Engine model questions

MountainMen1167

Hello,
I made a post on Reddit and was told to make a post here.
I have some question about the engine model of the Black square.

The product mentions custom engine simulation, but I find no evidence of this. The engines do not behave right.
I am sorry, but I find no evidence of anything being custom outside of a few animations and the tablet.

I made some observation over the past few days. I know its a lot, I hope you can clarify most of these findings.

Startup and on ground:

You can prime the engines for minutes, you will not lose any fuel from the tanks
engines can still be primed with empty tanks
Engines still try to start with empty tanks
Engine will not start with the mixture in cutoff
Tablet shows direct injection, these engines are port injected.
You can't prime the engine with the engine driven fuel pump.
After startup, rpm overshoots.
External view show engines at 600+ rpm when turning with the starter
No rpm increase as the engine warms up
Manifold pressure is perfectly linear with throttle position and only slightly varies with rpm
If you apply full throttle and pull the prop back, manifold pressure will increase past ambient.
wrong wastegate behaviour. The Westgate is normally fully closed or regulating, this happens when boost before the throttle body reaches the set maximum.
The engine cant reach full static rpm.
Shutdowns are instant the moment you reach 400rpm.
Starter doesn't prime the oil system
Oil warms up too fast
Engine stalls at idle if you are above 1500ft. yet starting at idle brings the rpm to around 1000 to then they drop down and die again.
Engine doesn't take damage if its cold and then firewalled
No vapour lock possible while the engine is running
Fuel boiling out of lines don't enter the cylinders
Vapour lock Starts do not follow real world logic: Throttle half open, mixture cut-off
When the engine stalls. there is no indication, shaking or anything, its just gone
engine will fire after exactly 1.5 engine rotations. This can be replicated with 100% repeatability by looking at the prop blades.
engine will not complete the start if the starter is let go below 400rpm
starter rpm way too low
CHTs and oil cool down too fast when on the ground
EGTS instantly go cold after shutdown
Oil pressure not affected by temperature
engine is cold when spawning on the runway
manifold pressure at idle is below the 10" stop on gauge, idle manifold pressure too low
Run accumulator dry on the ground, take off, and the engine will windmill restart even if the lever is left in feather and the prop is still feathered.
Engines instantly shutdown when shutting fuel off.
Fuel flow at idle too low
There is no difference in mag drops. 4 Identical mags
Perfectly matched engines and props, no difference between the two
No difference in performance with an engine at 100% or 1% health
engine not affected by air conditioning
engine not affected alternator load
too many "compression wells", should be 120° apart, are 60°.

In air:

Opening and closing the pressurisation system doesn't affect engine performance.
Mixture doesn't affect turbo performance(it does on the tablet, but manifold pressure stays the same)
In flight windmill restarts are just instant. also no priming required or anything
External view still shows the engine spinning at 300-500 rpm and it jumps around a lot when feathered. thrust is still being provided according to dev tools.
Windmill restarts magically prime the engine, even when mixture is cutoff
You cant spin a feathered engine with the starter, on the ground or in flight.
The compression physics are very, sticky. Its like a sharp detent.
Random manifold pressure dips when flying
Prop sync inop
Fuel and oil system work fine when flying inverted
After feathering there is still fuel flow with pumps off sometimes
Delay when transitioning in and out of feather
Propeller drag applied the moment of moving the lever above feather, even though the prop is still feathered
little to no rpm overshoot on prop governor
Weird airframe shakes caused by in flight shutdowns
Engine will still windmill restart with a dry accumulator
exhaust will still backfire even after shutting down with the mixture lever.
CHTs barely cool of after in flight shutdown
CHTs not affected by mixture
Reducing rpm increases oil pressure
Forced turbo failure in dev tools gets reset.
Leaning the mixture increases fuel flow
Feathering the prop is inconsistent, sometimes i have to cycle the lever to get it to feather
Starter cooling not affected by airflow?
Engine performance doesn't match manufacturers tables
EGTS cool down too much at low power settings
No stumbling or rough running with incorrect mixture

Ccitis

I mean, I get it... I am also super particular with these things but imagine the time and cost needed to address alot of these things. I will tell you as well, alot of this will be limitations in the sim (mixture behaviour for example). My suggestion (no joke) go fly a real airplane! You sound like you have good knowledge of airplanes. Maybe you are a pilot (I am too, although no medical now)

MatzeH84

You're not wrong. This is the part where A2A still has the edge in GA flying, although I feel their product has also more shortcomings compared to the older P3D releases.
However keep the price of the addon in mind. Most of the things you mentioned would require an engine model running externally from MSFS, which is a huge task. In daily ops BS is giving me more immersion than A2A.
Although I admit some points I would like to see updated as well: oil temperature and viscosity simulation, individual (slightly different) running engines and start on primer would be on top of my wishlist.

MountainMen1167

Most of these items are very basic.
The COWS DA40 gets all of these right and they do not use an external engine model.

MountainMen1167

Addition the the original posts

As long as the propeller lever is out above the feather position, the propeller hub should be pressurised by the accumulator.
Manifold pressure appears to switch between a "running" mode and "non running" mode when starting or windmilling.

