Project Dornier Do-17z2

the Do17 porpoising is slow and undulating,
but, still not acceptable for a good bomb run.
i could be mistaken, but, i think v7 had it, too.
i'll go back and double check earlier versions.
this is nothing, compared to the stock fw190.
that thing porpoises wildly and picks up speed.
in multi player, i would always get caught by a p51.
but, if i switched the ap on soon enough,
i could out run one very easily.
yes, it was kinda cheating,
but i didn't modify anything.

as i recall, many years ago,
Ivan did some air file work in that department.
danged if i know what he did, though.
patience, milo, patience.

as for the gauges, yes, i could do another panel,
with stock multi engine gauges.
true, they're not historically accurate,
but, they're more usable for us non german speakers.

are we limited to cessna gauges
or, are there other multi engines in fs98?
i guess it doesn't matter...
there are only cessna in the cfs gauge folder.
i was trying to keep it stock.
 
Hello Smilo,
Yes, in versions before, the porpoising was worse. I was trying to adjust AP for flight testing at different altitudes, and it only worked on some planes.

Looking further into CFS1, given the options in the menus, I don´t really think AP´s were meant to be used on this simulator, as combat flying would be done manually, but as it had to be compatible with FS98, they had to let an AP work. How, is another matter!

Then, basically, FSFS gauges for piston engines are all only Cessna ones, I´m afraid.

Cheers,
Aleatorylamp
 
true...in the early days, i often asked,
what's up with this auto pilot feature?
as you know, there are several options
in the controller settings section.
i never really used it until i got into the heavies,
were straight and level flight, along with
making minute course adjustments
were essential for making a precision bomb run.
ap is also valuable on those long waypoint missions,
especially in multi player, were you don't press x
to jump to the next waypoint.

i'll never forget a joint ops graduation mission.
i had to plan and execute a bomber escort mission.
i was flying top cover over france,
escorting the bomber at high altitude.
we were both using auto pilot.
at one point, i inadvertently, clicked off my auto pilot
and lost control of my p51. i was plunging earthward
and couldn't pull out...in the bomber, my instructor
kept saying, don't crash, don't crash.
fortunately, i was able to pull out near the deck.
much later, i came to understand, the ap altitude
was governed by elevator trim. when i kicked it off
at high altitude, the aircraft was still set at high elevator trim,
forcing me to loose control.
at least, i think that's what happened.
whew...live and learn.
 
Auto Pilot Stuff

Hello Smilo, Aleatorylamp,

It is still rather difficult for me to get time on the game machine to do any testing.
My development machine also has CFS installed but as you already know, it isn't cooperating.
My time is still not my own for a while longer. There are times when I have my laptop and have nothing else to do but that does not work for testing anything unfortunately.

Regarding the Dornier 17Z and Autopilot, I need to know a few things:

1. What is the Zero Fuel Weight (Weight in the AIR file that you are using)?
My intention is first to sanity check the Moment of Inertia values to see if they are appropriate.
I suspect that they are not. AI Air files tend to be screwed up in one way or another from what I have seen in the past.
I can do the MOI calculations as long as I have a laptop and access to the web to pull some dimensional specifications.

2. I had a chance to download V7 over the weekend and flew it for a few minutes.
I believe it is way too agile in Roll and especially in Pitch.
The sensitivity in Pitch is why I suspect that the MOI is too low but it is possible though less likely that the damping is too low.

3. Once those are "correct", then the easiest path is to adjust the trim effect per notch.
What to do depends on what the aeroplane is doing under autopilot.
If the oscillations are divergent (Get larger and larger) then the chances are that the trim effect per notch is too low.
That is because the autopilot can only change trim at a certain number of notches per second.
If the effect per notch is too small, then the autopilot cannot follow the changing situation and thus the pitch diverges.

If the autopilot cannot hold altitude (constantly goes slightly up or slightly down) then the probable reason is that the effect per notch is too high. It may be looking for a value in between where it can actually go.
If you turn the simulation speed up, you might see the aeroplane look like it is "shivering" because the notch changes happen quickly and have too great an effect.

