Controlling an Actuator

[jeffie3]

I am looking to install an actuator on my saw. Thinking about using an Arduino mini. Here is where I have a question. As the saw blade get dull. I need a way to adjust the feed on the actuator. It uses a hydraulic cylinder at this time. Would I use resistance to adjust the feed? As the resistance increases it would slow the actuator motor down. Or is there a better way?

Thank you,

Jeff

 

[jeffie3]

After a bit of reading. I came across this. INA219 High Side DC Current Sensor Breakout Board. Am I on the right track?

Thank you,

Jeff

 

[WHONOES]

Need more info. What actuator on what machine how do you plan to drive it etc etc.
Some comprehensible diagrams would be very useful.

 

[jeffie3]

Actuator is 24volt 5mm/s 250mm stroke max load 1600n. The saw is a 10 inch band saw. A Jet HBS-10HD. At this time it has a hydraulic cylinder that controls the down feed. You lift the arm, insert a piece of steel, tighten the clamp and adjust the cylinder to control the down feed. I am hoping to change it over to an actuator. Many reasons. To learn more about microcontrollers. To be more precise with the down feed rate to extend the life of the blades and make a better cut. And to let the actuator lift the saw arm. Age and a broken collar bone that can't be fixed. Missing bone. If I have to cut several pieces I pay for it for a few days.

What I am thinking is. As the arm is coming down and the load on the blade increases, from being dull or material not being consistent. The current load on the motor will increase. Let the actuator lift the arm a certain distance. Then slow the feed rate by a percentage and continue the cut.

Using a microcontroller I can program different material cut rates for a starting point. I can use this formula to find the starting point. (Feed Speed (feet per minute) × 12) ÷ ((Band Wheel Diameter (inches) × 3.1416 × RPM) ÷ Tooth Pitch (inches))
If I am missing something let me know.
Thank you for your help.

Jeff

 

[hevans1944]

I like the idea of using an actuator to lift the horizontal band saw above the work after making a cut. Not liking so much the idea of using an actuator to control the downward movement versus the present "leaky cylinder" mechanism. The optimal speed of the downward "cutting movement" depends on a LOT of things, among them the type of material being cut, the width of the band saw blade, the pitch of the band saw teeth, the linear velocity of the band saw blade, the amount (and type) of liquid coolant applied during the cut, and probably the phase of the moon... just kidding on that last one! However, dull blades should be replaced, rather than the "feed" adjusted to accommodate them. The time-honored method of adjusting the down-feed for this type of saw is by "trial-and-error," starting with almost zero down-feed for unknown materials, and then experimenting with linear blade velocity and increasing the down-feed rate until acceptable results are achieved.

While it may be possible, even desirable, to fit a linear actuator to replace the current mechanism, you will also need some means to determine how fast the linear actuator is moving down while cutting, as well as limit switches that remove power to the actuator at the extremes of actuator movement. Sensing motor current is not definitive. Too many variables affect the motor loading. What you need (for the downward motion only) is a position transducer, such as an LVDT (Linear Variable Differential Transformer), which would be my choice, or perhaps a "string potentiometer" which would also serve well in this application. You can easily make your own string potentiometer using a multi-turn potentiometer with a small drum affixed to its shaft to wind the string around. Some mechanical design is required, but if you study the "slide rule" type of tuning mechanism, found on some multi-band radios, you will get some ideas. Off-the-shelf string potentiometers have a built-in spring (much like a clock main-spring) that retracts the string and allows the potentiometer wiper contact to return to its starting position when string tension is released, so you may want to try that approach in a DIY design.

If you do decide to use a string potentiometer, you can make a simple tachometer or actuator velocity sensor by connecting a small permanent-magnet DC motor, with a double-ended shaft, in tandem with the potentiometer shaft... or just wind a few turns of the string around the DC motor shaft.

Even without a tachometer, your microprocessor can measure the changes in displacement at fixed intervals of time to determine downward actuator velocity, and then use a look-up table or an algorithm to determine optimal downward cutting speed. You won't need this when you need to use the actuator to raise the (heavy) band saw blade mechanism after a cut has been made. Depending on the actuator, you may just simply tell it to go "full speed ahead" until the blade mechanism reaches a limit stop.

