Engine (shock) cooling what's the deal

[piperboy84]

Last night I decided to climb up from my strip at 250 above MSL to 4000 to practice some stalls and minimum controllable airspeed etc, within a 4 miles radius above my strip, the temp at the field was about 12C on my departure at about 7pm, during the shallow turning cruise climb the CHT gauge was just over half way in the middle of the green which is where it normally is ( the engine is a lycoming 0360 c4f). Upon initiating my descent and heading back to the strip i noticed a cloud cover moving over at about 1500 to 2000 agl I decided to expedite my decent by basically pulling the power back a bit and pointing the nose down and dropping at about 1100 to 1300 FPM to get in under the approaching cloud cover. Levelling out at about 1200 agl to set up for landing i noticed the CHT was right at the very bottom of the green range which got my attention and I chastised myself for not doing a slower fpm circular decent to avoid such rapid cooling.

I understand that the type of decent I did was a dumb thing to to do even if the CHT never got outside the green arc. My question is that even if it did stay within the green arc can rapidly dropping from normal operating temperature to almost falling out of the bottom of the green operating threshold damage the engine?

[AdamFrisch]

For every person you hear talking about shock cooling being bad, there's an equal amount of people saying it's nonsense. Mike Busch, AOPA Pilots resident A&P, pilot and columnist thinks it's hogwash.

I don't know the answer. It's a bit like the rich or lean of peak argument - we will never get a definitive answer, I think.

[peterh337]

I am certain shock cooling is not a myth, but it is an issue only if the CHT is high enough to start with i.e. when the metal is not at its normal strength.

There is a well known article by a gliding tug operator who reported consistent cracked cylinders, until they did a bit of a low power section before the (rapid) descent, and it cured the problem.

A simple rate of change of temperature is unlikely to be a problem otherwise you would crack cylinders due to shock heating during takeoff.

Lyco recommend 60F per minute max rate of change, which is fair enough. The EDM700 warns if this is exceeded and it never is in normal ops at any significant value of CHT. It is easily exceeded just before landing but by then the engine is cool as a cucumber anyway.

When descending, I reduce MP by a few inches per minute at most.

The ROP v. LOP is no longer an argument, BTW ;) It is well established.

[piperboy84]

ROP v. LOP ? What is this

[B2N2]

Mixture settings:
ROP = Rich of Peak EGT
LOP Lean of Peak EGT

http://www.warmkessel.com/jr/flying/td/jd/images/pp18ms.jpg

Source: Pelican's Perch #18: Mixture Magic (http://www.warmkessel.com/jr/flying/td/jd/18.jsp)

[A and C]

Some years back the BGA did a lot of research into shock cooling, their work dramatically reduced the number of cracked cylinders within the glider tug fleet.

I highly recomend the people who doubt shock cooling is an issue take a look at the scientific report, it might change the way some view this issue.

[sevenstrokeroll]

many moon ago I worked for a place that flew bank checks and other high priority papers around.

we were told and checkridden using a procedure called: Stage Cooling. We were to reduce the manifold pressure one inch perthousand feet of descent, timed to arrive at the outermarker at approach power setting. ( lycoming engines on pipers)

now, in real practice sometimes it worked, sometimes it didn't...and many times we had to deal with the controllers strike in the USA and had to scud run (legally) using our wits. AND THAT MEANT cutting to idle and spiraling down through a big hole to a small airport or vice versa.

do your best, be nice to your engine and it will be nice to you...but you gotta fly it the way you see it and you can't SAVE the engine and lose the plane and pilot.

[Armchairflyer]

WRT the picture from Pelican's Perch #18: Mixture Magic (http://www.warmkessel.com/jr/flying/td/jd/18.jsp) shown above, a bit further below the same article states thatThis second and much more serious red area holds true through the entire spectrum of power settings, so it is much more serious. Virtually all factory big bore engines suffer from this uneven power distribution, which sets an artificial limit on just how lean we can run. Uneven mixture distribution causes the entire lean-of-peak-EGT region to become a red zone, not available for use!So actually at least the poor armchairflyer is solidly confused concerning the proper mixture setting (except for taxiing, where AFAIK any degree of leaning does no harm in that almost idle regime).

[peterh337]

A simple approach, which yields nearly all the benefits, is to forget LOP and just use peak EGT.

Around peak EGT, the combustion is stochiometric i.e. optimal, and since power comes from burning juice (gosh, really?) once you are stochiometric, there is no way (to a first degree approximation, anyway) of getting more power for the same fuel flow.

