Building my own old school "setworks"

[fluidpowerpro]

I decided to build my own version of an old school types of "setworks". It will use low tech limit switches that will run on some cams I machine. I know that there are more modern ways of doing this, but this technology is what I understand and can fix later on if I need to.
The set up will allow me to have 11 preset board thicknesses that I can select with a rotary switch. It will be used with the electric winch I have already installed on my mill.
Here are a few pics of what Ive completed so far. I'll update as I go.

 

 

 

 

[VB-Milling]

Now this is super cool. Looking forward to more updates

[Crusarius]

That is a super kool concept. never thought of doing it that way. 

Ever think about using a few scroll wheels off some old mice for a rotary encoder?

Very interested in seeing the cams you come up with to.

[fluidpowerpro]

I didnt know you could use a mouse wheel as an encoder so no, I have not. Either way, I wanted to avoid encoders or LVDT's, etc, simply because that type of system is beyond my comfort level. 
I'm building it at my "cabin workshop" so progress will go in spurts. Ill be working on the cams on my next trip.

[Southside]

So that's pretty awesome.  Mind explaining what we are looking at in the photos?  Can you keep the explanation in Techno Dinosaur speak?  When "mice" were mentioned above my first thought was "How would rodents help?"  :D

[DMcCoy]

So that's pretty awesome.  Mind explaining what we are looking at in the photos?  Can you keep the explanation in Techno Dinosaur speak?  When "mice" were mentioned above my first thought was "How would rodents help?"  :D

Another comment where I was glad my mouth didn't have coffee in it.  :D
I'm so outdated, if I can't see a loose wire and know that isn't why something doesn't work I'm pretty much done.

[Crusarius]

Rotary encoder is pretty easy. it just counts steps. some have course steps some have fine. once you figure out the length of each step coding is pretty easy. May be worth just playing with one to give you a true digital readout. May have to get creative keeping the area clean though. the only reason I have not attempted something similar is the amount of sawdust and debris that you will have to keep out of it to keep it working right. Since my mill lives outside I didn't think getting into a digital setworks at this time was a good idea. maybe on the next version.

[fluidpowerpro]

Here is an explanation of what your looking at so far.
On each side you see 2 linear bearings. The cam arrangement will mount to those bearings and whole assy will mount vertically to the side of the mill. The cams will slide up and down on the bearings as the mill head moves. If the head moves up, the cams slide down on the bearings. These will be connected to the mill head by a small cable. 
The micro switches are mounted to a plate that is mounted on 1/2 of a cheap X-Y milling table. This will provide the means to adjust the position of the switches to "zero them". 
The whole works is mounted to a 1/4" thick, 8" x 51.5" aluminum plate. When completed the whole assy will be enclosed in a rectangular box made out of 1/4" aluminum, 52" tall x 8.5" wide x 5" deep.
What you see now is missing 2 micro switches. My electrical circuit requires 2 limit switches for each position ( 11 positions/board thicknesses).

As far as mice go, the only opening to the enclosure will be less than 1/4" dia where the cable goes through, so that should keep them out... :D

[Ljohnsaw]

So is there going to be a plate with a bunch of divots or bumps at various intervals that the micro switches is riding on?

[fluidpowerpro]

Yes but not one plate. The cams will be 1/2" x 1/2" aluminum square stock. One for each thickness setting. That way if I want to modify one setting, I dont have to remake the whole plate. It will also be easier to machine them because I will just need to machine the divot across the 1/2" face the switch roller rides on.

[Mattjohndeere2]

This is awesome. Thanks for sharing. I've been thinking of a similar thing for the Hudson mill I have, except more of a digital approach. A rack and gear, connected to an encoder, read by an Arduino or similar hobby computer, and a digital display with a keypad for manual entry. It would use some relays to bypass the manual winch switch, and I could just program whatever cut types I'd want (1", 5/4, 1.5", etc...) And let it do the work of operating the winch for each next cut.

I'm excited to see how yours works, really cool!