If I ever learn how to weld, I will certainly need to purchase a horizontal band saw to cut stock to length. I wonder how much those puppies cost on the used market? And I wonder where in the world I would find space for it...

 

[WHONOES]

Assuming your blade motor is ac driven, you could use a current transformer on one of the power leads to the motor and rectify and smooth it to get a dc level that is proportional to speed. Then feed the result, suitably scaled, into an A-D port in your μP and use that to control the feed rate of the arm.

 

[Harald Kapp]

I'm by no means an expert in sawing, but I'd look at the pressure the saw exerts on the workpiece and adjust that. This way you depend much less on the speed of the saw blade and the sharpness of its teeth.

 

[hevans1944]

I'd look at the pressure the saw exerts on the workpiece and adjust that

Sensible idea, Harald, but a horizontal band saw applies a virtually constant downward pressure on the blade, sometimes adjustable by moving a counterweight. The "leaky cylinder" has a piston that supports the blade mechanism during its downward (cutting) travel. The cylinder has an adjustable "needle" valve that (when fully closed) supports the blade mechanism and is adjustable for a "controlled" more-or-less-constant velocity cutting descent. A dull blade, or even worse a blade with NO teeth left, will not cut anything no matter how much downward pressure is exerted on the blade, and no matter how fast the motor driving the band saw moves the blade. Umm... that applies to steel, non-ferrous metals, and most hardwoods. Your mileage (or kilometers) may differ if the band saw is used (for instance) in a dairy operation for cutting hardened butter into sticks for packagaing purposes. And it may be possible to cut Styrofoam with a toothless band saw blade, although I have found a hot-wire "cutter" works well for very little cost.

The idea of using the motorized actuator to lift the band saw mechanism after making a cut has considerable merit. Those darned things are heavy, even when well-counterbalanced. My 76 year-old arms would quickly tire if I had to manhandle steel stock through one of these every day! It was manageable, if all I had to do was a few "one off" cuts for a special project, but I wouldn't want to have to operate one as my "day job."

 

[jeffie3]

I like the idea of using an actuator to lift the horizontal band saw above the work after making a cut. Not liking so much the idea of using an actuator to control the downward movement versus the present "leaky cylinder" mechanism. The optimal speed of the downward "cutting movement" depends on a LOT of things, among them the type of material being cut, the width of the band saw blade, the pitch of the band saw teeth, the linear velocity of the band saw blade, the amount (and type) of liquid coolant applied during the cut, and probably the phase of the moon... just kidding on that last one! However, dull blades should be replaced, rather than the "feed" adjusted to accommodate them. The time-honored method of adjusting the down-feed for this type of saw is by "trial-and-error," starting with almost zero down-feed for unknown materials, and then experimenting with linear blade velocity and increasing the down-feed rate until acceptable results are achieved.

While it may be possible, even desirable, to fit a linear actuator to replace the current mechanism, you will also need some means to determine how fast the linear actuator is moving down while cutting, as well as limit switches that remove power to the actuator at the extremes of actuator movement. Sensing motor current is not definitive. Too many variables affect the motor loading. What you need (for the downward motion only) is a position transducer, such as an LVDT (Linear Variable Differential Transformer), which would be my choice, or perhaps a "string potentiometer" which would also serve well in this application. You can easily make your own string potentiometer using a multi-turn potentiometer with a small drum affixed to its shaft to wind the string around. Some mechanical design is required, but if you study the "slide rule" type of tuning mechanism, found on some multi-band radios, you will get some ideas. Off-the-shelf string potentiometers have a built-in spring (much like a clock main-spring) that retracts the string and allows the potentiometer wiper contact to return to its starting position when string tension is released, so you may want to try that approach in a DIY design.

If you do decide to use a string potentiometer, you can make a simple tachometer or actuator velocity sensor by connecting a small permanent-magnet DC motor, with a double-ended shaft, in tandem with the potentiometer shaft... or just wind a few turns of the string around the DC motor shaft.

Even without a tachometer, your microprocessor can measure the changes in displacement at fixed intervals of time to determine downward actuator velocity, and then use a look-up table or an algorithm to determine optimal downward cutting speed. You won't need this when you need to use the actuator to raise the (heavy) band saw blade mechanism after a cut has been made. Depending on the actuator, you may just simply tell it to go "full speed ahead" until the blade mechanism reaches a limit stop.

