Mailing List lml@lancaironline.net Message #25979
From: <Sky2high@aol.com>
Sender: Marvin Kaye <marv@lancaironline.net>
Subject: Engine Cool-Ada
Date: Mon, 04 Oct 2004 13:26:36 -0400
To: <lml@lancaironline.net>

To the impatient,

 

First, much thanks to George, Walter and Tim for an extremely informative presentation and facility tour.  Special thanks to Mike for organizing the un-organizable Lancair special interest community – those hot heads that want cooler heads.

 

The basic thrust of the presentation:

 

Under the assumption that enough cooling air is present and adequately departing the bottom cowl, the presentation focus was on making sure that the cooling air was passed thru the cylinder head fins.  The majority of the heat is generated at the top couple of inches of piston travel – thus one must pay attention to using all of the fins at the top (upper) portion of the cylinder to dissipate the heat generated in the cylinder.  Remember that there are conditions that affect how much heat is generated by the combustion event, FA ratio, peak pressure timing, power, etc.  Again, this discussion will be directed to making the engine baffling work to maximize heat removal by making the available air reach all the cooling fins.

 

The following also assumes that the cooling is top down with the engine in a tractor configuration.  A primary difference between Continental and Lycoming is that the valve push rod tubes are on the cylinder bottom for Continental and the top for Lycoming thus making some bottom baffling requirements more challenging for the Continental (in my opinion).

 

A) Many baffle setups do not completely route cooling air around the fins on the lower portion of the cylinder even though they look pretty from above.

 

B) Some arrangements for other uses of cooling air (intercoolers, oil coolers) may block or create low pressure areas that rob the cylinder head of adequate airflow.

 

C) Both Lycoming and Continental cylinders have a “flat” side that does not allow air to flow to the lower fins because of the baffling, except at inter-cylinder locations.

 

D) Some cylinder locations can be super-heated by infrared radiation from the exhaust stack and the impact is fourth order as the temperature of the exhaust increases.  This condition must be examined for appropriate shielding.

 

Yes, readers will have to remove both the upper and lower cowling to evaluate the performance of their baffles.

 

The slide below depicts measured cylinder head temperatures (I don’t have the notes on exactly what the conditions were) around the cylinder for various cylinders (Bonanza). Note the temperature spread around the cylinders and the difference between some.  CHT probes are located at the cyl bottom – seemingly the hottest location.  Also, note that more even cooling around and between the cylinders would be less stressful on the engine.

 

 

Specifics for (A):

 

The picture below (blurry that it is) shows a baffle addition with a fence to better direct air to some bottom fins.  It is your job to figure out what cylinder it is for.  This typifies some of the cylinder areas that have to be examined.

 

 

 

 

Specifics for (B):

 

The oil cooler below deflects a great deal of cooling are over the top of the cylinder behind.  One recommendation would be to put a “curved wing” about 3/8” above the top of the cooler to direct some air flow down to the face of the cylinder.

 

 

Another problem depicted below is the creation of a low pressure area near the cylinder because another air flow user is drawing off a great deal of air.  George Kool-Aide Braly tried a scoop (or roof) behind the cylinder to direct air down but says that did not work because of the low pressure.  Instead he built a vertical fence along side the intercooler that was baffle sealed to the upper cowling and extended just forward of the necked down part of the intercooler.  Cylinder cooling was restored.

 

 

 

Specifics for (C):

 

The above picture also demonstrates the problem described in (C).

 

Below shows the “flat spot.”  This is troublesome on the fwd right cyl and back left cyl on the continental engine and the fwd left and back right (??) on the Lycomings – at least it is on my 4-cyl 320.  Troublesome because air cannot pass from above the spot to below without some baffle work.

 

 

The picture below shows an additional hole to get air to the lower baffle although George recommends a “roof” (maybe even with sides) is used to direct the airflow down.

 

 

See the opening?  See it behind the B-lead terminal on the alternator?  Two pix back one could work that in below and behind the landing light.

 

I hade modified my 320’s cyl two baffle to better allow air to the fins at the lower front quadrant and cyl 2 is no longer the hottest.  Maybe I can get a picture of that mod posted soon.  It is time to fix the baffling behind cyl 3 to cool that one better also.

 

Specifics for (D):

 

Darn, no picture!  When you look underneath the engine, you will see that some baffle extension will be needed on fins adjacent to the exhaust port.  It is useful to enclose the extension with full side walls to ensure the air doesn’t spill out before cooling the fins.  Another benefit is that the side nearest the exhaust also becomes a radiant heat shield so the fins aren’t additionally heated by the nearby exhaust pipe.  The exhaust system should be examined for where radiant heat may affect components, including the cylinder head.  Some things can be shielded with aluminum on the “thing” or a stainless steel standoff shield may be attached to the exhaust pipe with a SS hose clamp.

 

It’s a beginning.  Good luck and stay cool.

 

Thanks again to George, now known to me as Mr. Kool-Aide from Ada.

 

Scott Krueger AKA Grayhawk
N92EX IO320 Aurora, IL (KARR)

Some Assembly Required
Using Common Hand Tools.
Requires hot and cold running air.
 
 
Subscribe (FEED) Subscribe (DIGEST) Subscribe (INDEX) Unsubscribe Mail to Listmaster