DCS: BO-105 3D/FM Announcement and Discussions

[The_Fragger]

BO-105 Development Report

During the last few months, the Miltech-5 team has been hard at work on developing a completely new set of functionalities for the BO-105 helicopter. Simulating helicopters is a long and complex endeavor, and throughout our journey the team has learned valuable lessons along the way. Development cycles often require taking a step back, and realistically assessing the needs and wants of the community. Your ever-increasing standards and expectations do not fall of deaf ears, and we have taken draconian measures to meet them. In concrete terms, this means we had to go back to the proverbial drawing board, take an honest look at what the team could realistically achieve with existing tech… and we concluded that we had to change the way we do things.

We needed to set a new benchmark in terms of helicopter flight dynamics fidelity and more accurately simulate various aircraft systems, devices, and components. “MEDUSA” is the name of our own internal software framework, specifically built from scratch to better simulate realistic aircraft physics, including behavior according to their mass and properties.

The traditional way to simulate a specific user action (for example, moving a switch in the cockpit) is to cause a chain of actions, which then are checked for all possible situations. Here is an example on how things were typically computed prior to using MEDUSA:

• Is electrical power available?

• Is the relevant circuit breaker pressed in?

• Is the device useable/undamaged?

• If all previous conditions are met, results of all connected devices are calculated.

Hardware Devices

The MEDUSA software considers individual devices/components as dynamic objects, which are in turn affected by environment arguments, and react accordingly. Any external change of arguments does not require a specific software module anymore. The reaction is an automatically computed result based on the combination of all device results. Electrical, fuel, hydro-mechanical and oil lubrication systems of the simulated helicopter are as close to an integrated 1:1 representation of the real helicopter as can be.

One of the most critical physics models to get right is the rotor assembly. Each rotor blade is divided into 10 sectors. Each sector computes:

• Sector speed

• Sector lift

• Sector drag

• Sector momentum

• Sector centripetal forces
   

 

All sector forces are added, which in turn affect the entire helicopter. The three images below show calculations in different situations. The green bars show the lift development, the red bars the induced drag. The first image has the collective in downmost position.

The second image shows the collective raised 7 degrees upwards.
The third image shows the collective raised in the uppermost position.


 

This approach ensures that special situations like autorotation or blade stall are dynamically computed and affect the whole flight model.

Special care was also taken to accurately model rotor blade stall, which can lead you into dangerous situations. If a helicopter is in a rapid descent (vertical velocity greater than 800 ft/min) with almost no forward movement, the inner blade sectors 1, 2 and 3 have a resulting angle of attack (AoA) greater than 18 degrees. This is partly caused by blade twist (8 degrees), collective position and the direction of streaming air caused by the abrupt descent. The physical position of the inner sectors affect the lift vector, which in this case is reduced to almost nothing) and the pilot has almost no possibility to recover. Why? Well, a natural reflex to decrease the decent rate would be to raise the collective further up… but the side effect of doing so further increases the blade angle, which is already stalling. The image below represents the stream of air in a blade stall situation.



The beauty behind this concept is that the stalling blade is recognized as physical result by the rotor device software.

In other news, the electrical and fuel supply systems are done. The turboshaft simulation still needs a bit more tuning since the modelling of the compressor and the combustion chamber are not quite where they need to be yet. The next step will be the integration of the gearbox and the rotor systems. In parallel, another team is working on integrating the module with the DCS SDK (Software Development Kit) interface.

The Miltech-5 team has already shown the advanced stages of our photorealistic replica of the BO-105’s cockpit and helicopter body, but we thought it was important to share what goes on under the hood when it comes to systems, flight model and component interactions. The BO-105 is an exciting challenge, and we are very excited to share further upcoming features as we approach the finish line. We will keep you informed continuously about the development progress.

 

Edited February 2 by The_Fragger

[Raven (Elysian Angel)]

Nice! If it works as advertised, MEDUSA sounds like the future of (helicopter) flight simulation  

[cloose]

A little hard to read but great info. Thanks The_Fragger and Good luck.

[rato65]

Thanks for the update, and for the details about internatl decisions you made to allow developing a realistic simulation. I know from different projects outside DCS that simulation development takes incredible much of time, and of course long-breath-motivation. Considering the very high level of functional realism given in DCS your way is understandable and appreciated with the Bo-105 project. However, I hope it will succeed and it will become available at some stage, sometimes it might be worth a consideration to provide a non-final technology for feedback from the community to see if everything is on the right path. But this is not easy, I understand.

Anyway, great to learn this is not dead, and I am very much looking forwrard to the future of the Bo-105 - as are many others I suppose. Good luck!

Edited February 3 by rato65

[unknown]

Great news, i can't wait to fly this beauty!

[Guille H. Mono]

Why I keep hearing that this module "will come in three versions, of which two are flyable and one is only AI"?

It is a good way to create expectations and then ruining them in the same sentence.

 

[zerO_crash]

Fantastic stuff, especially since helicopter dynamics are among the most complex to recreate in the digital environment. Such a methodical, yet accurate approach, is definitely the way to go.

 

It's also good to hear that you'll be creating more helicopters

[hermes7226]

How big is the team working on the module?