Sub physics - experts needed

[XabbaRus]

I asked the question once and was told no couldn't be done even if starting from a stupidly deep depth.

[Dr.Sid]

Originally Posted by XabbaRus

I asked the question once and was told no couldn't be done even if starting from a stupidly deep depth.

Are we talking about loop or roll now ? I thought you said loop and thought roll. So I used loop too. And now it seems you think loop after all.

[MarkShot]

Thanks for your answers. I personally know and have worked with quite a number of independent game developers. I think you have what it takes to realize your vision: drive, attention to detail, and the technical skills.

As an example, I would mention David O'Connor the lead designer/owner of Panther Games. (I've worked with him for over five years.) Over ten years ago, he set out to build a realtime engine of operational command. His peers in the military at the time were very skeptical and yet he kept at it with two essential focuses:

(1) Modeling real world processes as this actually occur.

(2) Making the AI component generic and as semantically analytical as possible of the terrain/objectives/force structure as opposed to going for the easy route of scripting.

In the end, he produced an amazing engine which captures the feel of military command better than 90% of the games on the market. Only a few have tried something similar: ProSim and MMG. This engine rivals what the actual US Army uses. I can say that, since we have team members who are veterans and/or consult to the General Staff College.

---

You raised the issue of "oversimulation". I think I understand what you mean, being that you have a programming background, I would think that this is perhaps one of the main risks that a finished product ever sees the light of day.

Back in the 1990s, I competed on flight sim ladders. Before the Web, I decided to build a utility for video game ladder administrators to manage ladders and publish updated status reports for BBS' and services like CompuServe and Prodigy.

I spent about six months on the project, but did not complete it. Why? Well, being in the industry I used all the best practices on my design. However, there weren't 20 people on the project; just me. After six months, I pulled the plug. I started from scratch, but the second time I cut corners. For example, I gave up on ERD database designs and compressed data structures and flattened them. I produced something very elegant and workable (from the UI) in about 3-4 months. However, internally it was very messy and did not have nearly the expansion possibilities of my first attempt. But I think you get the point.

Tailor the scope to match the level of resources available. Additionally, remember the longer you take to deliver the final project, the more backwards the graphics will look (especially 3D) when you finally deliver it.

Good luck!!!

[Dr.Sid]

I just can't stop playing with it. Here is another version

http://roger.questions.cz/other/sim3.zip

It has bank autopilot (B key) which tries to hold sail up.
It also has pitch angle and depth autopilot, just set desired depth or angle (both has shortcuts too, check the readme). Any manual input to controls will switch of corresponding autopilot (planes to angle or depth, rudder to course, bank to bank).

All autopilots reacts based on speed, they use larger elevations at slow speed. It's harder to loose control with them, but not impossible. All autopilots are very simple now.

Note that all autopilots now work only around 'normal' conditions, and only with forward movement. Anyway if you find some really strange behavior, let me know.

[Bubblehead Nuke]

Originally Posted by Dr.Sid

Send whatever you have Anyway for me the arrow keys works correctly. Left moves rudder toward negative values. Right toward positive.

To get some decent bank:
Press 0 for flank speed. Wait till speed hits 30. Ignore slight bank to the right due to screw reaction. Press ']' for full rudder right. It should perform something like barrel roll.
Now this isn't realistic from many points of view. You would loose the air from balast tanks when upside down. Screw and planes would not work above the water (I don't test it now).

Are you sure you have the sim2.zip version ?

When submerged there is NO AIR in the MBT's. You open the MBT vents and close them when you are sully submerged and the tanks are full of water.

On the 'Snap roll' as we called it it has to do with having the sail on the top of the hull and none below it. The center of rotation in all axes is the reactor compartment as that is the single most massive concentration of weight.

With that being said, when you toss the rudder over at high speed the ship rotates and the sail becomes becomes a HUGE plane surface due to a sudden change in the angle of incidence.