[Magicman]

I noticed that you are using an electric winch for head movement but I did not see where you mentioned braking or holding.  A shunt (direct short) across the motor leads when stopped will provide electric braking and should prevent overrun.

[fluidpowerpro]

The winch I am using has a brake built in and it seems to work good so far. I dont have many hours on it yet....

[Ljohnsaw]

Yes but not one plate. The cams will be 1/2" x 1/2" aluminum square stock. One for each thickness setting.

OK, now I see it.  You mentioned "cams" and I was thinking more like an engine cam shaft and couldn't figure out how you were going to get different spacing on a round cam and get it to come out even with every size... Doh!  Pretty cool idea/implementation.

[fluidpowerpro]

This is awesome. Thanks for sharing. I've been thinking of a similar thing for the Hudson mill I have, except more of a digital approach. A rack and gear, connected to an encoder, read by an Arduino or similar hobby computer, and a digital display with a keypad for manual entry. It would use some relays to bypass the manual winch switch, and I could just program whatever cut types I'd want (1", 5/4, 1.5", etc...) And let it do the work of operating the winch for each next cut.

I'm excited to see how yours works, really cool!

The way you were thinking is not as ancient as my design and probably the smarter way to go. I must admit that most of what I build isnt the latest tech, but it usually works. I say usually because every project is a prototype so there will be tweaks and refinements along the way. 

[D6c]

Nice looking project.  Hope to see a video of it in operation.
I considered (briefly) building a setworks for my LT40 using a ball screw and servo motor from a CNC mill but I'd have to do a lot of learning on the control end of things....might have attempted it If I didn't have a 100 other projects I'll never get done.

[fluidpowerpro]

I worked on the assembly this past week. 
Changes/Progress
1. The zeroing adjustment for the micro switches will need to be out one side of the enclosure so I added a right angle drive to the adjustment rod. The drive is just a low cost unit I got on e-bay. Its meant to be used as a right angle screw driver attachment.
2. After much effort I was just not able to make everything for 11 setpoints work. It was all just too tight and there was not enough room for fine adjustment of the cam positions. There is too much slop, side to side, on the micro switch rollers so unless the cams were perfectly straight, the rollers would come off the cam by the time the cam traveled its full 24" length. Tried putting sheet metal dividers between the cams, but the tolerance stack up on everything was just too much to manage. Decided to scale back and only have 7 setpoints. This allowed me to remove one micro switch between each set giving me 1/4" spacing in between. I will change from 1/2" wide cams to 5/8" wide. This will give me the leeway I need to fine tune the final position of the cams.
3. Added a spring to the zero adjust slide. The spring will pull the slide to one end removing a slight amount of play that was in the slide mechanism.

Here are some more pictures. The pics are with the 1/2" wide cams but those will be changed to 5/8" when I get material.


 

 

 

 

 

 

[Ljohnsaw]

Why two switches for each setpoint rail?

[fluidpowerpro]

I am going to have an indicator light to tell me when the cam is "zeroed". I could have used a single DPDT switch and actually purchased some, but those were much larger and both contacts didnt change state at the same time. One switched earlier than the other based on lever travel. 
In the end decided to use 2 single pole switches. 2 of these ended up being smaller and actually cost a lot less.

[Ljohnsaw]

So when the units moves over the notch and drops in, it stops the motor from running, right?  So you would be using the normally open contact (closed when riding the cam).  So you run power to the common lead and then to a power (solid state) relay for the lift motor from the NO contact.  When it dips into the notch, the motor relay goes "off" stopping the motor.  Couldn't you use the NC contact to power your light (LED)?  If using a mechanical "power" relay, then just some of the other contacts for your LED?  Then you only need one switch per setpoint.

[fluidpowerpro]

Yes, I'm using the N/O contact and its held closed by the cam until it falls into the notch. 
I dont have my circuit in front of me, its at my shop, but I recall that unless I had a separate set of contacts for the light, (there is only one "zero" light for all setpoints) that portion of the circuit would make all circuits on the output side of my selector switch common.
Each switch is only active as determined by the selector switch....
In my trade ( hydraulics ) there are always multiple ways to accomplish something. Some better than others, but not all bad. I'm sure electrical circuits are the same. 

