| IAT - preventing heat soak (difficult hot starts) - Click HERE for Original Thread |
| ae1969 |
Well this question spawn off from a comment that was made.
I have an issue with hot starts.
So I logged my drive home today.
My IAT (intake air temp) was hovering around 80-85 F. NOW when the car is shut of...... what happens the sensor begins to soak the heat up.
The engine runs at 180-200F and the heat from the engine and the turbo make the IAT temps go from 80-90F upto 140F! when it is shut off.
So when I start the car. The injector Pulse width is decreased (cause it thinks its damm hot). The car is leaned out and it lopes and has a rough starting until the air starts to flow and drops the IAT temp to at least 110.
SO the idea is to decrease the heat transfer to the sensor. But how.
Any ideas. |
|
|
| notaturbotalon |
| Never played with a megasquirt system before, but maybe you have something near what I have in the EMS. In my start up options there is a seperate graph where I can add or subtract fuel over a timeframe when the car is starting, then gradually reach the main fuel map. Anything like that with yours? |
|
|
| ae1969 |
quote:
Originally posted by notaturbotalon
Never played with a megasquirt system before, but maybe you have something near what I have in the EMS. In my start up options there is a seperate graph where I can add or subtract fuel over a timeframe when the car is starting, then gradually reach the main fuel map. Anything like that with yours?
Uhmmm yes I do. After start enrichment allows me to + - fuel for specified amount of engine cycles. Unfortunately it is based on engine temp.
I could set my 180-200F engine temps to subtract fuel on start up.........
It takes approximately 5-10 minutes for the IAT sensor to heat soak upto 140f. BUTIf I stop and start engine within 1-2 minuts the sensor has not had enough time to heat soak. The engine temp will be telling the ecm to use wrong amount of fuel which will probably make it hard to start as well. Unless I let it heat soak. (over the next 5 minutes)
Any other ideas? |
|
|
| fatbastard |
| This isn't a new problem with the MS. Do you have the GM IAT in an aluminum manifold? If so, try moving it around. Aluminum holds and transfers heat in the worst way. |
|
|
| newaccorddriver |
| im not sure how your porsche is setup, but why not try and move the sensor right after the intercooler and as far away from the engine as possible? |
|
|
| ae1969 |
quote:
Originally posted by fatbastard
This isn't a new problem with the MS. Do you have the GM IAT in an aluminum manifold? If so, try moving it around. Aluminum holds and transfers heat in the worst way.
DING DING DING.... correct.
I have read a few of these issues but have not seen a good solution.
You can see the sensor half way between the intake manifold and intercooler. It is on the hard pipe. All the pieces being aluminum. I had actually contemplated making a carbon fiber? hard pipe? but lets just say I am really stretching.
We are talking a 50 degree increase due to heat soak. Not good. |
|
|
| newaccorddriver |
| thats basically right over a hot and toasty engine... maybe you should try putting it considerably farther south right like in the intercooler or something as it would probably be a bit harder to heatsoak that area |
|
|
| ChromeDragon |
| I think the only realistic answer is liquid nitrogen cooling. |
|
|
| sparkycivic |
Just an observation:
are you absolutely sure that it's leaning out based on the IAT... or is it really rich thinking that the engine is cold due to ECT switch shorted or something. this sounds exactly like what happens to my buddy who'se throwing coolant-temparature-sensor-shorted codes on his golf, so it won't start or idle when it's hot unless the throttle is pinned, and does this untill it gets to closed-loop.
is there a test where you can double-check that on a MS? |
|
|
| Blaine B. |
| Thin phenolic spacer gasket between the AIT sensor and the surface it is mounted on.:dunno: |
|
|
| ae1969 |
quote:
Originally posted by sparkycivic
Just an observation:
are you absolutely sure that it's leaning out based on the IAT... or is it really rich thinking that the engine is cold due to ECT switch shorted or something. this sounds exactly like what happens to my buddy who'se throwing coolant-temparature-sensor-shorted codes on his golf, so it won't start or idle when it's hot unless the throttle is pinned, and does this untill it gets to closed-loop.
is there a test where you can double-check that on a MS?
