Slant Six Forum
https://slantsix.org/forum/

Placement of IAT sensor with turbo
https://slantsix.org/forum/viewtopic.php?t=45857		 Page 1 of 2
Author:  Sam Powell [ Sat Aug 06, 2011 3:26 am ]
Post subject:  Placement of IAT sensor with turbo
This si a copy of a post from another thread. It really belongs here.

The issue is hard starting after the engine is partly warmed up, but hot under the hood, and maybe close to Tstat opening. This means water at ECT is very hot, but engine is not yet. Also, it means IAT is going to be in a hot environment. MAtt, and Wizard both suggested placing the IAT somewhere else.

So here is my post in response. Any insights here?

Would someone discuss please, the impact of location of IAT on things? Why should the actual temp of the intake air right before the TB not be the best one to sample air temp? If the air there is indeed warmer,why would this mislead the ECU into thinking the engine is warmer than it is. I guess I do not understand how the ECU uses this information in the fueling strategy. Is it supposed to be sampling actual air temp as it enters the TB or temp of air outside engine compartment?

1. Can the sensor being in a warm environment create a reading hotter than the air itself?

2. Should not the sensor be placed AFTER the intercooler? It was suggested putting it in the air filter. Would this not then give a false reading any time it went into boost? I don't know. This is an honest question. I am not arguing for one thing or the other. I am just trying to understand the practice and science. I could go look at my Toyotas but they are NA, and not boosted.

Sam
Author:  Rust collector [ Sat Aug 06, 2011 4:37 am ]
Post subject: 
This is also a copy from the other thread, as you posted the same question here. :lol:

I am not an expert, but having the iat after the intercooler makes sense to me, as you are tuning the engine, and the engine gets air after the cooler...

Now, if it is closer to the engine, or the cooler... I would guess it would heat soak more if you put it on top of the engine, than it would on the 'cooler end of the pipe.

When the engine is running, I would think the air moves fast enough, and the temp. change while going thru the pipe would be small.

Well, it makes sense to me anyway, hope someone that knows their stuff wil correct me if I am wrong :wink:
Author:  Sam Powell [ Sat Aug 06, 2011 5:07 am ]
Post subject: 
Thanks for the thoughts. Question is, can the temp of the sensor itself effect reading? And, is the sensor supposed to be reporting outside air temp, or actual intake air temp?

Sm
Author:  Rust collector [ Sat Aug 06, 2011 5:24 am ]
Post subject: 
As far as I know, the temp of the sensor itself will affect the reading a bit, is it "open" on the sensor-ing end?

I believe most cars would use the intake air temp sensor to monitor intake air temps, but I know there are some that uses "outside" air.
I doubt "outside air" temps would be very useful in a turbo car.
Author:  Rust collector [ Sat Aug 06, 2011 6:11 am ]
Post subject: 
How about hooking up the computer, and see what happens when you heat up the sensor with a hairdryer, or something?
Author:  SlantSixDan [ Sat Aug 06, 2011 8:28 am ]
Post subject: 
…or unplugging the IAT and see what happens to the car's behaviour…
Author:  slantzilla [ Sat Aug 06, 2011 10:16 am ]
Post subject: 
My experience with the IAT is on my GTO. Those silly Aussie's thought the best spot for it was in the tube between the filter housing and the throttle body. All well and good, except they placed it right against the tank on the radiator so that it reads coolant temp instead of intake air.

However, it had no effect on starting, it just makes the car a total pig after it gets warmed up. The IAT tells the ECM to pull timing, and when it does the car really slows down.

Put your IAT in the place where it will stay the coolest.
Author:  MoparFreak69 [ Sat Aug 06, 2011 9:56 pm ]
Post subject: 
I would say the ideal spot for an IAT sensor would be as close to the intake valve as possible, while still leaving room for heat soak prevention from the engine heat itself. This way it is sensing what is actually going into the cylinder, instead of a sense of what is happening before the air enters the engine for the combustion process.
On my Magnum V8 engines the IAT temp sender is located in one of the intake ports about 6" before it enters the head. This way the least amount of differential between sense and actual air in cylinder temps is achieved. If I were building the system I would place it about 6" from the head in an intake port to avoid possible heat soak from the engine itself while still providing a pretty accurate reading to the ECM about what is going on inside the engine.
Author:  Sam Powell [ Sun Aug 07, 2011 5:24 am ]
Post subject: 
That is excellent for cruise conditions, but not so great for the start up situation, where the sensor gets hotter than it will be two minutes later. You would think the engine would respond well to having fuel withdrawn from the hot mix. But not the slant. If the O2 sensor were configured to correct fuel mixture at idle, maybe that would help. With a wide band, you could specify an idle a AF ratio which was good when warm.

With the Accel ECU there was a trim tab for manifold temp. You could adjust the fueling with this by setting a coefficient curve for various temps of the manifold. When the for each temp of the manifold you could specify a different value that would add or subtract fuel. Once I discovered this existence and purpose of this, I was able to tune hot start ups to work well. But there is no such coefficient with MS II version 3.57.

