Air Contamination of Cooling System 1

[Autotecheng]

Hello, my question is about air in engine cooling systems.

Coolant is formulated to resist boiling, and furthermore kept under pressure to increase its boiling point. I think this is to control its phase. In other words we must prevent coolant from boiling (vaporizing) because vapor does not conduct heat away from the engine very well, right?

What I want to know is how air affects the coolant's tendency to boil, and does it reduce its boiling point or reduce its ability to conduct heat?

There are tools available for draining and refilling the cooling system that work by drawing a vacuum of about 26inhg and "sucking" all of the coolant out, then using the vacuum to draw the new coolant in. Their manufacturers claim that cooling systems can have air pockets trapped in high points in the engine which lead to localized overheating. The highest points of the cooling system generally seem to be the water jacketed areas around the combustion chamber where cooling is most important.

My gut feeling is that if the normal expansion and contraction of the coolant pulls air into the system from a leak or from a low reservoir level that the air will dissolve into the coolant when the system increases in pressure, and migrate into the engine, where it will separate and dwell. My intuition says that air contamination will increase the tendency to boil.

Based on that I'm thinking I should pressurize the system and cap to its rating to check for leaks, but also drain and refill it with fresh coolant and check for leaks a second time under vacuum as the best maintenance practice.

What do you think?

 

[MikeHalloran]

I'm guessing the subject came up because of an overheating incident; you give no other clues about your motivation.

You can order a radiator on the Internet, and it will probably be there the next day. You also get a good price. ... or at least a better price than you get on the highway, so I'd go ahead and replace it now. It's probably partially clogged, and the tubes in a modern radiator are so narrow that you can't rod them out, and the stuff that clogs them seems pretty much insoluble in anything that won't also eat metal.

I'd replace the thermostat, too, just because they're cheap.
Also any hoses that are difficult to replace on the highway, like some short bypass hoses.
Don't overtighten the clamps. I just snug 'em up. ... and use Permatex #2 on the nipples.

Vacuum filling may or may not be needed for your car. If it is, a statement to that effect and/or some complicated bleeding procedure will be in your owners manual. I usually do it even on a car that doesn't need it, because it gets out the last little bit of air that takes days to float up with just a gravity fill, so it saves topping up after a change.

As for confirmation of the deleterious effects of air, I'd just point to the presence of overflow/recycle jugs on just about everything nowadays. They're not cheap, and they're not required by law, so there is a clear benefit to the manufacturer. Be sure to clean the gunk out of that while you have the coolant out of the system.






Mike Halloran
Pembroke Pines, FL, USA

 

[ivymike]

trapped in high points in the engine

Not to a significant extent in any engine I'm familiar with - the head and block guys go to a great deal of trouble to make sure air finds its way out.

 

[MikeHalloran]

Yeah, they do.
My '95 Z28 LT1 had two brass screws that had to be opened a little to bleed the air out of some odd high spots, or the 'reverse flow' cooling system would not work right. Pulling a vacuum at the pressure cap was quicker and more effective.

Luckily, Chevy abandoned the reverse flow fad after a few years, and got more power out of the LSx series engines to boot.

But if you decide to fix up something like a boat-tail Impala SS, it's important that you know how to get the air out of the coolant. You might get a good deal, buying from someone who doesn't.







Mike Halloran
Pembroke Pines, FL, USA

 

[MiketheEngineer]

Had a Le Car with about a 12 step process to get all the liquid in and air out. Got to love those French engineers!!

 

[BigClive]

The early model Oz Holden Commodore V6 needed to have a screwdriver (or similar) wedged under a radiator hose in an almost totally inaccesible spot around the back of the engine to bleed air out of the cooling system. I seem to recall that V12 Jags had quite a few bleed screws as well.

