Print

Please login or register.
1 Hour 1 Day 1 Week 1 Month Forever
Login with username, password and session length.

[NFMike]

The principle required is fairly simple.
Each turret is equipped so it knows where it is pointed relative to dead ahead and arcs where it cannot point.
A central control knows the ship's compass heading and is told the compass direction of the target. With a bit of math it works out the target bearing relative to the ship's head and sends it to all the turrets (and updates this whenever the ship's compass heading or target direction changes, or just continuously).
Each turret then rotates to the target bearing (or as near as it can) and decides whether it can fire at the target (not if it can't reach the bearing or is in a no-fire arc).

In the case of a micro-computer the turrets will largely be virtual objects within the program, with just the servos at the turret (assuming use of servos).
Achieving that isn't so simple of course.

[derekwarner]

Bob........don't  let the turret numbers 1 to 7 inclusive put you off .......

Our MBM member Geoff with his 1914 Iron Duke has mastered getting turrets A, B, Q, X & Y to point at the same target.....[and fire].......brilliantly achieved

http://www.modelboatmayhem.co.uk/forum/index.php/topic,43803.msg587567.html#msg587567

Derek

[Colin Bishop]

Derek, that photo you have posted shows that the forward turrets are firing on the port bow while the aft turrets are actually pointing over the port quarter. The other photos posted on Geoff's topic appear to show that the turrets only all bear in the same direction when firing at 90 degrees on the beam. Correct me if I am wrong but this would appear to be basically the 3 position electro mechanical system referred to earlier as opposed to the ability to point all the turrets in the same direction when the target is between 0 and 90 degrees from the ship's axis as would be feasible using an Arduino solution.

Plague has essentially repeated my draft spec for this in his post.

Colin

[Bob K]

Geoff's Iron Duke is a masterpiece IMO.  I loved following his simulated gunfire thread.   

A little lateral thinking on turret positioning.  How about if all guns could theoretically train the 360' specified by required bearing commands, but had microswitches to cut power to the servo when it attempted to enter the no-go zones of obstructing superstructure ?  Thus each turret starts at 0' (bow) and can only turn when the bearing value passes beyond its limiting zone angle.  Is an electro-mechanical solution possible ?

[Colin Bishop]

 Bob, if you are considering a computerised/electro mechanical hybrid then yes, something could probably be cobbled together but you would be substituting a lot of limit switches and fiddly wiring for maybe half a dozen lines of computer code – not a very good trade off.
 
It has become very apparent from the posts on this topic that while people may be prepared to accept that to achieve an all singing, all dancing turret control facility a computerised solution is necessary, it clearly takes many of them well out of their comfort zones and that is entirely understandable.
 
Our Arduino experts argue that Arduino is simple when you get to grips with it and for them it is. But any form of programming does require a new skill set and it is not one which is a natural extension from our normal modelling activities. Moreover, computer programming does require very precise and logical thinking and the ability to resolve problems (bugs) in the code as a mental exercise whereas we are more commonly used to looking at something visually on the workbench. Some people are better than this than others who may really struggle to make sense of it all.
 
This is why I suggested that there is a commercial opportunity for anyone able and willing to develop a generic prototype. OK, it might just be pin money but I think there would be a reasonable demand.
 
It doesn’t make sense to me to have a lot of people attempting to develop their own solutions. A generic system can be applied to any naval vessel from a single turret monitor to the seven turrets of Agincourt – it really is a one solution fits all situation.

So you could just buy the central control board with its GPS/electronic compass shield pre programmed and then one sub module for each turret (and maybe the gun director too!). The turret modules could come pre programmed to bear up to 90 degrees from the ship’s keel line and work straight out of the box but could also be individually programmed from a user friendly program on your computer which allows you to tweak the permitted bearing angles and traversing speeds.

A TX with a rotary control would be needed otherwise the R/C would be standard gear.

So, is anyone willing to take the plunge and have a go at doing this? If it is as easy as our Arduino experts tell us then it is just a question of their time plus the cost of the individual Arduino components needed to make one turret work.
 
Colin
 

[malcolmfrary]

Colin put his thumb spot on the main problem in an earlier post.  You do need to know what you want your system to do before you start designing it.  I always found that writing down what I wanted my program to do in plain English was a good start in writing the actual program in terms that the processor would understand.  If I couldn't make solid logic from the English version, I would have little chance in computer language.
Computers of any type do do exactly what they have been told to do.  This can be embarrassing.

[Bob K]

Hey guys, I accept the principles of what you are saying, but after researching all the links given I can't see where to even begin.  I have done PC programming before, mainly BASIC, Fortran & Assembler, but even 'simple' tasks like noughts and crosses can take a huge amount of time to get working.  I have already listed out my desired parameters of operation, to which I will add one more.  For any realistic operation you have to train turrets up to 135 degrees each side otherwise it will merely look like two separate manual control systems for fore and aft facing guns.  There will be a huge number of embedded conditional loops with slowing, stop and start somewhere in the centre.

ie:
Get GPS value
Get desired bearing value
Calculate relative angle
Can this gun achieve this relative angle, if so initiate start loop.
   Is relative bearing at least 6' for start plus stop loops?
         Run slow training.
               check relative bearing difference is still within current range
                    if bearing difference is within, say 3' of optimum start slow down loop
               check relative bearing difference is still within current range
         Run slow training.
Check firing bearing has been achieved
If yes go to get GPS bearing.

