Jump to content
robjahnke

EGT Puzzle - Rotax 912 - 2006 CTSW

Recommended Posts

On my 2006 CTSW I recently installed a data logger on my GRT EIS that I use to monitor the engine.
I can now see my EGT's graphically which revealed an interesting EGT puzzle.

At 4900 RPM, (on takeoff) the EGT's are all very close from side to side and front to back.
4900 RPM     EGT1: 1363   EGT3: 1397   EGT2: 1422  EGT4:  1405
                       (------------ Right ----------------)   (-------------- Left ----------------)

But when I throttle back to 4660 - 4750 RPM (for cruise) something interesting happens.
The EGT's on the front two cylinders goes down a little over 200F.
and the EGT's on the back two cylinders goes up about 25-50F.

4660 RPM   EGT1:  1238   EGT2: 1269   EGT3: 1421  EGT4:  1467
                     (------------- Front ---------------)    (------------- Rear -----------------)

It's a lot easier to see this on the graphs I have attached.

The PUZZLE:    So what is causing this phenomenon?      Is this normal?

I have some speculative ideas, but nothing I'm completely comfortable with.

So I won't poison your thinking by falsely leading you astray with speculation.

I look forward to ideas and discussion by the group.

Thanks, Rob


If the attached images don't appear here, you can find them on this link.
https://photos.app.goo.gl/PRi6wsDANnzYiiqq5

EGT 4900 RPM.jpg

EGT 4650 RPM.jpg

Share this post


Link to post
Share on other sites

I suspect the front cylinders run richer due to the design of the intake manifold. Evidence of this can be gleened from examination of the sparkplugs.

Share this post


Link to post
Share on other sites

There may be some influence where the probes are mounted on the exhaust pipes, with the front two receiving cooler ambient air from cowl opening around the radiator?  Especially in these artic winter seasons.  Having header wrap may reduce the variance?  WOT with lower airspeed in climb out would reduce this variance, compared to lower throttle setting / faster airspeed? 

Tom's suspicion is probably more of what's at play, but with most complex puzzles it's often not one simple answer but a factor of items.

Share this post


Link to post
Share on other sites

Did the aircraft speed vary with the rpm?  If you are going faster through the air, more air can get to the rear cylinders.  As you slow down airflow to rear cylinders decreases, which could cause temps to go up.  Did cylinder temps go up as well?  I don't know if you did this testing on the ground or in the air.  Tom probably has a better handle on it, but airflow might be a factor.

Share this post


Link to post
Share on other sites

The engine having the split like that may be exactly why Flight Design typically only has one probe per side on the rear cylinders.

I have had this discussion before with a friend of mine. We know of instances where someone has installed an engine monitor, only to find that one or more of the cylinders is running hotter than the others. Everything was fine before the monitor was installed, and in one case it had been fine for 50 years. Now with the new information they were chasing a problem that didn't really exist.

Share this post


Link to post
Share on other sites

OAT was 41F. This was a test flight on December 5th, 2020. 
(My friend just flew by my house in his 150 this morning - minus 7F) 😁

The attached graph shows IAS & ALT along with the temps. Don't worry, 300 on the graph =3000ft.

If the attached images don't appear here, you can find them on this link.
https://photos.app.goo.gl/PRi6wsDANnzYiiqq5

Thanks for the input so far 😀 

Rob

EGT_IAS_ALT.jpg

Share this post


Link to post
Share on other sites
1 hour ago, Tom Baker said:

The engine having the split like that may be exactly why Flight Design typically only has one probe per side on the rear cylinders.

I have had this discussion before with a friend of mine. We know of instances where someone has installed an engine monitor, only to find that one or more of the cylinders is running hotter than the others. Everything was fine before the monitor was installed, and in one case it had been fine for 50 years. Now with the new information they were chasing a problem that didn't really exist.

I have noticed my basic airplane with no engine instrumentation other than oil temp & pressure, coolant temp, and volts, seems to have fewer causes for concern than the guys with full instrumentation...  Ignorance is bliss!

Share this post


Link to post
Share on other sites

I only have CHT on Cyl 2 & 3. The CHT's are very stable and do not change abruptly anywhere, and particularly at the abrupt EGT change.

Cyl 2 CHT (middle front) consistently runs about 20F below Cyl 3 CHT (middle rear).

Here is the graph.

