[ethcnc]

Hello there,

I was trying to build a prototype that can make a Bernoulli spiral using paint. I want to use servo with encoders and I want to buy a plug and run set with a coding and path generator software. Anyone has any suggestions where to start and where to buy the parts?

Thanks,
MN

[Jim Dawson]

A few questions:

Budget?

Working size?

Accuracy required?

Pretty much any appropriately sized 3 axis machine will do what you want and there is lots of software available. For only a Bernoulli spiral, an Excel spreadsheet could do the math to generate the data points or you would write a quick program in VB or something to do it. You will have to figure out how to apply the paint, I have used a magic marker, but never tried to apply paint.

[ethcnc]

Thanks Jim,

The reason I wanted the software is because I have other inputs I want to add in the near future after I get the parts and test. I'm planning to use servo dispenser for the paint.The accuracy is very important too. Please help me get the parts with expandablle PLC. I was looking at Panasonic Minas 5 PLC but not sure if I can use it to run other type servos. Please give me some guidance. I will give you more info if needed.

Thanks,

[Jim Dawson]

Let's look at the process. First you have to create some kind of graphic or at least data points that can be converted to the G-code that the CNC program can understand. The CNC program then converts the G-code into something that the machine controller can understand to actually run the machine. With some purpose written software, it would be possible to go directly from data points to machine control, but as far as I know nothing like that exists on the market.

Panasonic Minas 5 does not seem to be a PLC, but rather it's a servo drive. You normally want to use a matched set of motors and drives, so if that's what you want to use, then you will want to use the Panasonic motors also. It can be controlled by a PLC according the product literature. For what you want to do with multi axis positioning, you are really going to need a motion controller. There are a few ways to do this, but for expandability, a Mesa or Galil controller would be your best options. Galil is my prefered controller.

It would be possible to run the system on Mach3 software, but given your need for custom functions it would be an interesting macro programming excersize to get it to do what you want. There are other options available. Are you capable of programming a PLC? There will be no off-the-shelf software to do what you want.

Have you actually designed the machine yet? That is really the starting point. Again, what is the working area that you need? This will pretty much dictate what motors and drives you need. The size of the machine determines the mass that the motors need to move and thus the size of the motors and drives. You could start with an inexpensive CNC router kit, then build from there. Or design a machine from the ground up.

I'll help you through this, but I will have more questions. Seems like an interesting project.

[ethcnc]

Jim, I don't know how to thank you!

I can make something to show you what I would like to build using Catia but since it is so simple I have tried to draw it and attach it here.

Attachment 368750
The vertical shaft stays at the center of a round table and rotates only at the center (rotating the horizontal actuator), the horizontal actuator will move on the radius one step per rev., and the dispenser (carried by the actuator) will dispense the paint or paste controlled by a servo.
The most important thing for me is that the system is integrated and to be able to add some other input such as temperature in future.

I thought of teaching myself how to program the PLC because I have experience in other programming software but let me know if I need more training to program a PLC. I wanted to say Panasonic motion module FPG-PN8AN because it has 8 axis but I'm not sure if I were mentioning the right part since I started browsing without expert guidance (I was looking at the FP Sigma family but a bit expensive). The main reason I wanted Panasonic was because I wanted to keep it when I build the real machine.

There is no load on any of the axis except for the weight of the shaft, parts, and actuator. To be honest, I wanted to start designing the machine after I get the parts but don't know if I can buy any of those built for CNC gantry. I was looking at those servo plug and run systems found on CandCNC.com but was not sure if they can do what I want. Can those servo plug and run systems do it? Do I need other programming or coding software? Mach3 is just to run the software? Can I control speed of each axis separate? If I buy the CandCNC kit, can I use it for this spiral motion (no gantry) using other cantilever type setup? Can I control the dispenser with the same system? If I write a program using Matlab, can I convert it to G code?