Something like this?  I particularly like the Arduino's with triple pins that you can plug servo leads straight into, otherwise a rats nest of wiring will be necessary.  Arduino compatible GPS units can be bought for under £13.
My ideal would be two identical units, each with half the channels (0-4) forward facing turret capable, the others (5-9) for rear facing turrets.  That way you can plug in servo leads according to turret orientation.

[dreadnought72]

So, is anyone willing to take the plunge and have a go at doing this? If it is as easy as our Arduino experts tell us then it is just a question of their time plus the cost of the individual Arduino components needed to make one turret work.

Colin, this is my plan for January!


I'll be using an Arduino Mega, a solid-state compass (don't need GPS) and a clatter of geared stepper motors. Plus lots of wire. I've already soldered up a controller for testing, which will replicate a TX. It's got a rotary pot (for bearing info) and two switches ('on', 'off', 'park'and 'track'). A dozen LEDs for turret status and feedback. Total cost is about eighty quid.


I'll set up a thread when I start programming. 


Andy

[Bob K]

Wow Andy, that sounds awesome,  I am sure this could be a highly popular solution as almost everyone with a turreted warship could benefit by adding realistic gun orientation.  I would be happy to contribute to development costs, especially as I introduced the problem at the start of this thread.

Reading up on stepper motors vs servo's I can see there are some technical advantages, but to maximise take up I suggest basing it on servos as modellers with existing warships could also buy into this to optimise maximum potential.  Also, any means to minimise wiring would greatly boost usage.  ie: Being able to plug servo connectors straight into the Arduino. 

I aim to pick up the hull in the Spring, fitting the running gear and bulkheads first, so absolutely no hurry.
Anything I can do to assist, just ask. 

[Colin Bishop]

Sounds great Andy, I think a lot of us will be following your progress with interest. Definitely a worthwhile project.

Colin

[John W E]

yes I think we will all be sitting ready - I have me jotter open and pencil sharpened already

John

[dreadnought72]

Reading up on stepper motors vs servo's I can see there are some technical advantages, but to maximise take up I suggest basing it on servos as modellers with existing warships could also buy into this to optimise maximum potential.  Also, any means to minimise wiring would greatly boost usage.  ie: Being able to plug servo connectors straight into the Arduino. 

The thing is, the geared steppers are good to about a twelfth of a degree of angle, don't have servo-like limits, and can be repeatably positioned to the same point for as long as they're powered up. If that's not enough, they cost <£5 off the 'Bay.


 


Andy

[John W E]

- any good suggestions of what starter kit to look at for  Arduino boards?  I have been browsing the web - and there does seem to be an awful lot about.

john

[derekwarner]

dreadnought72 ......Andy.....when you say...'geared steppers are good to about a twelfth of a degree of angle'....that is certainly greatly exceeding the repeatable train position for this model application.

I have little if any real knowledge of stepper motor characteristics,.....so just as one simple parameter that seems not mentioned to date , can a geared stepper provide a smooth [jitter free] output shaft movement of say 90 degrees in say 4 seconds?..........

Derek

[dreadnought72]

Yes. I've conducted a few tests with these geared steppers and that speed is achievable. Go too fast, though, and - while the Arduino can pump out the steps quickly enough - the stepper driver chokes on the digits and the motor stalls out.


Andy

[Colin Bishop]

John,

Before splurging the cash I would suggest reading up on the subject a bit. I bought the Kindle edition of John Boxall's book:

https://www.amazon.co.uk/Arduino-Workshop-Hands-Introduction-Projects-ebook/dp/B00CLHK0OQ/ref=sr_1_1?s=books&ie=UTF8&qid=1482878772&sr=1-1&keywords=john+boxall

It is very readable and very informative. It will certainly give you some ideas on what you might want to buy to get started.

I rather doubt if I will take it further at the moment as I don't normally build warships and the only working feature on most of my models tends to be lighting so the effort in getting to grips with the Arduino syntax and interfacing isn't directly relevant to my modelling. However I do find the Agincourt subject very interesting but more in terms of defining the specification, outlining the programming requirements and seeing what interfacing options are available to make it all work.

When you think about it, the possibility of being able to exercise proper 'director control' of the main armament on a dreadnought battleship model using just one R/C channel is pretty amazing really!

Colin

[Bob K]

A lot of mention of firing at 90 degrees.  Can we please please spec this for all guns training to a maximum of least 135 degrees as this would have been the more normal scenario for dreadnoughts firing at a target either before or abaft its beam as the battle lines converge.  I am not sure what the maximum training angle was but it was certainly a lot more than merely right angles to the centreline.  The more modern Iowa Class could turn and fire through a total of 300 degrees of rotation, so 270 would not be unreasonable for WW1 ships.


   Markgraf

[derekwarner]

Sorry Bob...