[Ljohnsaw]

OK, I just sketched it out for proof of concept.  It actually make for very simple wiring.

Positive to your selector switch (fused).  Selector switch to each micro switch common lead (7 wires now, 11 before).  Join all NC micro switch leads to the indicator light that goes to ground.  Join all NO micro switch leads to motor control (solid state) power relay positive lead, other lead to ground.  Main (fused) power to (solid state) power relay to motor to ground.

Power goes from battery, to selector switch to micro switch.  If riding on cam, motor runs.  Power back feeding along the NO contact buss goes nowhere else as the selector switch stops the flow.  Micro switch rides the cam and drops into slot.  Power now goes to the NC contact, lights the light and back feed on the NC buss is halted like before.

You just need a power reversing switch (DPDT) after the power relay and a on/off switch (on power relay input) to disable the motor when adjusting to "zero" with your crank (leave power on to selector switch so light can light up).  you would need a momentary switch to bypass the micro switch (selector input to NO buss) to start the up/down process.  There would also need to be a halt switch (NC in the power relay input circuit) at each end of your travel.  The bypass switch would allow you to jog back if set up right.

In use, I would expect to see motor switch off and then crank to zero - light comes on.  Make cut and return.  Set direction switch (probably down), set motor switch to "run" or "enable" and press bypass button.  Motor would start up, light would go out.  Continue until notch - motor stops and light lights up.  The light becomes more of an "on target" indicator.

Hmmm, might have to make one of these for my mill!

[fluidpowerpro]

As I mentioned before, I dont have my electrical schematic with me, but next time I'm at my shop Ill make sure and bring it home with me so I can share it with you. You can then look it over and if there is a better way, I'm all ears.

[fluidpowerpro]

you would need a momentary switch to bypass the micro switch (selector input to NO buss) to start the up/down process.

I plan to use a single shot time delay relay to get around this. It will be set just long enough to allow the micro switch roller to get out of the slot and will automatically reset before the next round of adjustment is needed.

[fluidpowerpro]

 

 

 

 

 

 

 

 

 

 

After much trial and error finally found a cam profile that works. I used a 1/4" ball nosed end mill, .050" deep. 
Also started to add sides and ends.

[moodnacreek]

I never understood 'automatic setworks'  I'm thinking you turn the movement on and a limit switch turns it off. So there is a limit switch for each pre-set or thickness. And an override feature jog around and get to a starting point to do a 'stack' of like thickness boards.   On circle mills this can be on a wheel or a vertical slide. It seems like a electric brake gearmotor would be needed but it could also be hydraulic .  I guess the endcoder/ computer set up is the modern way and that is really beyond me.

[fluidpowerpro]

Yes, t

I'm thinking you turn the movement on and a limit switch turns it off. 

Yes, thats what it will do. I manually initiate the saw to go down. It will continue down as long as I hold the switch. When the limit switch is tripped, the motion will stop.

[Crusarius]

Going through an awful lot of work to still have to hold the button. I would want to be able to hit the direction key and walk away. 

Another thing I planned on mine was to put a limit switch on a magnet and be able to place it wherever I want on the mill. That way when I am ready to load the next log I can set the switch to the height I want the head and hit go. Then by the time I get the log loaded the head is at the right position and should take minimal movement to get to my first cut point.

My head takes over 2 minutes from last cut to top of travel so the last feature would be very nice.

[fluidpowerpro]

Thats not a bad idea. I think for now I'll hold the button. If everything works reliably, it wont be hard to add some latching contacts later.

[moodnacreek]

Going through an awful lot of work to still have to hold the button. I would want to be able to hit the direction key and walk away.

Another thing I planned on mine was to put a limit switch on a magnet and be able to place it wherever I want on the mill. That way when I am ready to load the next log I can set the switch to the height I want the head and hit go. Then by the time I get the log loaded the head is at the right position and should take minimal movement to get to my first cut point.

