CHT & EGT per cylinder

Help creating logic scripts for Air Manager Instruments

Moderators: russ, Ralph

Message
Author
User avatar
Keith Baxter
Posts: 4674
Joined: Wed Dec 20, 2017 11:00 am
Location: Botswana

Re: CHT & EGT per cylinder

#41 Post by Keith Baxter »

Jacques,

I am sure this needs a lot digging to get the submersion we need. Just how far does one go??

There are many articles on aviation “pressure cooling” on the web.

Some insight i talk from here...

https://www.avweb.com/ownership/the-sav ... s-is-more/

"As engines grew more powerful and multi-row radials and horizontally opposed engines went into service, it became obvious that simple velocity cooling wasn’t up to the job. For one thing, cooling was uneven — front cylinders got a lot more cooling airflow than rear cylinders. For another, sticking all those cylinders out in the breeze created horrendous cooling drag. A better scheme was obviously needed.

That better system was known as “pressure cooling” and is the method used in all modern piston aircraft. Pressure cooling is accomplished by placing a cowling around the engine and using a system of baffles and seals to produce the volume and pattern of cooling airflow necessary to achieve even cooling with minimum drag.

What Do Baffles Do?
Our modern piston aircraft are powered by tightly cowled, horizontally opposed engines. Inside the cowling, a system of rigid aluminum baffles and flexible baffle seals divide the engine compartment into two chambers: a high-pressure area above the cylinders, and a low-pressure area below the cylinders and behind the engine. Cylinders are cooled by the vertical flow of air from the high-pressure above the engine to the low-pressure below it. Cooling airflow is top-to-bottom, not front-to-back".

"It’s important to understand that the pressure differential between the upper and lower chambers is remarkably small: A typical, high-performance piston aircraft generally relies on a delta-P of just 6 or 7 inches of water — about 1/4 PSI! Aircraft designers try to keep this delta-P to an absolute minimum, because higher delta-P means higher cooling drag."

Keith
AMD RYZEN 9 5950X CPU, Corsair H80I cooler, ASUS TUF GAMING B550-PLUS AMD Ryzen Mother Board,  32Gb ram Corsair Vengeance 3000Mh, MSI GTX960 4G graphics card 

JackZ
Posts: 2262
Joined: Mon Feb 22, 2016 1:02 pm

Re: CHT & EGT per cylinder

#42 Post by JackZ »

Not going that far, I have seen that with 4 cylinders engines, most of the time the cylinder #3 is the hottest, because this is also where the oil cooler is located, hence additional heat generated here. The numbering starting with #1 as the front left cylinder.

Same logic applies for turbo engines, with the turbo inter coolers located here near the #5 and #6 cylinders if I’m not mistaken. In this picture, the front cylinders are on the left.
CA8711A6-4540-43BE-85C5-1D70164CB998.jpeg
Jacques
Last edited by JackZ on Sat Mar 20, 2021 7:03 pm, edited 4 times in total.
My YouTube Chanel on the A320 (Real SOPs by an Airline Pilot IRL):
https://www.youtube.com/playlist?list=P ... 0Q6SBASRqJ

User avatar
Keith Baxter
Posts: 4674
Joined: Wed Dec 20, 2017 11:00 am
Location: Botswana

Re: CHT & EGT per cylinder

#43 Post by Keith Baxter »

JackZ wrote: Sat Mar 20, 2021 6:40 pm Not going that far, I have seen that with 4 cylinders engines, most of the time the cylinder #3 is the hottest, because this is also where the oil cooler is located, hence additional heat generated here. The numbering starting with #1 as the front left cylinder.

Jacques
Jacques,

Yes how far does one go is a good question. I do not know.

Is random changing the cylinder temp the answer. Not so sure. That is why I opted for a more static approach which is not exciting to the eye. Some variability, I admit, is needed. Just how intricate does one go?

Keith
AMD RYZEN 9 5950X CPU, Corsair H80I cooler, ASUS TUF GAMING B550-PLUS AMD Ryzen Mother Board,  32Gb ram Corsair Vengeance 3000Mh, MSI GTX960 4G graphics card 

JackZ
Posts: 2262
Joined: Mon Feb 22, 2016 1:02 pm

Re: CHT & EGT per cylinder

#44 Post by JackZ »

That was my whole point. Introducing a random hottest cylinder is probably not the best idea, as most if the time IRL, it is the #3 (for 4 cylinders engine) or the #5 (for 6 cylinders engine)
Having some variations on EGT /CHT is nice though, to emulate the behaviour of different cylinders with the flying conditions.

But the problem of the OP was how to deal with stock XP airplanes which had only one CHT for the whole engine instead of specific values per cylinder like their REP variants.
Was just chiming in to say that knowing what was the hottest cylinder is probably more deterministic than purely random for technical reasons. Having the #1 being the hottest cylinder seemed weird to me.

So we basically agree.

Jacques
My YouTube Chanel on the A320 (Real SOPs by an Airline Pilot IRL):
https://www.youtube.com/playlist?list=P ... 0Q6SBASRqJ

User avatar
Keith Baxter
Posts: 4674
Joined: Wed Dec 20, 2017 11:00 am
Location: Botswana

Re: CHT & EGT per cylinder

#45 Post by Keith Baxter »

JackZ wrote: Sat Mar 20, 2021 7:10 pm So we basically agree.

Jacques
Jacques,

I will look into this at some stage. From what I see, the Air-frame houses are doing exactly that. A random "EYE CANDY" representation.

@JackZero sorry to distract. I have not had the time to try out your code. Will do so during the week. What are your thoughts ?

