Engine spool-up is way too fast

[Tim-HH]

Hello,

everywhere you can read that the T-33 engine takes 15 seconds from idle to full power. This is quite typical for early jet engines. However, in your T-33 the engine takes only a few seconds from idle to full power.

It would be great if this could be adjusted. After all, it is one of the main reasons why the T-33 exists. Many P-80 pilots struggled with the engine's slow response during a go-around. After several accidents Lockheed decided to build a trainer and the T-33 was born.

A more realistic engine response would certainly improve the T-33 experience in the sim 🙂

Greetings
Tim  

[richboy2307]

Hi @Tim-HH,

Thanks for your feedback.

The throttle response in our T-33 is tweaked exactly to the specification and feedback from Greg (Ace Maker).

Per his guidance, the throttle response should be near instantaneous and it most certainly will spool up from idle within a few seconds. To the extent that the spool up needs to be carefully managed up to 70% RPM to avoid engine surges beyond 100% RPM, which we have modelled.

While we appreciate the additional information and feedback submitted by the community, we stand by the invaluable, verifiable and direct first hand data and input provided by Gregory "Wired" Colyer of AceMaker Aviation when it comes to the flight characteristics and behavioural model of the engine simulated in our rendition.

Thanks!

[Dudley Henriques]

5 hours ago, Tim-HH said:

Hello,

everywhere you can read that the T-33 engine takes 15 seconds from idle to full power. This is quite typical for early jet engines. However, in your T-33 the engine takes only a few seconds from idle to full power.

It would be great if this could be adjusted. After all, it is one of the main reasons why the T-33 exists. Many P-80 pilots struggled with the engine's slow response during a go-around. After several accidents Lockheed decided to build a trainer and the T-33 was born.

A more realistic engine response would certainly improve the T-33 experience in the sim 🙂

Greetings
Tim  

If I may.............
I'm presently "flying" the sim T33 and have flown the real one, (Long ago 🙂
I'm not noticing any spool up issues in the sim.
One thing about approaches in this airplane; Many pilots flying the early jets (including me) learned early on how to handle the built in engine spool up situation.
After the break coming off initial, I would apply speed brakes as I used g to bleed off the airspeed down to my gear and flap limits. I'd leave the speed brakes out through my entire approach. What this did is raise the drag index. This added drag allowed me to carry power well over 50% to compensate for the added drag and into the rpm range where power could be increased avoiding any spool up issues if added power was needed during the approach.
Hope this helps a bit.
Dudley Henriques
 

[Alan_A]

@Tim-HH and @richboy2307 -

I think it would help a lot to know which engine is being modeled.

If Greg Collyer's aircraft is the basis for the performance numbers (as it should be - it's the one that iniBuilds researched) - then the engine in the sim aircraft isn't an Allison J-33, it's a Rolls Royce Nene Mk. 10. Like many (most?) of the T-33s currently flying, Ace Maker is a Canadair license-built T-133 with a Rolls Royce engine.

John Terrell (aka Bomber12th, aka JohnnyT5000) pointed out in the MSFS forums that the start-up procedures and the performance seem correct for an RR Nene.

At 2:55 in this video, Ken Pacholski, at the time the owner of N133KK, describes the RR Nene in his aircraft as "more powerful, much more powerful" than the original Allison.

If all that is in fact the case, it'd be good if iniBuilds could specify the engine in the documentation.

@Dudley Henriques - I've been using that procedure in the T-33 (and in other early jets like the F-86 and the Fouga Magister) and it works well.  I think I must have learned if from you along the way. Thanks!

[Tim-HH]

@richboy2307 thank you very much for the quick reply! I appreciate it a lot. However, there are still some conflicting inflammations.

2 hours ago, Dudley Henriques said:

This added drag allowed me to carry power well over 50% to compensate for the added drag and into the rpm range where power could be increased avoiding any spool up issues if added power was needed during the approach.

@Dudley Henriques  Thank you very much for your valuable feedback. Do you think this is correctly simulated in the iniBuilds T-33? Because what you describe is exactly what I miss in the sim. There is zero lag in the iniBuilds T-33 when you set power from idle to full power. As @richboy2307 says "...[the engine] will spool up from idle within a few seconds". I just tested it again: With the engine at 20% RPM I can accelerate to 50% RPM within a second. Which is contrary to all reports including yours.