My head takes over 2 minutes from last cut to top of travel so the last feature would be very nice.

That's what time delay relays will do. You can have delay off or delay on and they are adjustable. That 2 min. up down travel time would be unacceptable here.

[fluidpowerpro]

Just finished another round of work on the assembly.

• Finished the cams and mounted them to the slide plate.
• Finished the enclosure with door.
• Because the saw structure, and where I want to mount it is shorter than what is needed to get full travel on the cams, the cams actually have to exit the enclosure out of the top. The red "hat" pictured covers the opening to keep dust out. It also has a drip edge to shield the top of the door.
• >
   
  
   
  
   
  
   
  
   
  
   
  
  
• Snow will be flying soon so I removed the saw from the bed to bring inside to mount the assembly. It will mount in the space between the existing electrical control box for the winch and manual scale you see in the picture.

Next will be to mount pulleys and cable that will move the slider plate up and down with the saw head, and then wiring.
*Note that I did do some testing with an ohm meter on the switches and cams. As best I can tell using a hand held caliper, I'm holding about .020". We'll see where it ends up once its mounted. I'm sure adding pulleys and the repeatability of the winch brake will make it worse. In the end I think consistency is the big thing I'm shooting for. Regardless of where the actual board thickness ends up, I want each board to be the same.

[fluidpowerpro]

Finished another round of work on the assembly.
I'm getting close to being able to test it. Just a bit more wiring to do.

This last round included:
1. Mounted the assembly into the saw frame. Had to relocate the current visual scale to make room.
2. Added pulleys and cable to move the cam plate up and down with the saw head. Pictured is with 1/8" cable but thats too stiff so will be changing to 1/16".
3. Changed the adjusting wheel for zeroing the cams to one that is a larger diameter. This one turns so much easier.
4. Added switches , etc to the existing electrical control box. The box originally had a screw on cover so I added hinges to better accommodate mounting things in front through the door.
5. Added a latch to the door of the aluminum enclosure.
6. Relocated the "max up" limit switch. This shuts off the winch when the upper limit of head travel is reached.




 

 

 

 

 

 

 

[Ljohnsaw]

Looking good.  Can't wait to see it in action. 

If not already planned, you will probably need some sort of external indication as to the position of the x-y slide so you know which way you should zero on a scale.  Otherwise you might find you end up at end of travel after repeated zeroing (multiple logs/cants).

My setup is completely external and a manual process to zero so its a no-brainer.  If it was hidden (like yours), it would be very difficult.

[moodnacreek]

Just finished another round of work on the assembly.

• Finished the cams and mounted them to the slide plate.
• Finished the enclosure with door.
• Because the saw structure, and where I want to mount it is shorter than what is needed to get full travel on the cams, the cams actually have to exit the enclosure out of the top. The red "hat" pictured covers the opening to keep dust out. It also has a drip edge to shield the top of the door.
• >
   
  
   
  
   
  
   
  
   
  
   
  
  
• Snow will be flying soon so I removed the saw from the bed to bring inside to mount the assembly. It will mount in the space between the existing electrical control box for the winch and manual scale you see in the picture.

Next will be to mount pulleys and cable that will move the slider plate up and down with the saw head, and then wiring.
*Note that I did do some testing with an ohm meter on the switches and cams. As best I can tell using a hand held caliper, I'm holding about .020". We'll see where it ends up once its mounted. I'm sure adding pulleys and the repeatability of the winch brake will make it worse. In the end I think consistency is the big thing I'm shooting for. Regardless of where the actual board thickness ends up, I want each board to be the same.

That picture of the notched aluminum bars, put together, looks like my scale board.

[mike_belben]

Im not fully understanding the notched scales and cam follower microswitches.   Is this a mechanical means to avoid a stepper motor, home switch and driver?

I dont mean it as a criticism, youve done a lot of nice work. 

[fluidpowerpro]

Yes, thats why I call it "old school". I know its low tech, but its what I understand.

[fluidpowerpro]

Looking good.  Can't wait to see it in action.  