I fly at peak EGT all the time, and don't bother with LOP. LOP is very sensitive to the exact balance of intake air distribution and other factors.

The exact setting is not critical because the power curve is quite flat around the peak EGT point.

Lyco authorise all (I think all) their engines for peak EGT at 75% of max rated power or less. So there is nothing dodgy about this.

Obviously the over-riding thing is all this is cylinder temperature management so one can use peak EGT only in cruise, at cruise speeds.

I wrote this (http://www.peter2000.co.uk/aviation/engine-management/index.html) up a while ago...

The benefit of LOP relies on second order factors... the lean mixture burns slower, which suits low RPM (less friction losses in the engine) because one is stuck with the fixed ignition timing in our engines. One gets about 5-10% more MPG at say 2200rpm than at say 2500rpm. But if one can fly peak EGT at 2200rpm then the gain is really small. I did very careful flight tests, where the TAS was held constant (i.e. constant thrust) and could never achieve any MPG improvement using LOP, over peak EGT. The measurement resolution was about 1%. Most people who fly LOP are getting great MPG gains because they end up flying slower :)

[maxred]

Graphic engine monitors, EGT/CHT analyzers, fuel-flow transducers and digital tachometers give exact readings for every engine parameter measured and can be great aids when trying to troubleshoot engine problems. However, a single-point EGT is the only tool needed to sufficiently lean your carbureted 0-470 engine. This is because fuel and air distribution in the engine's induction system is grossly inefficient and monitoring of single-degree changes and measuring fuel in ounces rather than gallons is unnecessary. Adjusting the fuel mixture for a particular power setting can be easily accomplished with a single-point EGT using TCM Service Bulletin #M89-18 as a guideline. Percent power and knowing where you are in relation to peak EGT is then all that's necessary for adjusting the mixture on an 0-470 engine. (Keep in mind that too much fuel is as destructive as too little fuel.)


That's Continental for you:cool:

[Pilot DAR]

Piperpoy,

I worked for an engine overhaul shop for many years. Shock cooling damage is real, and occurs quickly with careless engine operation. As has been said, 1 inch of manifold per minute, or thousand feet maximum rate to reduce power. Yes, the engines will stand more abuse, but not forever, and it only takes once to badly crack a cylinder or 6. Glider tow and paradrop ops usually have a good grasp of these things.

Think of it this way: You have a cast part with many intricate features, in particular variances in the thickness of the casting where there are fins, heavy areas, and thinner areas (between exhaust and intake ports) These all reach their intended operating temperature during normal cruise, and run hotter during a climb, where the engine is working harder, and airspeed (= cooling) is less. During the time available for the engine to heat up after start, these temperature even out. Normal operation keeps them normal.

You're cruising along, and decide to go down. You carelessly snap the throttle closed. The generation of heat pretty well stops, as the engine is no longer developing power. The cooling, however remains the same, 'cause you did not slow down first. So the effect of the cooling is unchanged, but no heat is being generated to balance it. The areas of the cylinder subject to the greatest effect of cooling, will cool very quickly. Other thicker or more shielded areas take longer to cool. Now you have vastly differing temps in the adjacent areas of the same casting, with some shrinking much faster than others (we agree that cooling aluminum causes it to shrink right?). If a whole bunch of crystals of cast aluminum shrink at different rates, some are going to pull away from others, and you get a crack.

I have seen lots of these cracks, and with some cylinders, you just know where to look. You can tell how well a pilot treated that engine. If you must pull power off quickly, pull the nose up lots first, and slow the plane down. At least the slower plane will not cool the engine so fast, and reduce the risk of damage.

[piperboy84]

Thank you DAR, I appreciate the detailed explanation.

[horizon flyer]

On a Lycoming 360 the magic number is 380F, below this temp research shows that power changes and sudden cooling does not harm the engine.

Above 380F is the danger area, only change one of anything that cools the cylinder heads. i.e. Do not reduce power, richen the mixture and increase airspeed together, will crack the heads over time.

Forget Lycomings max 480F cylinder head temp, this is an absolute red line
it may run, but wont last long.

Cruise temps should be 380f or less for long life and above 400f in climb are into the danger area. Best if in all modes of operation temps stay 380f or less.

Point of LOP is not just saving fuel, but keeping cylinder head temps low using air rather than expensive fuel. Matched fuel flow for each cyl is required to do this see GAMI injectors. Difficult to do with a carb.

