Adding pitch control to a model sub.

[cstranc]

Andy,
You know that 3/4"

Andy,
You know that 3/4" sounds like a good distance if you are moving batteries around. Of course if we use half the weight it will have to move twice as far. etc. So much to play with so little time.

I am really glad you mentioned the metacentric height. It's one of those terms you just need to know when dealing with anything that floats. But I have been to toung tied to mention it and describe what it is. For those that are curious check out http://en.wikipedia.org/wiki/Metacentric_height

Sorry it does not talk about submarines specifically. Such is life when 99.9% of water craft are confined to the surface. Still the link is to Wikipedia. Maybe we could edit the nice picture and put a torpedo in it. That should get some notice...

That is an excellent suggestion to disable the pitch control when under way. That's when the dive planes come into their own. I shall make sure it gets into my plans.

Bob, Andy you have inspired me. As soon as I finish this post, then do my day job that spills into my night, I will get some numbers on how slow that lead screw is... I really like the idea of fixing the ballast tank and putting a lead weight on the leadscrew. Having a servo would be the ultimate in simplfication. Out with the APC and 12V motor.

But curiosity has me and I am going to see how the system behaves with the moving ballast tank first. Maybe that way I can actually say "did it both ways, and this was better".

Neil,
It's funny you mention the pump idea. I was interesting in that too and found a post that mentioned a 12v fuel pump ( http://www.ripmax.com/item.asp?itemid=L ... tegory=080 ). I tried to get one to play with, but could not get a distributor that would talk to me

The other thing to think about is you may not reallly need two tanks. If you had one tank at the front that kept the model level when it was 1/2 full then you could pump water in to drop the nose, and take water out to lift it. BUT this would probably need you to have the air in the one tank at pressure. Two tanks sounds much easier if you have space.

I do tend to agree with Andy that it could be more difficult to get this to work accurately. I suppose you could move the tanks closer to the center of the hull so you need to move more water to have the same effect. Still I think it is in-herently less accurate than a lead screw. I don't know the size of hull you are thinking about but I'm guessing that you would need a fair amount of real estate for the two tanks and a pump.

Nuke!
This is my second week on the message board. I have been jealous because everyone has their own little message beneath each post and I could not think of anything... But now I know I wish to be known as the "over engineer" in your / our honor

Please don't expect daily updates. Most of the time I am a weekend (and every idle thought during the week) modeller.

But today I do have a little thing to add. Time to upload some more pictures...

[cstranc]

We all know that air

We all know that air loose in the hull is probably the cause for 80% of the "less than elegant" dives.

It is quite common that as a model sub dives it can trap bubbles of air against the top of the hull. If you think of a modern nuke shaped hull as it dives most of the air will escape though a vent in the sail but some will slide back allong the underside of the top hull and get stuck at the "highest" point around where the hull tapers down towards the stern.

So before we do all sorts of electronics pitch controlling stuff let us see what it takes to get those pesky air bubbles out of the hull... I know, it's not rocket sciene. We need holes in the hull. But are all holes created equal?

Time for a little experiment.



Here we have a length of 3" plexiglass tube (1/8" thick) with duck tape closing off then ends so it can trap some air. If you look closely you will see a 1/6" hole drilled in the center of the tube.

The plan goes like this.
- Make a hole in the tube.
- Submerge the tube 1" below the surface and measure the size of the bubble remaining in the tube.
- Move the tube 3" below the surface and measure the bubble again.
- Repeat with a different hole.

This is what I found.

                            Size of Bubble 
Hole Size         at 1" deep          at 3" deep
1/16"                  33mm               25mm
5/64"                  25mm               20mm
3/32"                  18mm               16mm
7/64"                  0                      0       (occasionally bubbles could stay)
1/8"                    0                      0       (bubble almost always go)

What else did I learn. Well it would help a little to use a countersink to increase the hole size on the inside of the hull. That way you could have a 5/64" hole on the outside of the hull and 1/8" on the inside.

I thought that square holes would release the bubble faster. They did not.

On my model I have a bunch of mooring points that are 3/16" holes with a bar going across the centerline. I was afraid the bar would prevent the air from escaping. It did not this geometry worked well.

Here are a couple of other hole things I tried.



The brass vent on the left is three 1/8" x 1/4" rectangles glued togeher. It removed all the bubbles.

The big chunk of mosquito netting could keep lots of air trapped in the hull.

That's it for my notes on the effect of the hole sizes.

