Maker Pro
Maker Pro

Actuator control for my nuclear missile complex

I always figure that will get more views. :)

So, this is actually for my former Titan II nuclear missile complex.

I'm making 4 cabinets that will each open and close with their own actuator.

Each of these will be independently operated so the quantity isn't relevant.

I'll likely use this actuator:
https://www.progressiveautomations....-speed-linear-actuator?variant=18277288706115

Here is what I'd like to achieve:
- Single momentary pushbutton switch
- Single push and release of the switch will open
- Single push and release of the switch will close
- Single push and release during movement stops movement
- Single push and release always changes direction (polarity)
- Actuator has built in limit switches but doesn't have feedback so I'm guessing I'll need to have limit switches.
- When the movement stops due to the limit switch, the next switch press and release should move the opposite direction (If it fully opens, the next push closes and vice versa)

I don't mind paying for help and / or giving credit on the YouTube channel. Speaking of the channel... not trying to promote but if you want some context on the project you can check it out here: https://www.youtube.com/channel/UCd50A5qLv8FemVufSvDgkCQ

Hoping this all makes sense. Thank you very much!

GT
 
You can't have all those single pushes of the same switch to do all those different functions.
You can have them sequenced. 1.open, 2.close, 3.stop, 4.reverse etc.
But all this will require programming a microchip.

Martin
 
You can't have all those single pushes of the same switch to do all those different functions.
You can have them sequenced. 1.open, 2.close, 3.stop, 4.reverse etc.
But all this will require programming a microchip.

Martin

Martin,

Thank you for your reply. While I don't know electronics (obviously) I find it hard to believe this can't be done programmatically. My apologies for questioning you especially considering your long standing here. If it can't be done I have some scrambling to do...

GT
 

(*steve*)

¡sǝpodᴉʇuɐ ǝɥʇ ɹɐǝɥd
Moderator
This seems like an eminently doable project.

As @Bluejets suggests above, you're going to need to carefully consider what a button press means. The same action under the same conditions can't mean different things.

Another issue I note is that the linear actuator you have specified notes that it must be operated under a minimum of 50% load to ensure correct operation. How will this be ensured?

Also you say that the actuator has limit switches, but then go on to say that limit switches will be required. I'm not certain what you mean here, there are alternatives to limit switches, but I'd always consider them a good idea. You need a way to detect a stall condition, where something prevents operation to the designed end point. This could be because someone has their body in the way. Depending on how powerful the actuator is, you need to protect what is preventing movement, the thing being moved, and/or the actuator.

For an electrically operated actuator you can measure current, limit the on time, or continuously monitor position (or actively detect obstructions). Depending on the circumstances, you may do more than one of these.

This sort of application cries out for a state machine solution.
 

Harald Kapp

Moderator
Moderator
I find it hard to believe this can't be done programmatically.
Of course it can.
You need sensors that tell you the position of the "door": open, closed, moving. You may be able to use the limit switches to detect open and closed, everything else is moving.
The direction you can either detect with another sensor or simply follow it programmatically by using a state machine. This may sound more complex than it is: Simply keep track of the commands you issued to open or close the "door" in a variable within your code to know whether the "door" is currently closing or opening.
You have now, as far as I can oversee, to trigger the required action once the button is pressed (btw: do not forget to debounce the pushbutton).
Speed is rather unimportant, I guess. An Arduino or a raspberry Pi can do the job. You'll need additional driver boards (shileds, hats,...) to drive the actuator from one of these minicomputers.
 
Seems with some programmable logic you can do all this. For each, except my understanding of the two above, teh current state determines what happens when you activate the push button.
.
The main one I didn't understand is the distinction between
- Single push and release during movement stops movement
- Single push and release always changes direction (polarity)
Seems if it is moving, an activation reverses and stops. What determines the difference?
.
So four cabinets, four pushbuttons, right?
.
Your actuator is 12 volts.
So you are happy with working on 12 volt equipment?

Datasheet says the fastest actuator. That's what's needed?
 

bertus

Moderator
Hello,

Looking at the datasheet, the needed current to operate is between 2 A (no load) and 9 A (full load).
Are the actuators driven one at the time or are there situations where more actuators are driven at the same time?

PA-15_specs.png

Bertus
 
Thank you for all of the responses! I didn't see a quote function so I'll answer all of the questions here:

*Steve* By limit switches, I mean simple microswitches that are set to activate at the full open and full closed position. As far as strength of the actuator, the most powerful one of what I'd like to use is 33lbs of force. While that wouldn't be comfortable I can't see it as dangerous.

Harald - You said "Speed is unimportant" - I'm not sure what this is referring to but these are cabinets / shelves so I'd like them to be somewhat fast. That's why I've chosen these particular actuators.

Nanren - Yes, four cabinets and four separate switches. Unless it is smart to do so, I plan on these being separate except for maybe the power source.

The cabinets are on very nice rails - think of it as shelves that come out of the side of a wall. So, the actuators will be under very little load unless they bind.

