EGT is not a "real" temperature

[Shadrach]

11 minutes ago, jlunseth said:

I say yes to point 1, I don't know enough about pipe metallurgy to answer 2 but I will go along with what you say, but on number three I beg to differ a little.  EGT is a temperature that is "related" to the softening point of the metallurgy, but the relationship is not a very accurate one.  I think the question about whether EGT is a "real" temperature, goes to whether it measures the tempereature of something other than itself, and it does not.  It eaqures the temp to which it (the probe) is heated by periodic puffs of hot burning gas.  TIT is interesting.  TIT is always higher than any of the EGT'S and the difference, in my engine, can vary from 50-125 degrees.  Being curious about this, I have asked, and learned on this site and from some friends who design engines and boosting systems, that there is a secondary event occuring in the exhaust system.  The exhaust gas, in other words, continues to burn in the exhaust system and it burns hotter than when it came out of the port, by quite a bit.  So EGT is really not a very good measure if you are looking to protect the pipes because the pipes are seeing significantly more heat than what your EGT(s) are telling you.  

I have run my engine now for a couple of hundred hours at LOP in cruise.  There is quite alot of variation in EGT between the cylinders.  To give you a general idea, the coolest EGT might be around 1490 and the hottest might be around 1550, while the TIT is brushing up against 1600.  I haven't seen any deterioration in the pipes.  There was some scaling of the tailpipe early on, but it has not changed in a couple of hundred hours.  The Bravo and Acclaim guys routinely run and even higher TITs.  But I guess what I am saying is that EGT is not a very good indication of the temp in the exhaust system, and it depends quite alot on which EGT you pick.  

The question I have heard raised about high EGT's is not the pipes though, it is the valves.  If you run LOP, EGT's will be about 100 degrees hotter than when ROP.  It stands to reason.  A good ROP setting in my engine is 125 ROP, and LOP I am about 20-30, so there is a rougly 100 degree difference.  One question that gets raised is whether the valves will have a shorter life because they are being exposed to hotter exhaust gases when LOP.  I haven't seen it in my engine but maybe I will when the engine is broken down in a couple of years for the overhaul.  Although I have heard the question raised I have never seen any evidence provided to answer it one way or the other.

APS is doing a Master Seminer in early June in Ada, for people who have already been through one of the live seminars.  I can't go, but maybe someone else can and will see if they can get an answer.

The combustion event can continue into the exhaust stroke, but it's not optimal in terms of BSFC. I can get a second EGT rise as I near 80-100LOP not a good place for efficiency (or engine smoothness). 

I've alway been told that the issue with TIT was not softening of metal, but thermal expansion. Prolonged operation above TIT redline can result in the vanes expanding to the point of contact with the turbo housing. 

[jlunseth]

I don't completely understand the nature of the secondary event that occurs in the exhaust, but it is not just an extension of the combustion event that happened in the cylinder.  I can tell you that something is going on, because TIT is always higher than any of the EGT's by a significant amount.

 

Edited April 1, 2016 by jlunseth

[Shadrach]

5 minutes ago, jlunseth said:

I don't completely understand the nature of the secondary event that occurs in the exhaust, but it is not just an extension of the combustion event that happened in the cylinder.  I can tell you that something is going on, because TIT is always higher than any of the EGT's by a significant amount.

 

TIT probe sees 3 exhaust gas pulses for every revolution of your 6 cylinder engine. The EGT probe sees 1 pulse every second revolution. That is significantly more heat over the same period of time,  all other things being equal.

[jlunseth]

Well, TIT does not see pulses as much as it sees flow.  The collector is about six or seven feet long from Cyl 1 (I believe that is the rearmost cylinder on the left side, opposite the turbo on the right), around the front, down the right side and to the turbo.  So the flow evens out as I understand it.  I am sure there are pressure pulses, but there is quite alot of mixing going on.  But as I have been told by experts, the reason for the higher temperature in the collector is a secondary combustion event in the tubes.  Don't ask me the mechanism, it has long since left my brain cells.  I have someone I can ask though, if it is important to someone.

Edited April 1, 2016 by jlunseth

[bradp]

Okay I was struggling with the relationship between flame front starting (i.e. ignition) to the time (i.e. crank angle) at which peak pressur occurs.  Google has led me back to Mike Busch.