Now that the hull is rotating due to this asymmetrical vector, the rudder now is no longer in a horizontal plane and starts to induce a vertical vector into the ships motion. This now causes that HUGE plane called a sail to further change is angle of incidence and continue the roll which further rotates the ship. Eventually the rudder is going to become more of a STERN plane than a rudder and lift the stern causing the ship to pitch down sharply. While this is going on the boat will CONTINUE to rotate into the direction of the turn. Yes, you could in effect perform a huge sloppy barrel roll, BUT, you will exceed crush depth before you could complete the roll. There would be other 'complications' as well before you got over half way thought it, but you would crush before then.

Here is something interesting aboout the depth excursions due this handling problem 688's have:

The USS Oklahoma City crew had a shirt made once that said:

"60 down, Rods down, Still Around"

They experienced this problem in sea trials and managed to save their butts because of a GOOD throttleman and a EOOW who had initative.

We do have operating procedures for high speed ops and certain system to prevent this, but I feel it is not appropiate to discuss them in an open forum.. I will add that you are on the right track with the limiting of rudder movements.

[Dr.Sid]

Nice .. so I have it right, at least in what is causing the effects, and in what direction. Now what about the amount ? I can seem to be able to roll the boat under crush depth, so it calls for little less strong effects.

What do you think BubbleheadNuke, I mean about the amounts of these effects in my sim ? Too wild ? Too tame ?

As for preventing system, all any other 'supposed to be secret' system .. you don't have to tell anything, that's perfectly OK.
I'd like to use different rule on such systems: if it is possible and usefull, it is most probably used

[Bill Nichols]

Originally Posted by Dr.Sid

Nice .. so I have it right, at least in what is causing the effects, and in what direction. Now what about the amount ? I can seem to be able to roll the boat under crush depth, so it calls for little less strong effects.

A hard rudder at high speed can easily cause a snap-roll that will quickly send you below crush depth, if no corrective action is taken. That's one of the first things they show you on the 3-axis simulators at sub school.

[Molon Labe]

Just a thought to throw in here...

Those of us who fly F4 are familiar with the expression "pulling on the limiter." The fly-by-wire system controlls the amount of elevator pitch rather than the pilot controlling it directly with the stick. This prevents overstressing the ordnance, overstressing the airframe, and helps maintain control of the aircraft. There is also an override switch to use in stalls, when the flying on the limiter won't get you any results.

I mention this as a conceptual approach for how to deal with users executing high-speed manuevers, which also might be rather consistent with the systems BH has alluded to.

[MarkShot]

BH Nuke and Bill,

You describe maneuvers with the sub basically behaving like a large aircraft; very interesting.

It makes me go back to all my online air combat days and start to wonder what AC maneuvering and concepts is analogous to subs:

(1) For example, energy (kinetic and potential) are often considered with the exchange being done through converting altitude. Of course, a sub is supposed to be more or less neutral and not a flying brick. However, BH Nuke has pointed out how blowing tanks can produce great speeds. So, is flooding or blowing ever used during evasive maneuvers to realize speeds in excess beyond propulsion driven speeds or acceleration?

(2) In flying, one is often concerned with making energy efficient turns as opposed to creating high wing loading situations. In evasive maneuvers does one try to not take too much speed off the boat by how one makes their radical maneuvers?

(3) Going back to radical maneuvers and evasion. A missile although a lot faster than an AC may for the same reason (high speed and momentum) be unable to maneuver as radically and turn as tight. (Of course, thrust vectoring and the lack of human G limitations do give a missile some advantages.) Thus, an AC can if maneuvered tightly at the right moment effectively dodge a missile in a turn (of course, there are proximity fuses for this). Is it correct to assume that none of this applies to subs and torps given that the relative speed difference is not that great? Or can a sub turn so radically as to create an overshoot situation with a fast moving torpedo?

(4) In flying, there are too things that happen to control surfaces that become very problem. First, if the angle of attack is too great, the surface could experience turbulence and effective cease to provide control. Second, if the speed is too great, the surface can experience compression and effectively become locked and/or have no affect on movement. Does any of this apply?

(5) Speaking of control surfaces, AC often have trim surfaces or trim applied directly to control surfaces to correct for various flight conditions. Do sub control surfaces have analogous features?