Thank you,

[Jim Dawson]

OK, the sketch helps. Now I understand what you want to do, but I still don't know how large the work area is. The size is important to determine the needed motors, and may dictate the optimal design for the whole machine.

One of the problems that I see with a rotary design is the cable management. Without a rotary connector, you are going to be very limited on the amount of rotation of the shaft, and rotary connectors are very expensive. An X/Y table/gantry will generate any path you want, at single digit micron accuracy if needed. Given the state of the current technology and the available equipment it is much simpler to generate an accurate spiral path with X/Y linear motion than it is to do it with rotary motion. A cantilever system will work fine, I actually have a 4x8x1 foot cantilever X/Y/Z system in my shop.

Here is a picture of an inexpensive cantilever X/Y/Z system similar to what I have. You would only have to mount the system on a table.



This would be a good base to start from. You would have to replace the controls and maybe the motors/drives, but the mechanics are there. Add some linear encoders and the system would be incredibly accurate.

Mach3 is a CNC controller software as is CandCNC. You would still need CAM (Computer Aided Manufacturing) software to generate the G-code from some type of vector graphic file. There are also converters that will take a jpg or bmp file and convert it to a vector graphic. You mention that you are using Catia, I think the math is already built into it that would generate a Bernoulli spiral shape. Then just export that as a DXF file and process that with any CAM program to generate the G-code.

If I were doing this, I would build or buy an X/Y/Z table of some kind, buy whatever motors and drives are appropriately sized, use a Galil motion controller, and use my CNC software to run the whole thing. Why the Galil controller? It is the easiest to program, and has a massive amount of PLC type functionally built in as well as being an industrial class motion controller. And compared to many of the industrial controllers available is relatively inexpensive. It will also take temperature inputs. Also has 40 to 80 uncommitted I/O (depending on model) for PLC functions that can be fully coordinated with the motion. The reason I would use my CNC software is because it is very flexible, and can be customized for specific operations. It would also be possible to use it to directly generate the G-code for a Bernoulli spiral internal in the program by adding a module with the proper algorithm. The trig module is already in the program, you only need to input the parameters to generate the spiral shape.

[ethcnc]

Jim, I can't see the picture for some reason.it s a 3 feet diameter table and I will have a solution for the cable. If I go with the Galil, how much is it? Can you help me buy all the needed components including the software? If we have a list all the parts, what are they? Is it possible to use servo motors that come with built-in encoders?

Thanks,

[Jim Dawson]

The picture seems to be gone. Perhaps posting pictures with a watermark is not allowed on this forum, dunno. Here is a link to something similar (this is not a recommendation, just for reference). https://www.ebay.com/i/391842955100?chn=ps&dispItem=1 I would post a picture of mine, but I would have to drag it down off of the shelf to get pictures.


Galil prices run from a couple hundred used on EBay, to around $6000 new from Galil for their latest 8 axis controller with built in 1200W servo drives. What you really need is a 4 axis controller; X,Y, Z, and A for the paint dispenser. This would run about $2300 new from Galil for their top line model, or maybe as little as $250 on EBay. The exact model is dependant on how you want to set up the computer. They supply two main models, PCI cards that reside in a desktop computer, or stand alone ethernet units that can be used with any computer including laptops.

Unless there is some overriding reason for using a rotary system, and since your goal here is to get paint onto a surface, I would recommend going with a standard X/Y table of some kind. Trying to create a rotary system as you describe will require some interesting engineering and software challenges, where a straight X/Y system can be built with off-the-shelf parts and software, and do exactly the same job. And as I said above, can have single digit micron accuracy if needed. I suspect that since the medium you are working with is paint, then +/- 0.1 mm accuracy would be more than sufficient, and is easily achievable.

Just as an aside, the paint dispense rate can be programmatically ''geared'' to the X/Y vector speed to dispense the correct amount of paint without regard to the speed of the individual axes so that the dispense rate will be consistent over the distance traveled.