Then you have to think about where the holes need to be. The funny thing about bubbles. You would think they would always just go to the highest point. There is certainly a pitch angle where this will happen. But for small angles they can just choose to stay still.

I'm pretty happy with the Sierra. It has lots of mooring points and vents marked all over the top surface. I will be taking great care to punch each one through the hull so they can discarge any bubbles that want to hitch a ride on my model.

[cstranc]

Well I think this is

Well I think this is the weekend / week to play with the pitch control. As people have pointed out there are several ways to implement this. I would like to test now, so that I can layout the components in the hull after the experiment.

The way I see it there are a couple of things to decide:

1. What to move?
- The ballast tank.
- The batteries.
- A lead weight.

2. How to move it?
I like the idea of using a lead screw because it gives such accurate control. I suppose I should think of other alternatives too...
So right now I am thinking curious if the lead screw be driven by a speed controller and a motor, or by a continuous rotation servo.

Other things to think about:
- Should it be driven by the APC, or not?

I would love people ideas on what to else to test.

This is the basic layout for the test... The hull is 68" long and just under 8" wide.



The hole has been cut out for the sail, but the sail has not been attached. I have drilled a few of the hull features out to let out the air bubbles.

These are the "moving parts" so far.



For a water tight container I am using a Pelican Case 1050 ( http://www.pelicanproducts.ca/industries/Marine.htm ).

I have taken one of these down 9 feet in a swimming pool. It works well, it opens and closes with a simple / secure mechanism. I am planning to use these cases to hold the motor and batteries in the completed model.

The other items]http://www.subcommittee.com/forum/In the container.jpg[/img]

I have two conductive posts mounted in the lid of the case to allow me to connect to the motor on the drive screw.

And this is what it will look like for at least one of the tests...



Now I have to fill the testing tank, charge the batteries, put in some lead weight and foam to simulate the rest of the hull contents...

Ohh what fun.

[Rogue Sub]

Nice Signature Chris. I see

Nice Signature Chris. I see you have found yourself. Good idea on the WTC. Did it come to you on your own. I know another guy that used the same type of thing from a company called GSI. Is yours made of lexan aswell. Is held up to 15-20 feet if I remember right. We were playing with it in a divers den and we could go deep enough to annoy them with no leakage.

[subshop]

PITCH CONTROL

CHRIS, first off i'm building a /32 scale fleet boat, about 10' long. my thoughts about pitch control: what if the bow ballast tank was controlled by a switched pump, the aft tank was controlled by another pump hooked up to a apc. now , as you dive you increase the ballast in forward tank and bow begins to sink while rear tank is being filled in response to the apc trying to level the boat with the bow. when you stop filling the forward tank the apc will also stop filling the rear tank when it levels out the boat. if at any point you use the bow planes to move up or down the rear tank will adjust as depth changes to level off again. when you level off the bow planes if the rear ballast tank is not trimmed correctly the pump will make that adjustment through the apc. when the forward tanks are pumped out to raise the boat the rear tank begins to pump out the water.
hope this makes sense . looks good on paper can't wait to try it out this summer. any input would be great. neil

[cstranc]

Nuke,
I made a model

Nuke,
I made a model last year using the 3" cylinders from Mike's Subworks. When the seal was installed in the end cap it was a bear to put the end cap on and remove it. Also I was a little frustrated that all the shafts (twin screws) cables, ... had to come out through one end cap.

I was looking for an alternative and planned to make a custom WTC when I bumped into a pelican case in an outdoor store. It's the perfect size for batteries, motors, ... But I am still using a custom WTC for all the radio gear. Just want the extra space.

Good to know these things can go deep. I have a suspicion that if I ever see the Sierra down below 10' I'll be diving in to retrieve it.

I shall look up the GSI containers. The Pelican 1050 is nice, but if I can get something bigger I might go for it.

Neil,
That must be some increadible boat! At 10' you'll need a semi trailer to move it. Or I suppose with a little renovation you could move down to the marina and live on board

I am guessing that the sheer size of the hull will help a lot in providing stable dives.

Can you help me to understand your setup a little more. Ok, you have seperate forward and aft tanks. I assume that you have a vent on the top of each tank to allow the air to escape (causing them to flood from ports on the bottom of the tanks). So I was puzzled how a switched pump would actually adjust the water level in the two tanks.

When you want the sub to angle down and pump water from the stern tank to the bow tanks would that not cause a higher pressure in the bow tanks and the water flow in (that is causing the dive) would reduce negtaing what you are doing.

Then when it's time to surface pumping the water around does not sound right either. You really want to control the volume of air in each tank. Would you be using the pump to move air or the water?

