reported On Vortex Ring State from active Mi-8 instructor

[Furlow]

Not a pilot so pinch of salt. Ive got thousands of hours in DCS Mi-8 and Ive got all the muscle memory for staying out of VRS downpat, and I thought Id throw something in that hasnt been mentioned specifically. VRS in the hip is the only part of the flight where you have to react to something before it happens, as opposed to getting a true warning or indication to respond before or as it happens, which I think is why it probably claims so many victims. When you start to learn the approach the first lesson the DCS Mi-8 teaches you is as you approach lower ETL limit you add twice as much collective as you think you need, then you double it, and then you wait to see what effect that had a second later as the VRS hands grabs you and tries to pull you out the sky. As is pointed out, its hard at first but it quickly becomes part of muscle memory and you crown yourself king of VRS.

 

My point here (if you can call it one), is the VRS fight always felt a bit out of place compared to the rest of the flight, again, the rest of the time you are responding to what the aircraft is doing after it does it, but VRS is a completely separate entity that has to be entirely pre-empted on every approach. You fly your response against VRS about a full second or two before it 'activates' because it seems to behave as a pre-programmed stumbling block, an obstacle in the road that must be driven around every single time, rather than a dynamically created event dependent on conditions. (This also helps to create muscle memory however, as right now VRS acts in the same way each and every time).

 

I completely agree with the hip being an uncertifiable deathtrap if it was like this realistically, I just wonder if the real reason for this type of modeling is possibly more to do with controlling sim pilots to look realistic rather than create an entirely realistic module. Sim pilots are a pretty enthusiastic and inventive fan base at the best of times, and will generally behave in an unrealistic manner (namely pushing the aircraft far beyond its safe limits) on a regular basis, and when you watch a track replay of such flying, it looks terrible and some arcadey arma type game. Such greatly magnified pre-set roadblocks like the hip VRS may simply be there to try to rebalance the realism lost to pilots who quite happily wreck the engine and transmission of a brand new helicopter every flight. So in other words the VRS on steroids is a poor mans substitute to a crew chief asking wtf have you done to my helicopter? I think if we had any kind of wear and tear modelling, plus somehow persistent aircraft, you would then be able to remove the magnified VRS and people would still fly fairly realistically looking due to preserving and not stressing the airframe, but without that kind of concern then pre-programmed stumbling blocks was the best they could manage to try and get sim pilots to fly in a semi-realistic fashion.

 

Sorry, rambling and kinda forgot where I was going with this, but while I definitely agree with VRS being modeled bizarrely, I think there might be an ulterior motive behind it, simply because real life values didnt look real when in the simulator when flown by non pilots in a non-realistic way. (Ironic).

[cw4ogden]

Excuse the formatting (copy paste) From MI-8 flight operations manual - seems the throttles are meant to be set all the way up for anything above about 3000' 

https://www.scribd.com/document/322000057/Flight-Operation-Manual-for-the-Mi-8-Helicopter
2.8.3. Take-off-landing operations and maneuvers at low speeds near the ground at airfields and sites, located at altitudes higher than 1000 m, are made with at least 93% revolutions of the main rotor for the purpose of ensuring reserves of pedal control in these modes.

 

Also from Mi-17 manual:
Reduction of the gas generator rotor speed below 85 to 88 % at an airspeed close to zero even at the main rotor speed within the acceptable limits, causes transition of vertical descent at a speed up to 20 m/s (vortex ring condition).

 

[AlphaOneSix]

93% Nr is the rotor speed for takeoff power. 95% Nr is for everything else. The aircraft should not be flown with the ECLs pulled up. If one engine fails, the other will automatically allow allow you to pull more power. Moving the ECL's up from their center detent should only be done for ground testing purposes, since moving the ECL upwards will limit collective travel.

[Viktor_UHPK]

Hey. It's an interesting discussion.

I am not a real Mi-8 pilot - but I have somewhat focused on the helicopter in the last years and studied the POHs and technique manuals.

 

The technique manual states:

"[on approach] after reaching a speed of 60 km/h and less, the required power increases and the helicopter tends to increase the vertical speed. Therefore, it is necessary to maintain the vertical speed constant (2-3 m/s) by gradually increasing the collective pitch of the main rotor, and then reduce it [the vertical speed] as the forward speed decreases and approaches the ground."

 

The visual / instrument patterns flown on the Mi-8 are designed for a descend at 2-3m/s on final. 