If I ever learn how to weld, I will certainly need to purchase a horizontal band saw to cut stock to length. I wonder how much those puppies cost on the used market? And I wonder where in the world I would find space for it...

Thank you for your suggestions. I did learn something new. I have never heard of a "string potentiometer" All the suggestion seem interesting. Some of it is a little complicated for me. But as with anything. You start with an idea and soon with discussion you come up with a solution. If at anytime I question something or keep asking because I don't understand. Bear with me. I really never mean to offend. With the saw. When cutting, your blade slowly becomes more dull. The more you cut the less effective it becomes. Once you break a blade in(you cut some soft material to remove burrs that could break when cutting harder material). The cylinder keeps it from falling too quickly. If your adjustments allow it to fall to quickly it will shorten the life of the blade and depending on the material you are cutting, cause the cut to "wonder". Leaving you with a slanted cut. If the cut is slanted to much, it will stall or break the blade. As you cut more pieces you have to keep guessing the cylinder adjustment in order to get the best cut.

 

[jeffie3]

Assuming your blade motor is ac driven, you could use a current transformer on one of the power leads to the motor and rectify and smooth it to get a dc level that is proportional to speed. Then feed the result, suitably scaled, into an A-D port in your μP and use that to control the feed rate of the arm.

I am a little confused. Yes it has an AC motor. My idea of using a microcontroller with the 1nA219 sensor came about by watching my Wire EDM. As it cuts you can watch the amps rise and the wire will slow down a bit, then if it rises too much it will backup a certain amount then go forward again at a slower speed. If the amps stay steady it will increase the speed until it is back to the programmed speed. I was thinking I might be able to do the same with the saw and an actuator. Start out at a programmed speed based on the material, speed and blade. Sample the load on the actuator and adjust the actuator speed. My thought is. If the current is rising on the actuator motor. The down feed is becoming excessive. Back it up a bit then slow the feed rate down and start lowering the arm again.

 

[jeffie3]

I'm by no means an expert in sawing, but I'd look at the pressure the saw exerts on the workpiece and adjust that. This way you depend much less on the speed of the saw blade and the sharpness of its teeth.

That is what I am hoping to do. The cutting calculation is a starting point. Optimal cutting feed rate. Then sample the load on the actuator motor and micro adjust the down feed using a microcontroller.

 

[jeffie3]

Sensible idea, Harald, but a horizontal band saw applies a virtually constant downward pressure on the blade, sometimes adjustable by moving a counterweight. The "leaky cylinder" has a piston that supports the blade mechanism during its downward (cutting) travel. The cylinder has an adjustable "needle" valve that (when fully closed) supports the blade mechanism and is adjustable for a "controlled" more-or-less-constant velocity cutting descent. A dull blade, or even worse a blade with NO teeth left, will not cut anything no matter how much downward pressure is exerted on the blade, and no matter how fast the motor driving the band saw moves the blade. Umm... that applies to steel, non-ferrous metals, and most hardwoods. Your mileage (or kilometers) may differ if the band saw is used (for instance) in a dairy operation for cutting hardened butter into sticks for packagaing purposes. And it may be possible to cut Styrofoam with a toothless band saw blade, although I have found a hot-wire "cutter" works well for very little cost.

The idea of using the motorized actuator to lift the band saw mechanism after making a cut has considerable merit. Those darned things are heavy, even when well-counterbalanced. My 76 year-old arms would quickly tire if I had to manhandle steel stock through one of these every day! It was manageable, if all I had to do was a few "one off" cuts for a special project, but I wouldn't want to have to operate one as my "day job."

You are correct. The cutting pressure comes from the weight of the arm. The cylinder allows it to fall at a certain constant rate. Here is where it gets a little unreliable. With all the conditions being perfect. Including the phase of the moon. I do wonder about that one some days. You can set the dial at a number. Make a cut. If any of the conditions vary. It will cause a problem. A tooth will break. A few teeth will become dull. It will snag and stall the saw. The down feed will stay steady. Soon you will have too much weight on the blade and it will cause other problems. It will not be to long until you are replacing a $100.00 blade. At times, I can cut 100 or more 2" diameter 1018. Then at times I can get 10 cuts of the same material, and I am having to replace the blade.