Thats the beauty of stand alone. Every sensor can be monitored.
I pulled the old multi-meter on the IAT as well just to double check.
But I am anal about these type of things. I sit with the Megatune (ecu software)/laptop on. With the car just shut off.
IAT temp is 80-90F......... engine coolant 180-200F.
5-10 minutes pass.
140F on IAT. Engine coolant still pretty hot... 180-190F
You can actually see the IAT climb on the laptop. I can even log this with the laptop. |
|
|
| ae1969 |
quote:
Originally posted by Blaine B.
Thin phenolic spacer gasket between the AIT sensor and the surface it is mounted on.:dunno:
You know I had been thinking about something like this. ...But I keep thinking the thread of the sensor still touches the pipes. It may still help? Not a bad idea. How much it helps I definitely would be able to log.
I was at home depot today...........looking at those PVC pipes wondering. Should I or shouldn't I.
It would look like shit but it would at least let me log if an improvement is possible or not......... I am assuming it is the heat conducting through the pipes........ and not the actual air that is heated by the engine/turbo still. |
|
|
| Ron@Revolution |
| I doubt a spacer is going to help since most of your heat transfer is porbably from the motor and not just through the pipe it is mounted on. I would mount it further towards the front of the car and on the opposite side of the pipe (away from the motor). Also if your car never see's engine IAT's temps that high you might try just adjusting the fuel correction % at the higher values like say 120 deg and up. My last sugestion would be invest in some DEI heat shielding for the sensor and possibly the pipe itself. I would say the pipe isn't your problem though since I use the same sensor closer to my motor in Aluminum piping without the problem of it heat soaking. My engine bay setup is also different with the TB being on the opposite side of the engine then the turbo etc.. Hope this helps. |
|
|
| chris f |
| On the srt's under hood tempitures are killer, from what i can see to combat the same problem you are having are, mount it in a plastic tube, somthing short. keep it low or in front of the engine. |
|
|
| fatbastard |
The problem is stagnant air, correct? The turbo is right underneath the piping/sensor? The problem is likely twofold, transfer from the aluminum is the main cluprit, but having the piping directly over the turbo/exhaust manifold is exacerbating the problem. Heat rises, and when the car isn't moving, thats causing a ton of problems. When it's moving you're pretty much stuck with just radiant heat, when it's not you get convection in there as well. Also add to the fact that there isn't air flowing through the pipe.
I think Ron hit the nail of the head, move the sensor closer to the front of the car, possibly in the intercooler endtank itself. I'd jsut head out and buy a pipe plug (I think the GM sensors are NPT? I can't remember) or possibly another sensor, drill and tap the intercooler. |
|
|
| Ron@Revolution |
| Yes it is NPT thread! |
|
|
| JustinL |
I think there must be a way to do it with a software solution as well. I think the afterstart enrichment table may be able to help us. It's just start-up that's the problem; once the air starts circulating past the sensor things sort them selves out quickly.
I would also agree with whoever was saying its a convection problem and not a conduction problem. It actually is the air in that pipe that gets insanely hot when the engine sits and not the pipe conducting all the way to the sensor. The air in the intercooler is the same temperature in these conditions.
Alex, we should do an experiment to see how much conduction we can measure through the sensor housing by heating it on the bench. |
|
|
| ae1969 |
quote:
Originally posted by JustinL
I think there must be a way to do it with a software solution as well. I think the afterstart enrichment table may be able to help us. It's just start-up that's the problem; once the air starts circulating past the sensor things sort them selves out quickly.
I would also agree with whoever was saying its a convection problem and not a conduction problem. It actually is the air in that pipe that gets insanely hot when the engine sits and not the pipe conducting all the way to the sensor. The air in the intercooler is the same temperature in these conditions.