If not already planned, you will probably need some sort of external indication as to the position of the x-y slide so you know which way you should zero on a scale.  Otherwise you might find you end up at end of travel after repeated zeroing (multiple logs/cants).

Yes, that did cross my mind but I'm going to go without for now. I have found that while zeroing, it works best to zero by moving the slide in the upward direction, against the spring. That way I know any play has been eliminated. I did put indication arrows by the adjustment wheel noting which way to rotate for up and down. In use, I will try to avoid working near the travel limits of the slide by always going down, and then adjust up until zero is achieved.

[moodnacreek]

I think Meadows used this system on one of their automatic setworks years back. Before that they and others had a chain driven drum that stopped the hyd. set motor through a solenoid valve. This was circle mills but other than the kerf it is all the same.

[fluidpowerpro]

Hi all, although I have worked more on this, no picture worthy changes to show. 
I was actually able to do some testing on the system and my electrical circuit worked as intended. While cycling up and down I determined that I need to scrap the small "micro" size micro switches and go to the "standard" size micro switch that is larger. The lever arms on the smaller size are just too weak and are easily bent. The larger size is more robust. As it goes while your building something, I went with the smaller ones because my original circuit used more switches so I needed the small size. The design then progressed, using less switches, so now I can fit the larger ones.
Also, while cycling up and down repeatedly for about 5 minutes I learned that my PMW motor control did not like that. It blew up! Yes, it actually blew up! loud noise and some parts flew!
I used that control while sawing most of last summer with no issues so it worked good enough for that kind of duty, but as long as I've got it in the shop, I'm going to go to a larger control just to be safe. This will require a larger electrical enclosure that I have on order already.

[Southside]

So the magic smoke didn't just escape - it erupted eh? 

[fluidpowerpro]

Yea, after looking at it, I think what blew up was a capacitor. 

[Ljohnsaw]

Were you switching directions quickly when it happened?  You MUST let the motor come to a complete stop before you change direction.  Otherwise you get a transient spike that pops the controller.  You might be able to add an inductor to the lines going to the motor to suppress the spikes somewhat.  I fried my controller for my power drive doing that.  You could make it fool proof by adding one of your one-shot relays into the power circuit.  Somehow it would need to break the power for 1 or 2 seconds when you move the direction switch from one setting to off.

[fluidpowerpro]

Thank you for the suggestion. I dont know if that was the case, but it very well could have been. My winch has a brake and it does seem to stop very quickly, but I may have tried to reverse faster than it engaged.

[fluidpowerpro]

Hi all, 
I finally finished the system and have it installed. 
A few of you asked for a video so here goes... Its the first video I ever made and posted on Youtube.

Not sure if I'm doing this right...

low tech setworks - YouTube

[Hilltop366]

Nice job on the build and video.

[fluidpowerpro]

For those of you that followed my build of my own version of setworks, now that I have some time using it, figured I could give an update. I'm happy to report that it's been working really well with no issues so far.
My last job was all 2" thick trailer decking. When I made the cams, I machined them 2.140 and the boards are ending up at a consistent 2.10. 
The main thing is they are consistent.
I can always make a new cam if I want to tweak the final thickness, so that's no big deal.

 

[fluidpowerpro]

Also wanted to shout out a thank you to ljohnsaw for helping me refine my electrical control circuit. Having a second set of eyes to look at my design was very helpful.

[Ljohnsaw]

Your very welcome.  I enjoy a puzzle/challenge!

[moodnacreek]

First let me say you do nice work and you did something I have long thought about. The part I don't understand is the 'zeroing in'. In my mind that would only get done once.              In sawing a log into boards I presume you make a cant into a determined width and then keep hitting your preset button down to the last board. Is this how it works?

[fluidpowerpro]

Yes, you only zero once. After you raise the head to return, each time you push the down button, the saw moves down to the next groove in the cam. For example, you make the first cut, raise the head, return, push it once, the head lowers to the last cut, push it again, it lowers to the next cut. The next cut, you will lower 3x, etc, etc,.

[moodnacreek]

Got it, thanks. Been out of range for a few days.