kurt_1

This may have been covered before but since cold-air-intakes are all the rage and intake air temperature is often hotly debated - especially when it comes to dyno-runs - I thought this might be of interest:

I was interested in the real temperature of the air entering the engine so I tried to measure it as close as possible to the heads.

A platinum temperature resistor was fitted to the tip of a tube (see pic, left side) and inserted into the intake via the vacuum connection at the side (right side). The resistor was placed about 1 inch down the trumpet to cyl #2 , not touching any metal.

A second resistor was attached to the side plate of the intake.

Then I went for a 1 hour drive that included autobahn and road stretches with few and only brief stops in between.

The result can be seen in the plot below. With this set-up it takes about 1 hour for the intake to reach a stable temperature of around 65 degC with some smallish variations due to load and speed. The air reaches about 50 degC after 30 minutes and fluctuates a bit more with load: high speed stretches reduced the temp down to 45 degC - and probably would have reduced it further but I left the Autobahn for a regular road.

Normal driving on out-of-town roads resulted in a stable 65degC of the intake and around 52 degC of the air.

I have to say that I didn't expect the air to become so hot and taking so much time to reach a stable temperature. With this in mind, it seems plausible that cooling of the intake before and during dyno runs can have a considerable effect on the max amount of ignition advance and thus the measured torque.

I don't know how this translates from my later intake to the earlier ones but I guess it s not too different.

Any thoughts on how to increase the cooling of the intake and effectively reduce this pre-heating of intake air?

_____
Kurt
90 S4 satin black metallic
928 OC Charter Member

SeanR

I think that you are not going to be reading intake air temps at that location as it is going to get heat soaked from the engine. I'd put the sensor in the path of the airflow in the intake tubes instead. Porsche did a pretty good job of making a cold air intake as it is stock.

kurt_1

I understand the difference between the air temps which I measured and the ones further upstream, e.g. in the intake tubes ( or in the airbox, where earlier cars had the temp sensor).

But to my understanding it is the actual temperature of the air entering the combustion chamber that is a critical factor and that temp was what I tried to come close to. Of course there will be some more heating down the trumpet and then probably some cooling due to the fuel vaporization after the injectors but I would like to have less heat put into it by the intake manifold.

_____
Kurt
90 S4 satin black metallic
928 OC Charter Member

John Speake

Jim Corenman has been running ST Alpha-N on his GT for some time. Alpha-N needs an airtemp measurement to determine air mass. His airtemp sensor is in the airbox. Same location pretty much as the 16v & early 32v LH2.2 engines 84/86.

The readings are usually only a few degrees above ambient as long as the car is moving. When stationary, temps rise quickly to 40degC or more above the ambient. If then stopped with the hood closed, the temps stay very high for a long time. Even with hood raised it takes 30minutes or so for the temp to drop. We use a fast temp sensor, the one used 84/86 is terribly slow.

As Sean suggests, all the indications are that your sensor is measuring the temperature of the metal surrounding it, the the air flowing over it..

What is the time constant of the sensor you used ?

worf928

Heat soak at any point in the intake tract is going to be inversely proportional to flow. The slower the flow the more heat is transfered from the engine to the air.

I suspect that reading the temp inside the intake will result in poorer temp measurement accuracy - or at least much slower response - at idle and part throttle: vacuum doesn't conduct heat very well.

But as a result, measuring temp at the air box may be best. It would be interesting if Kurt would repeat his runs with a second thermistor of the same type and calibration in the air box. Comparing the plots from the two sensors on the same run would be interesting.

kurt_1

my sensor is quite slow. I haven't determined the actual time constant but I guess it is in the order of several seconds, maybe even 10s depending on air flow.

I had done some measurement at the airbox,too, and got into the same ballpark figures as you mentioned from Jim.

I understand the purpose of the air temp sensor for determining air mass but since my interest was in terms of max ignition advance (and how much heat is already getting into the combustion chamber via the intake air), a fast response time of the sensor wasn't that important.

_____
Kurt
90 S4 satin black metallic
928 OC Charter Member

kurt_1

Absolutely, I could see the temperature rise quickly when coasting.

Agreed, this is not a figure to use as fast input for the engine management. But it wasn't intended to do that.

