MV DHC-2 Beaver released!

[bazzar]

Be careful not to confuse propwash effect with CoG effect with brakes on. If the contacts are slightly behind true CoG, even a fraction, and brakes are held, you can artificially create tail lift with judicious use of power. That is not propwash.

 

[cavaricooper]

FWIW-

The Beaver in its current iteration is good... but has potential to be GREAT. It it my hope that the FDE can be revisited and tuned specifically in re. the low airspeed regime and ground handling (tail) behavior. That and attention to the longitudinal axis trail in pitch/power application, would elevate this model to superlative standing.

The radio digit clipping I will fix (thanks again!).

Best- C

 

[henrystreet]

For what it might be worth, not just an "armchair" pilot here.

No specific aspersions intended. In fact, yours and Cavaricooper explanations have helped me understand bush flying a little better. My regime is typically MIL jets that are accompanied by whole manuals, dozens of pages, of performance data that is frequently ignored or even considered incidental to many simmers.

Again, thanks for the great explanation.

 

[Rudyjo]

I'm a little more impressed with this plane now.
I took all the cargo out of it and put 20 gallons in each tank.
With just a few degrees of flaps and the pedestal wheel full forward, I have no problem running the plane all the way down the runway on two wheels. Same on landing, as long as the speed is kept up it will stay on two wheels all the way down the runway.
Also much better handling on the ground, with a little speed, it will now turn almost within it's own length.
Getting the weight out of the plane has made a big difference.
I've only tried this with the Tundra model, will see how it works with the regular wheels.

 

[cavaricooper]

Thanks Rudy-

I will reinstall and check... deleted it after the first few updates, perhaps they tweaked the FDE since I last flew her...

Best- C

 

[stearmandriver]

The amount of speed required is the issue. Yes, with enough speed, the elevator becomes effective. The problem is just that 50kts shouldn't be required to get the tail up. This video illustrates a normal Beaver takeoff (or most other taildragger). Notice how aggressively the tail pops up, how soon in the roll, and at how slow of an airspeed (and with less than full down elevator deflection).
https://youtu.be/KHVfZx_MCoc

Don't mean to harp on this. She's still very fun to fly. Just want to be clear that the issue isn't a subjective one of weight and balance.

 

[cavaricooper]

Andrew-

Is this with the latest updates? If so, I shan’t bother reinstalling... I keep hoping that the FDE will evolve into something wonderful...

Best- C

 

[henrystreet]

It would be very easy to replicate the conditions in the video. 5NC2 RW20 wind from the south at 7. The unknowns, of course, are weight and balance of the aircraft.

EDIT: Another variable that has not been discussed is the realism settings in the simulator doing the testing. A video showing the behavior of the model accompanied by good documentation of weight/balance, simulator realism settings, environmental conditions, etc would likely go a long way to getting an FDE update, if the FDE deviates from the aircraft actual performance..

Has anyone tried this test with the JustFlight model out of curiosity? Again, using the same criteria and settings as discussed here.

Finally, I have no dog in this fight except to illustrate how quick the community is to criticize and how sloppy their "testing" can be. Real world experience is obviously very important in judging the accuracy of the FDE but are you assuming that the developer did not use real world pilots in their testing?

 

[Dimus]

Experienced STOL pilots even lift the tail by gently braking during the take off run. Just saying, not sure if it was done in this video.

What I can confirm is that the FDE was developed by a real world Beaver pilot with enormous experience, same like in the Otter. We all know how bad FSX is in ground dynamics, friction etc. Developing an FDE is always a big compromise and setting of priorities. If a possible shortcoming in this particular behavior is something that will keep one from enjoying all the rest that this model has to offer, of course it is up to them.

 

[Rudyjo]

The amount of speed required is the issue. Yes, with enough speed, the elevator becomes effective. The problem is just that 50kts shouldn't be required to get the tail up. This video illustrates a normal Beaver takeoff (or most other taildragger). Notice how aggressively the tail pops up, how soon in the roll, and at how slow of an airspeed (and with less than full down elevator deflection).
https://youtu.be/KHVfZx_MCoc

Don't mean to harp on this. She's still very fun to fly. Just want to be clear that the issue isn't a subjective one of weight and balance.

Thanks for the video, while I saw quite a difference with the weight reduced, it wasn't like the film.
I don't know anything about changing numbers in the fsx.cfg files, is it possible that there is a number in there that if changed would produce what I'm seeing now at 30kts instead of 50 kts ?
I had no idea that the real life Beaver was capable of taking off almost as quickly as a Cub, it shows just how right they got it when they designed it and why it was so popular in places like Alaska.

 

[vortex]

What I can confirm is that the FDE was developed by a real world Beaver pilot with enormous experience, same like in the Otter.

But surely if Milviz haven't modelled prop wash effects, it won't make any difference who helped develop it as it will never match one of the key features of the real thing?

