, cause I tuned.
Surely you are not referring to the reversed MAT correction curve?
Seems to me that low impedance (peak and hold) injectors could be better for this. However I don’t know the cons of these, or if any are still made. I know MS drivers could be set up for them in the past, and I think still can be.
To be honest, I'm just glad to see you elsewhere than the political thread! But, if you have the time before heading back I'd love to hear how you actually tuned around heat soak...
It's been a while, but i remember dealing with it a bunch.
so much so that i deteremined it wasn't really AIT heatsoak. I had, at one point, installed a second AIT sensor down into a wheel well and some sort of switch to toggle between them at first start to read "ambient" vs installed in the pipe. but they always read very close to each other -- unless the cooling fan was running and dumping hot ait on the sensor at idle when it was installed by the TB.
moving the ait to the IC endtank solved that little part.
going to seq. fuel and spark, clearing/tuning the MAT corrections table, then running adaptive timing for idle control, helped significantly after that, probably other tables/settings im forgetting. The worst i would ever lean out at idle was 15:1 and that was no issue for my 1.6L even on 460cc shitty rx7 injectors with a 1.4ms deadtime set. I could restart the car on a 100°F or 20°F morning and have no issues, so i called it day.
I was always convinced the PRC would be benefitical for this, and used to wire it up on some older setups, but no one ever got back to me on experimenting with it.
I'm not sure why everyone still kills MAT correction, I don't and with a non 1980's IAT sensor it's no big deal. We've already established that is not the cause of leaning out at idle. I ran the GM sensor on the endtank on my old 1.6, also pretty much nerfed it so hard that I may as well not even have bothered with it. Could have made a calibration curve that just read a flat 70 degrees and left it out completely. That's why I didn't even bother with one on the new miata, tried something different with the VW sensor/nissan MAF instead.
I also have one of these GM ambient sensors strapped up near the fog light, just to log for comparisons though, no switch.
Still, my injectors are more than twice as large as those old rx7's, even if I dropped to 43psi base. Regardless of what deadtime is set at, because it's totally irrelevant, the effect of it changing swings my tune a hell of a lot harder than it would on injectors half the size.
I never stall from lean, and am pretty amazed at how lean I can run without stalling. But I know things still change with heat. Can hear it on decel burbles fading the longer I keep it running, see it on hot starts, see ego constantly feeding fuel the longer it idles. It works, but it could work better. This is me trying to add the PRC, but in a different way I guess.
the built-in curve of the MS1 and MS2 and initial MS3 release was ******* bonkers; it caused hella problems idle or not. That's why I pushed Ken so hard to fix it back in like 2011.
When running EGO, I was using my full 15% tolerance just to fight the code from unnessecarily pulling fuel for no reason whatsoever. I tuned the curve by hand to where ego wasn't being used to compensate for extreme temps -- I don't believe any MAT air density correction is really nessecary between 60-120°F
On stock ECU, the air temp sensor is not used literally but more as a variable on the entire fuel modeling algorithm, which is a very far cry from the simple PV=nRT of the MS3. Ie the temp from the sensor is taken as a basic starting value, and a "true" IAT value is derived by combining things like VSS, estimated airflow, etc. Open VCDS on a VAG car and watch the IAT as you drive the car and you'll see what I mean.
I made a thing to do innovate serial in the JBpref format, plus a bunch of other stuff.
I've been looking, but there's no (officially supported) way to overwrite MAT with something off CANbus, right? I tried just poking things in, but Ken says the canbus code does address checking and only allows certain offsets to write to prevent corruption. Trying to get him to elaborate further is like pulling ******* teeth.
Scott, thanks for the reply. We all have learned a lot from your sorrows. I had not realized that you had stopped adding MAT correction at hot.
My Mat Air runs from 111% @ 0F (never verified in real life) to 100% at 70F to 93% at 160F. I tuned by looking at 100kPa at 3K RPM and bringing EGO back to 0, like it was at my tuning temp of 70F. This pretty much minimizes EGO for anything above 75 kPa, any RPM, any MAT (as read at the outlet of the FMIC).
I tune at 100kPa because little or no throttle expansion cooling (Joule-Thomson effect)... thus, the intake temp is approximately equal to MAT sensor temp. At throttled flows, there is cooling going on that cannot be accounted for.
With that approach, I think the MAT corrections are pretty real and pretty good. Not fully ideal gas law.... about 50% IIRC. Fully ideal gas was what Brain fought against years ago, that he mentions above.
And, yeah, 50*F - 115*F is 105% to 95%, so EGO can take care of it... But, Deezums is looking for that last 5%.
I was always convinced the PRC would be benefitical for this, and used to wire it up on some older setups, but no one ever got back to me on experimenting with it.
I'm not sure why everyone still kills MAT correction, I don't and with a non 1980's IAT sensor it's no big deal. We've already established that is not the cause of leaning out at idle. I ran the GM sensor on the endtank on my old 1.6, also pretty much nerfed it so hard that I may as well not even have bothered with it. Could have made a calibration curve that just read a flat 70 degrees and left it out completely.
Well this makes me feel alot better than I still run a zeroed out MAT curve on my similar setup. Tuned the car at 30F and anything above 80kpa or os is either good or pulling fuel from EGO. Below that, it's all over the place.
As far as lean hot restarts, I see up to 17afr idle when my normal is 14. The only thing I could figure out to do about it without adding extra sensors is to use the ASE taper based on coolent temp. It works since the temperature of the coolant is almost directly corrilated with how long the car has been sitting. <10min sitting and it restarts perfect. More than (about 170F) that and I have to add fuel down to about 130F restarts, after that it's fine again. The worst seems to be starting at ~150-160F.
Read the resistance of your injectors when hot, and when cold.
I got a difference of 3 ohms. I suspect that there’s some correlation there between fueling and heatsoak.