Depends what you want to see. Under comparable conditions, I think my measurement can be a useful benchmark when experimenting with insulation or cooling (if I find a way to do that) of the intake manifold.

At the moment I can't as I broke me second sensor during the tests. I may have to get another one.
_____
Kurt
90 S4 satin black metallic
928 OC Charter Member

Leon Speed

Kurt, I did something similar a few years ago. I simply took my digital weather station inside the car and placed the remote outside temp sensor on top of the air filter. I got about the same temp readings you got.

The Fixer

Kurt,

Thank you for the data.

Thumb ruler: Every 10 degrees of cooler air equals a 1.5% gain.

To answer one of your questions, Hood Venting will reduce engine bay temps
and thus air temps entering the motor.

I have noted a nice decrease in temperature overall in all conditions because i have routinely checked my motor temp (bad temp gauge) in the same locations with the same IR gun long before and after i vented my hood.

I can see all the hot blurred air escaping the bay while sitting at a light and feel the volume of hot air with my hands when getting out after a drive. It just keeps pouring out when parked.

When moving, this hot air is sucked from the bay at a higher rate because of the low pressure location on the hood. It works better than i had imaged it would. Remember Porsche also did this to the 1984 Daytona lightweight but made the venting area larger than i did.

So how fast have you gone on the Autobahn?

kurt_1

Thanks for the pics, that looks interesting and roughly what I had in mind!
I recall the graph of the pressure distribution over the car and the location you placed your vents is certainly ideal.
However, I doubt I can bring myself to cutting up my hood for experiments. Maybe I can get a cheap used one for testing
I also remember the pics in Project 928 showing the extended gaps between hood and fender on one prototype. But that would require even more modification to the fenders - and both of them!

I will first see if insulation of the intake has any effect. I am not sure but I had the feeling that the intake temp went up temporarily when the fans came on. That blast of hot air might have an effect - or it may not so I need to figure out how to do this NOT using flammable material...

The longer stretch was between 230 and 240 km/h, I tried to keep the load as constant as possible to reach equilibrium temperature, which didn't quite work.

This bit of Autobahn is my test track as it's close by and always pretty empty. But it is also two-lane only and a bit curvy so this is about the max you can do safely.

_____
Kurt
90 S4 satin black metallic
928 OC Charter Member

mark kibort

These vents are not vents at all, because this is in the beginning of the high pressure zone so, these vents are now intakes.
I don't think that you can lower the intake temps in this way. the intake is fed by high engine temps, but take solace in that the air going into the engine has very little transient time to heat up. if you do measure intake temps, in the air flow, you might be pleasantly surprised how cool it is. (at WOT ).

kurt_1

I am not sure about that. If I see it correctly in the pics, they are still in the negative pressure zone between Points 8 and 9 in the attached plot.

That was why I took my air temp measurement in the air flow and found the air temp only to slightly drop at WOT and then staying roughly constant (and un-cool) at the conditions most relevant to my driving - steady loads of 80-90%. So I would like to reduce that to for more ignition advance and less knock.

I agree that most likely the intake is fed with engine heat directly and insulating it from the air may not yield any improvement. I recall that in the days of carbs there used to be an insulating plate between the heads and the carbs . This would help here too, but I am not sure if it is possible to fit something between the intake and the heads. Maybe the earlier intakes with rubber sleeves are less prone to transfer so much heat ?

_____
Kurt
90 S4 satin black metallic
928 OC Charter Member

NickT

This topic is so persistent

https://rennlist.com/forums/11474091-post165.html

bureau13

The stock intake tubes do get air from in front of the radiator, thankfully, but the inlets seem perfect to be fed by some kind of scoop or NACA duct in the hood...I've never seen anyone do that, but wouldn't that provide a significantly cooler source of intake air?

NickT

Cruising under power with the stock tubes produces about a 10 degree differential between ambient and MAF intake, as show in the chart above... I can produce more charts on this than you ever want to see...

So the question is... will 10 degrees really make a difference? (no)

To Marks point above "if you do measure intake temps, in the air flow, you might be pleasantly surprised how cool it is. (at WOT )". I have done that above.. and yes, I am surprised at the results... Based on my measurements I see no point is changing anything, unless it's just to look cool...