With this applied you won't ever get into VRS.

 

[cw4ogden]

4 hours ago, AlphaOneSix said:

93% Nr is the rotor speed for takeoff power. 95% Nr is for everything else. The aircraft should not be flown with the ECLs pulled up. If one engine fails, the other will automatically allow allow you to pull more power. Moving the ECL's up from their center detent should only be done for ground testing purposes, since moving the ECL upwards will limit collective travel.

 

Good to know.  

 

@Viktor_UHPK I'm curious which manual is that you're referencing?  

I know, at this point how to work around the VRS.  My question is how accurate is it?  It's much harder to get into now that I've learned the module, but I still think there is something off with how it's working.  I get that following a shallow approach to land will keep you out of it, but in reality, you need to be fairly steep for it to be a problem.  

It's unfortunate VRS diagrams are notoriously non-aircraft specifc, but the way I read the diagrams, and the way we always flew IRL, unless you were coming in in excess of 30 degree approach angle, VRS was not part of the equation.  As I mentioned earlier, we even had a rule of thumb for flying: if your intended landing area is at or above your feet / pedals area, you were safe from the portion of the aerodynamic envelope prone to VRS.  You didn't have to consider rate of descent at all.  You just can't get into it on approach, in anything but a very steep approach (barring a large tailwind or high altitude operations near max G.W.).

 

[Viktor_UHPK]

2 hours ago, cw4ogden said:

 

@Viktor_UHPK I'm curious which manual is that you're referencing?  

 

The manual is called "ТЕХНИКА ПИЛОТИРОВАНИЯ Ми-8МТ" (Mi-8MT piloting techniques). I only have a Russian version unfortunately.

 

Additionally the Mi-24 technique manual states:

Quote

 

"typical mistakes on approach and landing: High vertical speed before hover.

The reason for this mistake is late application of collective pitch increase by the pilot - not taking into consideration the relatively slow reaction time of the TV3-117 engines. During a standard approach with a sinkrate of 2m/s the pilot has to start increasing collective pitch when reaching not less than 30m height and not less than 70kph. The rate of collective pitch increase has to be adjusted so that at a height of 5-8 meters a vertical speed of 0,5m/s is reached."

 

(sorry for the rough translation)

 

So maybe one additional reason that the VRS feels so "severe" is the slow spool-up. It definitely helps to be ahead of the aircraft and pull collective early in anticipation of lift / power demand.

 

 

Edited April 18, 2021 by Viktor_UHPK

[AlphaOneSix]

One issue with Western pilots that are used to aircraft is that they tend to wait until the bottom of the approach to start bringing in power. I'm no expert, but I think this has to do with the relative low inertia rotor systems that are easy to speed up, and more responsive engines that can spool up much faster, as Victor_UHPK has said. During transition training at my organization, one of the first things we teach new pilots (new to the Mi-8/17, not new to helicopters) is to get the power in early. I wonder if most VRS problems in DCS are caused by this failure to have enough power pulled in early enough during the final approach.

 

The manuals I have differ a little bit but they generally call for a descent rate between 2 and 4 m/s at around 40 km/hr, decreasing to 1.5 to 2 m/s by the time the aircraft is in a hover. Max descent rate at 40 km/hr is 4m/s and max descent rate at a hover is 3 m/s. Touchdown should be between 0.1 and 0.2 m/s.

[cw4ogden]

1 hour ago, Viktor_UHPK said:

The manual is called "ТЕХНИКА ПИЛОТИРОВАНИЯ Ми-8МТ" (Mi-8MT piloting techniques). I only have a Russian version unfortunately.

 

Additionally the Mi-24 technique manual states:

(sorry for the rough translation)

 

So maybe one additional reason that the VRS feels so "severe" is the slow spool-up. It definitely helps to be ahead of the aircraft and pull collective early in anticipation of lift / power demand.

 

 

 

And @AlphaOneSix It also possible my hip flying friend isn’t on the mark and the hip just really is a bit more susceptible to it than I’m used to.  For many of the reasons you both and others have cited.  
 

I can google it, but I want to, probably already should have taken a look at what the revs per minute of the hip main rotor is because if it is significantly less than the 47, which Iirc is 225 then it’s likely my source led me astray.  
 

it still perplexes me there isn’t more in the way of literature on it, but I can’t in good conscience say it’s wrong, and I’ve fleshed it out fairly well now, but I just can’t tell if it’s right or I’ve learned to live with it.  
 