 

[crerus75]

I have one of those boat-tailed Impalas (almost), a 95 Caprice 9C1 with the LT1 engine. It was a NJ State Police cruiser and was rode hard and put away sweaty a lot. I had a lot of trouble bleeding it, even when I followed the conventional bleeding wisdom for LT1s (warm it up, jack up the front end, open the bleeder screw on the thermostat housing to purge air). I suspect there was a lot of scale and sludge in the cooling system that would trap air bubbles. I could never get all the air out until I vacuum-bled it. I epoxied a Schraeder fitting to an old coolant tank cap. I had a stainless steel soda keg that I used as a vacuum/coolant reservoir, and a hermetic compressor from a scrapped dehumidifier as a vacuum pump.

The keg has an inlet and outlet with proprietary fittings, but I threaded them for NPT threads. The outlet has a dip tube that extends to the bottom of the keg. Connect the pump to the inlet and the pressure tank adapter to the outlet. Put a quart or so of coolant in the keg and pull a vacuum on the whole mess. This releases any trapped air bubbles in the system. When the radiator hoses collapse, disconnect the vacuum pump. The vacuum in the cooling system will draw coolant back out of the keg since the dip tube is immersed.

This is a home-brew version of a commercial vacuum bleeding system I'd seen. I just built it out of junk I had laying around. I use a similar system for brakes but instead of the keg there's a glass olive jar with no dip tube. I pull a vacuum on the brake bleeders (one at a time) and keep topping off the master cylinder until I get fresh fluid in the olive jar. This pulls all the old oxidized, heat-cycled brake fluid out of the system at the lowest point so you don't drag it through the ABS accumulator.

 

[mechanic6]

99% of all the engines i've ever worked on(lots) have never had problems with air in the cooling system.
so its not something i'd worry about.

i would worry about applying a hard (-26" HG) vacum to any cooling system.

 

[crerus75]

It wasn't a hard vacuum. At home I limited it to what the hermetic compressor or my old worn out carbon vane vacuum pump would produce (15 in hg.). Notice I said, "When the radiator hoses collapse, disconnect the vacuum pump." Collapsing the radiator hoses does not require a hard vacuum.

When I was a full time auto technician, we routinely used a commercial version of the machine I've just described on just about every cooling system flush and fill. It did not require the system to be burped and was faster than waiting for a thermostat to open. Never a problem, and we serviced passenger cars, light trucks, and commercial fleets of small (Ford E350 chassis) buses that ran stop-and-go routes through town. In other words, a variety of vehicles, some of which were run under conditions that would be listed as severe commercial service. Never had a problem with vacuum-bleeding a cooling system. If you haven't run into a problem bleeding a cooling system, especially a scaly one that sees a lot of short-trip driving or that is minimally maintained, one might suspect you haven't worked on as many engines as you think.

 

[ivymike]

maybe part of the problem is in the definition of what counts as success? It has been my experience from the equipment/vehicle design side that:
1) if an engine needs vent lines installed somewhere then there are ports at those locations (turbo water jackets, egr cooler, top of aftercooler, t-stat housing, etc) - but a naturally aspirated engine is commonly designed without any internal air traps at all.
2) there can be no external air traps (where the main or vent hoses have humps, etc - they should all slope slightly upward to the expansion tank or engine over the whole length, depending on the line)
3) in your installation design, if you have the vent lines in the right places and you fill the engine from a hose flowing 5 gpm, then after running the engine for a few minutes to burp it you add coolant to refill it and the total "make up" coolant is less than 12% of the total system volume then you're okay on air removal. If your make-up volume is >12% of total volume then you've not put your bleeds in the right places or they're undersized.

Having successfully done the above, there would be no need down the road for someone to use a vacuum gizmo during a coolant change provided he adds the coolant at less than 5gpm after the change, and runs the engine to "burp" it before topping off. Having a significant amount "scale" inside the cooling system (enough to block your vent lines or internal passages, for example) would be a problem of its own and should be addressed directly rather than accommodated.

 

[patprimmer]

I'm with ivymike on this.

