flyrotary@lancaironline.net Mailing List Archive

James,

As Jeff pointed out, we are essentially trying to do the same thing that Mazda did in the car with respect to removing air from the system. As I see it, there are two very different conditions under which we are trying to eliminate the air. Condition 1 is during the initial filling of the cooling system without the engine running. Here we are trying to eliminate the air that has accumulated in high spots of the system. Condition 2 is when the engine is in operation. It is unlikely that there will be large pockets of air just in high spots now. Instead, we are trying to divert a small fraction of the coolant flow from the rapidly moving stream (somewhere in the neighborhood of 15 ft/sec for a 1" dia passage) into the expansion tank. Here, we are trying to establish a low velocity and long residence time to allow the tiny bubbles of air to separate upward before the coolant is taken from the bottom of the tank and reintroduced into the main coolant stream again. The air will eventually all be trapped in the expansion tank, similar to the particle collection action of a bypass oil filter. Under this second condition, the locations of the connections of the lines going to the expansion tank are more dependent on relative pressures within the system such that the desired flow is obtained rather than being in the high spots of the system. The goal is to have a set of connections that works under both sets of conditions.

In Jeff's drawing, the water pump can be considered to be part of the engine block with respect to air removal when the engine isn't running. It is more difficult to come up with a drawing to describe pressures in the system under operating conditions, particularly if a bypass type thermostat is included.

I chose to locate the connection for the "air bleed" line in the water pump housing just below the thermostat. This is not necessarily a high point in the system, but works for condition 2. My Schrader valve at the high point above the thermostat takes care of condition 1.

It seems to me that Evans coolant is not often used in aviation applications because of its lower specific heat compared more common antifreeze mixtures rather than considerations related to thermal expansion rates.

Steve

From: Rotary motors in aircraft <flyrotary@lancaironline.net> on behalf of Jeff Whaley <flyrotary@lancaironline.net>
Sent: Friday, January 30, 2015 8:13 AM
To: Rotary motors in aircraft
Subject: [FlyRotary] Re: swirl / expansion tank configuration

Hi James:

Yes, we have discussed Evans Water and other items on this list before – the general consensus is there is no advantage to these products over 50/50 glycol/water.

I think what Mazda used in the car is very similar to what we are using in aircraft applications except we have modified our coolant systems (water pump outlet and inlets) largely because we use non-standard radiators and also in many cases to reduce overall engine height . The connections you have - we don’t, so we made our own.

The air-bleed line in my system is normally connected to the intake manifold and protrudes up out of the iron casting closest to the transmission. The other line in my diagram is a connection added to a custom filler port above my water pump outlet. Both lines are effectively doing the same job – removing trapped air from the block; in level flight they are both about the same height above the block but during climb and typically highest engine performance, the block connection is at the front and therefore the highest point. The stock Mazda water pump housing had a filler neck/pressure cap and overflow port built into it; I had to remove all that to turn the outlet port 90 degrees.

Jeff

From:	James R. Osborn <flyrotary@lancaironline.net>
Subject:	Re: [FlyRotary] swirl / expansion tank configuration
Date:	Thu, 29 Jan 2015 16:38:23 -0800
To:	Rotary motors in aircraft <flyrotary@lancaironline.net>

You guys are really confusing me! Granted all my rotary work so far is on my car. Once upon a time, my car had an AST (air separator tank). It was a small round job next to the radiator, maybe 3 inches OD, 3 or 4 inches tall. I’m thinking it served similar purposes to the swirl tank in your designs? Mazda made it of plastic and it had a nasty tendency to burst after aging in the heat soaked engine bay long enough. Most Rx-7 owners including myself, removed the AST system - others replaced it with aftermarket aluminum versions. As I recall it had a tube at the bottom that went downwards to an extra nipple on the bottom of the radiator, but I forget where the input came from. I am wondering how this system compares to the ones you are describing? Where is the pump in the last figure (Steve vs Jeff)? And where is the “airbleed hose” bleeding air from? Is the idea that it is just a high point in the system that continuously is drawing out any air bubbles that get created or trapped in the system?

I have to ask another related question - why not use Evans waterless coolant? I think we discussed it a bit before. It seems like the expansion/contraction is a lot smaller because it boils at a much higher temperature than standard coolant/water mix. I run this in my car now, but my car is a project car that hasn’t gotten enough serious abuse to have more opinions on the efficacy of Evans. But I believe that Rotax engines use it - so aircraft use is not beyond the pale. I am just wondering if all this (seeming) complexity in the cooling system arrangement would be needed if a waterless coolant were used.

— James

image001.gif

image002.gif