Thread: Machine Digital Control

I have seen that kind of thing on some top end thicknessers. I don't know of any plug n play type option. One thing to keep in mind is that it is not as simple as just telling the table to move the table to "100.0mm" for example. The electronics will need to move the table to ~105.0mm first and then back up to 100.0mm so that backlash in the thread is not a contributing factor in the accuracy. And ideally, if the table is already at 105.0mm and you want to move the table up to 100.0mm, the electronics need to know to just do it without the double movement.

The SCM thicknesser at trade school had this type of thing. Bloody thing needed to be re-calibrated each morning because the building's power was shut off each night.

Kuffy, I have seen it in top end machines as well but the A3 cost me enough cost me enough as it is. I have a sneaking suspicion it might be able to be done using one of the new Raspberry thingos or something of that nature but beyond that I know nothing. If I can't implement it I have a motor out of an electric seat and all the fittings for the motor to drive the existing mechanism to save me winding it by hand. I have sort of approached this problem before but I don't think I explained myself well enough and things change in electronics very quickly as we know.

I think that it should be a pretty basic electronics operation. And by basic, I mean like you'd probably need the equivalent of a couple of semesters of integrated circuitry courses and a similar amount of programming experience.

I think that you would probably want to try to think in revolutions. The ultimate output for the device will need to be some kind of numeric value that is transferred through something mechanical that makes a certain number, or fraction of a number, of rotations of the main height dial. This may involve some pulleys, and stepping things up or down. You'll probably have to have one wheel that just turns x number of times, and then that would have to translate to 1/x rotations of the actual dial on the thicknesser so that you can adjust to whatever the measuable "resolution" of the height adjustment actually is.

Again, I think it's doable with the right equipment (cheap) and experience (expensive/time consuming).

Good luck. Post progress.

Cheers,
Luke

FWIW in my post I'm talking about scratch building electronics and programming them to do what you want to do. I have no idea where to even begin looking for something like that already built and ready to go.

Hi Chris, the only simple way that I can think of, is by using your motor with a fwd, rev, jog switch and a DRO to give the read out. Fwd or Rev the motor while watching the DRO, then jog to get the measurement required, or wind by hand. Should get out of it for around the $100 mark at a guess.
Kryn

Yes, I am starting to think that. I can put the up/down controls where I can read the height gauge and be done with it but the original plan would be nice.

Originally Posted by Chris Parks

I have a combination planer/thicknesser, the table has to be manually raised and lowered quite a long way and I intend to power that operation, I have the parts to do so but I was thinking about putting more control in it. Is it possible to put a system in where I dial in a target height and press go and the table reaches that figure and then stops? Can I do this without spending the rest of my life's savings? I am a total dummy where this stuff is concerned BTW.

the only auto units I've used required the operator to keep their "finger on the button" whilst adjusting the table height. However there was no back lash or need to overshoot the mark and return.
OTH Altendorf saws have a programmable move function, just enter the off-set and press go -- buy you're talking really big $.

perhaps a cordless drill and nut driver?

The sorts of movement being discussed is standard protocol in 3D on 3D printers and CNCs.
If a $500 3D printer can do this in 3D I can't see why it should be that much more for a 1D movement.
The only real difference would be the need for a larger stepper motor and larger motor power supply
The software, sensors for measuring range, etc could all come from a 3D printer.
The device needs to interact with an operator so some sort of controller and I/O display panel would be needed and that could be an Arduino board and a small screen and numerical key pad.

And Bob has outlined my original thinking. A $50 no name tablet might suffice for the screen but it is the control side that I need help on. A large stepper motor is not big bikkies nor the Arduino itself. These days it surely must be possible to do this stuff at a reasonable given the knowledge which I haven't got. Could the keyboard be graphical on the tablet screen? Is my question more applicable to being asked in a CNC forum, would that be a better approach?

Originally Posted by Chris Parks

And Bob has outlined my original thinking. A $50 no name tablet might suffice for the screen

It could well be provided some control software that ran on that tablet was readily available or easily adaptable.
It might be easier to use something like an Adafruit touchscreen given its already adaptable to Arduino stuff.

A large stepper motor is not big bikkies nor the Arduino itself. These days it surely must be possible to do this stuff at a reasonable given the knowledge which I haven't

Yes its not expensive.

Could the keyboard be graphical on the tablet screen?

See above

Is my question more applicable to being asked in a CNC forum, would that be a better approach?

Yep although they tend to deal with far more complex stuff.

Here you go after a 3 second search
https://www.youtube.com/watch?v=zl0Mb5MISUw

found this thread after seeing the post in the CNC section.

You can do it, but there are a couple of issues.

With most cnc machines and 3d printers a lot of effort is put into reducing or eliminating backlash. My thicknesser has an awful lot of backlash, which would make things a little more complicated. The easiest way to think of backlash is when you change direction how far do you have to move the height adjustment handle before the move actually starts. On mine its about 1/8th of a turn of the handle.

The elimination of backlash has lots of benefits, but one of the biggest is that you can use an open-loop control system. That is where you tell the motor "turn this much" and you'll get an accurate movement. The other way of doing things is a closed-loop control system, where you tell the motor to move, and watch how far the machine moves and stop the motor when you get there.

if your just after big moves open loop will probably be okay - but just don't expect to dial in 62.5mm and get 62.5mm without some changes the drive mechanism within the thicknesser.

Anyway, there are lots of ways to implement this - it really just depends on what you want to do, and what your comfortable with. I'd probably start by looking at the camera slider posts (Camera slider ) from last year. That is basically a single axis control mechanism, done in a very minimal way. The original poster from that series had a lot of trouble as he wasn't a programmer and didn't really want to learn how to program enough to do what he wanted either. If you don't want to do programming, I'd probably go with a grbl style controller and use a raspberry pi or something to send the commands. That system used a stepper motor as the drive mechanism, and it was all open loop.

There is zero backlash in the machine as the Hammer height readout depends on it, is accurate to thou in practise and everyone who has one raves about the accuracy so the drive fundamentals are already in place. I will look at the link a bit later on and thanks to everyone who has replied. If it can be perfected there are a lot of users out there who can take advantage of it.

This idea pops up a bit, I have thought of exactly the same thing for my table saw, but have been too lazy and time/money deprived to do anything about it.

I have put a bit of thought into it though. The best solution I can think of is to use an existing CNC controller, in my case I would use a Beaglebone Black running LinuxCNC. Main reason is I have one and used it with much success on a 3D printer. Also it is reasonably cheap - about $120.00 bucks with a Xylotex port that makes things easier.

It is really overkill for this but it makes things easier for a few reasons :

The machine will need to be homed every time you power it up to know its position, this is simply a button on screen.

You need to scale the stepper/servo to suit your particular machine, this is a number in the initialising file that only needs to be done once, slightly tricky but with a small amount of swearing and minimal alcohol it can be done.

You won't have to learn C programming or anything cutting edge (pun)

Only other parts are a screen, keyboard, mouse if you like, a stepper or servo drive with motor and a power supply.

Once it is up and running you simply type in a move command and it moves to that position.

Again it is overkill but probably the simplest option as far as the learning curve goes.

A custom controller could be built to replace the Beaglebone but by the time everything is taken into consideration it wouldn't be any cheaper I don't think. Although it might be better in that you could use a keypad and simple LCD screen it would still require a bit of work to get going.

Cheers.

Russell.

nice solution epineh. Any reason for the beaglebone as opposed to a pi3?

whups, never mind - I actually went to look at the Xylotex site, which answered my question