I would also add, that if the top hose has a bump that traps air, it can mostly be removed by pushing the bump by hand or squeezing to burp.

I also just use a trolly jack to raise the top hose to header tank position in difficult cases. I have never had to do more than that.

Regards
Pat
See FAQ731-376 for tips on use of eng-tips by professional engineers &

http://eng-tips.com/market.cfm

for site rules

 

[crerus75]

Mike,
I agree with you on the design aspect. However, having seen it from the other side, there will always be a few (not many, but a few) cars that don't "burp" easily-- or possibly at all. Yes, scale should be taken care of. So should sludge, which tends to restrict heater cores and other small passages and can lead to trapped air pockets.

In the real world, however, you will always have customers that don't want to do it the right way. "Just put a new water pump on, and please reuse the coolant. I just changed it last year, thank you very much." Do you refuse? If a small but not insignificant portion of your customers are either poor or fixed-income elderly and have to keep old heaps running (I've worked at a garage like that), maybe it's not in your (and their) best interest to always insist on doing it "the right way." If you work in a higher-end specialty shop (again, I've been there) you can perhaps pick and choose your work more fastidiously.

Or how about a customer that's keeping an old cooling system together using Barr's Stop Leak or some such? I've seen heater cores and radiators plugged with this stuff. The reduced flow can trap enough air that the car won't burp in a conventional way. Do you sell the customer an aggressive cooling system flush which may remove all the band-aid-in-a-can that keeps their system from leaking like a watering can? Or do you just vacuum-bleed it and send it on its way? As long as the customer understands the risks inherent in each choice, which one benefits them the most?

What happens if the cooling system is badly designed from the factory? What happens if the customer modified it in some way, perhaps adding an aftermarket thermostat housing that points the upper radiator hose at some ungodly angle that Archimedes himself could not have envisioned? What if someone added an oil-to-water cooler and the lines run uphill just a little? What if it's a motorhome and the coolant lines are long enough or badly oriented enough that burping is a colossal PITA? Will a vacuum bleeder alleviate or solve these problems? Maybe.

Ultimately, a vacuum bleeder is not necessary in most cases. Yet they are handy to have, because every once in a while a cooling system can be a real dirty name to get purged of air, for whatever reason. Since we don't always know which ones will be difficult, and since the vacuum bleeder is (or can be) faster to use, we tend to use it quite a bit. Is it necessary? No. Is it nice to have when you're trying to make book time, when you're working on a clapped-out "honey wagon" with a cooling system salvaged from the wreck of the Bismarck, or when or the customer's waiting in the office? You bet.

 

[patprimmer]

crerus75

You infer a good point not previously mentioned I think.

When replacing coolant or doing a fresh fill, the heater should always be turned full on.

Regards
Pat
See FAQ731-376 for tips on use of eng-tips by professional engineers &

http://eng-tips.com/market.cfm

for site rules

 

[jcd06]

My feeling is that air will rather migrate being mixed with the coolant flow and will separate after engine shut-off, possibly getting trapped in some highest points.
How easily it will bleed away through some clever placed passages or how easily it will get flushed away when coolant flow re-starts depends on engine design.
I completely agree with ivymike that this has been thought of, at least when considering “modern” engines.
Modern engines were not necessarily designed in “Modern Times”

As mechanic6 and crerus75 already warned, I would be very prudent with vacuum.
Engine gaskets in contact with the cooling system are not designed to withstand vacuum.
Especially o-rings because they are lodged in a groove that is wider then their core.
They got pushed inwards or outwards depending on the direction of the pressure difference in normal working conditions.
They might start leaking if you suddenly force them to squeeze in the other direction.