(6) As mentioned previously by BH Nuke, speed changes things. How significant do speed considerations impact the behavior of sub control surfaces? Is there a sweet spot of optimum control? Are there problems at the high and low ends of the speed range?

(7) With AC the fuselage itself also functions to provide lift and handling. Does the sub's hull provide a similar function?

Thanks. (feel free to not answer as appropriate)

[Dr.Sid]

All what he asked ..


7) Sub's body does much more lift then the control surfaces. It's just matter of area. I guess sub can use front and rear surfaces together in same direction to produce lift, but it will be much less lift then whole body can produce. Maybe this is used for handling in PD. To experts: is it ?

And in turns whole trajectory change is because of body lift (I mean lift to the side .. is there any special word for that ?). Rudder just turns the body around. Water hits the body from side, producing side-lift, moving the boat in direction of the turn. With aircraft it is different. Body has small area compared to the wings, so it's easier to change vector of wings lift to do the turn.

[MarkShot]

(8) Someone mentioned roll induced by the torque of the prop. With AC such torque tends to make turning in one direction easier than the other. Is such true of subs that don't have counter rotating props? Also, in AC rudders aren't perfectly straight to offset the prop torque. Are subs truely symmetrical?

---

Dr. Sid,

You asked about variable pitch props. In my current readings, such things were tried in the early 1900s. There is also discussion of various clutches too. Pressure clutches were ultimately abandoned as unreliable.

My second book just arrived ... US Subs post 1945 ... it's hard as I tempted to jump into the second book, since I am only up to reading about the year 1910 and US subs used for harbor defense; but I am trying to have some self-discipline. Also, the book points out that US subs rated for harbor defense versus European subs rated as ocean going was largely semantics. US subs of the period had similar capabilities to European subs, but given the closeness of European coast lines, they considered their subs capable of an 800nm range to be ocean going. Whereas given the USA's relative isolation, the same boats could only be used for harbor defense.

Initially trying to get boats on the West Coast and the PI was very problematic. There was no way to transit them there (they were too bulky to transport by ship) and West Coast build facilities didn't initially exist.

[Bill Nichols]

Originally Posted by MarkShot

BH Nuke and Bill,

You describe maneuvers with the sub basically behaving like a large aircraft; very interesting.

It makes me go back to all my online air combat days and start to wonder what AC maneuvering and concepts is analogous to subs:

(1) For example, energy (kinetic and potential) are often considered with the exchange being done through converting altitude. Of course, a sub is supposed to be more or less neutral and not a flying brick. However, BH Nuke has pointed out how blowing tanks can produce great speeds. So, is flooding or blowing ever used during evasive maneuvers to realize speeds in excess beyond propulsion driven speeds or acceleration?

(2) In flying, one is often concerned with making energy efficient turns as opposed to creating high wing loading situations. In evasive maneuvers does one try to not take too much speed off the boat by how one makes their radical maneuvers?

(3) Going back to radical maneuvers and evasion. A missile although a lot faster than an AC may for the same reason (high speed and momentum) be unable to maneuver as radically and turn as tight. (Of course, thrust vectoring and the lack of human G limitations do give a missile some advantages.) Thus, an AC can if maneuvered tightly at the right moment effectively dodge a missile in a turn (of course, there are proximity fuses for this). Is it correct to assume that none of this applies to subs and torps given that the relative speed difference is not that great? Or can a sub turn so radically as to create an overshoot situation with a fast moving torpedo?

(4) In flying, there are too things that happen to control surfaces that become very problem. First, if the angle of attack is too great, the surface could experience turbulence and effective cease to provide control. Second, if the speed is too great, the surface can experience compression and effectively become locked and/or have no affect on movement. Does any of this apply?

(5) Speaking of control surfaces, AC often have trim surfaces or trim applied directly to control surfaces to correct for various flight conditions. Do sub control surfaces have analogous features?

(6) As mentioned previously by BH Nuke, speed changes things. How significant do speed considerations impact the behavior of sub control surfaces? Is there a sweet spot of optimum control? Are there problems at the high and low ends of the speed range?

(7) With AC the fuselage itself also functions to provide lift and handling. Does the sub's hull provide a similar function?