 

[stearmandriver]

It would be very easy to replicate the conditions in the video. 5NC2 RW20 wind from the south at 7. The unknowns, of course, are weight and balance of the aircraft.

I would encourage you to test this out if you have the MV Beaver. Give yourself every advantage. Nice and light, medium headwind, CG as far forward as you can stuff it. If you can get the tail off the ground anywhere near that quickly and positively, with that small amount of forward yoke (you can see the small amount of elevator deflection in the video), I will have to assume there's something wrong with my install. For me, with the model loaded as stated, with full forward elevator deflection, the tail *grudgingly* starts to ease up around 50kts; a speed at which the aircraft would actually fly. This is simply not correct.

EDIT: Another variable that has not been discussed is the realism settings in the simulator doing the testing. A video showing the behavior of the model accompanied by good documentation of weight/balance, simulator realism settings, environmental conditions, etc would likely go a long way to getting an FDE update, if the FDE deviates from the aircraft actual performance..

There's truth to this. There are obviously limitations to any sim engine. I'm using maxed realism settings on everything, and I guess I just assume everyone else reporting this issue is as well, since they tend to be more experienced sim or RW pilots who would notice the issue in the first place.

Has anyone tried this test with the JustFlight model out of curiosity? Again, using the same criteria and settings as discussed here.

We're talking about the Beaver here. JustFlight did an Otter, right? I don't have the Otter from either dev so can't comment on that. I'll say that, comparing this FDE to the FDE of the best Beaver we had in the sim until MV came along, the Aerosoft Beaver X... The MV FDE is superior in a general sense, though the Aerosoft was still very very good... and the Aerosoft Beaver can pop her tail off the ground just as sprightly as any taildragger pilot would expect, unless she's loaded with a very aft CG.

Finally, I have no dog in this fight except to illustrate how quick the community is to criticize and how sloppy their "testing" can be. Real world experience is obviously very important in judging the accuracy of the FDE but are you assuming that the developer did not use real world pilots in their testing?

I only have a dog in it in the sense that I'd love to see this model live up to its potential. It's already very very good... it could be damn near perfect. My comments aren't geared to "fight", exactly; I'm aware people love to argue for its own sake on the internet, and sometimes they have a vendetta against a particular dev. None of that applies to me. I continue to respond just to rebut the other side of the coin: people who defend the dev because they assume any criticism must stem from user error or lack of knowledge/experience. I'm sure that's often true... but not always.

I'm not assuming the developer did not use RW pilots... but I'll admit, it's occurred to me to wonder just how much wheeled Beaver experience their guy has. If, for instance, he spent the majority of his time on skis or floats, it could affect things. In trading correspondence with MV, I have only heard from programmers who lose the thread of the conversation quickly when it turns to aerodynamic matters and operational reasons why it's not only possible but sometimes important to get the tail off the ground. They seem to have been told that 3 point takeoffs/landings are the norm and so normal takeoffs and wheel landing performance just aren't a priority for them. ::shrug::

I'll note again the statement from the (somewhat arrogant, I find) head of MV: he told me propwash is not modeled, and I was unreasonable to ask about it, because (direct quote): "we aren't doing accusim at this price point." Now, first of all, the MV Beaver with addon (which is required to equal the number of configurations that come with, say, the accusim Cub) is well into accusimmed price range... but secondly, it would seem to me to acknowledge they're aware of this and just don't care to address it. Which is, of course, their prerogative... but let's be honest about it!

 

[cavaricooper]

Well said, Andrew... I would echo your sentiments... it's good, and has potential to be GREAT.

I personally wish they'd get there because I'm hopelessly infatuated with this airframe and bush lore... This one has the looks, but...

Best- C

 

[fliger747]

I did the FDE programming and as a matter of fact have over 1000 takeoffs and landings in the Actual Beaver. My personal airplane, a Supercub, I can stand the tail up at a full stop, not so with the Beaver and in fact it is quite reluctant to lift the tail. The main landing gear, just as with the Otter, is placed quite a ways forward. With full power it is possible a ways into the takeoff run to get the tail up a little, which is important in protecting the tailwheel on rough fields and reduces the drag on a ski takeoff, but it's a ways. My belief is that the forward gear location is designed to resist nose over tendency on soft, rough and short bush strips where there may be a necessity for strong braking. Certainly uneven deep snow on skis the lane resists nosing up well, much better than say a 185.

Flight simulator flight dynamics programming is lacking in several areas. The elevator action is certainly one of the most egregious in that it does not directly model the airflow over the stab and elevator surfaces, but applies a rotational moment to the aircraft CG based on indicated airspeed and elevator displacement. Many taildraggers can be in a three point position scooting down the runway and application of up elevator will lift the mains off the runway. Think about this one a bit, in actuality the plane cannot increase it's lift unless the AOA is increased, which is impossible when the tail remains on the ground.