About the main question how air affects the coolant's tendency to boil I have no clue to be honest.
I think that different aspects play a role.
Well designed coolant liquid will be formulated to dissolve as less as possible gasses and I expect the amount that actually gets dissolved so small that it will have a negligible effect.
Up to the boiling temperature of the cooling liquid, the amount of air –dissolved or not- will expand more than a same amount of liquid. This is not going to make cooling any better.
Depending on their position, formation of air pockets can cause hot spots that promote coolant boiling.
But that’s probably not the whole story…

When we all think alike, no one thinks very much. ---Walter Lippmann

 

[crerus75]

A properly designed cooling system should be able to withstand a vacuum. One of the functions of the radiator cap is to allow the hot, expanding coolant to flow out of the radiator and into the overflow tank, and to conversely allow coolant back into the radiator as it cools off and contracts. If the radiator cap does not allow coolant to be pulled back in from the overflow tank (which happens if the cap is defective or the line to the overflow tank is plugged), the system will be under considerable vacuum. This can be enough to collapse the upper radiator hose, a fact that surprised me the first time I saw it. The cooling system must (should) be able to withstand this, as it is (should be) an anticipated failure mode.

 

[JayMaechtlen]

Not all heaters control coolant flow, though some may still.
Some applications merely divert air around the core, so the coolant can flow unobstructed.
Takes one failure point out, anyway.
fwiw
Jay

Jay Maechtlen

http://www.laserpubs.com/techcomm

 

[jcd06]

Perhaps this contains the information that you are after:
Luftabscheidung aus Kühlkreisläufen von Fahrzeugmotoren (Rüger, F) - Automobiltechnische Zeitschrift volume 73 (1971)
There's one on Amazon.de but not cheap


When we all think alike, no one thinks very much. ---Walter Lippmann

 

[turbomotor]

Hi Autotecheng:

You may want to reconsider the idea that boiling coolant is not good or desired. See attached info on nucleate boiling. Nucleate boiling (as opposed to film boiling) typically improves heat transfer, possibly due to the convective flows created by the boiling, and possibly due to the phase change that occurs at the hot spots in an automotive engine head. The vapor may recondense elsewhere in the coolant system where the heat flux is much reduced, thus increasing the bulk temperature of the liquid phase in a cooler area of the system.

I think many of the modern high bmep automotive diesel engines with aluminum cylinder heads require nucleate boiling to cool the hot spots in the head at maximum load.

See attached articles that can explain much better than I can.

Regards,

Dick

http://cdn.intechopen.com/pdfs/1335...ng_flow_in_automotive_engine_applications.pdf

http://en.wikipedia.org/wiki/Nucleate_boiling

 

[rmw]

I didn't read the last few posts so forgive me if someone already said this.

You forgot to ask Henry about it. Henry's law deals with the soluability of air (actually gasses) in water. Automotive coolants are somewhere like 50% water in most cases and the part that is water is still just that, water so it acts like water.

So Henry's law says that water when it is cold has more ability and propensity to dissolve air in it and when it heats up it wants to give up that air - the air wants to come out of solution. Keeping the water under pressure retards that release.

Therefore, any air trapped in the engine cavities will dissolve into the water to the extent that Henry will permit for the temeprature and amount of air already dissolved in the water (or not.) In other words, if it is not already saturated with air for its temperature, it will absorb more air.

As the engine begins to run and the water heats up it is longing to give up the air but prevented by the pressure created by the water pump and when it comes across the thermostat (pressure drop point) it does so and the released air travels downstream to the radiator - and mother nature does her thing and it finds its way to the top.

Simultaneously the volume change of the water (plus any volume of released air) is forcing water out of the radiator cap into the reservoir where the air at the top of the radiator leaves with it and is released. The water that is cooled in the radiator now has the capability to absorb more air once it enters the engine and will do so to the extent of the temperature and pressure (google for tables), especially when the engine cools.

Long story short, as the water works its way around the loop repeatedly, it gives up its air on the hot side and unless and until new air is introduced into the system, it will remain fairly well deaerated. The surface of the water in the reservoir tank is always absorbing air, so some does get back in, but it is my belief that after a while the amount of air in the system overall is nil and the water is fairly deaerated.

rmw