- Single push and release during movement stops movement
- Single push and release always changes direction (polarity)
Seems if it is moving, an activation reverses and stops. What determines the difference

With the exception of fully closed or fully open, every time a press occurs the mode changes in a simple sequence:

Plus sign = movement one direction | minus sign = movement other direction
Move+|Stop|Move-|Stop|Move+|Stop|Move-|Stop|Move+|Stop|Move-|Stop|Move+|Stop|Move-|Stop

Yes, I'm good with and assumed this system would be 12v.

Bertus - each system will only have one actuator. So even though there are 4 doors, think of each of them as an independent system.

Hope those answers help. Thanks everyone!

GT
 

(*steve*)

¡sǝpodᴉʇuɐ ǝɥʇ ɹɐǝɥd
Moderator
I spent some time watching most of your "first season". Whilst I would normally both change the topic and remove your youtube reference, I agree that they are actually relevant to this thread.
 
I spent some time watching most of your "first season". Whilst I would normally both change the topic and remove your youtube reference, I agree that they are actually relevant to this thread.

Thank you Steve. My apologies if I broke rules. I very much appreciate the help here!

GT
 

(*steve*)

¡sǝpodᴉʇuɐ ǝɥʇ ɹɐǝɥd
Moderator
The rules are pretty tight on advertising and promotion. However in this case you're clearly here for assistance rather than self-promotion. The link to your channel gives both context to your request and also answers any doubting Thomas who may not believe your choice of residence.
 

Harald Kapp

Moderator
Moderator
I didn't see a quote function
In the menu bar left of the [sub sup] Icon there is an "insert" icon. Drop it down and select "Quote"
Or simply mark the text you want to quote with the muse. A "reply" button will appear that generates a quote in your edit box.
Harald - You said "Speed is unimportant" - I'm not sure what this is referring to but these are cabinets / shelves so I'd like them to be somewhat fast.
I meant computing speed of the controller. Sorry for not being precise enough.
 
ou need a way to detect a stall condition, where something prevents operation to the designed end point. This could be because someone has their body in the way. Depending on how powerful the actuator is, you need to protect what is preventing movement, the thing being moved, and/or the actuator.
Safety issues noted:
So what more is required?
Do you have any experience in PLCs?
Interesting that you want the one-button control. Some people hate the things wanting in preference dedicated buttons, eg a momentary, open, or close, or stop.
 
Safety issues noted:
So what more is required?
Do you have any experience in PLCs?
Interesting that you want the one-button control. Some people hate the things wanting in preference dedicated buttons, eg a momentary, open, or close, or stop.

No experience with PLCs. I'm a tech guy (Wi-Fi is my specialty) but I'm not a programmer. The one button was my design. It's a very clean modern system so I didn't want to clog it up with buttons and labels.

As I mentioned before, I don't mind paying for / giving credit for the work if someone can spend time helping me out on it. Thank you again!

GT
 
Stall detection still to be sorted out.


WiFi?
So are you familiar with state machines?

I suppose I should have stall protection to protect the actuators. I've lost a few before b/c they were stuck and I didn't know it (thought they were up against the built in limiter)

Wi-Fi (802.11) has a state machine but I'm not sure that will help me here. I'll do some reading on them so I can do my best to be conversant.

Thanks!

GT
 
Is it something like this?
Events along the top rows, current state down the left column, state transitions in the table.

upload_2020-5-12_14-17-3.png

upload_2020-5-12_14-17-37.png

Just one choice of states.
Could also have a representation two interconnected state machines, one for the direction and one simply moving or not.
Could also fill in idle-forward + forward limit --> idle/backward & its pair the other way.
 
Is it something like this?
Events along the top rows, current state down the left column, state transitions in the table.

View attachment 48168

View attachment 48169

Just one choice of states.
Could also have a representation two interconnected state machines, one for the direction and one simply moving or not.
Could also fill in idle-forward + forward limit --> idle/backward & its pair the other way.

Nanren,

Thank you for that! It makes total sense to me and the graphical state machine looks correct. In the table version, the two question marks... that would have to be some kind of error... correct? The only situation I can see it happening is if it's moving forward and the backward limit switch is somehow triggered (of course the entire opposite of this sentence also applies). I'm not sure how the limit switch would get triggered. But it does enforce that I'd need to make sure the limit switches couldn't easily be accessed.

What is the next step?

Thanks!

GT
 
Sort of depends on your technology of choice.
As a separate thought, though. More than once, I've seen posted here a solution, and then someone else suggests something even easier.
It might be worth waiting to see if this gets any comment, if you are not in a hurry.
.
Next?
Custom electronics is a possibility, but PLCs are sort of made for this sort of thing.
PLCs are very common as quite simple ready to install boxes, enclosures with terminals for connections. They take inputs and have outputs and can be programmed in a few different ways. For simple things ladder-logic is quite reasonable. I guess you have four buttons, stall detectors, and eight limit inputs.
You may be able to combine to use one PLC from more than one cabinet. Not sure of the best way to implement stall detection.
9A full load for each actuator. Need to check PLC output limits for whether you need some sort of extra drive/relay &c.
Perhaps there's a PLC expert here?
 
Top