Lycoming did issue a service bulletin to reduce timing on 360 from 25 to 20 degres this reduces cyl head temps and does not affect power as over timed anyway.

Note Running at Max EGT is not the max power point, it is 125f down on the rich side of peak.

Lycoming say at 65% power no damage can done by any cooling or mixture changes. But Lycoming do say that any damage to one of their engines is down to the operator not their design and manufacturing faults which are many.

[peterh337]

Note Running at Max EGT is not the max power point, it is 125f down on the rich side of peak.

Correct; and you pay for that. Approx 10% more fuel for the same speed.

Lycoming say at 65% power no damage can done by any cooling or mixture changes.

I haven't seen that. Where is it? It cannot be simply true anyway because you could cruise climb at 65%, heat the pots up to say 450F just nicely, and then cut the power.

I don't like the old "cruise climb" institution; it doesn't achieve anything at all, while delivering high CHTs which are hard to avoid.

I did flight tests, climbing from ~1000ft to ~7000ft, using various cruise climb profiles, using the constant-EGT profile (the best way), and using a full-rich profile. The fuel burnt during the climb was pretty similar for all of them, but the cruise climbs took longest to get there and produced the highest CHTs (actually pretty high values; manageable only with an EDM700 or similar). The constant-EGT climb performance was quite similar to the full-rich climb, but the latter would not work for high altitudes e.g. over 10000ft because the engine would be way over-rich.

I think "cruise climb" is an artefact of the goode olde days when you had no instruments and when also "over square" was not allowed because nobody understood that it is/was a meaningless concept in engineering terms :)

[piperboy84]

My plane is a fixed prop 0360 (vy is 90mph) so what is the best method of climb up to 5k do you think? i.e. cruise 2300rpm at 100mph or full power,full rich at VY

[Big Pistons Forever]

One of the problems with this discussion is the question seems to be binary
Yes shock cooling is a big deal or No it is nothing to worry about.

My personal opinion is for the simple 4 cylinder engines fitted to your typical club trainer/tourer it is pretty much a non issue. The flying school I help out at has permission to go to 3500 hours before overhauls on the C 172 engines (0320 D2J with 2000 hr manufacturers recommended TBO). They live a hard life with lots of idle power approaches frequent power changes in the circuit etc etc and in general no procedures to prevent shock cooling) Most engines make it to 3500 hours with no cylinder work at all.

However I also fly a Cessna C 421 with the 375 hp Geared turbocharged Continental GTSIO 520 engine. The engine manual cautions going abruptly from cruise power to idle power in flight will reduce the engine life by 50 hours :ooh:

These I think represent the extremes. I think you can divide engines into 4 broad groups

1) Simple 4 cylinder less than 200 hp where precautions to prevent shock cooling while nice to practice are not required.

2) Non turbocharged 6 cylinder more then 200 hp engines where prevention of shock cooling is desirable and will likely extend cylinder life

3) Turbocharged engines making up to 350 Hp where prevention of shock cooling will definitely extend cylinder life and is essential to make TBO

4) High power turbocharged engines of 350 hp or more where prevention of shock cooling is absolutely essential to prevent cylinder damage from occurring over quite short time frames and the engine has no hope of making TBO if regularly shock cooled.

However even for the simple engines my philosophy is to be nice to the thing that is keeping me in the air. So I let it warn up before applying power, watch the oil temp and CHT (if fitted) and especially on warm days, avoid high power low airspeed operations to avoid overheating and try to make all power changes slowly and smoothly.

[abgd]

Something that's always bothered me is the effectiveness of short engine 'warms' during pfls and descents.

I can see the point of regularly making certain that the engine isn't icing up and will provide power when required, but does a 4 second burst of power really keep the engine significantly warmer than doing nothing at all? Or perhaps descending with half an inch of throttle (rather than idle) and some sideslip?

[Aerozepplin]

I think "cruise climb" is anartefact of the goode olde days

A relic possibly of radial engines? The power reduction after takeoff in a flat 4/6 is pretty well established in the industry, but have yet to hear a reason other than avague “to be kind to the engine”. I’ve yet to see temperature figures that suggest it is.

I’ve seen quitea few replies to this question that are similar to Pilot DAR’s, an engineer who has found cracked cylinder, and attributed it to shock cooling.

