Topic: Rudder Control (Read 17728 times)

Bunkerbarge · « **on:** June 11, 2009, 12:51:05 pm »

I was wondering whether any of you clever electronic chappies could provide any suggestions for a set up I want to organise.

I have a model with a total of six rudders, two normal behind the props and four flanking rudders in front of them. I have one available proportional channel and one available switched channel so I want to use the switched channel to change over from one set to the other and use the proportional channel to control which ever set of rudders is selected. I already have one servo to control the normal rudders and one servo to control the flanking rudders.

One obvious idea is to use the switched channel to change over a three pole, two way switch which changes the signal from one servo to the other but I was wondering if there was anything a little bit tidier.

Any thoughts very much appreciated.

Proteus · « **Reply #1 on:** June 11, 2009, 01:33:01 pm »

go for a tx that is used on aircraft, a lot of the new futaba high end ones on 2.4 or a ff9 you can get a 2.4 module for it then you can mix channels as you wish turn mix on and off. its up to you then what you want, and you have a nice tranny.

Proteus

Bunkerbarge · « **Reply #2 on:** June 11, 2009, 02:53:29 pm »

I've already got and using a Spektrum DX6i can this be configured by mixing to do what I want? Time to get the book out!!

OMK · « **Reply #3 on:** June 11, 2009, 04:03:28 pm »

I'm not famiiliar with the DX6i, but it's possible that it might have an appropriate built-in/user-defined mix already.
Failing that, Bluebird has asked a same/similiar question. He's looking to do something along the same lines. Take a look at the sketch. In essence, channel 5 will operate either his gun turrets or searchlights. Exchange his turrets/searchlight for your normal rudder/flanking rudder and you should see the similarities.
Bluebird's version uses four channels total, but no reason why you couldn't have a cut-down 2-channel job.
The PIC software is being written as we speak............

portside II · « **Reply #4 on:** June 11, 2009, 04:21:21 pm »

when you figure out your mixer controls will you let me know how you did it , i have had a computerized radio (hitec eclipse 7)since they came out and still havent worked all of it out

.
daz

OMK · « **Reply #5 on:** June 11, 2009, 04:38:37 pm »

Me, I'm a Futaba nut. Nope - not berating any other make/model, but merely highlighting the fact that Fut's FF9 has more mixes than you can shake a stick at. The FF9 is dead easy to program a mix for which you're looking to do, so maybe it possible that the DX6i will have the same.
On the other hand, if no such mix available, I could soon cobble together a simple non-PIC system for you, albeit a little less elegant than the PIC version. It means you would have to manually centre the set of rudders to neutral (for obvious reasons) before swapping over to the other set. So if you can live with that...............

Bunkerbarge · « **Reply #6 on:** June 11, 2009, 06:50:16 pm »

All I seem to be able to come up with as far as the mixers are concerned is to have a percentage of one channel operating with a percentage of another channel. I don't seem to be able to get it to switch from one to another servo. When you think about it it isn't surprising because the switching function has to be after the Rx output to change from one servo to the other.

I've come up with some nice little rotary switches but that would be another servo so PMK's thoughts look interesting. I obviously want to operate either one servo or another depending on a switched channel. Which channel in the sketch is controling the channel five action and hence determining whether it operates either the guns or the searchlight?

Centralising the rudders is obviously not a problem so you may well have the very thing there PMK. Thanks for your thoughts.

flashtwo · « **Reply #7 on:** June 11, 2009, 06:54:32 pm »

Hi PMK / Bunkerbarge,

Looking at the rudder requirement has made me revisit my general purpose PIC board that I use to control my steam plant servos.

Currently the board receives a single channel and operates a forward-neutral-reverse servo when the input moves away from a central neutral deadband with the three output positions all adjustable by the pots.

Looking at the code I now realise that to handle two PWM inputs would be quite easy. I would, however, need to know the relative pulse positions of the two inputs, since the two PWM output pulses would need to come in the gap between the input pulses, i.e. if the inputs were concurrent, then the outputs would have to come after the last input.; its all to do with timers and interrupts as you may have guessed.

PMK -I take it that you're a bit of PIC fan.

All the best
Ian

here is a photo of my servo PIC board.