Life would be easier if the tail feathers were directly modeled as to the airflow from prop wash, flap extension and a myriad of other items that have effects on the plane with power and speed change. There are various workarounds, but one thing changes another and some compromise is necessary in an inexpensive desktop simulator. Overall I feel as an experienced Beaver pilot and some other 25,000 flight hours that it pretty closely replicates the experience of flying the DHC2.

 

[Rudyjo]

Stearmandriver and Cavaricooper:

I just finished reading one of the best flight simulator reviews I have come across regarding the Milviz Beaver, it's a very long review.
The author says just what you are both saying.....It is one of the better planes released for flight simulation with a few negatives, one being the very long time to get the tail wheel off the ground.
The easiest way to find this article would be to Google "Milviz Beaver", one of the first links is " Avroliner project, Review of the Milviz DC-2".
Besides the part about the Milviz Beaver, it also goes into the history of the DC-2, if you have an interest in the actual plane, it is very interesting.

EDIT... the post from fliger came in while I was writing this, I had not read it before posting this.

 

[cavaricooper]

Tom-

I am MOST gratified you posted here- THANK-YOU! I appreciate your experience and the work you have put in. As far as the FDE, I believe your statement and summary is succinct and accurate. I am fond of most of it except for the ground handling and the pitch power lateral stability where there seems to be a hard crossover point in re. the yaw axis. In descending and ascending turns, with power application and removal, there is a hard point in the yaw behavior/curve that shows up readily on the ball indicator.

I would plead you continue to press the case to obviate your next sentence... "Life would be easier if the tail feathers were directly modeled as to the airflow from prop wash, flap extension and a myriad of other items that have effects on the plane with power and speed change".

This is TOO GOOD a model to just let this go. Please, please plead with Colin and revisit this issue when possible. I WANT TO LOVE THIS SIMULATION, but with a fair bit of tail wheel time myself, I just cannot bring myself to do so in it's current iteration.

Best- Carl

 

[Daube]

Nice to read about the technical details involved in the Beaver, directly from the mouth (well, fingers...) of the FDE developer. It's also cool to read that the FDE dev has actually some real experience about the plane itself.

Concerning the tail-raising behavior, I believe A2A was able to get some kind of prop-wash modelled into their aircrafts. But since the simulator engine can't handle that, they are probably modelling it through the external Accusim module, then. Perhaps Milviz can implement some similar external module in future, which would allow them to model much more things the sim can't do ?

 

[vortex]

With full power it is possible a ways into the takeoff run to get the tail up a little, which is important in protecting the tailwheel on rough fields and reduces the drag on a ski takeoff, but it's a ways.

First, thanks for responding on the forum. I'm in no way questioning your experience on the Beaver, but based on your comment above I'm curious to know how you explain the performance of the real Beaver in the YouTube link above (https://youtu.be/KHVfZx_MCoc). Within just a few seconds of starting to roll, and with only a small amount of elevator deflection, the pilot's able to easily level the fuselage during the takeoff run. This does seem to be at odds with what you've said.

 

[stearmandriver]

Tom,

Thank you very much for chiming in. I can only echo what Carl has said; I REALLY hope that now that the Otter is out, you have the opportunity to revisit the Beaver.

I understand your points about the far forward position of the mains , and I can certainly see how that would cause her to resist tailing up a bit; but then, I also feel like the relatively long arm (at least with a forward cg) of the decent sized elevator would go a ways to making up for that. I know she's no super cub, and I certainly don't mean to come off as argumentative and apologize if I am... it's just that I can also see that her tail comes up a good bit better than the MV rendition, both on videos and when I watch a wheeled Beaver take off in reality. She also seems happy to make wheel landings, something the MV rendition struggles with.

I have no Beaver time, but a good bit of time in various other medium sized taildraggers (Howard DGA, gull wing Stinsons, Stearman etc), and this just doesn't "feel" right. It wouldn't be so glaring if the rest of the flight envelope DIDN'T "feel" so right, and so really is a testimony to the great work you've done with this model.

Personally, I would pay for a custom flight dynamics environment / upgrade that can overcome the limitations you mention (which I understand are very real), the way other developers have done. Of course that doesn't mean enough others would that it would be cost effective or time effective for MV to build it, but I'll join Carl in pleading - I'll even BEG - for any improvement.

Thanks again!

 

[cavaricooper]

If you are referring to the half digit showing in the standard radio stack, edit file 'radi_d.xml' in base Beaver 'panel/C310_XMLGauges'

Change 6.3 to 5.2 in these two lines:

<String>%((A:COM1 ACTIVE FREQUENCY, MHz))%!6.3f!</String>

<String>%((A:COM1 STANDBY FREQUENCY, MHz))%!6.3f!</String>


WS

Based on Andrew'ss recommendation, I just reinstalled the latest installer and the FDE mod... the panel clipping issue is still uncorrected. Thanks for this tip! I'm off to fly her now...

C