For best control and accuracy, servo motors with encoders would be my choice. My prefered method is to use linear encoders on the load, this compensates for any inaccuracy in the drive system and is the way all of my machines are configured.

Once the design is nailed down, then is the time to create a list of materials. Not quite there yet.

[ethcnc]

Hi there,

sorry it took me this long to respond. If I buy the x/y/z table can i use all axis for just rotation and to control the linear actuator? If that is possible, then I just want to go ahead and start looking at a full kit. I was looking at

https://www.candcnc.com/product-cate...servo-systems/ for a while but not sure if I can program it for rotational and to control the dispenser. The design is going to be started after getting the parts. For the rotation I am planning to use slip ring to avoid complication of wires. Do you have any suggestions?

Thanks

[Jim Dawson]

Hi there,

sorry it took me this long to respond. If I buy the x/y/z table can i use all axis for just rotation and to control the linear actuator? If that is possible, then I just want to go ahead and start looking at a full kit.

You can create a perfect circle or any other arc shape with only X/Y linear motion, no rotational axis is required. I have done so to single digit micron accuracy. It's all done with trigonometry internally in the motion controller. You do not need a rotary axis to generate a Bernoulli Spiral or any other arc path. I could easily do this on my 3 axis CNC router or milling machine.

I was looking at

https://www.candcnc.com/product-cate...servo-systems/ for a while but not sure if I can program it for rotational and to control the dispenser.

I have no idea if the CandCNC is adaptable to running a rotary axis with a linear axis and another rotary axis (paint dispenser) attached. To achieve an even distribution of paint, the paint dispenser is going to have to be programmatically geared to the vector axis. The vector axis in this case would be a function of SQRT(A^2+B^2), where A is the rotary axis and B is the linear actuator. The C axis (paint dispenser) would be programmed to rotate (dispense) at some factor of the vector axis velocity. I don't know if the CandCNC is programmable to do this, but it might be possible to do it all in G-code by controlling the X/Y/Z/A axes. In that case, CandCNC should do what you want. Not sure what CAM program would be used to generate the G-code.

The design is going to be started after getting the parts.

You don't know what servos and drives to buy until you design the mechanical parts of the machine. You need to know the mass and inertia of the system to be able to properly size the motors. There is no way to know this without designing the mechanics and doing the math. And since you want to use a rotary system, it's going to be some interesting Calculus problems. Maybe Catia can do it. You have some engineering to do before you start buying parts.

For the rotation I am planning to use slip ring to avoid complication of wires. Do you have any suggestions?

Thanks

I did a quick Google search. Something like this might work, looks like they are designed for your application. I have never seen them before, thus no experience at all..

Senring SNZ030 series, serve for 8pcs servo motors -Senring slip rings - Senring slip ring

My biggest concern is electrical noise and missed encoder pulses, thus affecting the accuracy and reliability of the machine. Maybe these work OK, I dunno.

[ethcnc]

Jim, thank you for your educative response. I really don't know how to design it since I don't know much about the motors. All I know is that I want smart motors and linear actuator just like I showed you on the sketch. I don't think here is any load or drag on the motors since one servo is mounted on the linear and the only thing the actuator is carrying is the paint hose and the other motor is carrying nothing but mounted on the vertical shaft to spin the shaft, last and third motor is also mounted on the actuator to dispense the paint and no load or drag force.

The slip ring you posted is exactly what i want to use and that will solve many complications. The electrical noise you mentioned... is it from the slip ring or from other parts?

Jim, do you sell any of the CNC parts?

Thanks again!

[ethcnc]

Can I relate what I am thinking of to lathe machine where the table is stationary and has a programmable coolant nozzle?
Haas Video Gallery - Tip of the Day | Haas Automation

[LeeWay]

Well, can you modify an ink jet printer?

[Jim Dawson]

Jim, thank you for your educative response.