OMK · « **Reply #8 on:** June 11, 2009, 07:16:08 pm »

"Which channel in the sketch is controling the channel five action and hence determining whether it operates either the guns or the searchlight?"

If I understand your question, the answer is channel five. It equates to the non-proportional (toggle) channel on your radio. Flick it on or off to select from normal rudder to flanking rudder. Your spare (proportional) channel will take care of the direction of whichever set of rudders is selected... if that makes sense.

Holler if you need such a unit. I could probably shorhorn the whole caboodle on a board about the same size as a standard Rx.

OMK · « **Reply #9 on:** June 11, 2009, 07:29:01 pm »

"PMK -I take it that you're a bit of PIC fan."

Sort of. Assembler language can be a pain in the gregory at times, so these days I tend to cheat.... I use a BASIC complier.
I think Wombat is the man to know for the real programming, though.

Regarding two inputs, I just read the value of both inputs, store those values in two variables, then manipulate the variable to whatever it needs to be before it gets to the output.

Bunkerbarge · « **Reply #10 on:** June 11, 2009, 08:31:54 pm »

PMK, Yes I'm with it now. Channel five would switch from the flanking rudder servo to the normal rudder servo, with both servo's being controlled in my own case by channel four.

If you can put together such a device I would be very interested. I'll send a PM.

John W E · « **Reply #11 on:** June 11, 2009, 08:35:07 pm »

hi all

Reading with great interest - I think what PMK has forgot to say is - the original idea of mine is a patent of mine

aye, the patent price is - for copies

= 10 bottles of cider to PMK

= 10 packets of jelly babies sweeties for me

Bunkerbarge

quick idea for you, you just require a micro switch and a Y lead to acomplish what you are after. All you need to do is switch the signal from one servo to the other servo via the micro switch - if you want a scribble - give a shout. Channel 5 servo would operate the microswitch.

aye
john e
bluebird

Bunkerbarge · « **Reply #12 on:** June 11, 2009, 08:58:38 pm »

Hi John, yes, completely agree. I was just investigating other possibilities without the use of another servo but I might well end up back at a micro-switch or a small rotary switch operated via a micro servo.

Thanks for the help.

OMK · « **Reply #13 on:** June 11, 2009, 11:58:19 pm »

It's true, it's true. Bluebird is the original brains behind the thinking.

I'm not a greedy man. Nine bottles will do.

OMK · « **Reply #14 on:** June 12, 2009, 12:22:48 am »

The story so far.....

I've just emailed B'Barge, no reply as yet, so before this gizmo goes into orbit, just one itsy-bitsy question first...........

Bunkerbarge, before replying to the PM, consider this first: Assume that your two main rudders are operative and the four flanking rudders are where you left them before flicking your channel 4 switch (i.e: all at their neutral position). At this point you have to remember that no signal pulse will be reaching the flanking servos because it's been diverted to the main rudder servos. In short, no pulse is reaching the flanker's, which in turn means that no pulse is there to hold them at a steady neutral. So would there be a danger of any undercurrents throwing the flanking rudders from their neutral positions?
If that is the case, then perhaps it might be better in opting to go with the PIC-based version, whereby the inoperative servos stay locked at neutral while you're using the other ones.

The reason for asking out loud and not via PM is because millions of members are just as curious about them undercurrents.
Not only that, I've got a sneaking suspicion that DickyD is earwigging.

Bunkerbarge · « **Reply #15 on:** June 12, 2009, 03:22:14 am »

Hi PMK, I understand your point but my thoughts as regards the more mechanical solution would be to use a "Y" lead from the Rx to the two servos and switch the signal lead only via a micro-switch. So the question there is would the power cable remaining intact to the unused servo keep it in neutral? I suspect the resistance of the gearbox would maintain the neutral position anyway but experimentation would prove either way.

By the way all four flanking rudders are mechanically linked and operated from a single servo and the two normal rudders are also mechanically linked.

I have found a sub micro-servo (3.7g) that could operate a micro-switch making the change over mechanism very small and neat. From what you were saying in your PM the PIC route could actually be a bit bigger.

Interesting comparissons that may also help others in their thinking!!

OMK · « **Reply #16 on:** June 12, 2009, 05:33:57 am »

"So the question there is would the power cable remaining intact to the unused servo keep it in neutral?"