 

Edited April 19, 2021 by cw4ogden

[cw4ogden]

And looks like it’s not readily available if either know what the mi-8 main rotor spins at in terms of revolutions per minute.  
 

how is RRPM set in the hip?  I’d assumed it was fixed by the maintenance pilots with fuel flow and track, as it seems not adjustiable by the pilot, or am I missing a switch etc?

Edited April 19, 2021 by cw4ogden
Spelling

[Viktor_UHPK]

13 minutes ago, cw4ogden said:

And looks like it’s not readily available if either know what the mi-8 main rotor spins at in terms of revolutions per minute.  
 

how is RRPM set in the hip?  I’d assumed it was fixed by the maintenance pilots with fuel flow and track, as it seems not adjustiable by the pilot, or am I missing a switch etc?

 

 

192 RPM - You can adjust it with a spring loaded three position switch on the collective pitch lever.

[cw4ogden]

5 minutes ago, Viktor_UHPK said:

 

192 RPM - You can adjust it with a spring loaded three position switch on the collective pitch lever.

I gotta look at what to see what it’s called but I’m assuming it’s being controlled by my slider which is how I go from idle to flight rrpm.
 

Thanks for the replies.  That’s a fairly significant difference in rrpm at takeoff settings.  That could certainly be a cause I’d thus far not considered.

 

It may be the 47 is just really better than average similar bird regarding VRS because in any given profile that extra RRPM and lack of rail rotor would and could be the difference.  
 

And it’s the lack of upflow that feels off, as I stated somewhere up there in the post.  
 

But that much difference in RRPM on its face, might be the difference I’m feeling.  There would be noticeably less upflow.  And at the end of the day that’s a decent answer for why at least.  
 

it would also explain why tweaking the throttle levers works in the sim to make it feel more “right” to me, even if it is incorrect procedure.  

[AlphaOneSix]

Each collective has a spring-loaded INCR/DECR switch to adjust the rotor rpm. It’s set by increasing the collective pitch angle to 3 degrees on the blade pitch indicator and then using the switch to set the rotor rpm to 95%. At full INCR the rotor rpm should be between 96-98% and at full DECR it should be between 89-93%. It’s not meant to be used in flight, but rather just prior to taxi. 

[cw4ogden]

47 minutes ago, AlphaOneSix said:

Each collective has a spring-loaded INCR/DECR switch to adjust the rotor rpm. It’s set by increasing the collective pitch angle to 3 degrees on the blade pitch indicator and then using the switch to set the rotor rpm to 95%. At full INCR the rotor rpm should be between 96-98% and at full DECR it should be between 89-93%. It’s not meant to be used in flight, but rather just prior to taxi. 

Roger, same as 47 before we got fadec controls. We had Incr / dincr to set but rrpm but it was governed by some fairly simple flywheel governors and manipulating fuel control.

[AlphaOneSix]

2 minutes ago, cw4ogden said:

Roger, same as 47 before we got fadec controls. We had Incr / dincr to set but rrpm but it was governed by some fairly simple flywheel governors and manipulating fuel control.

Yep same in the Mi-8

[randomTOTEN]

6 hours ago, cw4ogden said:

unless you were coming in in excess of 30 degree approach angle, VRS was not part of the equation.

My experience in the DCS: Mi-8 is the same. VRS really isn't a factor in forward flight.

 

I think what would be productive would be for you to fly the Mi-8 into the conditions which you believe results in unrealistic VRS, and save that simulator session as a replay ".trk" file. And then upload it here for all of us to directly view. Please be mindful of the forum attachment limits, the shorter you can make the track file, the better.

It's a lot better than this beating around the bush that it feels like. You can directly show us what you're talking about.

 

I also think it speaks extremely highly of the simulation that we are considering VRS in this much detail.

[cw4ogden]

On 4/18/2021 at 9:58 PM, randomTOTEN said:

My experience in the DCS: Mi-8 is the same. VRS really isn't a factor in forward flight.

 

I think what would be productive would be for you to fly the Mi-8 into the conditions which you believe results in unrealistic VRS, and save that simulator session as a replay ".trk" file. And then upload it here for all of us to directly view. Please be mindful of the forum attachment limits, the shorter you can make the track file, the better.

It's a lot better than this beating around the bush that it feels like. You can directly show us what you're talking about.

 

I also think it speaks extremely highly of the simulation that we are considering VRS in this much detail.