 

[hevans1944]

Some folks have tried to get around some of the disadvantages of the horizontal band saw by using a circular abrasive "cut off" disk. We had one of those in our shop for awhile, but someone decided to use it without coolant and burned up a disk. Apparently the original installation of the machine did not include providing a recirculating coolant system, so this uninformed end-user must have decided cooling was unnecessary.

I am an electrical (electronic) engineer, but I dabble enough with mechanics to know that cutting tools generate heat, and removing heat is essential to extending tool life. The burned up disk adhesive, that had held the abrasive on the disk, created a most awful odor. That was enough to discourage me from using it, because management had no interest in providing a coolant recirculating system that would have (hopefully) prevented burning up cutting disks.

The big advantage the band saw (horizontal or vertical) has over circular abrasive disks is stock capacity and speed of cutting. The abrasive disk limits stock diameter to whatever the distance is between the arbor diameter and the outer diameter of the cutting disk, often as little as one inch. IIRC, that's the maximum diameter I can cut with my hand-held, un-cooled, diamond-abrasive, cutting disk tool. The abrasive disk cutter we had at my day job was good for about two inch diameter stock, and sometimes you just need a diamond abrasive to cut some materials... hardened tool steel for example. I don't know if band saw "blades" are available with diamond-impregnated bands instead of teeth. The closest I have ever seen to that is a semiconductor wafer saw that uses a diamond slurry that is picked up and carried on a loop of thin string to make thin slices on very brittle wafers. That one was a complicated and messy mechanism that was very slow, but it could slice very large boules. No doubt there have been some improvements since the one I observed in the 1960s.

As @jeffie3 noted in post #12 above, adjusting the down-feed of a horizontal band saw is a tricky, unstable, trial-and-error procedure for most saws. In our shop we had an el-cheapo model that I would use mainly to cut aluminum extrusions. Even with this relatively soft metal, I often had trouble because the previous user had dulled the band saw blade. Still, the horizontal band saw was better than using a hand-held hack saw, even if the hack saw had a new blade.

 

[WHONOES]

I have a circular saw that cuts metal as well as wood.

 

[jeffie3]

Some folks have tried to get around some of the disadvantages of the horizontal band saw by using a circular abrasive "cut off" disk. We had one of those in our shop for awhile, but someone decided to use it without coolant and burned up a disk. Apparently the original installation of the machine did not include providing a recirculating coolant system, so this uninformed end-user must have decided cooling was unnecessary.

I am an electrical (electronic) engineer, but I dabble enough with mechanics to know that cutting tools generate heat, and removing heat is essential to extending tool life. The burned up disk adhesive, that had held the abrasive on the disk, created a most awful odor. That was enough to discourage me from using it, because management had no interest in providing a coolant recirculating system that would have (hopefully) prevented burning up cutting disks.

The big advantage the band saw (horizontal or vertical) has over circular abrasive disks is stock capacity and speed of cutting. The abrasive disk limits stock diameter to whatever the distance is between the arbor diameter and the outer diameter of the cutting disk, often as little as one inch. IIRC, that's the maximum diameter I can cut with my hand-held, un-cooled, diamond-abrasive, cutting disk tool. The abrasive disk cutter we had at my day job was good for about two inch diameter stock, and sometimes you just need a diamond abrasive to cut some materials... hardened tool steel for example. I don't know if band saw "blades" are available with diamond-impregnated bands instead of teeth. The closest I have ever seen to that is a semiconductor wafer saw that uses a diamond slurry that is picked up and carried on a loop of thin string to make thin slices on very brittle wafers. That one was a complicated and messy mechanism that was very slow, but it could slice very large boules. No doubt there have been some improvements since the one I observed in the 1960s.

As @jeffie3 noted in post #12 above, adjusting the down-feed of a horizontal band saw is a tricky, unstable, trial-and-error procedure for most saws. In our shop we had an el-cheapo model that I would use mainly to cut aluminum extrusions. Even with this relatively soft metal, I often had trouble because the previous user had dulled the band saw blade. Still, the horizontal band saw was better than using a hand-held hack saw, even if the hack saw had a new blade.