Good question, but the simple answer is no. I see what you're saying but it doesn't quite work that way. Even if the servo is supplied with power from the battery it still needs the third wire (the pulse signal) in order to know where neutral is.
When you said that your model has four flanking rudders I automatically assumed the model was already tried and tested. I was waaay ahead of myself - didn't realise you are still at the suck-it-and-see stage. In which case, your guess is as good as mine as to whether the servo gears will hold up to it. What would be the worse scenario if the flanking rudders were to move by any undercurrents? Do you think the model would still be maneuverable/manageable by the main rudders if the flanker's were, say, as much as 30 or so degrees from neutral?
Me, I'm brave enough to take a stab that the servo bearings alone would provide enough torque to counter the thing from moving, and you probably know yourself already just how much force is needed to turn a servo horn with fingers alone. So maybe I'm putting too much emphasis on undercurrents.

"From what you were saying in your PM the PIC route could actually be a bit bigger."

Nope, not necessarily. Now that I know you only need two outputs means that a lot of real estate is saved: Just 2 x servo leads, 2 x on-board 3-pin servo connectors and 1 x PIC micro would yield an overall smaller circuit board... nowhere near the size of a fag packet.
Maybe a moot point, however, because I still reckon something equally as elegant could be done if you take the mechanical route. Reading between the lines it's pretty obvious that the mechanical route is where you're heading already. Doubtless, you'll get more fun from watching a servo operate a microswitch than you would from staring at a computer on 18 legs.
At the risk of sounding like I'm trying to teach how to suck eggs, watch out for the high-voltage spikes when the microswitch changes over. You might need some capacitive suppression there in order to counter the servos from twittering?

FullLeatherJacket · « **Reply #17 on:** June 12, 2009, 08:10:07 am »

In my experience of making and using show demonstration boards, it's not a good idea to use a mechanical switch to change the signal pulse from one device to another unless you have absolutely no other option. However, this applies mainly to ESCs and R/C switches, so I couldn't say for definite how a servo array might be affected.

PMK's concern about undercurrents moving the servo centres is perhaps a bit extreme (poor twittering old thing) - I think the mechanical advantage in the servo gearboxes would prevent that unless the rudders were the size of barn doors and/or you habitually sail in Tsunamis.

Suit yourself.

FLJ

wombat · « **Reply #18 on:** June 12, 2009, 09:13:55 am »

PMK old dude - why don't you just get the latest version of MPLAB - comes with freeby C compilers (for PIC16 and lower at least)

As for the problem in hand.....

Much as I would like to recommend something fancy, I suggest a single channel switcher with a relay to switch the signal lead - do not go for one that uses transistors or FETs for the switching. However, and this is a big however, you need to do a bit of investigation. You need to get a low current switcher (1A rating or less) rather than a high current switcher - this is because you neeed gold plated contacts as the "wetting" current will be very low. Ideally you neeed a relay that meets a signal switching rating such as BT47W (used for switching telephone lines) rather than power switching - bigger relays relay on the inevitable arcing to give a good contact so will be unreliable for signal switching. You could always ask FLJ about his switcher relays.

You can then switch the pulse signal to the servo - when the pulse signal goes, the servo will hold its position. Hitec and Futaba servos stop driving the motors in this condition so they consume less power. You may find that when you switch you get some "runt" pulses or noise into the signal input on the servo, but this should not have too much effect as the time this will occur over (2-5mS) is so small with respect to the response time of the servo that there will be no real response. Only thing is you will need to be careful when you switch off the flanking rudders so they stop in the neutral position - bit of care might be required

Wom

OMK · « **Reply #19 on:** June 12, 2009, 09:49:01 am »

Okay, okay - MPASM it is. And maybe I were a tad over-cautious regarding undercurrents.... which means a simple circuit would do the job after all. An earlier proto' version consisted of a small on-board relay, worked fine, no problem at all. The only downside was the hefty current draw. So instead of the relay, I'm toying with the idea of using a quad bi-lateral chip (4016). But why do you suggest not to use FETs or transistors?

malcolmfrary · « **Reply #20 on:** June 12, 2009, 12:06:17 pm »