Keith
AMD RYZEN 9 5950X CPU, Corsair H80I cooler, ASUS TUF GAMING B550-PLUS AMD Ryzen Mother Board,  32Gb ram Corsair Vengeance 3000Mh, MSI GTX960 4G graphics card 

User avatar
Sling
Posts: 5237
Joined: Mon Sep 11, 2017 2:37 pm
Contact:

Re: CHT & EGT per cylinder

#46 Post by Sling »

Jacques is spot on. I agree with everything he said here on this subject. I have similar experiences. If you really want to do something realistic I’d take a close look at the TorqueSim SR22. It probably has the best custom engine modelling there currently is. As I understand it this stems from an extensive model of the engine and all the variances at play. I guess unless you have very accurate real world development or in service data then this is very difficult to achieve.

My current approach is to structure the instrument code as if data for all cyclinders were available. My hope is that the next version of Xplane will have realistic per cylinder datarefs such that any current instruments can be easily converted. I’m sure you are all doing something similar.

JackZ
Posts: 2262
Joined: Mon Feb 22, 2016 1:02 pm

Re: CHT & EGT per cylinder

#47 Post by JackZ »

@JackZero Not to dismiss your idea completely, but I think that your idea to update constantly the EGT/CHT with random values for each cylinder is nice but not so much realistic IMHO.

In reality EGT and CHT values are quite stable at a given moment, then do not fluctuate all the time, especially CHT.
- CHT is a cylinder Head temperature. The probe is mounted on the aluminum block and as you know the temperatures in metal vary quite slowly. So CHT takes always some time to be updated.

EGT is measured at the exhaust via a hole and the probe is mounted in the exhaust gases stream. From that the EGT is way more reactive, almost instant and that’s what is used for leaning/searching the peak EGT.


So if you want to vary the EGT/CHT for each cylinder accordingly, I’d probably distinguish two cases: above Rich if peak (around 1450 F for EGT) and below rich of peak.
From there I would have three cases to make the EGT and CHT for the remaining cylinders apart from the hottest one (in my example here, the #4) shift according to the Fuel Flow.
You would then add or substract Delta values so that EGT would vary in a logical manner. In the middle section (case B), to avoid each cylinder to reach peak temp at the same time, you can pick a lower delta3 negative value in order to keep each cylinder temp lower than the hottest one you chose.
It’s overly simplified of course but somewhat realistic behaviour.
Same applies to CHT.
B7689128-875F-47B1-9927-201F6728B95B.jpeg
Jacques
My YouTube Chanel on the A320 (Real SOPs by an Airline Pilot IRL):
https://www.youtube.com/playlist?list=P ... 0Q6SBASRqJ

JackZero
Posts: 47
Joined: Tue Dec 29, 2015 12:31 pm

Re: CHT & EGT per cylinder

#48 Post by JackZero »

Hi Jacques, Hi Keith,

I think your right that a completely random distribution might not be the optimum solution. I might instead add another user property where the users can define the base offsets for the different cylinders, either as a absolute value (-15 F) or as a percentage value (98%) that will then only have a rather small variance put on top.


I also had the idea of fixing that every cylinder will peak at the same moment by shifting the curves for the different cylinders, but this would be rather complicated, as we only have access to the data that we currently get from the frame and all values that we've yet stored.

The user starts the engine and EGT/CHT rise to a specific, low value (as the engine is running idle) --> we'll need offset the EGT/CHT for the other cylinders by a few degrees to get a bit of variance, but at that point the instrument will still know only little about the engine...
Then the engine might be leaned at low power for taxiing, so we might even detect a first, very low peak (or not, depending on how much the engine is leaned).

Then the user sets power for the mags check... the temperatures will rise but we still have a incomplete picture... after this, the pilot might or might not lean the mixture for best power (hot & high operations) before takeoff, where we'll have another peak... then the plane will takeoff, so airflow increases and therefore EGT/CHT will decrease due to airflow (especially CHT), although the mixture wasn't touched...

Then the engine will be leaned again, once we're in cruise... either starting from a full rich climb or maybe from a best power takeoff/climb...

The math to moving one EGT curve a bit behind another one might be simple... but detecting the correct behavior in accordance to the limited picture that we have and creating an algorthm that will produce at least plausible results in all situations is what makes this difficult..

Regards
Florian

JackZ
Posts: 2262
Joined: Mon Feb 22, 2016 1:02 pm

Re: CHT & EGT per cylinder

#49 Post by JackZ »

Well I get your point but the idea here is to reuse the existing work from XPlane regarding one cylinder, then extrapolate the CHT/EGT for the other cylinders by shifting the curve.
To simplify the calculations (see my previous graph), one can assume that EGT/CHT follow the same two sloped lines each side of the peak EGT. And for each cylinder the delta (temp diff) is either >, < or slightly less than the reference depending on the Fuel Flow , with some variations.

According to that for each cylinder we have three (x2) fixed delta values, EGT_delta1[x], EGT_delta2[x] and EGT_delta3[x] where x is the #cylinder. Same for CHT.

Then it’s simply a matter of knowing on which side of the EGT peak we are, then add the appropriate delta value to the ref EGT to have the cylinder EGT/CHT.

No complex calculations here and I think it should work pretty close. All the heavy lifting is done by XPlane according to the actual conditions.

Jacques
My YouTube Chanel on the A320 (Real SOPs by an Airline Pilot IRL):
https://www.youtube.com/playlist?list=P ... 0Q6SBASRqJ

Post Reply