ziad

iTrader: (1)

actually not sure about the USDM cars but that spot is good as there is a big vent on all model evos stock front bumpers (7-9)... there is a plastic cover that hides most of the vent but i think there is still enough air through it.... if u are not using an under tray then the area could get dirty or wet.

mrfred

iTrader: (50)

Capping the table should be a good interim fix. Please document the high IATs after startup though.

Jack_of_Trades

iTrader: (17)

My buddy who bought my Evo is switching to SD very very soon so I will have him put the IAT bung down in that location and see how well it does in that spot.

l2r99gst

iTrader: (2)

I was thinking if you make a fix for this...maybe instead of a fixed value we can adjust a decay rate from, maybe a fixed value subtracted from the measured MAT temp, then decayed at an adjustable rate. I think that may be a bit better for different seasons.

Are you asking for some logs/data showing the phenomenon? Or just get some data for the heat soak values before I cap my tables? Either way I can get some data.

As mentioned above...I did some quick testing today. MAT temps were about 90ish driving, about 100 idling in my driveway for a while. After about 2 minutes with the car off, they were about 113. Then I came out 15 minutes later and they were about 135. I started the car and it came back down to the 100 range in about 30-60 seconds and everything was fine again.

Upon the restart the coolant temps were still like 190 or so, so closed loop was active. STFT was +25 for those 30-60s until the MAT came down and idle was rough. After that, STFT was bouncing around 0 again and idle was smooth.

mrfred

iTrader: (50)

I guess all that's needed to prove the issue is to apply an appropriate ceiling value to the MAT table, and see if the car behaves better. Was just getting to excited about data. :-)

l2r99gst

iTrader: (2)

Another option, just throwing it out there, might be to wire in an ambient air temp sensor into the maf plug.

I think SD 2.0 was going to use a baro sensor for startup...why not just add in a temp sensor as well, for people who have their MAT wired into the fuel temp ADC? I don't have spoolinup's SD harness, so I'm not sure how he has it wired (I think he just has the MAF IAT input going to a plug for a MAT sensor), but maybe it would be a possibility?

I think that might be too hard to code up, though, since some people are using the IAT input for SD MAT already. But, just wanted to throw it out there, just in case, since a SD 2.0 might be on the horizon.

I think the easiest and simplest solution would be to have some kind of configurable MAT startup (maybe subtracted from measured MAT) and decay rate, as we mentioned above.

Jack_of_Trades

iTrader: (17)

I think just coding it to run a default value for the first 30 seconds or so would make sense and be the easiest. Kind of like open loop startup.

mrfred

iTrader: (50)

That was actually the first option that crossed my mind, but I quickly concluded the same thing a J_of_T.

03whitegsr

iTrader: (8)

Locking it to a default value isn't going to work for anybody that sees large temperature swings. Might work for somebody on the coast with no elevation change, but I can easily see -10F and 60F temps in the same day with a 15 minute drive.

Jack_of_Trades

iTrader: (17)

Hmm, seems like the only real way to alleviate the issue is to relocate the IAT to a proper location then? That's about as real-world/live data you'll probably get I'd imagine? I am just about to dive into SD but I don't have any experience yet so forgive my ignorance on certain situations I may be chiming in about.

03whitegsr

iTrader: (8)

Honestly, it's not that big of a deal. It is annoying, but the car starts and runs fine within 20-30 seconds. I just blip the throttle a couple times and it settles down pretty quickly to normal AFRs.

I would say the baro issue on start up is a lot more important. I was BARELY able to even get my car started going up to 9000' in cold weather. That is not acceptable, IMO.

I think I'm going to bump the SD 2.0 thread so we can renew discussion on these issues.

l2r99gst

iTrader: (2)

That's exactly why I am suggesting a configurable value subtracted from measured MAT temp, also with an adjustable decay rate.

That way, on a hot summer day, where your MAT temps are like 100F cruising and then you sit for 20 minutes, the MAT might heatsoak to say 150F. Well, if we have a configurable value of measured MAT -50 (let's take -50 for an example), then the car would start with a MAT temp of 100. Then have that decay 1*/s and 50 s later, you are back to normal MAT temps, where the real measured MAT temp can now be taken from the sensor.

On the flip side of that, let's say that you're driving in the winter at 0F. Let's say your MAT temps are like 40F during cruise. You park your car for 20 minutes. That MAT isn't going to heatsoak up to 150, but maybe it will some, like to 90F or so. So, our configurable -50 will start the car with a MAT of 40F, decaying again.

That seems like it would work fine. This way you don't have any fixed MAT temps for starting, which I agree wouldn't work. We need a fixed correction from the measured heat soak. That way it is a relative number, not an absolute fixed number.

Maybe also only have this correction if the coolant temp is above a certain number, where heatsoak will be occuring. If the engine is completely cooled down and has been sitting for a few hours, we don't need the correction. It's only when the coolant temp is warm enough to keep underhood temps high enough to heatsoak the sensor. So, maybe something like only apply correction when coolant temp > 100F or something like that.

03whitegsr

iTrader: (8)

Subtract the difference between coolant temp and IAT?

Example: Hot day, 20 minutes after shutting off
100F Ambient
180F coolant temps
140F IAT

Estimated IAT = IAT - (180F-140F) = 100F

Cold day an hour after shutting the car off
40F Ambient
120F coolant
80F IAT
Estimated IAT = IAT - (120F-80F) = 40F

Cold day several hours after shutting the car off
40F Ambient
60F coolant
50F IAT
Estimated IAT = IAT - (60F-50F) = 45F


FWIW though, measured ECT will drop pretty quickly since it is measuring it in the thermostat housing that can cool off quickly. You'll see ECT climb very quickly on start up on a warm motor as the motor cycles the hot coolant to the sensor.

l2r99gst

iTrader: (2)

Your suggestion sounds like a viable solution.

I was simply suggesting to subtract a fixed value from the measured heat soaked MAT. But only apply the correction for coolant temps above a certain value (assuming the engine bay would be hot enough to heat soak the sensor). But doing it your way takes that into account, so I like it.

l2r99gst

iTrader: (2)

Thinking about your solution a little more....it may not work given enough time where the heat soaked MAT reading approaches the coolant temp reading. Say the MAT gets heat soaked to 150F and the coolant temp is 150F. Then there would be no correction.

I guess we have to think about it a bit more and see what will work the best. Either way, I think we have a couple possibilities. A simple correction from measured MAT applied only over certain coolant temps may be the easiest solution that covers the most situations.