The transistors usually want to be switched with respect to the power lines, a switch doesn't care. The 4016/4066 will do a good job, but why not just go for NOR gates? Even cheaper than bilateral switches. For the ultra cautious, the "unused" servo could be fed with a "neutral" position signal.

wombat · « **Reply #21 on:** June 12, 2009, 01:53:14 pm »

Quote from: PMK on June 12, 2009, 09:49:01 am

Okay, okay - MPASM it is. And maybe I were a tad over-cautious regarding undercurrents.... which means a simple circuit would do the job after all. An earlier proto' version consisted of a small on-board relay, worked fine, no problem at all. The only downside was the hefty current draw. So instead of the relay, I'm toying with the idea of using a quad bi-lateral chip (4016). But why do you suggest not to use FETs or transistors?

Well I was thinknig more of PICC Lite which is shipped with the latest versions of MPLAB.

Reason I suggested relays rather than FETs is that it is a simpler solution - if you use a transistor, you either need a pass transistor setup which means something akin to an analogue switch, or a pull-down system which requires some series resistance which could have all sorts of problems in a noisy environment.

Wom

Bunkerbarge · « **Reply #22 on:** June 12, 2009, 03:05:29 pm »

Well gents first of all I have to say how much I appreciate the input from you all, it's threads like this that make the Mayhem forum the valuable resourse we can all benefit from.

Of course most of this after about the first half dozen posts went straight over my head and i'm just about clinging on with my finger nails as regards understanding what is being said. To put things into context I have no idea what a PICC lite is, what MPLAB is or what on earth a FET is. I think I'm OK with a transistor though.

Just to give everyone the complete picture with this a bit of background may be usefull. I bought the model from a member here in Hull. It is fitted with two very large motors and the aforementioned six rudders. I have used it on a number of occassions with the two sticks of the DX6i used forwards and backwards for independent speed control and the two sticks used side to side, one for the main rudders and one for the flanking rudders. I have the two switched channels as spares. I have now purchased a Model Solutions sound generator so i was going to use the two switched channels to operate two seperate sounds. Then I realised that if I used one of the switched channels to change the rudder servo's over I could still have proportional control over the rudders, although only one at a time, but it released one of the proportional channels so I can then have four controlled sounds from the sound module.

Botom line, can any of these suggested devices compete in size with a 3.7 g servo connected to a two way micro-switch? I don't think that the flow of water over the rudders will be enough to overcome the gearbox ratio so I'm not too worried about the ones not being in operation moving.

roycv · « **Reply #23 on:** June 12, 2009, 04:00:42 pm »

Hi BB interesting thread! The model sounds like an American Tow boat, have you looked at Towboat Joes website, every combination of rudder control there!

Just a thought but you only need the rudders working with the prop wash on them, so why not use a servo in parallel with the ESC to decide which set of rudders to use, i.e. the forward or the astern set, this would then switch in automatically and you still have your spare channel to play with.

I can see that the servo working the micro switch would have to operate it quite early in its travel and maybe it should be biased to forward.
May be one of those kits for switching lights etc could be used (again in parallel with the ESC) and then set to operate almost immediately the ESC direction is changed.

kind regards Roy

Bunkerbarge · « **Reply #24 on:** June 12, 2009, 07:43:33 pm »

Hi Roy, I'll dig out a couple of pictures later and yes I'm very familiar with Tugboat Joe's site. Very interesting it is as well.

I'm afraid your suggestion is going to get even more complex as there are effectively three rudders to each propeller so the rudders would have to be split up to operate in conjunction with each motor and it's respective ESC.

Strangely enough I actually do most of the manoeuvring steering with the two motors and only use the aft rudders when the model is as speed (OK faster!) The flanking rudders are only really of any value when towing to direct the prop thrust to assist in barge handling and actually have very little if no affect in steering when under way, whether ahead or astern.

News:

Author Topic: Rudder Control (Read 17728 times)

Bunkerbarge

Proteus

Bunkerbarge

OMK

OMK

Bunkerbarge

OMK

OMK

Bunkerbarge

Bunkerbarge

OMK

OMK

Bunkerbarge

OMK

FullLeatherJacket

OMK

Bunkerbarge

Bunkerbarge