mx5cartalk.com

Following on from a conversation in a few threads it may help others with the same issue, getting cool air to the ITB's.

http://mx5cartalk.com/forum/viewtopic.php?p=806193#p806193

gslender wrote:

Coolant oscillates between 88.8 through to 91

Coolant oscillates between 39.3 through to 39.8

(From my other post)
What my data is showing that there is roughly the same separation between the external air temp (located in the mouth) and the sensor on the ITB's. These sensors are the same make so any error would be the same. The main issue appears when the car is not moving and there is heat soak however the coolant is falling due to the fans whilst the car is stopped.

Therefore it could be argued that the hot air from the fans has no where to go so is causing the heat soak when the car is not moving. My data supports this.

In addition the data shows that the outside air temp is also increasing when the car is stopped so the air from the engine bay is also coming out the front mouth!

Green = Water temp, Blue = speed, Purple = air temp, red = intake temp, orange = No1 cyl exhaust gas temp.

First session, note the air intake temp. The car was idling at the start for about 5 mins.
QR TA First Session on Flickr

This was about 30 mins later. There is some heat soak from the first session, plus again about a 5 min idle time.
QR TA Second Session on Flickr

End of the day
QR TA Last Session on Flickr

This is based on the last session. The data between 5 and 55 should be treated as outliers and ignored as there is very few data points. It shows the heat soak when the car is not moving. However there is an interesting pattern in the middle between 110 and 140 that needs a little more investigation. The full data dump is attached as a zip file. Note the oil temp data needs to be taken with a grain of salt as the sensor appears to be showing hotter than the oil actually is plus there appears to be a problem with the wiring as it comes on and off.
Innovate Graph on Flickr

Magpie wrote:Therefore it could be argued that the hot air from the fans has no where to go so is causing the heat soak when the car is not moving. My data supports this.

The 100s of people before you doing the same test have had the same results.

I'm going to first try just directing the radiator wash away from the cold side of the engine.... And then review the logs. If that doesn't bring down the temps, next option is to build an airbox and duct from the front directly, which will work 100% but is a lot of effort.


Sent from my iPad using Tapatalk HD

Let us know how it goes. A quick look at the data after a short run today shows that my new intake has dropped the temp difference between the mouth and the intake.

I'll be working on a radiator front duct during Sep/Oct hopefully I can work something out on the inside as well. I may also swap between the GV Bonnett and the OEM to see what the temps are between the two.

project.r.racing I agree that many people have done the same test and proven the same thing, however few people make available on these sort of forums their results/proof.

With ITB's that get air from the engine bay it is more of an issue. Further, at least for us RHD cars we are restricted in trumpet and filter size unless the brake package is redesigned to be boosterless. In addition, there are different ways that people have installed the air intake for ITB's, sailaholics is different to mine (pipercross dome filter) and whilst gslender is using Jenveys he has chosen to use sockfilters.

Again with the setups of sailaholic, gslender and myself we all have different radiators so again, hence if other people are interested in installing ITB's it is hoped (at least by me) that this thread will assist in making informed decisions.

The purpose of this thread is to exchange ideas and solutions to a problem, in fact some of gslender's ideas I will test namely directing the radiator hot air away from the ITB's (and post the test data). More data will be collect on the NACA duct that I have installed at Time Attack on 15/08 and see if it is a good idea or a silly one as % change from other sessions can be compared.

Other tests that could be done are differential pressures on/under the bonnet to allow an informed decision on the best place for ducts (unlike what I did and put it where it fitted). Yes this has been done 100's of times but how many people have access to CFD software to do something like this http://hanchagroup.wordpress.com/2012/12/05/mazda-miata-consultation/? Again, you could argue that Autospeed http://www.autospeed.com/cms/A_2159/article.html did it without CFD but is it MX5 NA specific, hence repeating the Autospeed on a MX5 NA would be informative.

Radiator ducting inside engine bay suggestions:
http://www.dsmtuners.com/threads/my-ducted-radiator-setup-road-racing.386716/

http://www.miataturbo.net/race-prep-75/aerodynamic-discussion-thread-70612/page19/#post1026958

I'm also looking at the possibility of creating a duct or semi-airbox (but leaving the socks on - but covering up as much as possible) but drawing air from the cowl / firewall and taking cool air from the low pressure area at the window base. I've removed the clutch reservoir and I think I have enough space for a 3 inch hole and it might be better than trying to grab cool air from around the radiator (or through it). All just ideas at the moment.

G, filter socks are about a 30% drop in efficiency I've been told over a remotely mounted panel filter due to them disturbing the airflow at the inlet and negating flow properties of nicely curved trumpets

I have gone through a bunch of data I had from my ITB setup and found that there is no proportional correlation between IAT and engine temperature. IIRC my IAT sensor was set up on the passengers side of the car, near the shock so it was more an under bonnet temperature reading and out of the direct path of the fan flow.

And yeah, filter socks are sh*t.

plohl wrote:I have gone through a bunch of data I had from my ITB setup and found that there is no proportional correlation between IAT and engine temperature. IIRC my IAT sensor was set up on the passengers side of the car, near the shock so it was more an under bonnet temperature reading and out of the direct path of the fan flow.

Ok, so aren't you saying that you measured engine temp and air temp, but the air temp sensor was located in an area that didn't represent where the air being sucked in was located? If so, that's not relevant and largely useless data or am I missing the point?

As stated elsewhere, I'm seeing a strong correlation between an oscillation in engine temp that matches the oscillation seen in the air temps. I'm also seeing a rise in air temp (from ambient) that begins when the engine reaches thermostat temp (ie it doesn't rise with engine temps, it waits until the thermo reaches its opening temp of 88 deg), which then pushes hot water to the radiator and then washes the engine bay with warm air... so that's a compelling connection between the two yeah?

Yes, I've heard repeatedly that socks are crap blah, blah.... but heaps and heaps use them and still get good results, so for now, they'll do until I decide what other option I wish to proceed with.

Pretty graph showing how most of the intake temp rise occurs after the engine gets to temp...



Air/CLT starts at 15/13 deg respectively.

It take 5 mins for engine to hit 88 deg, where air is 20 deg.

Then 10 mins later the air temp is at 40 deg.

I then see oscillation between thermo opening/closing and air temp oscillation (ignoring times when the engine is idle and the whole thing heat soaks)

G

G my data possibly shows the opposite, that is as coolant temp rises intake temp decreases (albeit a very small amount), however it also shows the effects of heat soak slowly building up. The decrease in intake temp could be simply a product of speed and load on the engine causing the coolant temp to increase.

In the picture (End of the day) of which I have posted the data for, in the middle it shows an increase in intake temp and the speed at zero. This was caused by a spin... and whilst the intake temp dropped a little the heat soak persisted.

The cool down lap shows the coolant temp decreasing, but not the intake temp decreasing.

I will do a nice country road run and see if your pattern can be duplicated, plus it could also show when, at least for me when heat soak has stabilised.

Oh come on G.
Lots of people use motec/haltec etc with good results but it doesn't stop you saying the megasquirt is a better option.

Just because everyone does it doesn't make it right.

sailaholic wrote:Oh come on G.
Lots of people use motec/haltec etc with good results but it doesn't stop you saying the megasquirt is a better option.

Just because everyone does it doesn't make it right.

True. Better quickly get rid of my air socks and build that custom air box !!!

G

I have a spare Pipercross similar to Sailaholics you could buy