It is also possible that you cannot find a notch value that solves both problems.
If not, then go for one that is slightly too small. If corrections are too great, the pitch will diverge, but that can be "solved" by turning the autopilot off and back on quickly where you expect the trim to stabilize. Keep doing it and the aeroplane will hit a point where the adjustments are not so great that the autopilot cannot correct in time with trim notches that are too small.

4. If it still seems that nothing fits together, think about minimizing the changes in trim from low speed to high speed.
I noticed that the V7 seemed not to be very well adjusted for neutral pitch trim.
That in itself is not a problem because many real aeroplanes are not well adjusted there.
Consider that this is what happened in real life with Jimmy Leeward's racing P-51 (Galloping Ghost) crash in 2011.
His aeroplane was running at around 530 MPH with full nose down trim to keep it going level.
When the trim tab fell off, all of a sudden, he hit somewhere between 11G and 17G.
It collapsed his seat and probably rendered him unconscious immediately.

With CFS, we actually can adjust the flight model for minimal trim changes with speed and the trim changes become much easier for the autopilot to deal with.
One can also adjust the AIR file so that it flies with neutral trim (pitch only) at a particular speed and altitude (and fuel load).

Yet another solution is to determine the trim settings required for a particular flight condition such as during the bomb run and list them in the checklist.
I do this for the projects I build.
It is tedious to do, but it is useful information for the virtual pilot.

Hope this helps. If not, I will see what I can do about the AIR file in a couple days.

- Ivan.
 
Hello Smilo,

I almost forgot to address the situation you brought up.
First of all, the deal with trim on the FW 190A as you probably already know is an interaction of a few factors that are just plain wrong.
The flight model is garbage anyway. I actually created a very similar situation with my A6M2 flight model with the intentional oscillation.
I wasn't trying to create such a situation but all of a sudden, I had the Zero under autopilot going about 40 MPH faster than it would normally.
I didn't explore any further but I do know it happened.

Your situation with the P-51D should not happen in real life, but does in the simulator (I believe) because trim notches are too large.
It makes the aeroplane unable to hold altitude exactly, because the trim setting is never stable.
Instead the autopilot is constantly switching it back and forth by a couple notches.
The problem is that when your autopilot goes off, YOU as a human cannot react fast enough to correct for a setting that may be very far from where it should be for equilibrium. This is what I believe happened.

In real life, most fighters did not have any autopilot which meant that the pilot had to manually trim the aeroplane for straight and level flight.
.....So there really was no autopilot to be lost.

BTW, your Dornier 17Z panel looks great.
Your issue with the twin engine gauges is exactly why I am working on gauges myself.
I am also working on plans to set up an experiment to help in figuring out how to do things but won't be able to do that either for a few days.

- Ivan.
 
weight in .air file

Hello Ivan
Thanks for your detailed info and instructions on autopilots. I´ll see if I can remedy the flaw.
Everything you say should come in useful!

Re. your question on the weight for the Do-17z2:

Empty weight:..................11486 lb
Crew of 4:...........................800 lb
Guns: 6x115 lb ....................860 lb
-----------------------------------
Dry weight in .air file:.........13146 lb

Standard Fuel:409 USG........2454 lb (no extra fuel tanks)
Bombs:.............................2205 lb (20 x 110 lb, although for long range half the number were carried)
Ammo: 4x1000 rounds...........113 lb (there are only 4 guns defined in the Dp files. 6 would be 169 lb).
-----------------------------------
Operational normal weight: 17918 lb

Thanks a lot for your time and help - I know you are quite busy with other stuff these days...
Cheers,
Aleatorylamp
 
Pitch and Roll moments better

Hello again, Ivan and Smilo,
Pitch and roll sensitivities are more correct now:
I had the elevator pitch moment at -1800 and aileron roll moment at -180,
and tuned them down to -1200 and -120.

Now it´s somewhat better:
If oscillations are more or less "normal" at around 100 fpm they get weaker and disappear.
If they are stronger, they won´t, but I suppose it´s normal.
Changing elevator trim from -0.25 to 0.20 didn´t help, as the more normal oscillations wouldn´t stop.

