Turboprop engine modelling
Guest
Posts: n/a
Turboprop engine modelling
I am in the process of creating a mathematical/physical model of a turboprop engine. The intention is not to get accurate performance but rather to get reasonably accurate spool up times, torque readings etc for demonstration purposes.
Has anyone seen anything similar done before? Any recommended reading? Just before I reinvent the wheel...
Cheers,
/ft
Has anyone seen anything similar done before? Any recommended reading? Just before I reinvent the wheel...
Cheers,
/ft
Guest
Posts: n/a
ft,
Best of luck ... it has been done before. The main problem with turboprops relate to their not easily suiting closed form solutions - pistons and jets are much easier. Try looking at some of the standard flight test manuals or engineering texts for information on the problems ....
Best of luck ... it has been done before. The main problem with turboprops relate to their not easily suiting closed form solutions - pistons and jets are much easier. Try looking at some of the standard flight test manuals or engineering texts for information on the problems ....
Guest
Posts: n/a
Thank you, John. I guess I could be more precise about what I need. I'm looking for ways to calculate things like the correlations between fuel flow, gas generator RPM and gas generator acceleration, the twisting moment generated by a propeller at various RPM/pitch (no IAS, runups only), the twisting moment generated by the power turbine...
Yes, I could go the long route. Calculate my way through the engine with pressures, mass flow, fuel/air ratio etc etc. Done that enough to be sick of it. BTW, that doesn't take long, does it? But since it has to be realtime and with limited computing resources, I want to cheat as much as I can get away with. I'm basically working on figuring out where to cut corners here.
Do you remember where and when you've seen it done? A lot of corporate work has been spent on this obviously but that doesn't do me much good - I need to find information in the public domain.
Cheers,
/ft
Yes, I could go the long route. Calculate my way through the engine with pressures, mass flow, fuel/air ratio etc etc. Done that enough to be sick of it. BTW, that doesn't take long, does it?
But since it has to be realtime and with limited computing resources, I want to cheat as much as I can get away with. I'm basically working on figuring out where to cut corners here.Do you remember where and when you've seen it done? A lot of corporate work has been spent on this obviously but that doesn't do me much good - I need to find information in the public domain.
Cheers,
/ft
Guest
Posts: n/a
ft,
This is sounding to me a bit like an undergrad mini-thesis exercise ? Might we know the task purpose, if that be not confidential ?
The sort of things you appear to be after would probably require the assistance of the engine manufacturer's data package and that is unlikely to be forthcoming for other than an airframe manufacturer.
Apart from that, the only option I could see would be to make a first pass model from AFM and planning data. Acceleration data could prove a difficulty to resolve but you might be able to get some in-service data from a helpful operator and then try to massage that into some sort of order.
This is sounding to me a bit like an undergrad mini-thesis exercise ? Might we know the task purpose, if that be not confidential ?
The sort of things you appear to be after would probably require the assistance of the engine manufacturer's data package and that is unlikely to be forthcoming for other than an airframe manufacturer.
Apart from that, the only option I could see would be to make a first pass model from AFM and planning data. Acceleration data could prove a difficulty to resolve but you might be able to get some in-service data from a helpful operator and then try to massage that into some sort of order.
Guest
Posts: n/a
Correct, it's the final part of my AE education. Hopefully, the end result will be a CBT showing the 'front office' side of running a modern turboprop.
Yes, I've been looking for the OEM engine data but, as you pointed out, it's not the kind of information they seem to be falling over themselves to make available.
I am lucky enough to have assistance from someone who has done more engine runups than most people so if I can't find anything I will reverse engineer something. No emergency, just a lot less elegant.
Cheers,
/ft
Yes, I've been looking for the OEM engine data but, as you pointed out, it's not the kind of information they seem to be falling over themselves to make available.
I am lucky enough to have assistance from someone who has done more engine runups than most people so if I can't find anything I will reverse engineer something. No emergency, just a lot less elegant.
Cheers,
/ft
Guest
Posts: n/a
The way this kind of thing is done in the "real" flight simulator business in real time with limited computing power is to use lookup tables. If you can work through the math in non-real time and construct a number of tables then all you have to do in real time is enter the right table and interpolate.
"The proof is left to the student"
"The proof is left to the student"
Guest
Posts: n/a
Yup, but in this particular case a physical model would really be a lot more suited for the task I think. Correct behaviour is wanted throughout the operating envelope and also in abnormal conditions such as hung/hot starts. Not to mention that creating the tables without access to complete OEM engine data is quite a bit of work!
The way the prototype is coming along, it's going to be a much simplified physical model, assisted and adjusted by tables and formulas here and there as needed.
Thank you both for your suggestions.
Cheers,
/ft
The way the prototype is coming along, it's going to be a much simplified physical model, assisted and adjusted by tables and formulas here and there as needed.
Thank you both for your suggestions.
Cheers,
/ft