I personally think that shock cooling has been massively overplayed, and that the 1” per minute or the like is silly over complication. However, I’d very much like tosee some real data (I’m dreaming I know) relating engine cooling rates (downloaded from an engine monitor) compared to cylinder life. If there is a connection, I would suggest it is related more to engines being cooled rapidly from a temperature they shouldn’t have been operating at, rather than engines being cooled rapidly from a sensible starting temperature.

Without numbers,the arguments could continue forever. The problem is that cylinders crack, because unless you park you plane in the shed the engine is going through hea tcycles, but how the nature of those heat cycle affect this… that’s the argument. There are well known clusters of cylinder cracking between certain serial numbers on some brands. I think pilots take the rap too often.

[AdamFrisch]

Big Pistons - I also fly geared engines, but the GTSIO are especially highly strung. I always quiz my old Commander mechanic (he knows serial numbers by heart after 30 years of only doing Commanders) if the old GTSIO-powered 685 would be something I should step up to at some point, fully knowing it's his least favourite of them all. Always gets him going. He says they rarely make TBO and often have to have some top work done before hitting 1000hrs. Sluggish take-off as well, even though they have all that power. They do perform like turbines up high, though.

I think the part of the recommendation in the manual you talk about is probably due to gearbox limitations rather than a cooling thing, no? The gearbox chatters and doesn't like to be pushed by air. I always try to fly mine with positive drive and have been cautioned to never pull to idle ever, except in the flare. It sometimes means she's little harder to slow down, but if you dump gear early, it's normally not a problem.

[Tinstaafl]

I manage a PA31-325 Navajo with the Panther upgrade (so a Lycoming TIO-540 350HP engines instead of 325HP). I've also flown C421s (Continental GITSO-520 with 375hp) & a PA31P (also geared, but a Lycoming GTIO-540 with 425HP).

In all cases I never slam anything open or closed. As a rule of thumb my pax. shouldn't hear much of an obvious change of engine sound in flight from one moment to moment. Descents are done with cruise MP/RPM until 'x' number of minutes prior to arriving in the circuit or at the FAF. 'x' is the difference between cruise MP and the MP appropriate in the circuit or at the FAF. The MP difference, if you like. From the 'x' point on I reduce the MP difference to match the minutes to go to the circuit or FAF. For example, in the Navajo, I need 25" in the circuit/FAF but usually cruise at 33" MP LOP. I descend while maintaining 33" so at 7 minutes prior to ETA I reduce MP to 32", at ETA-6 minutes I reduce to 31", at -5 minutes MP is reduced to 30", and so on. RPM stays at cruise setting. On short final, when the prop is on its low pitch stop I set the RPM levers for max RPM.

I had the owner of the Navajo install a JPI EDM when I took over managing it. As far as I'm concerned an EDM should be standard equipment. Using it I can watch every cylinder. For example, the aircraft's POH specifies to lean at climb power to between 28-32GPH. The CHT on a couple of cylinders rises to an unacceptable level at anything less than 32GPH but that's not shown on the factory gauge. Still within the factory limit 460 or 480 deg F. but damned if I ever want to see CHTs that high. Using the EDM, I've found that 32GPH per engine works well to keep CHTs around 400 in the climb.

Leveling off I leave the cowl flaps open while accelerating, then to half then closed, then set cruise power. I use the CHTs as a guide for this.

Cruise at 65% ROP most CHTs run over 400 deg F, a couple as high as 430. LOP most CHTs are under 400 deg, and only one or two at 400-410. I know I'm comparing equal power production because the TAS is the same. I gain on fuel consumption. I'd rather use excess air (ie LOP) to keep the engine cool instead of excess fuel (ROP).

In descents I can see the CHTs very gradually reduce until by the time I'm in the circuit they're quite cool. I do constant power approaches at reducing speed, adding drag as necessary so there is little in the way of major power changes on the approach and by the time of the last power reduction to flare I'm at threshold speed ie slow.

I taxi in at, or near, idle power which gives the last bit of cooling and lets the turbos spool down still with oil pressure. The couple of minutes taxing in corresponds with an EDM TIT 700. Using that as a cluebat, for short taxi ins I wait at idle for the TIT to reduce to 700 deg. In the same vein, if I have to use a higher power settings to manoeuvre then the cool down period restarts. It's really easy with an EDM - all the CHT & TIT information is right there in front of you.