This seems to be coherent with your comments.
Then, I tried increasing the pitch MOI but it makes it worse, so I left those.
I could try further reducing pitch sensitivity, but maybe the plane will loose too much manouverability.

Cheers,
Aleatorylamp
 
Hello Aleatorylamp,

Just a couple comments....
I don't believe the Dornier 17Z really was all that maneuverable.
It was a fairly underpowered bomber though its aerodynamics apparently were not too bad.
I had noticed that it seemed to be a bit overly sensitive in pitch or at least such was my impression.
The roll rate also appeared to be way too good; There were no issues in doing a complete aileron roll.
I did not time it, but you might want to do that to get an idea of how it compares to a typical fighter.

I can't recall if I mentioned changing the pitch damping.... I don't think it will cure your autopilot problem but it will address some of the pitch sensitivity.

I am fairly certain that your ammunition load for the MGs is way off for weight.
The value I would use here is 1.12 ounces per round.
This is for a disintegrating link belt and that is not necessarily what was used for this aeroplane, so the number might be a touch off.
Now keep in mind that the actual guns used in the Dornier 17s were loaded with a 75 round saddle drum and we are not taking into account the weight of the magazines.... But the magazines are not thrown out the window when empty either.......
If in doubt, just look up what a single complete cartridge of 7.92 x 57 weighs with about a 175 grain bullet.
I don't remember the actual bullet weight for their aerial guns, but I am betting it is pretty close to that.
Don't use SAAMI numbers because that is only for US commercially loaded ammunition.
I probably have a few rounds of this caliber in the basement but that is likely to be with either 153 grain bullet (Chinese) or 196 grain heavy ball (Portuguese) and with things all over the place, I don't even know where to look at the moment.

Your weight summary seems to be working on the assumption that the loaded weight (is it just a normal loaded or maximum take-off weight?) is with full fuel AND full bomb load.
Do you know this to be the case? Usually it is not.

I haven't been able to find a copy of the spreadsheet I use to calculate MOIs but that isn't a hard process.
It is documented pretty well by various sources and is pretty much an estimate anyway with as little information as I could actually find on the aeroplane.

Hope this helps.

- Ivan.
 
Hello Ivan,
Thank you for your input, help and suggestions! I´ll see what I can do to make pitch and roll more realistic.

Then, I´ll check the ammo weights. I´d looked them up in oz. and put in 1000 rounds per machine-gun, with 4 being in the Dp files, but 6 in the .air file weight. MTOW is 19482 lb, but normal operational weight wasn´t Maximum take-off weight. Adding up standards fuel, crew, ammo and bombload, I got to 17918 lb. I don´t really want to fly a fully laden plane with the extra long-range fuel - I´d have had to reduce bomb load anyway. I thought it would fly better being 1500 lb lighter.
Anyway, I´ll check the weights again.

Cheers,
Aleatorylamp
 
Aircraft Weights

Hello Aleatorylamp,

Tonight I went back to where I was storing various aircraft information and found a data sheet I had written back in 2010.

Regarding Weights:
Empty weight according to my notes is 11,484 pounds.
The book "Fiegende Bleistift" by Manfred Griehl gives 5200 Kg which is 11,464 pounds.
Both numbers are in pretty good agreement with your value of 11,486 pounds.

Where I believe you are missing data is your lack of a listing for "Empty Equipped" weight.
US Aircraft often call this a "Basic Weight".
The empty weight only includes basic airframe, engines, fuel tanks and very little else.
Armour, radios, gun mounts, gun sights, guns, emergency equipment such as a raft, flares / flare guns, charts, etc. are not typically included.
The value I have listed for Empty Equipped is 13,145 pounds.
To me, even this seems a bit on the light side.
By the way, the MG15 was a relatively lightweight machine gun. I believe they were only about 20 pounds or so each.
Even the .50 Cal Browning was only about a 75-80 pound gun.

To this we add the 4 man crew.
Germans must be a bit heavier than others because the typical German aircraft documentation allows for 100 KG per crew member.
That would add (for simplicity's sake and to allow for personal belongings) about another 1000 pounds.