Sorry for the delay.  I got burned out on this thread.
 

Indeed, the fact we are arguing about it speaks volumes of the simulator, as does the fact about half of the responders or more won’t take at face value, the multiple real life pilots telling the community it’s wrong.  
 

It’s a great sim.  I’d go the trackfile route, if it seemed there was any enthusiasm for fixing it, but it feels like the only enthusiasm is to find a way to impugn what several of us are saying, and that is, it isn’t correct.  

There is a tiny window in landing profile where you can be descending directly in your own downwash.  You have to be in that downwash descending vertically or near vertically for this phenomenon to occur.

 

I did try a few tests with smoke on the ground and I was hoping the airflow was modeled in the smoke, so I could see the vortex, unfortunately it doesn’t seem to be.

You have to find a very precise attitude and descent angle to encounter VRS because by definition if you are not going directly down into your downwash, the vortices are washed away.

 

I’ve also rewatched the VRS accident videos and will offer two more pieces of anecdotal evidence from viewing them:

 

There appears to be no attempt at corrective action, indicating VRS is rare in the hip, and not some boogeyman waiting for the slightest inattention to rate of descent.

 

if it was as bad as depicted in DCS, it would be known in the community and they’d have reacted with some form of corrective action.  I can see none.  

The fact there are recoded accidents points, ironically, to it not being crazy deadly.  I.E. we haven’t seen an R22 mast bumping crash in years because it is a known and well fleshed out hazard.  

 

In my estimation, they crashed not knowing what was wrong with their aircraft.
 

Secondly, both accidents happened in essentially vertical profiles and one actually was drifting backwards.

 

thats what it takes to get into VRS, you have to be going straight down into your downwash, who’s angle is dictated by the surface winds and your relation to them.  
 

With a 15 knot tailwind I can get into VRS at 15 knots, but otherwise, no, you just can’t.  Because I can’t simultaneously be going both fifteen knots forward and in a vertical descent.

 

Its not unless you find that one perfect pitch, power and airspeed  setting and deliberately hold it there, and even then you have to descend too fast.  
 

And I’d just reiterate, I would like to test this, but I’m burned out fighting with people, developer has yet to chime in, and I largely feel stupid for thinking I could break through the nonsense, when so many have already failed.  

None of that is directed at you, it’s just explaining why I’m accepting defeat over learning how to do trackfiles and conducting more extensive flight testing.

 

No one cares, and to many it’s a badge of honor they can fly the “deadly” hip.

Edited May 15, 2021 by cw4ogden

[bazza772]

@cw4ogdenthanks for your considered professional insight into this module and how it is modelled. As a 30+ year ex military FI and current civil fixed wing pilot it is good to hear how closely or otherwise certain aspects of the flight model are simulated from somebody with significant experience on type. I have zero experience of real world helicopter flight, apart from being winched out of the sea one sunny afternoon, and appreciate your expertise when discussing VRS.

Cheers.

[Fri13]

1 hour ago, cw4ogden said:

None of that is directed at you, it’s just explaining why I’m accepting defeat over learning how to do trackfiles and conducting more extensive flight testing.

 

Track files are automatically recorded. The last flight is in temporary file and after each flight when you exit to menu, you get option to save the track with own custom name.

 

The hard part really is to attach that file by finding it first from windows user directory (default is example: c:\Users\<username>\Saved Games\dcs.openbeta\tracks I think the name changed since 2.7 update) and then just add it to the post.

 

The time consuming part really is to start the mission, fly the wanted profile/sample in shortest possible manner and then save the track.

 

1 hour ago, cw4ogden said:

No one cares, and to many it’s a badge of honor they can fly the “deadly” hip.

 

I would be interested to know more. Espexially that what developers say.

 

 

 

[NineLine]

if there is still a bug here, please upload a track and a brief description. Thanks.

[cw4ogden]

17 hours ago, NineLine said:

if there is still a bug here, please upload a track and a brief description. Thanks.

@NineLine

 

Here are a couple of track files.  

My beef is it seems you can get into VRS, in the MI-8 without one critical element.  But it is the most important element, vertical or near vertical descent.  

If I had to guess, I'd say the VRS algorithm is looking for parameters of flight associated with a Pilot's rule of thumb for avoiding VRS.  
In other words, if the manual says use caution below 40 KIAS, that doesn't mean the phenomenon can happen at 40 knots.  They throw in fudge factors for tailwinds etc.  A big grey area to stay out of.