When the a sub is submerged people normally have their ballast tanks empty to prevent bubbles shifting. I suppose this should not be as critical when you have seperate forward and aft tanks.

If you really want to control the dive by controlling the amount of water in each tank you could consider using two pumps, one for each tank. Instead of having them free flooding you would force the water into them. But that would get really complicated really quickly too. If the tanks are meant to keep a head of pressure then you would need a check valve to prevent back flow. If the tanks are not meant to be unter pressure then you need to coordinate the vent valve with the pump.

Hey I'm getting confused just thinking about it. Do you have any plans that you can post. A picture is worth a thousand word (if your not stuck in a dark hole)

[subshop]

pitch control

chris, yes there would be two separate pumps, no vent holes in the tanks and the pumps are peristaltic, hence no need for any type of valve. the peristaltic
pump provides a good seal. i figured that if the boat is trimmed so that decks are just above the waterline then there will not be much additional ballast needed to submerge the sub . the pumps are rated at 12 oz. per minute so it should give the boat a realistic looking dive . i hope ! same for surfacing.
i should correct myself there will be valves on the tanks that will be part of the failsafe system. i will have a air backup supply that when activated will introduce compressed air into the tanks and the valves will allow water to escape. neil

[Rogue Sub]

Chris,
Ill post the web site

Chris,
Ill post the web site when i get home. The company makes some pretty large ones and they are all constructed of lexan.

[cstranc]

Neil,
Last year I built

Neil,
Last year I built my first sub (so I am NO expert). I used a gear pump and a check valve to pressurize a ballast tank. Any gear pump has a stalling pressure, where it cannot force more water. That's not a sudden thing. The pump can do 12oz per minute with zero back pressure. By the time it has doubled the pressure in the tank it may be down to 6oz per min, ... I found my pumps would essentially stop when they have compressed the air to 1/3 it's normal volume.

Actually this is what it looked like:


No speed controller it was either on or off in either direction. I had three micro switches activated by a single servo. One switch for the check valve, another for the pump motor, and the last to control the direction of the pump (by triggering a relay).

The microswitches were mountes side by side, and the servo pushed a plate that had ridges / grooves in it to activate each micro switch at a different time.

Below the legend you can see when the switches activated based on servo motion. You can see the direction is on half the time (causing the motor to fill the tank) and off half the time (causing it to empty the tank).
When moving the servo to fill the tank the motor would switch on first (allowing it to build pressure) then the valve would open. When moving the servo to empty the tank it would open the valve first (allowing water to flow out forced by the pressure alone) then at the end of the travel it would turn the pump on to aid the discharge.

Looked good on paper. I did not like the implementation. Always a little afraid the ballast tank might leak/rupture , leaving me with insufficient air to surface. Was a damn pain getting a hose clamp for the pressure side of the pump. In the end it was prone to leaking. The solenoid was fine because it used those copper compression fittings.

Then there was the overall performance. I had a 66" 688 class sub. The ballast tank took up all the space I could give it and because I always had the 1/3 of the tank filled with air it did not give me the delta in bouyancy I wanted.

The size of tank should not be an issue for you since you can mount it below the waterline. Then you don't need to take on water to compensate for the air in the top of the tank.

It's interesting. I was using a gear pump, and I certainly had back flow through it when it was off. I would test out your peristaltic pump, just to be sure it behaves as you expect through the full pressure range.

I am really curious how the thing would behave with a speed controller. It may not be what you expect. When the tank is empty putting the speed controller to 1/2 throttle may start pumping. By the time the tank is 1/2 full you may find the 1/2 throttle does not generate the torque needed to move the water against the pressure in the tank.

My Sierra is far more traditional. Free flooding ballast with compressed air or propell to flush it out.

Nuke,
Please do post the info. I was testing today and noticed two drops of water in my WTC! Shall have to track that down since I was less that 2' deep at all times.

[cstranc]

Well Environment Canada issued a

Well Environment Canada issued a "winter storm warning" for today. The weather did not amount to much (light snow / rain) but it was the perfect excuse to get into the workshop...

This was the time to test out the pitch control systems! Many hours of work to get to this point, finally to put the sub into it's natural element.

Being terribly methodical and an over engineer, I had to plan out the day. It was to go something like this...