Here are some tests I performed yesterday
https://streamable.com/aztjqu
https://streamable.com/yivhk2
https://streamable.com/rcos3v
https://streamable.com/3o1hso

Black Square

@MountainMen1167 Hello! Thank you for posting your findings here. Please excuse my delay in responding, I wanted to make sure I gathered my thoughts before writing. I hope you don't mind my lengthy reply, but I thought the amount of work you invested deserved an equally invested response.

Before we dig in to your observations, I wanted to say that I am always happy to hear from fellow pilots, engineers, and aircraft owners. Your suggestions always help me design better aircraft! These aircraft are passion projects for me, so my primary goal is to give my users unforgettable moments of virtual flying that have only recently become possible with MSFS.

I must admit that I found your accusations of "no evidence of anything being custom" and "a simple repaint of the default aircraft" to be undeniably hurtful. I've dedicated my life to these aircraft since MSFS 2020, as my friends and family can attest. I always try to do my best, and support the community with new features and custom code for their projects. While many of your critiques are perfectly valid, I hope that my extensive manuals might serve as evidence of the many custom features I prioritize during the roughly one year it takes to bring each aircraft to life.

I'm sorry that the aircraft did not match your expectations. I'm very sensitive to claims of false advertising, so I'm particular about what I write in my descriptions. I always publish my manuals online and provide the aircraft to video and print reviewers before each release so they can share their honest commentary. This gives the community the opportunity to learn about my aircraft before making a purchase decision.

I never claim to be the best, or anywhere close, but I always try to do my best and be honest about my limitations as a one-man-team.

With that said, I would much rather discuss the why's and how's of flight simulation from here. Thanks for bearing with me.

In theory, you've provided me with an exhaustive list of improvements that I can start working on immediately; however, you will see that a few challenges persist. I hope you won't mind if I don't address every single point by name, because most of them fall into three basic categories:

Some of them are oversights in my latest release (after only one hotfix), like positive fuel flow with empty tanks, the starter not circulating oil, the oil pressure while exercising twin engine propellers, and number of compression wells. I have already fixed these oversights for the next update. I've also recently put a lot of time into my own oil temperature modifications to much better approximate the real aircraft, which you can read about here. The unrealistic mixture behavior is the result of deficiencies that have existed in Flight Simulator for ~30 years, and were fixed only recently. I've implemented these new fixes in my Commander 114, and plan to port the changes back to all my other aircraft in future updates. This is my typical workflow, so I can safely test major changes on new aircraft while continuously improving my fleet for existing users.
There are also some features that have been on my backburner, and should be pretty simple. Often, it only takes the right suggestion from a user for me to implement these pending features. A few of your other points, I believe, are currently working in the aircraft, though I will have to test them later. For instance, engine performance being degraded by damage, propeller sync, and feathered propellers being affected by apparent wind.
As you noted in the bulk of your points, I am not simulating a complete internal combustion engine and all of its accessories to the point where resistance is applied from alternator loads, or the fuel system will falter while flying inverted. While each of your points might seem small, I'm sure you can imagine how combining them into a standalone simulation that plays nicely with the parameters required by the simulator is no small task. Just to achieve most of the custom features you see in my aircraft today, I've had to wage war with the underlying simulation. The originally incorrect turbocharger mixture control, or the lack of fully stopped propeller feathering are two good examples. With each of these features come secondary and tertiary effects that can be hard to predict without the simulator's source code. Aircraft are incredibly complex systems of interconnected systems, as you know!

Now, that doesn't mean any of your requests are impossible! I often say, "anything is possible with software." In fact, I would argue that they would all make my aircraft undeniably better.

From here, it's only a matter of practicality. I am very lucky that I get to develop these aircraft as my full time job, but I hope you understand that I must consider the economics of development if I want to continue supporting my family. As I often say here, it breaks my heart when I am approached with wonderful ideas for new features that I would love to implement, but the realities of developing these systems by myself often hinders these aspirations. In short, spending a year to develop a feature that might only matter to 1% of simulator pilots could spell the end of Black Square entirely.

What can be done about the engine simulation shortcomings you list here? It appears there are essentially two approaches. One can either develop an external simulation and inject the solutions back into the simulator, or you can develop on top of Asobo's platform to add a substantial layer of realism. There are tradeoffs to each method, but the latter is the one I have chosen to specialize in myself. This is the fundamental difference between myself and at least one of the other developers you mention.

From the categories above:

I have already fixed some of your points for the next updates to my aircraft.
I have taken note of your features from this category to explore during upcoming work.
Last are the features that would be very difficult to implement without fully separating my aircraft from the underlying simulation. These are the ones that break my heart, because I cannot currently justify the software redesign to bring them to fruition based on how the majority of my users interact with my aircraft. However, I always say that I can be convinced to undertake just about any amount of work if a substantial amount of the community make their desires known. I have also recently been speaking with a third party who might be able to contribute some of the exact features you mention here.

I hope this explains some of what you are experiencing, and provides a window into aircraft development for MSFS. While I'm working on some of your points for my future aircraft and updates, I hope you will stay in touch with any further suggestions you might have, because I always want to hear what my users are thinking.

Nick C.