Rob

EGT_CHT.jpg

Share this post


Link to post
Share on other sites
1 hour ago, robjahnke said:

I only have CHT on Cyl 2 & 3. The CHT's are very stable and do not change abruptly anywhere, and particularly at the abrupt EGT change.

Cyl 2 (middle front) consistently runs about 20F below Cyl 3 (middle rear).

 

 

20°F seems like a rounding error in EGT temps.

Share this post


Link to post
Share on other sites
7 hours ago, FlyingMonkey said:

I have noticed my basic airplane with no engine instrumentation other than oil temp & pressure, coolant temp, and volts, seems to have fewer causes for concern than the guys with full instrumentation...  Ignorance is bliss!

I have a friend who flies behind an IO-520. He is meticulous on maintenance and does an oil sample every time he changes oil. One came back with some elevated numbers. After talking to the testing company and several trusted A&Ps, he pulled the engine and had a reputable shop tear it down. They found nothing out of the normal range. He had them zero time the engine and reinstalled it. So after $25,000 and a lot of downtime, he no longer does oil samples. Sometimes too much information is not a good thing.

Share this post


Link to post
Share on other sites

I don't think too much information is the problem. Accurate information is the first requirement. How you react to the information is where the problems begin 😬

Rob

Share this post


Link to post
Share on other sites

Honestly, I'm not sure what good EGT sensors do in a 912-powered airplane.  You can't do anything about them without a mixture control.  Just seems like another thing to obsess over.

I'm not saying having good engine info is a bad thing, just don't get carried away with micromanaging a very simple engine.

Share this post


Link to post
Share on other sites
9 hours ago, FlyingMonkey said:

Honestly, I'm not sure what good EGT sensors do in a 912-powered airplane.  You can't do anything about them without a mixture control.  Just seems like another thing to obsess over.

I'm not saying having good engine info is a bad thing, just don't get carried away with micromanaging a very simple engine.

I do find them helpful during troubleshooting.

Share this post


Link to post
Share on other sites
17 minutes ago, Tom Baker said:

I do find them helpful during troubleshooting.

Sure, I think if you are trying to fix a known engine issue than all the info you can produce is great to have.  But for in flight engine management I question the usefulness of EGTs on a Rotax engine.  Basically all you can do to manage EGTs is terminate the flight and park the airplane, and if that becomes necessary you will probably get plenty of other clues like engine roughness, etc.

Of course you can get a HACman and manage your mixture manually, but that leads to its own headaches, and Roger Lee will lose all respect for you. :D

Share this post


Link to post
Share on other sites

It looks like everyone has commented who wants to.
So I will share my own hypothesis and rationale. This may or may not be the "Solution" to the Puzzle. Only Rotax designers know for sure (hopefully).
Thanks everyone for tossing out ideas. 

The PUZZLE:    So what is causing this phenomenon?

I think the first comment by Tom is the closest to explaining this phenomenon.
Tom said: "I suspect the front cylinders run richer due to the design of the intake manifold. Evidence of this can be gleened from examination of the sparkplugs."

When I first saw the split in EGT's I looked at the intake manifold and noticed that the path to the front cylinder is much longer than the path to the rear cylinder. But the rear intake has a sharp bend. The point is that the intake manifold is different between the front and back cylinders. At WOT the manifold must provide uniform flow to both the front and back cylinders as evidenced by the nearly equal EGT's. BUT ... when the butterfly at the exit of the carburetor starts to close, we significantly disturb the airflow. This disturbs the flow pattern in the intake manifold and results in unequal manifold pressure at the inlet of the front and rear cylinders.

This may or may not result in running rich or lean of peak. The fuel air ratio at the exit of the carburetor should be the same for front and back, so my guess is that the reduction in EGT is due to reduced manifold pressure at the front cylinders, which lowers the power developed by those two cylinders, which results in lower EGT's on the front two cylinders. As Tom said, the spark plugs may help to determine if the front two cylinders are running rich (or lean). The plugs were new 40 hours ago, so when I do the annual this spring, I will take a look to see if there are any additional clues. I typically cruise at 4500 RPM so most of the operation has been with the front EGT's 200-250F below the rear EGTs.

The CHT's are driven first by how much energy is being converted to heat in the combustion chamber, and second, by the cooling flow (in this case, air & water). The fact that the middle front cylinder is running about 20F below the middle rear cylinder at low AND high RPM tells me the overall cooling must be better on the front than the rear. 