 

 

Therefore - it seems that at a given MP and RPM (static), moving the mixture to a full rich (what was discussed above) or LOP setting will change the peak pressure to crank angle relationship such that peak pressure occurs later, increasing the detonation margin.

 

Given a static FF and MP, the effect of decreased RPM is to cause the peak pressure to occur at a lesser crank angle (after TDC) - and a reduced detonation margin.   Slower engine rotation means fewer degrees between ignition initiation and peak combustion event / peak pressure.  I suppose this is where the adage of not running an engine "over square" came from...

 

In conclusion - a smart engineer among us should develop a CHT bayonet that incorporates some sort of transducer that can mark peak pressure, and someone else can mod JPI's software to display engine crank angle as inferred by tach (i.e. by piggybacking an induction lead like a car timing light to mag leads).  Then when I'm trying to avoid ham fisting my engine, I'd have some idea of when peak pressures are occurring relative to crank angle.   This wouldn't get around issues of variability fuel metering, etc but might be a move in the right direction.   This is fun stuff to learn about.  I can also conclude that my car's engine is about 70 years more advanced than my lycoming.

[Shadrach]

13 hours ago, jlunseth said:

Well, TIT does not see pulses as much as it sees flow.  The collector is about six or seven feet long from Cyl 1 (I believe that is the rearmost cylinder on the left side, opposite the turbo on the right), around the front, down the right side and to the turbo.  So the flow evens out as I understand it.  I am sure there are pressure pulses, but there is quite alot of mixing going on.  But as I have been told by experts, the reason for the higher temperature in the collector is a secondary combustion event in the tubes.  Don't ask me the mechanism, it has long since left my brain cells.  I have someone I can ask though, if it is important to someone.

In terms of thermodynamics, pulse versus flow is irrelevant. The TIT probe is subjected to more heat (energy) for a given time.

[jlunseth]

Go to the APS live seminar.  They will run an engine on a test bed and it has ICH probes, so you can see the pressure nicely.  One thing that stands out is how hard it is on an engine to run at peak at high power.  The ICH's get very spikey and start to show signs of detonation (jaggedness in the spike).

The "where to run my engine" question that you raise, is the "red box" concept that APS uses.  It is narrower for your NA engine than for my turbo.  The problem with someone coming up with a measured red box is manifold.  One, it requires a very costly test stand.  APS has one and has run lots of engines on it, but they are mainly NA and Beech guys, so they do not have a ton of good data on the TSIO-360-LB that I drive.  In fact, none at all, just extrapolation from other engines.  Another issue is that the box is going to change with atmospheric conditions.  What is not "red box" one day, might be when the next heat wave moves in.  About all they give are general guidelines.

I have their book and my notes at home in a pile.  I will check on the "red box" parameters for NA engines and report back unless someone comes up with it in the meantime. 

I can tell you one axiom though.  At 65% HP or less you can run your engine anywhere you want without causing excessive ICH's.  So if you go up to a cruise altitude for your J where 65% is the best you can do, you don't have to worry about it, you can set fuel flow wherever you like.

The APS guys advocate WOTLOPSOP.  Wide open throttle, lean of peak, standard operating procedure.  It actually works really well for me in my 231, although technically "WOT" is not what I do because my turbo engine can be overboosted that way.  But for your NA, the way it works is you put the MP in all the way and leave it there.  You control percent power with the fuel flow, not with MP.  You pull the red knob out, straight pull, at a moderate rate, until you feel the engine slow down.  Then you adjust fuel flow to make the percent HP that you want.  I can't tell you what the magic multiplier is for your NA, that is in my notes too and I will look it up.  For my 231 it is 13.7. but the 231 engine has lower compression than your NA so YOU CAN'T USE MY NUMBER IN YOUR PLANE.  I will try to get you the right number.

The whole thing works like a charm in my plane, so I cruise along at, say, 11,000 and 160 kts with 11.5 GPH and a happy engine. 

[jlunseth]

6 minutes ago, Shadrach said:

In terms of thermodynamics, pulse versus flow is irrelevant. The TIT probe is subjected to more heat (energy) for a given time.

Yes, I agree, it gets "full heat" from all six cylinders and not just one. 