We are in agreement for fuel. The wing tanks could take 1550 liters or 409.52 US Gallons which would be 2457 pounds.
2454 pounds is good for 409 US Gallons.
This presumes that the ferry tank in the forward bomb bay is not installed.

Bomb load is 1000 Kg as specified so 2205 pounds is good.

I have the armament listed as 5 x 7,92 mm MG15 and a MG151/20 20 mm cannon as the swivel nose gun.
It sounds like the nose gun could vary which would explain why the nose framing was not always the same where the gun was mounted.
Assuming 6 x MG15 machine guns, we would have 6000 rounds of ammunition
Even though your DP only has 4 guns, it makes sense to include the full 6000 rounds.
The ammunition came in 75 round saddle drums carried on racks so it wasn't necessarily dedicated to a particular gun anyway.
At 1.12 ounces, that would give 420 pounds for 6000 rounds.

So what I am getting is
14,145 pounds Zero Fuel Weight
420 pounds Ammunition
2205 pounds Bombs
2457 pounds Fuel

for a total of
19,227 pounds for full fuel (no ferry) and bombs.

The number I have for an "overload" weight is 19,481 pounds which agrees with your value of 19,482 pounds.
Manfred Griehl's book lists the Fluggewicht as 8860 Kg which is 19,533 pounds which is only a bit above our maximum weights.

So with the exception of equipment weight, we are in pretty fair agreement.

- Ivan.
 
Hello Ivan,
OK!, it seems I was doing my arithmetic starting out with wrong numbers then.

So, just to understand it: First, I´ll use the 13145 lb empty equipped weight to include the guns, radios etc, increased by 1000 lb to include crew and their stuff (800+200 x 4), giving 14145 lb, and after that, add 420 lb ammo, and the normal fuel and bomb weights.

Then, the 6000 ammo rounds in the Dp file would be 4 guns with 1500 1.12 oz rounds per gun, one of the forward ones being a cannon. Shouldn´t all 4 swivel?

That, with increased MOI´s and lower pitch and roll sensitivities should then do the trick.

I already had the sensitivities lowered, and I´ve just done the rest. Now, aircraft movement around the three axes is more fitting with the added weight and lower MOI´s. Funnily enough, the unwanted the peaks in the performance curve are now smaller! So that´s great too!

Now, the only pending thing is the Autopilot oscillation, although, for the moment, it CAN be controlled with able damping joystick movements to reduce it to within + - 100 fpm, and then it will settle to near zero after a short while.

Attached, please find the new .air and Dp files.

Thanks a lot for your indefatigable help!
Cheers,
Aleatorylamp
 

Attachments

  • New Air and DP files.zip
    5.2 KB · Views: 0
Hello Aleatorylamp, Smilo,

Here is a little more of an explanation for the Empty Equipped and additional 1000 pounds for the Air Crew.
In reality, the Germans specified each crew member and equipment as 100 Kg.
I joke about Germans being heavier than others, but that is the number they chose to use.
Americans typically used 200 pounds per crew member but not always.
Japanese typically used 75 Kg per crew member.

(For the very special P-39 Airacobra, I have actually seen a reference where the weight of the pilot was specified as 150 pounds and that was an American plane.)

So, if I know the crew of the Dornier 17 is 400 Kg or 880 pounds, why did I use 1000 pounds???

My belief is that the Empty Equipped weight does not account for trapped fluids such as Engine Oil, trapped Fuel, etc.
A few gallons of oil easily make up the difference. There are also other little personal items which would suggest that 120 pounds is probably a low estimate. Perhaps some of the personal items are included in the crew weight?
It is hard to know for sure.

To get an idea of typical numbers, check out this thread on the FW 190A for which I had a very good listing of equipment weights.
http://www.sim-outhouse.com/sohforums/showthread.php/84748-Focke-Wulf-190A-Revisited

I SHOULD finish up that project and re release it. It is one of my favourites.

- Ivan.
 
Hello Ivan,
OK, Thanks. I´d already thought it was along those lines! Will check the link!

Hello Smilo,
I meant I was glad that your sim machine was working...