The WIKI people suggest that the Iowa Class 16" mounts could train to approximately 300 degrees [total], however even the WIKI people [untrained journalists] fail to understand or mention any comment on non-pointing zones as these were technically classified

With my work, [Department of Defence  - Navy - Above Water Weapons Engineering] a few colleagues & I had the privilege of a 45 minute personally guided tour of BB63's 16" gun and gun control systems.....all I can suggest is that non pointing [non firing zones] were introduced to prevent shock wave damage to the vessels superstructure

I understand this is a different vessel, however this portrays some of the inaccuracies  in WIKI reported specification etc..so research & cross referencing research detail is certainly required

Derek

[Colin Bishop]

Turret training arcs are often given in the major reference books such as Burt and Parkes. Collateral damage to the firing ship didn't just depend on the angle of the turret but also upon the elevation of the guns. When the battleship Rodney's third turret initially fired abaft the beam at full elevation it knocked bits off the bridge structure and some of the occupants senseless!

As has been said earlier in the topic, we don't want to get ahead of ourselves in over specifying the basic system but if it can be implemented along the lines of my draft spec then for those builders who wish to incorporate gunfire effects, this could be done quite easily by programming each turret module with its pointing arcs (permissible firing zones) so that the guns would only fire when the turret was pointing within those arcs. The data could simply be held in an array and referenced against the axis of the model. A very simple additional feature I think.

Colin

[Bob K]

Fair enough Colin and Derek.  All I want to see is turrets collectively training beyond the 90 degree broadside configuration.  In a model on lake scenario having turrets training thus will graphically illustrate the single central fire control system is not just multiple controls being transmitted individually.

IMO the ability to train both forward and aft facing turrets together will really make the overall visual effect truly awesome. Not sure how much we want to over complicate the programming with no fire zones, maybe just limiting rotation beyond a nominal 90 degree broadside would be more than enough.  I am thinking of actual engagement orientations at Jutland with battle lines converging.

This is getting quite exciting.  I hope the numerous calculations such as bearing computation, can this gun move to that bearing?, slowing rotation, and maybe even start and stop loops to work out will not make it over complicated.  Once the program for one turret is developed it can be copied to other channels with small timing variations edited in to reduce any ballet synchronisation effect.

Are we asking too much here ?

[Colin Bishop]

I think you can take training the turrets and the other things you mention as a given Bob. Each turret will 'know' what its limitations are as these will be individually programmed and, by the sound of it, small variations in training speed can also be accommodated to add extra realism.

As you say, quite exciting! Somebody has to be first.

Colin

[unbuiltnautilus]

I think eliminating the over coordinated 'ballet' is a good point. On one of our past models ( HMS Invincible Battlecruiser ) the system used for extreme firing angles, would result in the turrets arriving at their designated firing angle at differing times. This, I think, looked much more effective than the geared together look that can be seen on some models. The arduino route seems to allow this effect to be dialled in to the design. I am watching with interest. I shall also dig out a couple of our  turrets and take some photos, you still need a mechanism after all!

[NFMike]

Further to the above I'd suggest a refinement of the rotation control in addition to the speed setting, in the form of a slight delay in starting. It was mentioned earlier that the turrets are controlled individually in response to commands, so I'd suggest that part of the program would (effectively) delay the response of each turret module by a random amount between say 0 and 1 or more seconds. This is recalculated each time so it isn't always the same turrets being first, second, etc. The upper value can be an adjustable variable; setting it to 0 turns the effect off.

[kinmel]

I have a Raspberry Pi 3 and an Arduino Uno and so I have added the servo control board to an order..  http://www.hobbytronics.co.uk/raspberry-pi/servo-pro-12ch

The page describes the options in detail.

Those members with either of the Microprocessors can contribute to the project by creating scripts to try various options and we can run your version for you and post the result on Youtube.

The Datasheet can be downloaded from.... http://www.hobbytronics.co.uk/datasheets/slave_servo_pro_datasheet_v01.pdf

The instructions and examples of Code can be downloaded from... http://www.hobbytronics.co.uk/tutorials-code/tutorials-microchip/hi-tech-c-i2c-master

Who is up for a joint effort ?

[John W E]

Colin, one of the reasons I asked for a ‘beginners guide’ to this product was that I hoped those who endorse this particular product – i.e. Arduino boards – would come forward with their suggestions in layman’s terms for those who don’t understand and are frightened of electronics so that the majority would be able to follow. 

On this topic we may have lost approx. ⅔ of the reading members of this Forum due to them not understanding.  I believe we have a duty to cater for those who don’t grasp electronics initially when they first read the article.

Some people may not realise that this particular item can be used for other devices on the model, such as raising/lowering anchors.   Operating towing winches/using individual motors.  Also switching lights on/off in different sequences.

My opinion is, if the reader can be made aware of all the options open to using this board and how easy it can be – we must remember when we put pen to paper, those who are in the know, must take into account those who don’t.   People may be easily put off as I have said in an earlier post.
However, if we can convince them that one of these products can be as easy to wire up as an electric motor with careful guidance from a well written document by one of our members then we are on the way to helping a lot of people into the hobby.

John