There are a few ways beside the band saw. I have an abrasive cut off saw. I can cut up to 8 inches. It is used mostly for hot roll steels used in fabrication. It is very portable. It is designed to be used without coolant. The blade wears away exposing new material to cut with. If you do get it to hot and melt the edge of the blade it will not cut well. You have to break the melted part. As cheap as the blades are, most just get a new blade. And yes I also hate using it because of the dust and smell. It also disrupts the grain structure of steels(heat). That is why I only cut metals that I am going to weld and that are difficult to put into the band saw.

I also have a power hacksaw. It is a very old saw. It is getting harder and harder to get blades. I used to use it a lot more. But when I was able to pick up the larger band saw, used of course. It is a very interesting machine. Uses a hydraulic system to control the down feed. With each stroke it makes a cut, then it lifts the blade according to the adjustment to return the blade, then lowers it for the next stroke. You also have a pressure adjustment. If the blade has too much back pressure, which you can set that also, it will release the pressure, then attempt it again for the next stroke.

A saw I would like to have is a cold saw. They range from 8 inch blades to the largest I have personally seen 24 inches. I'm sure they are larger custom ones out there. The cold saw uses a carbide tipped circular saw blade. They are far better than all the above. They are not as fast as the band saw, but the cuts are better quality.

We also have a small band saw. Mostly for little quick cuts. Not to hard 1" inch diameter and below. It will cut up to 6 inches. It is just not very good at it.

When cutting harden steel. They do make carbide and diamond tipped blades. In many cases they cut the steels annealed, then have it heat treated. If is a made part. Some get it annealed cut it, then heat treat it again. But if you do have to cut it hardened the cold saw is a better saw to use. At times, we are contracted to cut hardened parts mostly titanium and carbide. We do it with our Wire EDM. The EDM is a cold process. No distortion with friction and heat.

Lastly. Yes and machining operation running coolant is alway better. Reducing heat and clearing away shavings. Aluminum loves to stick to the blade and load it up.

 

[jeffie3]

I have a circular saw that cuts metal as well as wood.

A DeWalt? They make a nice circular saw for cutting steel or wood. A customer of mine bought one and had to show it to me. He cut a 3-inch diameter 304 stainless steel bar with it. He had an A/C company. They are nice saws, just expensive. It is called a cold saw.

 

[hevans1944]

@jeffie3, thank you for the very nice overview of different sawing methods and machines. I would be willing to help you get a motor-driven linear actuator attached to your horizontal band saw for the purpose of lifting it after the cutting is finished. We might even, together, figure out a way to use it to control the descending cut. That will clearly be a more difficult task than a relatively simple "ratchet down and lift up" sort of implementation. A load cell under the work to be cut would help measure the downward pressure, and (as I mentioned before) a string potentiometer is a good way to measure linear distance of slow-moving objects. PM me if you like, or we can continue the discussion here in the Electronics Point forums.

Several cutting options no one has mentioned yet include water-jet abrasive cutters, which use ultra-high pressurized-water (30,000 to 90,000 psig) to carry an abrasive stream through a hardened nozzle; plasma cutters, which are cheap and portable, but leave a pretty large kerf; and laser cutters, which are large, stationary, and expensive but very fast and accurate.

I am not at all sure that covers ALL the ways to cut materials. Det cord might also be included, but this stuff requires special handling and experience most hobbyists (including me) simply do not have and cannot easily acquire (assuming we even wanted to, which I don't). Binge-watching seven years worth of Burn Notice re-runs on Amazon Prime isn't sufficient training.

 

[jeffie3]

@jeffie3, thank you for the very nice overview of different sawing methods and machines. I would be willing to help you get a motor-driven linear actuator attached to your horizontal band saw for the purpose of lifting it after the cutting is finished. We might even, together, figure out a way to use it to control the descending cut. That will clearly be a more difficult task than a relatively simple "ratchet down and lift up" sort of implementation. A load cell under the work to be cut would help measure the downward pressure, and (as I mentioned before) a string potentiometer is a good way to measure linear distance of slow-moving objects. PM me if you like, or we can continue the discussion here in the Electronics Point forums.