The EGT's are primarily driven by engine combustion. When the throttle is reduced, the lower EGT's on the front two cylinders tells me the combustion temperatures in those two cylinders is lower, and therefore less power is being produced by that 2 cylinder engine up front. 

The PUZZLE:     Is this normal?

I think so. It appears to be inherent in the design of the intake manifold /carburetor system. 

So what is the data telling me I should do? Nothing from a maintenance standpoint. But it tells me something about the operational characteristics. The "Word on the Street" and from Lockwood is that if you burn 100LL, you should run high RPM (5000RPM) at cruise to reduce lead deposits (and use Decalin) . But I have never really seen an explanation why. The fact that the EGT's are high and uniform at 5000 RPM might be one reason. I am lucky enough to have a supply of Swift Fuel 94UL which is exactly the same as 100LL without the lead. That means I don't need to worry about lead deposits and I don't need to worry about auto fuel gumming up the carbs when I let it sit too long.

I typically cruise at 4500 RPM ( why hurry to get nowhere 🙂 ). That means that my front 2 cylinder engine is producing slightly less power than my rear 2 cylinder engine. This might lead to more vibration, but maybe not because I am running at lower RPM, and the difference in power is not large. I can explore this in the summer. I have a prop balancer that I can leave attached while I fly,  so l will look at the vibration levels vs RPM to see if there is anything worth exploring further.

So is this too much data? NEVER. You can never have too much data in my opinion, but with a lot of data, you need to be very careful how you interpret it, or you will spend a lot of money chasing ghosts. Looking for changes from normal operation is the most helpful for everyday operations. And if you are trying to diagnose a problem, the data is invaluable.

In my opinion, this is not a simple engine. It IS an elegant engine, but needs careful attention to maintenance and adjustments. Much more so than your average Lycoming or Continental.

Happy flying,

Rob

image.png.0c590a799a7c4e24a666bcdbea284f93.pngimage.png.6dbb8a8dbc9ef24865f023a44e39f2d9.png

image.png.6dbb8a8dbc9ef24865f023a44e39f2d9.png                                                                                                                                                             Courtesy of Brian Carpenter - Rainbow Aviation

 

 

 

image.thumb.png.13e28c11c58f76e310607df2110b6d57.png

image.thumb.png.10c5d0fc286525fec746cc34f2c6cb6b.png

Share this post


Link to post
Share on other sites

Several of the guys on Rans Clan who run 912ULS have fiddled with porting and with designing new intake manifolds.  I don't know their conclusion.  I think Hal Stockman of Zipper Big Bore fame has done something in this area but I'm not current on what.

My sole point here is that this subject has been addressed but I don't know of any consensus.  It may be there is information online if one knows where to look.

Share this post


Link to post
Share on other sites

Rotax 912 ULS has been running just fine for 30 years and over 5.5 million run hours. Why mess with it? Many have over 3k-4K hours on them. Don’t make a problem just because you think there is one.

Share this post


Link to post
Share on other sites

Roger, people are trying to address a fuel mixture mis-match which is a fact.  Most of us tolerate it and some would like an improvement.  Continental and Lycoming had most of us convinced ROP was the only to fly until George Braley and a few others proved them wrong.  Now even those two companies concede LOP  has it's place.  I'd very much prefer a carb with fuel mixture setup if I could install one without too much fuss.

Remember, John Deere built 2 cylinder tractors for over 50 years before they figured out that it's not the best way to go.

Share this post


Link to post
Share on other sites

Rob,

I don’t know. I mean I really, really don’t know. But:

Your basic contention, as I understand it, is that higher velocities inside the intake manifold will even out the air flow to each cylinder. That doesn’t seem right to me. Higher velocities mean greater pressure drops throughout the system and, I believe, greater differential pressures between the two legs; certainly in absolute terms, but I suspect in relative terms as well, considering drag increases with the square of velocity.

Furthermore, the velocities in this inlet manifold appear to be extremely low. I calculated around 0.02 Mach (about 15 kts) in the separate cylinder legs and 0.04 Mach (about 30 kts) just downstream of the carbs. That’s assuming both have internal cross-sectional areas equivalent to a 2-inch diameter round bore and that the wide-open carburetor delta-p is negligible. It’s based on the engine’s 82.6 cubic inch displacement and 5000 rpm. This velocity seems so low that unless there’s some significant blockage or vena contracta, I don’t think there would be any measurable pressure drop at all in this manifold.