My pleasure

I really don't know how to design it since I don't know much about the motors.

It just a matter of sizing the motor(s) to accelerate the load at the rate needed for the job. The greater the mass of the system, the more power you need. Without knowing the mass of the system and the speed requirements, it is impossible to properly size the motors. In this case, the rotary axis will be turning relatively slowly, so the motor can be geared down dramatically, maybe in the range of 90:1 or so. But this is going to require a rather expensive zero backlash gearbox to maintain accuracy. As the motor is geared down, the torque requirement of the motor itself will be reduced by a factor of the gear ratio.

All I know is that I want smart motors and linear actuator just like I showed you on the sketch. I don't think here is any load or drag on the motors since one servo is mounted on the linear and the only thing the actuator is carrying is the paint hose and the other motor is carrying nothing but mounted on the vertical shaft to spin the shaft, last and third motor is also mounted on the actuator to dispense the paint and no load or drag force.

While there is very little load on the motors once at speed in this case, only some minor friction load of the system. But there is a lot of load on the motor during the acceleration phase, this in turn will affect the accuracy of the system. To complicate this further, the linear actuator will present a variable load to the rotational motor as a function of its distance from the rotational axis center. That means that the rotational axis motor needs to be sized to handle the maximum acceleration at the maximum extension of the linear axis, and taking into account the additional mass of the paint dispensing axis. This is calculated by the formula F=MA (Force = Mass x Acceleration) and the moment arm also has to be applied to the calculation. I haven't reviewed the formulas, but as I remember the force required is a function of the square of the distance. I'm going to have to dust off one of my old physics books It is going to be an interesting bit of engineering.

The linear axis motor can be quite small as can the paint dispenser motor.

I will propose one more alternative design: Would you concider rotating the workpiece rather than the paint dispenser? It would vastly simplify the system.

The rotary operating design that you want to use is proabaly the most complicated and expensive way to accomplish your goals. But since you insist on building a rotary system, let's proceed in that direction.

The slip ring you posted is exactly what i want to use and that will solve many complications. The electrical noise you mentioned... is it from the slip ring or from other parts?

The electrical noise would come from a connection that is not 100% positive in this case. There could also be ''cross talk'' between the power and encoder signals in the slipring unit. It would seem that these problems may have been solved by the slip ring manufacturer. Only some testing would tell the true story. There are also ways to electrically reduce the noise by filtering the output to the servo motors and using proper cabling. It is also possible to use servo motors that have an internal closed loop so that the drive is internal in the motor. This would reduce cabling requirements and make should make the system more robust. This eliminates the need to pass the encoder signals through the slip ring, but you are still passing control signals.

Jim, do you sell any of the CNC parts?


Thanks again!

No, I buy CNC parts. I design and build specialized machines for many applications. I also write the operating software for the systems.

Can I relate what I am thinking of to lathe machine where the table is stationary and has a programmable coolant nozzle?
Haas Video Gallery - Tip of the Day | Haas Automation

I need to think about that for a bit, but I don't think so.

[ethcnc]

Jim, thank you for the response.

In this case, the rotary axis will be turning relatively slowly, so the motor can be geared down dramatically, maybe in the range of 90:1 or so. But this is going to require a rather expensive zero backlash gearbox to maintain accuracy. As the motor is geared down, the torque requirement of the motor itself will be reduced by a factor of the gear ratio.

Yes, I was looking at some Panasonic parts a while ago but was not sure if that is all I need and the price was a bit expensive for prototype. Here is what I sent to the dealer last year to ask if I need more parts such as power supply, cables... to complete my prototype but didn't get any help. Parts are from Digikey.
In my case I want the vertical shaft to rotate as slow as 30 RPM and when fast 300 RPM (if possible).