Several cutting options no one has mentioned yet include water-jet abrasive cutters, which use ultra-high pressurized-water (30,000 to 90,000 psig) to carry an abrasive stream through a hardened nozzle; plasma cutters, which are cheap and portable, but leave a pretty large kerf; and laser cutters, which are large, stationary, and expensive but very fast and accurate.

I am not at all sure that covers ALL the ways to cut materials. Det cord might also be included, but this stuff requires special handling and experience most hobbyists (including me) simply do not have and cannot easily acquire (assuming we even wanted to, which I don't). Binge-watching seven years worth of Burn Notice re-runs on Amazon Prime isn't sufficient training.

Hello, sorry it has taken me a few days to get back. A job came in to make 40 wrenches for a customer, so I have been consumed with making them. The saw upgrade would have been nice to have. The wrenches are 1-1/2 inches in diameter. So 41 saw cuts. Was a bit tiring. I always make 1 extra. If anything goes off while manufacturing the tool. I have an extra. This way I don't have to start over to make just 1.

I would like to do this in the open forum. Others might find it useful to convert their saw. I'm sure from time to time I will embarrass myself with my lack of knowledge or understand a thing or two. But my quest for understanding will out weight the embarrassment.

Next I hope you will be understanding. I do have one flaw that bothers others. If I don't understand why or why not the project is moving in a certain direction. I tend to beat a dead horse until I understand.

Lastly. As with this break. At times, I might be away from it for a few days or even longer due to projects. I need to keep the lights on.

Yes there are many other ways to cut materials. I have two plasma cutters, two wire EDM's. I also use a 12-inch compound miter saw I use to cut aluminum. I use parting tools to cut things off in the lathes I have.

If this is all agreeable. I will write up a flow chart or pseudo code to get things off to a start.

 

[hevans1944]

It might be a good idea to get the mechanics worked out first, since whatever you decide to use for the actuator will be specific to the band saw you have. Maybe post some drawings, or even hand-sketches, of how you will attach a motor-driven actuator (there are LOTS of these to choose from!) to the band saw. What stroke does this actuator need? How fast do you want it to lift the blade? How far? What is the maximum downward (cutting) speed? What axial load must the actuator accommodate?

Investigate possible locations for a compression-only load-cell to measure downward cutting force, preferably located under the work piece. What is the maximum downward force that can be exerted? Is this in addition to the force provided by the weight of the saw arm? If so, how much more?

Determine where a string-potentiometer (or other distance measuring sensor) will be mounted and how it attaches to the band saw. Is there a provision for attaching a PM DC motor to the string to act as a velocity sensor? Can the string-pot be ordered with a velocity transducer already installed? Or is it possible to measure (very slow) velocities by successive position measurements, with known time intervals between measurements (the usual rate = distance / time calculation)? The answer to that last question is "yes," but if you can get the answer from a hardware sensor that is almost always a better solution.

It is somewhat tricky to obtain accurate time interval measurements unless the time markers are generated by hardware interrupts, and the interrupt latency is known and accounted for in the time interval measurement. This is the so-called "real time" processing environment that generally requires a dedicated processor to assure timing accuracy. Of course, if the time intervals are long (on the order of seconds), being inaccurate by a few milliseconds won't adversely affect the accuracy of the rate calculation. So, if the cutting speed is, say, an inch per minute, then taking distance measurements every five or ten seconds (twelve or six times per minute) should be often enough to achieve an accuracy of a few tenths of an inch per minute without worrying about extreme time interval accuracy. Just be aware that more often measurements, with well-defined measurement intervals, will usually yield more accurate results. But this isn't brain surgery, so only design for "gud enuf" not for the Nobel Prize.

So, that's where I would start. Measure (or calculate) the downward speed and force required, the retracting speed desired and how far to retract. Then design the attachment mechanism and select the actuator. Maybe test this using manual control before spending any more money on force (load cell) and displacement sensors (string pot or LVDT or whatever).

 

[jeffie3]

O.K. the actuator is 24volts 3 amps. 1600N or 359.69 lbs. Speed is 5mm/sec with a 250mm stroke. It has limits switches built into the actuator. As far as attaching it. I just need to make a clevis. It will fit where the hydraulic cylinder is at.

As for a load cell. It could be placed under the material in the saw vise. The problem I see with that. The weight is in constant change as you make cuts. The material weight variations are a lot. Will add more later.