Cheers,
Aleatorylamp.
 
right...actually, it always has worked.
the problem has been the main monitor
of a two monitor system.
i like to use a 48" led tv for the main.
it puts the panel into a more realistic perspective.
not to mention, enlarges everything for these tired old eyes.

the problem was, my old lg 48" slowly crapped out.
the right side had trouble staying focused.
i recently removed and replaced it with a used samsung 48".
it, too, is starting to fail, but, at least for now,
it seems to warm up in a few minutes.

i test flew the do17 with the new dp and air files.
made the mistake of engaging ap while climbing.
it was like a bucking bronco...whahoo, hot fun.
i disengaged ap and settled her down to straight and level,
then reengaged ap. she seemed to want to fluctuate altitude,
plus or minus ten feet or so.
i left for a bit to go have lunch.
when i returned, the sim was at the free flight introduction screen,
which is an indication that the plane crashed.
i'll look into it more later, i've got some chores to do.
 
Hello Smilo,
Thanks for testing! The AP thing is a bit difficult to fix.
At 4x sim speed, the oscillation wanes to near zero if it´s not too strong, but of course that´s no good anyway.
Let´s see if I can get it any better.
Cheers,
Aleatorylamp
 
Armament Expert....

.....I am not. Certainly not as far as the Dornier 17Z goes.

Hello Aleatorylamp, Smilo,

I will still make a few comments along those lines though as if I were "The Expert" (Note the Caps!)
Some of the aeroplanes apparently had a MG151/20 20 mm cannon.
Most did not. It isn't even listed anywhere that I can find now.
For the CFS DP, I would just presume we have the standard version with 6 x 7.92 mm MG15s.

Now the problem with doing this exactly the way it looks is that the Dornier had a very ineffective armament layout in my opinion.
6 MGs sounds like a pretty heavy armament and it probably was HEAVY, but the arrangements were not logical.
The top three guns good coverage for angles but the problem was that you have one may in a tight spot moving between three guns.
We already agreed that this is best simulated with a single swivel gun perhaps with extra ammunition, but this would be way more effective than the real guns were with their 75 round magazines and inability to track passing targets even with a wide field of fire.

Now I have already commented on this before, so why am I making the same comments?
..Because it was asked whether it would make sense for 4 swivel guns.

I think it would not for the simple reason that there were only three gunners.
There is no way that four swivel guns could be operational at the same time.
The Bombardier if he was doing his job would most likely be in the nose acting as observer / bombardier / gunner.
If he was next to the pilot, the swivel gun in the nose would not be functional.
My idea is to convert one of the swivel guns to a fixed forward firing gun in the DP.
Of the two bombardier guns, the least effective is probably the one next to the pilot.
With the asymmetric windscreen, if the bombardier were manning the gun next to the pilot, he would actually be a bit forward of the pilot and blocking the view to the right half of the instrument panel AND he would not be doing his job.
That is why I didn't think it made sense to put the bombardier in the cockpit.
Of the two bombardier gun, the cockpit gun has a fairly poor field of fire anyway..... But it does look cool.

I think it would have made more sense for the top rear and side guns to all be replaced by an unpowered turret type position.
At least then, the gunner would have some hope of tracking a target instead of having to jump back and forth between guns.

Regarding Autopilot, I will see if I can get it to work on my computer tonight.
Perhaps I will have better luck here.

- Ivan.
 
Hello Ivan,
Oh dear, I´m so sorry! Silly question on the 4 swivel guns. I´d completely forgotten about the decision about the armament and the reasons for it. Yes, now I remember that one of the forward guns would be a fixed one.

So similarly, maybe the dorsal one could be on a swivel, and the ventral one fixed then? Both rear firing guns being fixed is probably a bit too limiting, I´d think.

Thanks very much for trying to fix the autopilot later!
Well, it´s past midnight here, and I´m off to bed.

Hello Smilo,
Thanks a lot too, for doubling up your crews for the Do-17 panel + testing, AND for the Ad2k modified CubeFly tutorial!
It keeps one on one´s toes a bit, doing two things at the same time, I suppose. Anyway, this doesn´t happen all the time!

Good night, guys!
Aleatorylamp
 
You must log in or register to reply here.
Back
Top