VRS can happen, and can only happen, when the helo is descending vertically into it's own downwash, and with sufficient rate of decent.  You have to be going straight down into your rotorwash.  That appears to be the piece that is improperly coded.

It should be absolutely impossible to encounter VRS with an approach angle less than about 32 degrees as indicated on the VRS diagram.  

I believe on track 2, the first signs of VRS start occurring in the 30 knots range, and it is fully developed with 20 knots of forward speed.  

 

1.trk 2.trk

Edited May 26, 2021 by cw4ogden

[cw4ogden]

 

12:40 seconds:  There is no way that flight profile should result in a VRS accident.  The Vortices would be shed because he is not descending into them.
 

 

 

 

 

 

 

Here you can see the difference.  At about 1 minute in.  Notice how the bird flies fine, even backwards in a descent.  It isn't until he tries to descend vertically into his own rotorwash does VRS happen.

 

Edited May 26, 2021 by cw4ogden
readability

[sniperwolfpk5]

I am not a real world pilot but I think MI 8 is easy to get in VRS. During landing You have to give complete focus to VSI otherwise you will get into VRS. If your Heli is heavy then you must give extra precaution to prevent VRS, otherwise you will meet your friend called death

[cw4ogden]

25 minutes ago, sniperwolfpk5 said:

I am not a real world pilot but I think MI 8 is easy to get in VRS. During landing You have to give complete focus to VSI otherwise you will get into VRS. If your Heli is heavy then you must give extra precaution to prevent VRS, otherwise you will meet your friend called death

I'm not sure your meaning?  The DCS Mi-8 is easy to get into VRS or the actual aircraft is easy to enter VRS.

If you scroll up to the start of the topic, you will see:  I started the thread by asking that very question to an active MI-8 instructor friend of mine.
"Is the Hip more prone to VRS than other helicopters?"  His answer was pretty definitively "No."

I'd ask you to read through the whole thread, where I wage a fairly lengthy war on why VRS isn't realistically modeled in DCS, the Hip specifically, and upon what experience base I draw those conclusions upon.

I'm a retired U.S. army heavy lift standardization helicopter pilot who used to demonstrate VRS.  And I know it sounds self serving to come here and tout credentials, but I am an expert on rotary wing aerodynamics and VRS in particular and because I want to make the point:  it's not even required to be demonstrated to new pilots. 

 

Virtually no one gets a demonstration of VRS going through flight school, or did at least, unless it was because their instructor went out of his or her way to teach it.  It is for most pilots purely book knowledge.  I certainly didn't receive a demonstration.   Then I got into it once in real life, survived the test, to learn the lesson if you will, and it became a pet peeve of mine, that we didn't teach it.

I went out of my way to demonstrate it because, although exceedingly rare, VRS can absolutely kill you, and very quickly.  

So, in addition to being an expert, It falls into the category of one of my pet peeves, I went out of my way to demonstrate to pilots that it is deadly, even though it wasn't required.  But how it works in DCS is sketchy at best.  

VRS kills people the way sticking a fork into the toaster kills people.  And that is only if they don't know about the hazard or are operating in extreme high G.W. high D.A. situations and get careless.

I'm glad they tried to model it, but I also think it's worth the time to get it right.
It's also why mine is the umpteenth thread about VRS, from other real pilot's who say the same exact thing.
 

Edited May 26, 2021 by cw4ogden
spelling

[NineLine]

Thanks @cw4ogdenI will run this by the team.

[cw4ogden]

23 minutes ago, NineLine said:

Thanks @cw4ogdenI will run this by the team.

Thank you.  I think the video linked offers better evidence than my track files.  I've learned to fly the module as it is, and I really had to try to force it in the track files.  

I don't know if that's me just adapting, or evidence the modelling is pretty darn close, but is still off enough to feel quite wrong to someone familiar with the phenomenon.   

I've really tried to kick around is this broken, or am I not an objective observer, and I can't quite say anymore.  As I mentioned anecdotally somewhere in the post, my first encounter with VRS in the HIP I had to google what was trying to be modeled, because it felt so absolutely foreign.  

I think the moddelling of the phenomenon is quite good, Induced from an OGE hover it is spot on.  But the parameters where you are susceptible to it need the relook.  I have experienced and seen far too many DCS Mi-8 crashes that the flight profile just isn't enough for the ensuing disaster that follows.