[rapidshot]

I wonder why

Car engines start at the first turn but the aircraft engines dont most of the time, years ago i had a renault 12 with a choke on it, it showed the same behaviour with my 2008 aircraft engine

Car engines i am sure have the same shock cooling, why is it a problem for aircraft engines,

Why change oil every 50 100 hours for aircrafts while this period in cars is around 1 year or 10000 kms,

I really wonder why, is it a problem of additional weight and cost?

[Tinstaafl]

Properly maintained aero engines start just fine with the correct technique. I've had plenty of car engines that were mongrels to start.

Car engines *don't* have the same shock cooling issues - they're liquid cooled with a thermostat to control coolant flow into the engine so that the temperature is relatively constant.

Car engines don't operate at full power (or some high percentage of full power) then 55%-75% for hours on end. Car engines rarely see full power, let alone operate continuously above 55%. My car specifies 3000miles between changes. Conservatively choosing 30mph average speed (and for easy calculation) that's 100 hrs of use - and not even above 50% power for any of it.

[Big Pistons Forever]

Big Pistons - I also fly geared engines, but the GTSIO are especially highly strung. I always quiz my old Commander mechanic (he knows serial numbers by heart after 30 years of only doing Commanders) if the old GTSIO-powered 685 would be something I should step up to at some point, fully knowing it's his least favourite of them all. Always gets him going. He says they rarely make TBO and often have to have some top work done before hitting 1000hrs. Sluggish take-off as well, even though they have all that power. They do perform like turbines up high, though.

I think the part of the recommendation in the manual you talk about is probably due to gearbox limitations rather than a cooling thing, no? The gearbox chatters and doesn't like to be pushed by air. I always try to fly mine with positive drive and have been cautioned to never pull to idle ever, except in the flare. It sometimes means she's little harder to slow down, but if you dump gear early, it's normally not a problem.

Unlike the geared lycomings which uses a planetary reduction gear the Continental GTSIO 520 has one just massive spur gear to provide the reduction in prop RPM. It is pretty unbustable, the problem with this engine is the cylinders. They are basically the same as the 285 hp I0 520. Pulling another 90 hp (Cessna 421) or worse another 115 hp (Commander 685) is asking a lot. High CHT's or shock cooling will just kill this engine. However the last 421 I flew was sold with 1400 hrs on both engines and both had all original cylinders. It will hold up just fine if you treat it right.

[Piper.Classique]

I was flying tugs at Lasham when the cylinder cooling studies were done. Shock cooling does exist, and the changes made to the flight profile made a big difference to our maintenance costs. Does anyone perchance have a copy of the original study, or a link?
Thanks for thinking about your engine!

[peterh337]

My personal opinion is for the simple 4 cylinder engines fitted to your typical club trainer/tourer it is pretty much a non issue.

I think the reason for that is elsewhere:

Most flight training (UK, anyway) is done with full-rich mixture all the time, so the engine doesn't get that warm to start with.

Also many of the higher performance aircraft have less air going into the cowling, because air going down those holes means drag.

And the slower aircraft have less potential for shock cooling because the air at 100kt has a much smaller cooling effect than at 150kt+.

So I don't think it is exactly engine type or size related. For example the 360 is the same as a 540; just 2/3 of the length. It has the same cylinders, AFAIK. Maybe the 0-200 type engines are more robust, due to smaller cylinders but the metal is the same thickness as the bigger cylinders?

[Jan Olieslagers]

I am surprised to see no mention of the cowling vanes (or what do you call them?) ; I always understood Lycosaur-powered planes have them, and proper technique includes closing them during descent (sic!) for the exact reason of avoiding, or at least reducing, shock cooling?

[Piper.Classique]

Jan Olieslagers
*
Join Date: Jul 2010
Location: Near EBGB
Posts: 967
I am surprised to see no mention of the cowling vanes (or what do you call them?) ; I always understood Lycosaur-powered planes have them, and proper technique includes closing them during descent (sic!) for the exact reason of avoiding, or at least reducing, shock cooling?


Cowl flaps, and it depends. None on the 152/172, Pa 18 supercub, and many others.

[AdamFrisch]

Unlike the geared lycomings which uses a planetary reduction gear the Continental GTSIO 520 has one just massive spur gear to provide the reduction in prop RPM. It is pretty unbustable, the problem with this engine is the cylinders. They are basically the same as the 285 hp I0 520. Pulling another 90 hp (Cessna 421) or worse another 115 hp (Commander 685) is asking a lot. High CHT's or shock cooling will just kill this engine.