Alex, we should do an experiment to see how much conduction we can measure through the sensor housing by heating it on the bench.
If we could use two sensors that would be ideal. :)
One over by the air filter for start up. That would allow the car to run for the first 30 seconds or so....... and then the other sensor taking over.
COuld ask the megatune guy to see if he could do that? |
|
|
| fatbastard |
A good way to figure this out is to get a thermocouple and start placing it on different parts of the engine to see where it's getting the hottest. One of those laser temp thingy's would work well too.
I'm willing to bet that the intercooler endtank is significantly cooler than right above the turbo. Also, check the intake manifold itself.
Leave the hood closed, let everything heat soak for a bit, open the hood and start taking some readings. |
|
|
| ae1969 |
quote:
Originally posted by fatbastard
A good way to figure this out is to get a thermocouple and start placing it on different parts of the engine to see where it's getting the hottest. One of those laser temp thingy's would work well too.
I'm willing to bet that the intercooler endtank is significantly cooler than right above the turbo. Also, check the intake manifold itself.
Leave the hood closed, let everything heat soak for a bit, open the hood and start taking some readings.
I have one of those laser temp thingys... I will check today and report. |
|
|
| ae1969 |
So the car sat for at least 15 minutes
125 F - Engine body, frames, plastic pieces etc.......and HOOD from the outside
130 F - Hard pipe, intercooler and the pipe leading to the air filter.
135 F - Hard Pipe right next to throttle body
145 F - Hard Pipe coming off turbo going to intercooler
155 F - BRASS ON THE SENSOR ITSELF
160 F - Intake
180 F - Engine block
300 F - Turbo housing
Now I am wondering if I should try and insulate the sensor from the outside?
 |
|
|
| Ron@Revolution |
| That is where the heat is coming from move the sensor and your problems should go away. |
|
|
| fatbastard |
Ok, well, that shoots that theory all to hell. I'm surprised the intercooler is just as hot as the pipe above the turbo.
I remember trying to set the damn cold start enrichments on my MSnS-E on the colt, I never could get it right. There's no way to compensate for intake air temp on startup, it's all coolant based.
You're running MS-Edis, correct? I seem to recall there being two programmable outputs on it, that were configurable for intake air temps. I wonder if there's a way you could program the output to kick a set of switchable contacts (works like a relay, but instead of on/off, it's goes from one wire or the other) to get it to switch between IAT sensors? The issue would be to know when to calibrate it so it doesn't switch between the two sensors while the car is in motion.
All this talk makes me want to put it into the auto Talon and see what'll happen. Damn you! |
|
|
| ae1969 |
quote:
Originally posted by fatbastard
Ok, well, that shoots that theory all to hell. I'm surprised the intercooler is just as hot as the pipe above the turbo.
I remember trying to set the damn cold start enrichments on my MSnS-E on the colt, I never could get it right. There's no way to compensate for intake air temp on startup, it's all coolant based.
You're running MS-Edis, correct? I seem to recall there being two programmable outputs on it, that were configurable for intake air temps. I wonder if there's a way you could program the output to kick a set of switchable contacts (works like a relay, but instead of on/off, it's goes from one wire or the other) to get it to switch between IAT sensors? The issue would be to know when to calibrate it so it doesn't switch between the two sensors while the car is in motion.
All this talk makes me want to put it into the auto Talon and see what'll happen. Damn you!
The car gets pretty toasty.... I wanted to fry an egg on it like they did on a ferrari back in the 80's......
I am running MS2 Ver3......... EDIS is built in on this one.
No programmable outputs on IAT. I think I will wait for the winter. I am going to change my intercooler and maybe find a new home for the sensor.
I tried shielding the exterior of the sensor today (insulated). No go. It still heat soaked. It has to be mainly from the internal air. |
|
|
| Blaine B. |
| Metal is a much better conductor of heat than air. |
|
|
|
|
|