Step 0]http://www.subcommittee.com/forum/A nice work surface.jpg[/img]

And after]http://www.subcommittee.com/forum/Hours of fun.jpg[/img]

I built this testing tank for the last model. After using it I found that a couple of sheets of plywood and you have an awesome table. Great access from all sides and big enough for the model. But now it was time to fill it with that cold Manitoba water. Filling was uneventfull except when the hose got blocked with ice... At least it felt like that when I had to retrieve things from the bottom of the tank...

Next setup item]http://www.subcommittee.com/forum/Ballast tank assy.jpg[/img]

The plan is to remove the threads at the desired end points for the ballast tank travel. So we marked the end of travel (for the nut attached to one end of the tank)]http://www.subcommittee.com/forum/Mark travel end points.jpg[/img]

And then filed away the threads...



I am using a screw as a guide post for the file so I get a nice consistent location. I made the gap without thread a single file width. This is about 1.5 times the width of the nut.

This is what the lead screw looks like after modification.


Before assembling it all back together again I had to add some foam to the ballast tank.



The logic built into the system goes like: "If you want to raise the bow of the sub move the ballast tank forward". Obviously this would work while the ballast tank is full of air. I had to make sure it would work when the ballast tank is empty also. Otherwise if the tank itself was heavier than water when it shifted forward it would sink the bow, not raise it.

Ok, almost there. All I need to do is add some weight and floatation to the hull. I am not ready to put in all the components yet (motor, batteries, propell tank, ......) but I need to have some weight in there to make this a little more realistic. So I cut up a 3lb diving weight and distributed it in the hull.

Thanks to Dad for pointing out the cheapest way to buy lead is at a diving store. And also for making this nice little jig to melt the lead and pour it into handy bars that are great for trimming a sub. Too bad I misplaced all those bars and spent hours cleaning the workshop looking for them. Still it's always nice to have a clean workshop...

So you are probably thinking "How cheap lead is lead at a diving store?" It's gotta be pretty cheap, have you ever noticed how all the divers end up at the bottom of the lakes? Ya just can't have enough lead. (it's about $1CDN per pound).

This is 3lb of lead, and just enough foam to float it.



This gave me a rough idea how much foam I had to install in the hull. So I started doing my initial ballasting with the sub looking like...



Note this is just a test to play with the pitch control system. So I am holding everything in place with tie wraps. Don't try that in the open ocean folks. Believe me it will look more elegant when the Sierra really does set sail.

You can also see the external control of the ballast tank through some fuel line. I had a compressor I could use to fill the tank, but mostly I just blew air into it manually.

Now the steps I had planned for the ballasting went like:

Get the hull to neutral bouyancy when the ballast tank is flooded.
Keep dropping the hull into the water and flooding the tank. Add/remove floation as needed to make the hull just float a little.

Level the hull while submerged
So I have enough foam and weight, now shift it around so the hull sits level in the water while submerged.

In truth this step is kinda merged with the first step. Normally you say "hmm the bow is too high" and then remove foam from the bow which also gives you the correct balance.

Level the hull while surfaced
Empty the ballast tank and see how the sub sits in the water. At this point if the bow rises too high it means the ballast tank is too far forward. You need to move the ballast tank back, but you also need to move some of your other floatation foam forward to preserve the balance you had while submerged.

At this stage you start doing all three steps together untill all is good.

It might look a little like this...

It floats!


It sinks like a wallowing pig!

(as in it did not dive evenly)

It sulks at the bottom of the tank...


Ack. So much for a gracefull undersea vessel. Still I hope we can expect every model to behave this way the first time it hits the water. It's time to adjust.

Now initially I thought that I wanted to be able to see the pitch control mechanism at work when I was testing, so I left the upper hull off. But seeing the Sierra like this was just too horrible. As they say in the Navy - half a sub usually sinks. So I decided that I would rather assemble and disassemble the entire sub for this test, even if it's tougher to see what's going on inside.

Before I assembled and floated the full sub I decided to modify the sail a little...


First, a little foam (just where the escape pod is, I loved the irony). In the ideal case we want the model to be slightly positively bouyant. Now if all the foam is down below the waterline I thought it would be difficult to say "stop here at the top of the sail, this depth is perfect". But if I drop a little foam there it give a little natural pause.

Second I drilled two holes. One to let the ballast tank control hose out, the other to let the air from the hull escape.

So now we are going to be ballasting this...

I feel better already, now it looks like a sub, at least in dry dock.

It's first attempted dive...


You could hear the crew screaming as she went down....

On the surface it sat nice and level, and there is lead on the keel, so I assumed it would stay nice and level. But I had done a chain reaction goof...