Interesting. One learns every day.:ok::)

[Pilot DAR]

Car engines i am sure have the same shock cooling, why is it a problem for aircraft engines,

As Tinstaafl mentions, car engines are liquid cooled. The cooling air flows over the radiator, rather than the engine. The engine maintains a much more uniform temperature throughout, and that temperature changes much less quickly when conditions change. If something shock cools, it would be the rad, not the engine. This is a happy characteristic of liquid cooled aircraft engines. Having done 64 three minute climbs these past four days, in a DA42, with the required descent following each climb, I was delighted that the liquid cooling allowed somewhat more rapid power changes.

I always understood Lycosaur-powered planes have them, and proper technique includes closing them during descent (sic!) for the exact reason of avoiding, or at least reducing, shock cooling?

Yes, if you have cowl flaps, their proper use it very important. Any aircraft which has cowl flaps will also have a cylinder head temperature indicator. As mentioned by Big Pistons, some lower powered aircraft engines are less sensitive to shock cooling, and thus do not have cowl flaps. It is still and excellent habit to treat those engines similarly gently with power changes. It is still better for the engine, and forms a good habit, which will be reassuring to the check pilot when you are hoping they will sign you off in the bigger engine plane. That said, I have see cracked O-200 cylinders...

[Unblonde]

Car engines don't operate at full power (or some high percentage of full power) then 55%-75% for hours on end. Car engines rarely see full power, let alone operate continuously above 55%. My car specifies 3000miles between changes. Conservatively choosing 30mph average speed (and for easy calculation) that's 100 hrs of use - and not even above 50% power for any of it.


---
My bike specifies 10,000 km between changes, is a small, light engine producing 160 HP, runs at ~50 - 60% power a lot of the time (on motorways) and gives no problems... but it is new technology, fly by wire throttle, electronic injection, etc.

it is possible with airplane engines too - sure the new rotax engines will prove very reliable for instance... just need more competition and volume in engines to spur the manufacturers on - what's the point in adding 20kg to make an engine that only needs an oil change every 200 hours, when regulations will make you inspect it every 50 anyway.

Alex

P.s. how do you quote in a reply?

[Jim59]

At the club where I tow our procedure, after the glider releases, is to take 15 seconds to go from full climb at 65-70 kts, typically, to the descent airspeed of about 110 kts whilst maintaining more or less constant RPM (not constant speed props) and then a further 15 seconds to reduce to 2300 which gives about 1000' a minute rate of descent. At about 500' the tug is levelled off, without moving the throttle, allowing the speed to reduce into the white flap band and the rpm will of course drop further. Power is gently reduced on final. Since this procedure has been in place the number of cracked cylinders has become negligible - before it was not.

[India Four Two]

The descent procedure for the Scouts (180 HP Lycoming) at my club in Alberta was to throttle back to 2100 rpm, maintaining the tow speed ~70 mph and dump full flap. After one minute or CHT below 300, reduce to 1700 rpm (below the resonance zone) and descend at 90 mph with full flap.

I'm not aware of any of the Scouts at Alberta gliding clubs ever having any cracked cylinders.

[Tinstaafl]

"A lot of the time..." is still not the same duty cycle as a typical aero engine. An aero engine can typically climb at, or near, full power for some time then cruise at 65 or 75% for several hours. It only experiences power at less than 45% during the approach & landing.

As an example, I flew a Navajo yesterday for about 6 hours. A single leg of that flight consisted of full power for take off, 85% for 25 mins followed by a continuous 65% for another 2 1/2 hours then finally low power for the las 5 minutes for the approach & landing. It's quite capable of doing that day after day, hour after hour. I'd have no qualms about doing a 5 hour sector in it, with the engine running continuously at that 65% for the cruise. I wouldn't like my car to have to do that.

The piston engines in most aircraft don't need an ancillary electrical system to run, unlike cars with their battery, alternator, coil & distributor (or electronic equivalent of the coil/distributor) system. If the engine is turning the ignitions (there are two) will provide spark without any involvement of the ancillary electrical system. Hell, one whole ignition system can fail and the aero piston engine will still run with only a minor loss of performance.

Is it possible to wring greater economy or performance? Of course! But always at a price. Cost, weight, complexity, reliability & increased maintenance issues all get a look in. One thing I'd like to see is magneto timing adjustment system. It could be electronic or mechanical but I'd want it to have a failsafe reversion to a simple fixed timing magneto if the whiz-bang bit went wrong.