Thanks. (feel free to not answer as appropriate)

Quick response:

1) No.
2) Yes. The general rule in torpedo evasion is "speed before maneuvers."
3) Modern torpedos generally have much higher turn rates than their targets, so once they have acquired, don't count on out-turning them.
4) High-speed compression effects don't apply underwater because water is an incompressible fluid. Control surface 'stall' at high AoA: I never encountered it and it was never mentioned during training, so my supposition is that although theoretically possible, it is not a in reality not a problem. One control surface effect that does happen (and doesn't in aircraft) occurs at very low speeds: the force created by the planes isn't enough to cause a change in sub body angle, so the hull doesn't generate 'lift'. In practice, this means if you 'pull up' on the stern planes while at very slow speed, the generated force pushes the stern down more than the hull can create lift. Thus, instead of going up (as expected), the boat instead goes deeper.
5) Some sub classes have 'trim' devices on their planes and rudders, I've heard they are more often used for high-speed operations than for trim purposes. (At high speed, plane/rudder motions can cause violent responses if the operator isn't careful. By using the 'trim tabs' only when at high speed, this problem is mitigated).
6) I think I've already covered this, above.
7) Hydrodynamic 'lift' created by the hull, created when the boat takes an AoA or 'slip', is a major (if not THE major) contributor to submarine maneuvering. As diving officer, we were taught the mantra 'planes, angle, speed'. This means, first use the planes to control depth when the boat is 'heavy' or 'light', if this isn't enough then put an angle on the boat, and finally (as a last resort) increase speed. A good diving officer will keep the boat in trim so that only small plane motions are needed to control depth.

As an aside, when operating near the surface the hull can create significant lift even with no AoA (due to the Bernoulli effect). To the operator, it seems as if the boat is being 'sucked' up to the surface. This can make depth-keeping at periscope depth difficult (especially in boomers, which have a large, flat deck that exacerbates the 'suction' effect). One trick often used is to flood extra water into the aft trim tanks, to allow maintaining periscope depth in a 'stern down' condition to reduce the 'suction' effect.

(Surface ship drivers are familiar with the Bernoulli effect when conducting side-by-side underway replenishment: in this case the effect is to 'suck' the ships into each other.)

[fatty]

Who needs a subtle hand on the planes when you can do this:

[Dr.Sid]

Bill .. thanks for nice sum-up. So there are 2 ways of settings planes ? For lift (front and rear same direction) and for pitch rotation (opposite direction) ?
How is this done ? I mean .. I know 688 has 2 controlling wheels like in airplane .. which wheel controls what surfaces ? How is bank controlled ? Are there any more controls like pedals ? (I bet they are).

And Bill .. what is your opinion about screw reaction ?
Fatty .. I guess you'll face mutiny on your boat soon !

[Bill Nichols]

Originally Posted by Dr.Sid

Bill .. thanks for nice sum-up. So there are 2 ways of settings planes ? For lift (front and rear same direction) and for pitch rotation (opposite direction) ?
How is this done ? I mean .. I know 688 has 2 controlling wheels like in airplane .. which wheel controls what surfaces ? How is bank controlled ? Are there any more controls like pedals ? (I bet they are).

And Bill .. what is your opinion about screw reaction ?
Fatty .. I guess you'll face mutiny on your boat soon !

Traditionally, there are two planesmen. The bow/sail planes are controlled by one, the sternplanes and rudder are controlled by the other. Thus the two control wheels. If I recall correctly, the sternplanes/rudder were controlled from the left seat, the bow planes from the right seat. However, control could be swapped between the two. SOP was for the stern planesman to control the boat's angle, and the bow/sail planesman to control depth.

Nope, no pedals, and no control of bank. 'Snap rolls' are prevented by going easy on the rudder at high speeds; the tendency to pitch down in turns is counteracted by the stern planes (thus, the helmsman, who controls both the rudder and stern planes, needs to be experienced).

Don't know about how screw 'torque' is counteracted in sub design (other than the rare counter-rotating screw, and two-screw subs). It was never a problem afaik.