Inspiration! Maybe? Since the IAT temp sender sends a 0-5 (or maybe 0-1?) volt signal to the ECU, one could install a rheostat in the circuit and dial up a desired AF ratio from the comfort of your driver's seat. How would one go about designing such a circuit. Would a simple Rheostat do the trick, or would there need to be other components on a small board of some kind? I am pretty much an electronic dolt. I have no idea what all those pretty little round and square things in electronics do. So maybe someone could add to this discussion.

Sam
'
Author:  GunPilot [ Sun Aug 07, 2011 7:22 am ]
Post subject: 
I think the IAT needs to be close to the intake manifold, in the airstream entering the intake, and isolated from heat sinking from other components. I use the open element GM sensor with a plastic end on it. I do think the sensor will change readings based on the temperature of its base, at least until some air is flowing past it. So the change would take a minute or two and I guess it could affect starting.

All the sensor is, is a temperature sensitve resistor. It does not send voltage, it changes resistance based on temp. So to 'bias' it a variable resisor (VR, rheostat) could be used in parallel with it. No other parts needed.

I haven't tested it but you should be able to connect one wire of the VR to the wire leading to the IAT, and the other wire of the VR to ground. Fire up your tuning software for MS and the coolant temp should change as you wick the VR back and forth.

I'm not at home or I would whip up a quick experiment to test. I did do some testing of the GM sensors a while back and as I recall the cooler the sensor, the more resistance. Or maybe it's the other way around? :oops: :D A quick shot with a meter will tell.
Author:  Sam Powell [ Sun Aug 07, 2011 10:53 am ]
Post subject: 
I am going to prove my earlier statement of my ignorance here. :wink: So you use the rheostat to shunt part of the voltage to ground, on the wire coming from the ECU? Is there any danger of damaging the ECU by creating too direct a path to ground? I would have thought you would put the rheostat in series with the sensor itself. I told you I am a dummy, and do not understand the most basic concepts of how components interact.

I was under the impression the resistance of the sensor goes down as it heats up. The thing is, if this can work, and Lou is right about the IAT sensor effecting the fueling formula, then you can trim the AF ratio with this simple nob on the dash. I have been told pilots do this as they fly, because the air temp varies so much. Actually I was told this by an aircraft mechanic.

Sam
Author:  GunPilot [ Mon Aug 08, 2011 1:26 pm ]
Post subject: 
Yes, exactly. You won't hurt the computer. The computer sends a reference voltage to the pin the sensor is on. Current is very low. Based on the amount of resistance the sensor has, electricity can flow. How much flow is then indicated to the computer as a temperature based on the calibration table. Remember calibrating the sensor, or at least (as I did) choosing the GM sensor as it was already the default calibration? That's what's happening here. That's how you can use other sensor brands (Mopar, etc) as long as you recalibrate the MS to use it.

So, yes, the IAT value does affect fueling as Lou says. Yes, you can affect the mixture by altering the IAT value the computer sees. A reheostat configured as I said will act as a mixture knob, at least to the limits that IAT has authority within the fueling equation.

If you place the rheostat in series, it can only add resistance. If you place it in parallel, it can only subtract resistance. I checked the ultimate authority, the internet, and it looks like more resistance = cooler temp. So yes, in that case, you would want it in series. Looks to me like you would want a part that would give you a range from about 0 - 3000 ohms of resistance.

Think about it this way too - using the CLT sensor - you can use that knob to fool the computer into thinking its warmup enrichment should be on - you can monitor the WUE LED on the front panel for that. Wick the LED on to help start, then turn it back off when running. Just like a choke.

I know for a fact that some aviation EFI controller applications use exactly this approach for hot start priming.

This is what make the DIY EFI fun ;)
Author:  SlantSixDan [ Mon Aug 08, 2011 2:05 pm ]
Post subject: 
I'm still really curious to see what happens if you unplug the IAT altogether to see how much or how little it's contributing to the hard-start problem.

Then I'd be curious to see what happens when you put various fixed resistors across the IAT socket instead of the IAT, thus quickly and easily determining what effect a different IAT reading will have on the symptom.
Author:  GunPilot [ Mon Aug 08, 2011 2:06 pm ]
Post subject: 
Sorry, in the above I was using CLT (coolant temp) instead of IAT (intake air temp) in my example. You would rig CLT to turn on WUE, not IAT. May be an approach for curing a starting-only issue. MS does use IAT in the fueling equation, to what percent of authority I'd have to research. My post above edited.

As an aside, the reason for the adjustments to mixture for airplanes is due to differences in air density as the aircraft climbs or descends. Not so much for temperature, although temperature does affect air density, with is the factor that our IAT sensor is concerned with. The MS does compensate for barometric pressure on startup, but doesn't use it again unless you have a dual MAP sensor setup like a MAPdaddy.
Author:  Sam Powell [ Mon Aug 08, 2011 6:47 pm ]
Post subject: 
Thanks a bunch GunPilot. I will get a couple of rheostats,a and try a few of these approaches.

Sam
Page 1 of 2 All times are UTC-08:00
Powered by phpBB® Forum Software © phpBB Limited
https://www.phpbb.com/