[Unblonde]

Thinking about it, when I took the bike on a track day, when the profile more closely fits your useage, the engine gave up half way through... so not sure i'm on a winning argument here.

On the commander I fly, I've been taught to set the engine monitor (EDM700) to show shock cooling - it shouts if cooling exceeds 50F/min anyway, which makes life easier (albeit meant a descent at 160Knots - cruise is 130) to keep in enough engine power to avoid cooling too quickly.

Thanks,

Alex

[Chuck Ellsworth]

Shock cooling is only one side of the issue.

The subject is better described as thermal shock which will cover the expansion / shrinkage of metals as temperatures change.

For damage to occur during the cooling process you had to increase the temperature first, a byproduct of power.

Therefore common sense dictates thinking before using throttle movements.

[mad_jock]

I don't know much about engine design but as a general engineering description its more about the temprature gradient across a lump of solid.

You can also get shock heating.

Basically when you get a temprature difference across an object there will an associated stress/strain gradient set up by the differences in the expansion of the material at the different tempratures, if this stress or strain is more than a number of engineering events the material will fail.

Now even if you don't exceed the value which would cause a immediate failure you might very well be into an area of the stress / cycles of a fatigue graph which was never designed for by the engineer. So even if it doesn't fail you have used up the life of the material. The amount used will have a none linear quanity related to the amount of temp gradient.

And example of shock heating is pouring boiling water into a old style pint glass and it shattering

And an example of shock cooling is a compressed cyclinder exploding when the contents are expanded in a none controlled manner.

[peterh337]

modern cars specify at least 10,000 mile oil change intervals.

A large part of that is due to water cooled engines (close piston clearances), and better air filtering (aero engine air filters generally drop only about 0.5" MP across them, or less).

[abgd]

Perhaps it's a silly way of looking at it, but 10,000 miles is only about 75-100 hours cruise on many GA types.

[peterh337]

Not silly at all. Aero engine service intervals are not massively different to car engine service intervals, and the oil in the latter benefits from a cleaner environment.

One other factor I forgot to mention: in a car engine, after you park the car, it is harder for water to get into the engine and condense there, into the oil. Aero engines have big breathers, which is one reason why corrosion due to infrequent use is such a problem. Another reason is that the large amount of blow-by past the loosely fitting pistons (air cooled engines) dumps a lot of corrosive muck into the oil, which will accumulate if the engine does only short flights.

[Immortal]

There is a well known article by a gliding tug operator who reported consistent cracked cylinders, until they did a bit of a low power section before the (rapid) descent, and it cured the problem.

Can someone point me in the right direction for the article? I can't find it. I've read it a few years ago, but did not bookmark it. It's some kind of scan of a magazine article. Your help is appreciated!

[peterh337]

This is one (http://www.peter2000.co.uk/aviation/misc/shock-cooling.gif).

[TomNH]

Marine engines see use profiles much closer to aircraft engines than cars. Until relatively recently this wasn't too relevant as weight concerns weren't a major factor in the marine business so the engines didn't have much in common with aviation. For good safety reasons marine engines are almost all diesel except where power to weight was a major concern. Cooling can simply be managed through using a thermostat to dump the heat in the primary cooling circuit into the sea.

However this is changing: modern high performance, turbo and/or supercharged, diesels have appeared over the last decade or two. I've some experience with the steyr straight six monoblock engine that's got variants in the 2-300hp range. Our use profile is very demanding, flat out for ten mins or so then to idle and worst case to cold before starting again, and our problem was with shock heating (presumably 'cos the thermostat and liquid cooling smoothed out the cooling). We needed a steady supply of spare engines as we'd kill them regularly with only a couple of hundred hours on them, if that. Happily the boats are twins... Anyway the problem was massively reduced by running shore powered electrical block heaters. The primary cooling circuit is now kept at running temp and things are much happier. I read somewhere that the new six pot diesels being played with by Austro are based on a steyr engine.

[Pilot DAR]

I don't think that there is much of a basis to compare liquid cooled marine engines to air cooled aircraft engines in terms of cooling. The massive cylinder head of a marine engine has the thermal mass to resist/absorb temperature changes which result from changed operating conditions, so they cannot occur rapidly.