3 x servo motors ............................................... ..............MSME041S1S
3 x Drive (amplifier) .................................................. ...... MADHT1107B01
1 x Motion Control Module .................................................. .....FPG-PN8AN-ND
1 x PLC CONTROL LOGIC 16 IN 16 OUT 24V ..............................FPG-C32T2H

I will propose one more alternative design: Would you consider rotating the work-piece rather than the paint dispenser? It would vastly simplify the system.

If it is possible, I really want to stick with rotating the shaft rather than the work-piece. Rotating the work-piece will be an issue on future design phases. Once you help me size and pick all the parts, the calculus to program the motion will not be a problem.

The paint dispenser will have its own motor. The Leader in Meter Mix Dispense Systems and Valves for Adhesives, Sealants and Lubricants but didn't check the price yet. If too expensive, do you think we can design one easily? The dispensers on the link are more for viscous materials but I believe that I can use it for the paint.

No, I buy CNC parts. I design and build specialized machines for many applications. I also write the operating software for the systems.

You are exactly who I want to talk and who can help me. Please let me know if there is any more info I can give you to decide what parts are needed for this weird idea.

Thanks.

[Jim Dawson]

Jim, thank you for the response.

In my case I want the vertical shaft to rotate as slow as 30 RPM and when fast 300 RPM (if possible).

Well that adds whole other layer of complexity, I was thinking in terms of maybe 1 to 10 RPM. At 30 to 300 RPM the balance of the rotating system is going to be critical or it will tear itself apart. This is going to require a servo controlled counterweight to keep the rotating mass in balance as the linear slide moves out. This might also be done mechanically, need to think about that a bit. The speed also adds a number of safety issues to the design. The design will have to be such that as much of the rotating mass will be as close to the center of rotation as possible, and anything that sticks out beyond the center has to be a light as possible.

It might be possible to keep all of the servos outside of the rotating mass by using a coaxial mechanical arrangement to drive the linear slide. In that case it would only require a rotary union to transfer the paint, the slip ring would not be required.

This is going to get interesting. You stated in an earlier post that this is just a prototype machine, and I assume much smaller than the final version?

Yes, I was looking at some Panasonic parts a while ago but was not sure if that is all I need and the price was a bit expensive for prototype. Here is what I sent to the dealer last year to ask if I need more parts such as power supply, cables... to complete my prototype but didn't get any help. Parts are from Digikey.
3 x servo motors ............................................... ..............MSME041S1S
3 x Drive (amplifier) .................................................. ...... MADHT1107B01
1 x Motion Control Module .................................................. .....FPG-PN8AN-ND
1 x PLC CONTROL LOGIC 16 IN 16 OUT 24V ..............................FPG-C32T2H

In all cases you will need power supplies, cables, and other accessories.

Given your design speeds, I think it's going to require a lot more than a 400W motor to control the rotary axis. Notice I said ''control'' not turn. It all has to do with accelerating the mass, a 400W motor would eventually be able to get the machine up to speed but the acceleration time would be unreasonably long, and any subsequent speed changes would also be unreasonably long. A motor in the 3KW range might be more appropriate, but this is just a bit of a guess at this point. It really depends on how fast the speed needs to change.

Which leads to the system power requirement. What do you have for available power to run this machine? I have not seen any servos in the greater than 1.5 KW range that don't require 3 phase power.

For the rotary axis, it might be possible to make a ''servo'' system using an encoder equipped standard 3 phase motor and a sensorless vector VFD to drive it. These are available to operate on single phase up to 3 HP (2.2KW). These are off-the-shelf parts, and readily available. Not quite a purpose built servo, but pretty close, relatively inexpensive, and can be controlled with a motion controller. It's amazing what can be done with modern electronics. Should be more than accurate enough for this application, micron accuracy is not needed.

The servo that drives the linear slide can be much smaller, so the 400W or less should work, not sure about the paint pump yet.

There are a lot of servo motors/drive available, we'll pick the best for this application later.