I am going to call my engine friends and see if they can explain the secondary event to me - again.

PS I am a seeker on this, I don't know enough about engines to know the real answer.  But part of the puzzle is that long collector run in the 231.  The collector is sitting exposed to the airflow in the engine compartment just like the cylinders are, so between Cyl 1 and the TIT probe several feet away, there should be a lot of cooling going on, not heating.  Yet TIT is higher.

For anyone interested, I just ran across a similar discussion about EGT's on the Pilots of America Board:  http://www.pilotsofamerica.com/community/threads/the-science-of-egt-vs-tit.346/

Edited April 1, 2016 by jlunseth

[Shadrach]

1 hour ago, bradp said:

Okay I was struggling with the relationship between flame front starting (i.e. ignition) to the time (i.e. crank angle) at which peak pressur occurs.  Google has led me back to Mike Busch.

Therefore - it seems that at a given MP and RPM (static), moving the mixture to a full rich (what was discussed above) or LOP setting will change the peak pressure to crank angle relationship such that peak pressure occurs later, increasing the detonation margin.

Depending on where the mixture is set, this true 95% of the time!  At low altitude in the winter, I can run lean enough past peak to yield cooler CHTs than full rich. From that I can deduce that the engine is producing lower ICPs and therefore has higher detonation margins than at full rich.

Given a static FF and MP, the effect of decreased RPM is to cause the peak pressure to occur at a lesser crank angle (after TDC) - and a reduced detonation margin.   Slower engine rotation means fewer degrees between ignition initiation and peak combustion event / peak pressure.  I suppose this is where the adage of not running an engine "over square" came from...

Negative, this is not true!  Think about it. If peak pressure (PP) is occurring at say 22° ATDC and you reduce the piston speed with all other remaining equal (this is theoretical), PP is going to occur before 22° and therefore closer to TDC. This will decrease detonation margins and increase internal cylinder pressure.  It's a bit like lugging an engine up hill in too high a gear, as the engine slows the piston speed slows. At some point the ignition cannot retard the timing enough to fend off detonation and you get the dreaded ping.

Edit - I've just noticed that the graph you posted actually illustrates what I've said above.

In conclusion - a smart engineer among us should develop a CHT bayonet that incorporates some sort of transducer that can mark peak pressure, and someone else can mod JPI's software to display engine crank angle as inferred by tach (i.e. by piggybacking an induction lead like a car timing light to mag leads).  Then when I'm trying to avoid ham fisting my engine, I'd have some idea of when peak pressures are occurring relative to crank angle.   This wouldn't get around issues of variability fuel metering, etc but might be a move in the right direction.   This is fun stuff to learn about.  I can also conclude that my car's engine is about 70 years more advanced than my lycoming.

GAMI has been working on the PRISM system for quite sometime. I believe it that it has been engineered to take pressure info from each cylinder. It is not yet certified and will likey be quite pricey if it ever is.

 

[jlunseth]

I read what Ross just said and he is right.  Just don't get the idea that you will hear the "dreaded ping" because you won't.  At APS they ran an engine to detonation just to show us it is inaudible.  They were right.  Big spikey things on the screen and no change in engine sound.

[bradp]

6 minutes ago, Shadrach said:

Given a static FF and MP, the effect of decreased RPM is to cause the peak pressure to occur at a lesser crank angle (after TDC) - and a reduced detonation margin.   Slower engine rotation means fewer degrees between ignition initiation BTDC and peak combustion event / peak pressure relative to TDC (e.g. from 22-deg ATDC to 18-deg ATDC) .  I suppose this is where the adage of not running an engine "over square" came from...

Negative, this is not true!  Think about it. If peak pressure (PP) is occurring at say 22° ATDC and you reduce the piston speed with all other remaining equal (this is theoretical), PP is going to occur before 22° and therefore closer to TDC. This will decrease detonation margins and increase internal cylinder pressure.  It's a bit like lugging an engine up hill in too high a gear, as the engine slows the piston speed slows. At some point the ignition cannot retard the timing enough to fend off detonation and you get the dreaded ping.

I think we're saying the same thing here.  I inserted two things and bolded one to clarify what I was thinking.