Mixing is probably a bigger issue. As a rule of thumb, you would want a straight section with a length of 10 diameters to assure reasonably good mixing in a tube. That would leave the carburetors sticking out the instrument panel. So, Rotax probably had to design the manifold to split an unevenly distributed fuel flow as evenly as possible between the cylinders. But the distribution of fuel at the exit face of the carburetors is unlikely to be consistent across the full range of power (complex, real-world stuff like this tends to be non-linear). So, assuming they had to design the split for one power level, they probably picked full power, accepting the fact that the fuel/air ratio in each cylinder would vary more widely at lower powers. This may be the difference you’re seeing, if in fact, what you are seeing is normal.

Another possible issue is with the exhaust gas temperature measurement. From the photo you sent it looks like the temperature probe is clamped to the outside of the exhaust manifold. That’s not really measuring exhaust gas temperature. It’s measuring the heat transfer from the exhaust gas to the outside of the manifold (including the forced convection on the inner surface of the pipe and conduction through its thickness) minus the heat transfer from the outer surface to the environment (including both forced convection with air circulating around under the cowl and radiation to cooler surfaces). If the fore and aft probes are not mounted symmetrically, or the air around them is at different temperatures or velocities, or their radiate environment is different; they will read differently even when the gas temperatures are identical. The symmetrical issue embraces not only the distance from the exhaust port but also the side they are on (facing the hot engine or cool cowl) and even differences in the downstream cooling of the pipe (faster downstream cooling will conduct heat away from the sensor faster). If there is an inherent difference in the sensor readings due to these measurement errors, it will be reduced somewhat at high power because higher exhaust flow rates will inner heat transfer (heating the sensor) more than outer heat transfer (cooling it). So again, this may be the difference you are seeing.

By the way, I haven’t taken them apart, but it looks like my exhaust gas sensors are inserted into the gas flow through welded bosses on the exhaust pipes. This may provide a more accurate gas temperature measurement (though still not perfect).

Mike Koerner

Share this post


Link to post
Share on other sites

Mike,

The typical EGT probe is inserted into the gas flow inside the tube. A small hole 1/8" or so is drilled in the exhaust pipe and the probe inserted. The clamp is just the method for holding it in place.

Share this post


Link to post
Share on other sites

Thanks for your thoughtful comments Mike. Just a couple of notes. The manifold in the photo  is not my airplane. It's a photo I found online and I used it to show the shape of the intake manifold. TOM is right about the clamps. On my airplane, there are threaded bosses and a screw in style EGT so the probe is inserted into the gas flow inside the exhaust manifold. But you are very observant, and if the EGT's were installed as you said, they would not be accurate.

Thanks for the velocity calculations in the intake manifold. My contention is the butterfly valve at the exit of the carburetor is what causes extra turbulence in the manifold when it is partially closed at 4600 rpm. At WOT, the butterfly is perfectly aligned with the flow and therefore not introducing any additional turbulence that might alter the flow distribution in the intake manifold. A partially closed butterfly could certainly disturb the flow inside a tube at 30 kts.

But I'm with you, I really really don't know either. That's why I posted the puzzle .... so other people could think about it and offer suggestions. Your idea about the possibility of different flow patterns changing the fuel/air mixing is relevant. If the mixture is way off between the front and rear cylinders, I may see evidence when I look at the plugs. But if it's only slight, I may not see it there.  I don't see it as a problem at this point. But it IS a puzzle so it's worth exploring.

Thanks for the input,

Rob

Share this post


Link to post
Share on other sites

The huge majority of engines with multi cylinders will not have perfectly even EGT's and it will always vary with throttle. The EGT's aren't a fixed number and they are a range to fall in, but shouldn't exceed 120F separation. Since the advent of digital split hairs panel info we tend to look at them as absolutes and  should realise it's okay and normal to not be an absolute, but between ranges and then we try to fix things that really don't need fixing. Our EGT's are just that and they are only one indicator of what the engine is doing. We care usually only when they are way too high or low or have a significant split that wasn't there in previous runs. Some things are designed certain ways for a reason of which you will never know and designed for longevity. Plus of course playing with these things may only cost $18.5K if you screw up. Just a drop in the bucket. This has been played with now for a few decades and it remains the same. Just a new batch of people.

Share this post


Link to post
Share on other sites

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.

Guest
Reply to this topic...

×   Pasted as rich text.   Paste as plain text instead

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.

Loading...

×
×
  • Create New...