The author of the AOPA pilot report also flew a CT133 Mk3 Silver Star with the Rolls Royce Nene engine (the same airplane type as Greg's Acemaker) and wrote the following things:

"Do not advance the throttle too rapidly at the beginning of takeoff. This can destroy the engine because of an inherent limitation of the centrifugal-flow design. It cannot process inlet air fast enough at low airspeed to prevent the excessive exhaust-gas temperatures that can result from rapid throttle advancement."

"Because of limited power available at low speed, accelerating out of slow flight using power only (and not diving) is agonizingly slow. You move the throttle fully forward and not much happens. As airspeed builds and additional ram air enters the engine, power and acceleration increase."

"The problem with this is that when the engine is at idle, it can take 15 seconds to spool up to maximum-available power in the event of a go-around"

None of this can be replicated in the sim with the iniBuilds T-33. It basically behaves like a modern jet engine.

@Alan_A That is very true. We don't have much information about the modeled engine. However, from what I've read the Allison and the RR Nene are basically identical. The Nene has around 10% more power but otherwise they share the same design principles and face the same limitations of a jet engine with a centrifugal compressor. So I would expect the throttle response to be very comparable to the J-33. 

Greetings
Tim

[GregP]

Very interesting discussion here and it occurred to me that maybe some of the difference in opinion on how it should operate comes down to the difference between actual power output and RPM?

Is it possible that our simulated plane shows the RPM ramping up relatively quickly and in accordance with what Greg has observed, but actual power output lags a bit behind as some of the testimonial comments above seemed to indicate, due to not enough ram air pressure until reaching higher air speeds?
 

I guess this could be sort of tested by advancing the throttle quickly and seeing how fast the simulated one accelerates, but it would be admittedly difficult to quantify what would constitute 'fast or slow enough' acceleration. 

[richboy2307]

Hi everyone,

16 hours ago, Tim-HH said:

Do you think this is correctly simulated in the iniBuilds T-33? Because what you describe is exactly what I miss in the sim. There is zero lag in the iniBuilds T-33 when you set power from idle to full power. As @richboy2307 says "...[the engine] will spool up from idle within a few seconds". I just tested it again: With the engine at 20% RPM I can accelerate to 50% RPM within a second. Which is contrary to all reports including yours.

Yes, we believe it to be a faithful representation of the expected behaviour, to the extent possible without resorting to custom engine models running externally to the sim. Please see the explanation and video below that elaborates on this further.

14 hours ago, GregP said:

Very interesting discussion here and it occurred to me that maybe some of the difference in opinion on how it should operate comes down to the difference between actual power output and RPM?

Ok I think "GregP" above may be on to something here, and I can see where the confusion may be stemming from. The RPM % is not a direct measure of the available usable stable power output in all instances.

21 hours ago, richboy2307 said:

Per his guidance, the throttle response should be near instantaneous and it most certainly will spool up from idle within a few seconds. To the extent that the spool up needs to be carefully managed up to 70% RPM to avoid engine surges beyond 100% RPM, which we have modelled.

Per Greg (Ace Maker), there is no discernible delay between throttle input and engine RPM% change. As a result, you need to be careful in your throttle input (upto 70% RPM) to avoid engine surges that do not translate into a direct increase in usable/stable power output. 

This is also what is being referred to by the AOPA PIREP that you are quoting:

Quote

"Do not advance the throttle too rapidly at the beginning of takeoff. This can destroy the engine because of an inherent limitation of the centrifugal-flow design. It cannot process inlet air fast enough at low airspeed to prevent the excessive exhaust-gas temperatures that can result from rapid throttle advancement."

So what is this engine surge we are referring to?

Here watch this video of our T-33 where throttle is applied rapidly. (Observe the timer on the transponder - I start the timer as I advance the throttles, and stop it when engine RPM% stabilizes. The input viewer window indicates my physical throttle lever input with the white bar and "THR:XX" numeric values shown for reference.)

 
You'll note that, yes, the RPM% is increasing to beyond 50% within 1s but in fact taking ~4s to stabilize at ~50% RPM, and thereby a state where its providing a reliable increase in power output. This is the engine surging before stabilizing. 

You'll also note the EGT and fuel flow rise rapidly during this surge. Again, this is what the AOPA PIREP quoted above is advising to avoid. The surge will be more severe if you advanced throttles from idle to full power for example, increasing the risk of RPM% and EGT overlimit events.

This is one of the "delays" you ought to actively manage - i.e. the delay in achieving an increased power output from the engine when advancing throttles from an idle state rapidly. The actual increase in power output does lag behind the surge both in reality and in the sim.