-Brad

[Shadrach]

1 minute ago, jlunseth said:

I read what Ross just said and he is right.  Just don't get the idea that you will hear the "dreaded ping" because you won't.  At APS they ran an engine to detonation just to show us it is inaudible.  They were right.  Big spikey things on the screen and no change in engine sound.

It's true that you likely won't notice a thing in detonation other than temp spikes. However, with my Bose A20s and the heater full open, I can here the valve train. Maybe, just maybe one might hear some pinging. Whether you could pick it out of the rest of the clatter is another story!

[Shadrach]

3 minutes ago, bradp said:

I think we're saying the same thing here.  I inserted two things and bolded one to clarify what I was thinking.

-Brad

We are...I think I misread your post!

[Shadrach]

44 minutes ago, jlunseth said:

I am going to call my engine friends and see if they can explain the secondary event to me - again.

PS I am a seeker on this, I don't know enough about engines to know the real answer.  But part of the puzzle is that long collector run in the 231.  The collector is sitting exposed to the airflow in the engine compartment just like the cylinders are, so between Cyl 1 and the TIT probe several feet away, there should be a lot of cooling going on, not heating.  Yet TIT is higher.

For anyone interested, I just ran across a similar discussion about EGT's on the Pilots of America Board:  http://www.pilotsofamerica.com/community/threads/the-science-of-egt-vs-tit.346/

There's not a lot of cooling going on because there is a thermal barrier between the exhaust gasses and the pipe walls that acts as an insulator. This barrier exists in the cylinder as well, if it didn't, aluminum pistons would melt in short order. Under detonation the flame front breaks the sound barrier. Pinging is the sound of repeated sonic booms on a small scale. Those booms create a shockwave that disrupts the thermal barrier in the cylinder and viola the piston developers a hole in seconds. My question regarding secondary combustion is if ROP, where does the O2 come from to support A 2nd combustion event in the exhaust. If LOP, where does the fuel come from?

 

[carusoam]

EGT vs adiabatic flame temperature...

Six flows combining into one gives good logic for higher average temperature at the TIT gauge.

We probably see this same phenomena at the ship's 3:1 EGT gauge as well.

The actual flame temperature for gasoline and air is close to 3500°F.  

So, if we are reading 1,500°F EGT there is a lot of cooler temps over time for it to read a lower average value...  The EGT reading is an average from seeing flames as the exhaust valve opens to seeing nothing but leftover gasses until the next exhaust stroke.

Running ROP, there is Not a whole lot of excess O2 left for a significant secondary combustion event in the exhaust system.

One of the early anti-LOP arguments was related to hot excess O₂ corroding internal exhaust hardware including the muffler.  A good reason for some expensive SS metallurgy....

Other thoughts....

1) at the moment exhaust gasses stop burning, they have reached their maximum temperature.

2) expanding gasses cool greatly by the expansion.  Ideal gas law in thermodynamics PV/T relations.  All three are changing at the same time. pressure is decreasing and volume is expanding, 

3) pressure in the exhaust system is always decreasing from the point the flame stops.

4) restrictions to flow in the exhaust system keep higher pressures upstream of the restriction, lower pressures below the restriction...

5) a turbocharger provides a pretty good restriction compared to NA engine's open exhaust pipes.  This is the reason the Bravo has important guidelines for the exhaust clamps on the turbocharger exhaust system. Improper re-use of clamps can allow hot exhaust out driven by high pressure...

6) all that combines to simply say EGT is not a real temperature! 

These are PP thoughts that come to mind,  I am not a mechanic...

Best regards,

-a-

 

[N201MKTurbo]

Just a word about overheated turbine wheels. I've heard talk about erosion and expansion, I had to replace a turbine wheel once that got overheated. It never failed, I found it by inspection, it was still working fine. 

I stuck a bore scope up the tail pipe and found the tips of the wheel missing. They were all very neatly and evenly melted off.

[jlunseth]

Yeah, we don't mess with those clamps.  George Braley told a story about his brother at the APS seminar.  He was flying a twin turbo Piper of some kind and noticed odd readings on his engine monitor, so diverted to land.  He apparently never made the airport and had to put it in a field somewhere but was able to do that with his family aboard.  One of the clamps had come loose and flaming exhaust gas was escaping into the engine compartment.  The NTSB investigator told him they had investigated a few similar accidents and he was happy to meet one of the pilots.  George's brother said something about how it was not that big a deal.  The investigator said, you don't understand, you are the first pilot we have been able to talk to.  George tells it better.