- Installed a single 7.2V battery off the centerline of the hull.
- Noticed it did not sit right when the sub was surfaced so I moved foam over to the "deep" side.

Now as the sub started diving with more foam on one side then the other it started to get more bouyant force on one side then the other, so it keeled over.

I corrected this in several ways.
- Move a lead weight to counterballance the off center battery.
- Review all the foam and ensure it was cut symetrically from side to side. This is me duplicating the outline of one edge onto the other edge.


If you find your sub tends to rock to a side when it dives it could well be that "slightly different foam profiles on the two sides of the hull.

Of course the quick thinkers among us will probably point out that all the foam should be below the waterline when surfaced, so this should not happen. they are right. My model is under weighted so the foam must be above the waterline. I promise to do better next time.

Speaking of next time. A few more times out and into the water and...



The hull is level in the water, and slightly bouyant.



You can see just to top of the sail is above water. Also notice the little bit of foam. I am sure that's why the model sat so nicely at this depth.

And when we blow the ballast tank...


Ha! Russian subs. 2/3 of the upper hull is out of the water. In truth this model is sitting high even for a Russian sub. That means the ballast tank is too large, but that's ok. The ballast tank was sized so that it could lift things like the bow planes assembly and torpedo tubes out of the water. I think I shall need that bouyancy later...

That's it for the first ballasting of the hull. I shall have to add more tomorrow and describe moving the ballast tank around.

Scanning my notes for the day I see:

- It's probably good to have multiple points in the hull to add floatation and weight. Adding foam closer to the center will lift the hull more evenly that at either end, so depending on what you need...

- Shap of the ballast tank can be important... My ballast tank looks like:

Note how it is raised in the center of the hull? I thought that would guide the air there when it starts rising. What I found was: the towed array pod would not drain fast enough causing a large weight to lift out of the water and flop to one side. Then the shape of the ballast tank did not create a strong restoring force. Eventually it does right itself, but both the pod and the ballast tank will get a little extra care before the next session...

- Make sure your foam is below the waterline (while surfaced)...

- I was actually amazed how well the model sat flat. Still I was looking for excuses to change things. I think I shall split the foam in the sail and put half at the front and half at the back. You can figure out why.

[Rogue Sub]

Heres the link as promissed
http://www.gsioutdoors.com/Products/Lex

Heres the link as promissed
http://www.gsioutdoors.com/Products/Lex ... Stuff.html

[cstranc]

Great. There is even a

Great. There is even a supplier here in Winnipeg. Another thing to hunt down.

[cstranc]

Nuke,
Thanks again for the

Nuke,
Thanks again for the GSI post. Funny thing is they were sitting on the next shelf over from the Pelican 1050 I purchased earlier.

Here is my new improved capacity WTC.


$19 bucks, and it's dead easy to open and close and I can poke control rods and electrical connections out of most of it's surface area.

Darn good thing the Sierra is almost 8" in diameter

Of course there is a place for both. Back where the tail tapers down the 1050 will still fit.

[Rogue Sub]

Oh yeah that thing looks

Oh yeah that thing looks as hard as nails! Cant wait to see all the electronics crammed in there. Makes me wanna go buy one to play with except i have nothing to stick it in. Could make an rc Lexan GSI container whoopiiii

[cstranc]

Ok on to the testing.

Ok on to the testing. I had done the static trimming of the hull. It now sat level and slightly bouyant when the ballast tank was flooded, and it sat level when the ballast tank was dry.

This was the moment I have been waiting for... Moving that ballast tank around. To simplify things for the first test I took the APC out of the picture. Channel 2 (the dive planes) went strait to the speed controller that drives the motor that moves the lead screw that moves the ballast tank.

Time to attach an angular measuring device so I can quatify the results and away we go.

Here we are, submerged with the ballast tank fully forward.


And with if fully reversed...


I know. It's a good thing that I put the protractor on there so we can see the amazing 5 degree change in angle as the ballast tank moves 7" allong the hull.

To think I was worried about it being over sensitive...

Does this make sense? I suppose it does. The sub will adjust it's angle so the center of mass is below the center of bouyancy. And if you remember I only added a little bouyancy to the ballast tank. So moving that little bit of foam around does not have much of an impact.

That's easy to test & fix, just attach more foam to the ballast tanks and see what happens.

For some reason I forgot to measure the change in angle moving the ballast tanks when the sub was on the surface. I shall detail that the next time I test.

Hmm. Round #1 and I have to say I like the idea of moving weight around in the hull more and more....

But before I get to the second round of "automatic pitch control" testing it's time for a commercial break.