 

16 hours ago, Tim-HH said:

"Because of limited power available at low speed, accelerating out of slow flight using power only (and not diving) is agonizingly slow. You move the throttle fully forward and not much happens. As airspeed builds and additional ram air enters the engine, power and acceleration increase."

"The problem with this is that when the engine is at idle, it can take 15 seconds to spool up to maximum-available power in the event of a go-around"

So how do you ensure a proper increase in power output of the engine without causing a surge? You advance the throttles slowly and that is the other form of "delay" in response that you ought to manage.

If you keep the throttles spooled up, say at 50% RPM as Dudley mentioned, you only have to advance the throttles slowly for a shorter period of time till you reach 70% RPM, than if you were coming up from idle. Per Greg's feedback and data, beyond 70% you do not need to be as careful because the engine is far less likely to surge from rapid throttle advancement.

At the risk of sounding like a broken record, we do appreciate the additional information but as stated above, we're going to stick with the verifiable and direct first hand data and input provided by Greg, over some second hand reports found elsewhere. He is vastly experienced on the type and able to provide direct feedback to our team regarding all aspects of the T33 flight model and handling (including the powerplant) that is available to all within this sim.

His endorsement of the iniBuilds T-33 is a great source of credibility and confidence for us that this product, in its current form, is a faithful representation of its real world counter part.

Thanks!

[Dudley Henriques]

22 hours ago, Tim-HH said:

 

@Dudley Henriques  Thank you very much for your valuable feedback. Do you think this is correctly simulated in the iniBuilds T-33? Because what you describe is exactly what I miss in the sim. There is zero lag in the iniBuilds T-33 when you set power from idle to full power. As @richboy2307 says "...[the engine] will spool up from idle within a few seconds". I just tested it again: With the engine at 20% RPM I can accelerate to 50% RPM within a second. Which is contrary to all reports including yours.

 

Greetings
Tim

Probably the reason I'm not experiencing your issue is that I handle my throttles (Prop and jet both) in the sim exactly the way I handled them in real life.
ALL my power adjustments are made smoothly. I anticipate power increase AND decrease and apply throttle accordingly as needed.
Although true that you can in jets move throttles faster than when flying props I simply prefer the way I interface with my throttles which is to always adjust power with the objective being to avoid rapid changes in temperatures and pressures..
All I can say really is that "flying" the T33 (sim) using my method of handling power I'm not experiencing any issues at all.
Dudley Henriques

Edited September 12 by Dudley Henriques
Fixed formatting for readability

[Tim-HH]

@richboy2307 and @Dudley Henriques thank you very much for the feedback and the detailed information! I appreciate that a lot 🙂

[CF104]

Hi All,

I used to work on the CT-133 about 30 years ago and will provide as good an insight into the Nene 10 operation that I can remember well. One has to realize that the Nene isn't like a modern FADEC engine or even something like a J79 with a full hydromechanical fuel control. The Nene is very simple in terms of engine fuel control. It has 2 fuel pumps with a barometric pressure control and acceleration control unit, a metering valve, a throttle valve and a pressurizing and drain valve. The only thing controlling engine speed is the position of the throttle valve and the overspeed governors in both fuel pumps. The acceleration control unit (ACU) does exactly as it says. During rapid acceleration the ACU will actually reduce fuel pump output to maintain optimum fuel/air ratio to avoid high jet pipe temps and surges during acceleration. The key word is avoid. It will not prevent an idiot from toasting an engine by firewalling it. 

Given the above info one just can't firewall the throttle and expect the Nene to behave. To avoid over-temp you have to use slow deliberate throttle movements as it doesn't take well to ham-fisted operators. This is a 1940's technology turbine engine and the pilot is part of the engine control equation. Just think of yourself as the engines brain or FADEC. You control the acceleration, deceleration and over-temp control. 

Cheers,

John   

[einherz]

as Dudley Henriques said, i wanted said same but another form - real pilots are bad testers for simulator, because they have bad patterns - does everything right. i'm not real pilot, but long time ago in other simulator i flew p-80 me262 horten and stuff like that, and from that time with engines like those bird had, i use them gently slow. however i don't really know how the engine was tested, probably testers moves engine faster than they do in real, for found if that works right in the wrong way too:) in this t-33 i once kill engine because move throttle forward too fast... don't sure it was after or before i start to use alternative version of engine, but that was on touch and go... for be sure, i guess, is better be some one from real world that tested it wrong way too test again and say it works as should or power really can go more fast than real. i have that feeling too, the throttle too responsible, but i really have no clue how real it is.