Air cooled engine cylinders are thin and light by comparison, and when operated carelessly, can change temperature hundreds of degrees in a few seconds. Worse, this temperature change will occur in some places in the cylinder, but not so much in others. This caused stresses within the cylinder structure, and is when the damage can occur.

[phiggsbroadband]

Hi All, the mechanism of the shock-cooling is a little complicated to decide.

At first, you would think it was due entirely to the cold high altitude air cooling the cylinders fins from the outside.... However that has remained constant even whilst under power.

The only real change (dT/dt.) has been to the internal surfaces of the cylinders having their source of heat (fire) removed and replaced by a cooling flow of non-combusted gas.

Also the metal mass of each cylinder / cylinder head contains areas that have different temperatures when in normal ops.... The cylinder walls are subject to the heat of combustion; The Inlet Valve area has the very cool inlet fuel/air mixture, with the Exhaust Valve in the hot exhaust gas flow.

So the temperature gradients are quite high in these areas. Any sudden attempt to bring them back to a uniform temperature is likely to result in shock-cooling, and not necessarily in the areas of the fins cooled by the airstream.

Just as an aside,,, In the motor-cycle world, Racers tend to run their fuel mixtures for max power, whereas owners of Concours machines run richer mixtures to prevent heat discolouration or rusting of the exhaust pipes.!

Pete

[TomNH]

I guess all I was trying to point out is that they are converging. The steyr 'marine' block will be on airframes soon and it certainly can have some thermal management issues - even if they are heating rather than cooling, in my experience.

I terms of the original question: even relatively heavy liquid cooled engines can suffer thermal management issues so I'm certain that light, air cooled, aero engine can.

[Crash one]

P.s. how do you quote in a reply?

In the reply text box along the top is an icon third from the right (Wrap quote tags)
Click on that , it will put [quote] [quote] in the box. Copy & paste what you want to quote between these.

[Immortal]

This is one.

I knew I should have looked on your site! Thanks!

[Maoraigh1]

Marine engines see use profiles much closer to aircraft engines than cars.
I thought most marine engine use profiles were even farther from light aircraft use profiles than cars. They are started and run for days on end - sometimes weeks. (Yacht engines probably have the use profile of Hangar Queens - and I don't mean just sailing yachts.)

[TomNH]

The hangar queen point is fair, especially since the demise of red deisel, but most small high performance marine engines get used for a few hours at a time: mainly at very high power settings. My point was that unlike cars they are routinely at 70%+ for several hours at a time.

I think I'm going to give in now!

[astir 8]

Do air cooled engines also have a problem with the airflow coming from the front causing differential cooling from the front of the cylinder to the back side? Presumably that could cause oval bores and stresses.


Would it be more serious on radials than flat or inline air cooled engines? I seem to recall reading something about more even cooling being one of the reasons for the early flirtation with rotary engines (as opposed to radials)

[Big Pistons Forever]

A system of baffles dirrcts air to ensure cooling on the backside of the cylinders. One thing that has not been mentioned is the Importance of baffle condition in controlling CHT's. I have seen many privately owned aircraft with cracked and bent baffles and/or tattered and torn flexible rub strips. Poor baffling will allow localized cylinder hot spots and increase inter cylinder CHT difference . This is one area where saving money on maintenence is a false economy.

In particular if you do not have the red silicone baffle rub strips, you should IMO fit them at the next annual.

[peterh337]

I have just been re-doing mine, using this (http://www.mcfarlane-aviation.com/Products/?CategoryID=198&) material. Interesting claims about less vibration transmission than silicone, and I quite believe it.

The gaps visible in the pics do actually close up perfectly when the top cowling is on - this was verified with photos.

The CHTs dropped a good 10F, so that was well worthwhile.

I have thus far re-made just the top baffles (http://www.zen74158.zen.co.uk/aviation/baffles/); the bottom ones will need a hangar.

[ppl_fresher]

Hmm, that looks quite interesting! my upper cowling visibly vibrates about 2mm in flight...

Do you reckon this could be installed on a G-reg TB10?

[peterh337]

Definitely. It's a "cosmetic" repair only.

BTW if you really have 2mm of vibration, something is wrong. Perhaps a part of the engine assembly is touching the cowling.

[Obi_Wan]

Big Pistons Forever's summary on page 1 dated 22nd June is about the most accurate and simple summary of engine care required I've come accross in decades.
From those I hear from that do this kind of thing for a living (engine development sorts) spam can engines can endure more abuse than it's possible to throw at them during normal (and accidentally abusive) use.