I am still going to recommend a Galil motion controller for this project. They are motion controllers and PLCs in one package. They have plenty of inputs and outputs (up to 80) for your application, including a provision for temperature sensing inputs.and the programming software is free. And I have 20 years of experience with them so I can help you with the programming.

If it is possible, I really want to stick with rotating the shaft rather than the work-piece. Rotating the work-piece will be an issue on future design phases. Once you help me size and pick all the parts, the calculus to program the motion will not be a problem.

The rotating shaft design can be built but it's not going to be cheap. But given the parts you have been looking at, it seems that you have an adequate budget for this project. As far as the programming, that is really not a problem. It's just a matter of syncing all the moving parts together programmatically and then telling them what to do.

The paint dispenser will have its own motor. The Leader in Meter Mix Dispense Systems and Valves for Adhesives, Sealants and Lubricants but didn't check the price yet. If too expensive, do you think we can design one easily? The dispensers on the link are more for viscous materials but I believe that I can use it for the paint.

The ViscoTech servo dispenser pumps that I use for one of my designs are about $7000 each, these are about 10 inches long and about 1.5 inches square. These are low volume, progressing cavity, very accurate dispenser pumps. But for paint I think a small servo driven gear pump or small progressing cavity pump would be a good answer. Maybe an electric automotive oil pump would work as a basis for a servo pump, they are designed for low shear. I have never tried to pump paint with a gear pump, but I have done so with a progressing cavity pump. There are other pump types that are designed to pump paint, like dual diaphragm pumps. These are available in both air and mechanical drive.

To dispense paint at 300 RPM at the end of a 1.5 foot radius is going to require a pretty high volume of paint to get coverage. That is over 2800 ft/min surface speed, over 30 MPH! Imagine driving a car down the road and trying to apply paint to the road at 30 MPH, that's a lot of paint The trucks that apply the stripes to the road only travel 10 - 20 MPH or so when painting and they have huge pumps.

You are exactly who I want to talk and who can help me. Please let me know if there is any more info I can give you to decide what parts are needed for this weird idea.

Thanks.

Being semi-retired gives me the opportunity to only take on the weird and fun projects. I have made a career out of designing and building stuff that nobody else would touch or ''couldn't be done''. Sometimes you just have to think outside the box a bit.

It's time to get the mechanical design nailed down. This has to be done before we can spec the motors and control systems. Right now you have an operating concept, but no concrete design.
Just a few random questions: What is the machine frame going to look like? How are you planning on hanging onto the rotating shaft? Is on-the-fly height adjustment needed? Right now we have no idea what the linear slide system is going to look like, but I'm thinking a single arm with a rack & pinion coaxial drive. What does the paint nozzle(s) look like and how does it attach to the linear slide? What is the weight of all of these parts?

The other question that comes to mind is what tools and equipment do you have available to you? There is going to be a fair amount of fabrication and machining involved in this project. Also, where are you located? I'm in Sandy, Or. about 25 miles east of Portland, Or

[ethcnc]

Thanks, Jim.

Well that adds whole other layer of complexity, I was thinking in terms of maybe 1 to 10 RPM. At 30 to 300 RPM the balance of the rotating system is going to be critical or it will tear itself apart. This is going to require a servo controlled counterweight to keep the rotating mass in balance as the linear slide moves out. This might also be done mechanically, need to think about that a bit. The speed also adds a number of safety issues to the design. The design will have to be such that as much of the rotating mass will be as close to the center of rotation as possible, and anything that sticks out beyond the center has to be a light as possible.

I was thinking of worm gear but I'm not going to have the luxury of trial and error or flexibility as I can with the electronics. If all fails or gets too complicated I can turn my face that way.