There is an inspection door on my K, I open it and check the turbo and the clamp, pushing on the exhaust a little, during every pre-flight.

[jlunseth]

OK, I talked to my engine design friends, and by the way they are also pilots.  First, I was told that variations in probe placement create significant variations in EGT.  It is not just distance from the exhaust port to the probe, but the depth of insertion of the probe into the pipe and the shape of the pipe at that particular point.  A curve in the pipe, for example, can cause a spot where gases are cooler, and if the probe is at that point the probe will report a cooler EGT.  A probe that extends a quarter inch into the pipe will probably report a cooler temperature than one that extends to the center of the pipe.  The TIT probe is likely to report a higher value simply because by the time the exhaust gases have reached the TIT they have had an opportunity to mix and there will not be cool spots in the flow. 

There is also residual combustion in the exhaust.  Whether the engine is operated LOP or ROP, there will be some amount of unburned fuel that enters the exhaust system, where it will find more available oxygen and continue to burn.  In my inquiries I expressly brought up LOP operation and was told that even LOP, there would be some residual fuel, and it does not take much to continue combustion.

I did not ask this of my engine friends, but it occurred to me after the fact that there is a moment in time when both the intake and the exhaust valve are open.  In theory, the incoming mixture helps drive out the combusted mixture.  This process is not perfect and some of the intake mixture goes across and out the exhaust port, which is probably part of the source of the secondary combustion. 

So if you put it all together, EGT is an inaccurate measure of the absolute gas temperature coming out of the cylinder.  Further, there is a secondary combustion event in the exhaust system which raises the exhaust gas temperature in the tube, so EGT is not a good measure of the temp in the tubes. 

If you are an NA guy and you believe you are limiting the exhaust gas temperature to which the exhaust tubes are being exposed by using EGT, you are off by a hundred degrees or so.  The same is true for us turbo guys, except that we have the TIT to believe, and that will be a truer measure of what the tubes are seeing. 

[Shadrach]

But what you describe is not a secondary combustion event, it's just the tail end of the initial combustion event.  At night, it can be seen on radials with really short exhaust stacks (in daytime as well under conditions/settings). Secondary implies a second ignition and flame front.  I do not believe that happens. EGT can be affected by the speed of the flame front this is why if you see a second peak in EGT if you are able to lean far enough. In that situation, the flame front is so slow that it is still developing during the exhaust stroke. 

[jlunseth]

OK, but I left some stuff out to simplify.  There is adiabatic expansion when the exhaust gases enter the exhaust system, and also mixing of O2 available in the exhaust system.  In my mind, mixing of O2 would be greater in an NA system than in a turbo system, but I am assured there is still O2 mixing that occurs.  "Secondary combustion event' were the words someone else gave me when this whole thing was explained to me some months ago.  I am not engineer enough to know what precise terms apply, only reporting on what I am told is happening. 

That is what I know or can find out on whether EGT is a "real" temp.

Ooo, that reminds me, I know one other possible source.  I will email him and report back if I hear.  

[cnoe]

EGT can be affected by the speed of the flame front this is why if you see a second peak in EGT if you are able to lean far enough.

Here's what that event looks like with my IO360.

I'd been told that the 2nd peak (just prior to flame-out) occurred due to extra unburned fuel as mis-firing begins. As a student of the subject I can't offer expert testimony but am simply curious about what exactly happens.

[jlunseth]

I emailed one of my "Professors" from APS and got some information back several days ago.  I tried to post it, but the site was broken, so here goes:

I was reminded that the most important reason why EGT is not a "real" temp is that leaning out the engine (whether further LOP or further ROP) changes the effective timing of the engine, causing peak temperature to occur later in the combustion stroke, and closer to the point when the exhaust valve opens, making the EGT hotter.  But EGT is not combustion temp.  There is no redline limit on EGT.  My professor also said that TIT is a more real measure of exhaust gas temp.

The redline limit on TIT for continuous operation is universally 1650 (that's also what's in my POH), and that is to protect the turbo, not the pipes.  Excursions above that for short periods are allowed (my POH says no more than 1725 for no more than 30 seconds).