It is not exactly 30 to 300 but when I apply this process manually it takes 3 to 4 seconds. I apply it on a 14" diameter plate. I rotate my hand to 3 times with a wider width spiral (nozzle is almost 3/4" if not mistaken) to cover the plate pouring the paint in spirals. sometimes if I miss a space I return and fill that but the max is 4 seconds. So I thought if I improve this manual process 3 time it is going to me 180 RPM if I keep it at least same, it is 60 RPM. That is why I said 30 to 300 but the vertical is rotating with almost a constant velocity or with a constant velocity depending on the outcome. The reason I said 30 to 300 is because I wanted to try at high speed and low speed but not planing to increase the speed from 30 to 300. As I mentioned above, if I can improve the manual process by some time I will be glad and that is the reason I wanted to try this prototype. I think now I have given you a clear idea of what I am trying to do. how to draw an Archimedean spiral

What is the machine frame going to look like?

It can be mounted like milling machine and the plates will pass under it on a conveyor belt but it can also be mounted on the side of the conveyor belt ( cantilevered) and apply the paint.

How are you planning on hanging onto the rotating shaft?

if it is a cantilever and with a bearing at the end of the cantilever and the vertical shaft will be inserted in the bearing also the motor mount will be at the end of the cantilever.

Is on-the-fly height adjustment needed?

No, the vertical shaft rotates with constant angular velocity (I wanted to increment for different tests).

What does the paint nozzle(s) look like and how does it attach to the linear slide?

Here is where the xy table becomes convenient. I was thinking of using a hollow shaft to pass the paint hose through the shaft and locate the paint bucket above the cantilever and use swivel elbow joint.
Swivel Joints (Swing Joints) from CSE - Industrial Products Group

What is the weight of all of these parts?

If your question is about the weight that can interfere with the motor, then I would say not more than 20 pounds if we are using aluminum.

I can get tools but not sure what type of tools. If your question is electrical, then I say none.

I am located in Atlanta, GA.

Please let me know if you want me to answer the above questions more clearly or if you have any more questions or suggestions.

Thanks

[Jim Dawson]

OK, I think I have a pretty good picture in my head now of the overall system layout. The reason I was asking about your location and available tools is that there is going to be some needed machining and fabrication. I have a fully equipped machine and fabrication shop but I am too far away from you to be of much practical help in that regard, at least for the big stuff like the frame.

I have attached a quick concept sketch pdf to see if we are on the same page.

Here are my thoughts on the system:

The goal is to simplify the system to the degree possible, reduce the rotating mass, and to eliminate the need for a very expensive slip ring.

The nozzle is going to need a valve incorporated into nozzle to control the paint flow or else you will have a mess. That valve could be air activated and using a dual rotary joint that could have air and paint running through the same device. Probably going to have to invent a valve that will work, I'm not aware of any available device that would do the job.

The linear slide and counterweight can be driven coaxially from an external servo motor. A simple rack & pinion arrangement will work fine for that. The counterweight shaft and nozzle shaft would move opposite each other to maintain balance.

In order to make the coaxial pinion work, the pinion servo would be synced to the main shaft rotation, then by slowing down or speeding up the pinion servo relative to the main shaft the linear shaft would extend or retract at a controlled rate commanded by the controller.

The main frame is made out of 4 inch, 1/4 wall square steel tubing. The main shaft is 2 or 2 1/4 inch x 3/8 inch wall DOM steel tubing.

The main shaft will be supported in mounted bearings that are attached to the main frame.

The main drive motor is an encoder equipped, 1200 RPM, 3 phase induction motor, driven by a sensorless vector VFD, HP to be determined later. Timing belt drive at a 4:1 ratio, that will allow the proper operating range.

The pinion motor is a 750W servo, just to insure holding torque at higher RPMs Timing belt drive at about a 6:1 ratio.

There are still a lot of details that need to be worked out, but it looks like this is completely doable.

[ethcnc]

Jim,

Ok, lets se how the Galil motion controller works and I think you are winning on the xy table because once I grasp the idea I can go for the more complicated one.
Can we start with sizing the motors?

Thanks

[ethcnc]

Sorry I replied before seeing your last post.