Re-Build: AcornSix Plasma/Drill/GasTorch Proposed Retrofit

[HAL9000]

This is an existing plasma cutting machine that is a few years old but never ran. When the control was initially powered on a message appeared informing the operator that a control software update was available. After some hesitation, the update was applied and that was the end of the control. The update crippled the control and there was no support. (The machine was imported directly from China)
As can be seen in the photos, the original control was built by somebody who had no idea about electrical systems. Consequently, I have been asked to get the machine going using a Centroid control. While I've been using Centroid controls on all sorts of machines since the mid 1990s I have never built a plasma machine or used any of the components required to do it. So, I am looking for some guidance here, please.

The idea is to re-use as many of the existing servo components as possible where they are compatible with the AcornSix. The machine is fairly standard in its configuration: Two servo motors (Yaskawa SGM7J-08AFC6S) on the X axis, One Yaskawa SGM7J-08AFC6S on the Y Axis. These motors are only 750W running through a 40:1 planetary gearbox onto a 6mm pitch rack. At 3,000 rpm, this should give the machine a feedrate of around 12.6m/min.

The machine also has three vertical ballscrews. Each one is for plasma torch, oxy/actyl torch and a drilling head.
The plasma torch ballscrew is 5mm pitch and direct coupled to a Panasonic AC Servo Motor (Model: MHMF042L1U2M) This is a 400W motor rated at 3,000rpm with 1.27N-m of torque.
The machine is using the Hypertherm Powermax 125 plasma unit.

The oxy/acetyl torch is on a 5mm pitch ballscrew but this is direct driven by a 24VDC motor. This is one problem to resolve. How to control the speed of the DC motor with the AcornSix probably via a DC variable speed drive of some sort. Maybe analog volt or 4~20mA? It may be easier to swap the DC motor to a stepper. The original control had a CAN bus system with some smarts that could control the speed of the motor with a manual over-ride pot.

The drill ballscrew is also a 5mm pitch direct coupled to a stepper motor TIANYA 450B coupled to a DM860H-C driver. This will probably be OK. The Delta MS300 inverter is 200V. The spindle motor is 400V so the inverter will need to be replaced.

So that I can get the Centroid components ordered, I would like to clarify what is actually the best approach on this project.
Some questions then:

1. Because the Yaskawa Servopack can use the quadrature step and direction signal communication protocol (AFAIAA) then would it be advisable to use the AcornSix RS422 Differential Drive Output and Scout Expansion Interface Board?

2. What would be the best way to control the oxy/aceyl ballscrew if the existing 24VDC is not used?

3. The table size of the machine is approx 5 x 2.5m
Are the existing 750W motors going to be big enough?

Proposed list of Centroid components:
1. #15136 AcornSix
2. #15149 THC TX
3. #15321 THC RX
4. #15130 AcornSix RS422 Differential Drive Output and Scout Expansion Interface Board
5. #15023 CNC NUC (panel mount)
6. #14612 SCOUT CNC Operator Control Panel/Pendant (maybe)

This is the Google Photo Album link:

Any insightful comments would be much appreciated!

Thank you!

[centroid467]

Hello and welcome to the forums! This seems like it will be an interesting build.


1. Yes, those Sigma7 models should be able to use differential quadrature signaling for positioning. You will be able to assign whether or not an axis uses quadrature or step/direction signals per axis from the Wizard. Additionally, using P/N: 15130 does not preclude the use of the open collector or 5V line driver outputs.


2. Should the oxy/acetylene torch need to be position controlled? If so, your best bet would be to switch it to a stepper or servo.

3. What is your concern with the size of the motors? Considering this is a plasma machine and the spindle is used for drilling, I can't imagine they would see any more load than the machine itself. Do you suspect they will not have enough torque for good acceleration?

[HAL9000]

Thanks for your reply centroid467

I don't think the gas torch needs to be position controlled as such but rather have some speed control. As it is, the speed is too fast when the motor is just powered with 24VDC via a relay, as in ON/OFF. Probably the easiest way to solve the problem is to use a stepper motor.

The concern I have with the axis motors is that they won't have the power to Accel/Decel quickly enough. On 3 axis mills with an 800 x 400mm table on linear guides, we've always used 1kW motors direct coupled to the ballscrew. These tables have nowhere near the mass of the gantry on this plasma machine that is powered by a couple of 400W motors. The cross rail that has negligible mass also has a 400W motor. While my idea of motor sizing is only empirical, this setup doesn't seem right. Anyway, we'll find out once it's running whether it works or not but it's annoying to have to do the job twice.

One other thing that isn't yet clear in my mind is how the vertical axes will be labelled and controlled. The plasma vertical axis will be Z but what should the drill and gas torch be called? How would they be moved manually with the Jog Panel? We would have X, Y, Z plus two more axes.

[centroid467]

Thanks for your reply centroid467

I don't think the gas torch needs to be position controlled as such but rather have some speed control. As it is, the speed is too fast when the motor is just powered with 24VDC via a relay, as in ON/OFF. Probably the easiest way to solve the problem is to use a stepper motor.

The concern I have with the axis motors is that they won't have the power to Accel/Decel quickly enough. On 3 axis mills with an 800 x 400mm table on linear guides, we've always used 1kW motors direct coupled to the ballscrew. These tables have nowhere near the mass of the gantry on this plasma machine that is powered by a couple of 400W motors. The cross rail that has negligible mass also has a 400W motor. While my idea of motor sizing is only empirical, this setup doesn't seem right. Anyway, we'll find out once it's running whether it works or not but it's annoying to have to do the job twice.

One other thing that isn't yet clear in my mind is how the vertical axes will be labelled and controlled. The plasma vertical axis will be Z but what should the drill and gas torch be called? How would they be moved manually with the Jog Panel? We would have X, Y, Z plus two more axes.

Yes, I agree that the easiest way to have better control over the gas torch will be to use a stepper motor.


I'm not an expert on this but you have a 40:1 reduction from the motor to the rack pinion, correct? So, the speed is divided by 40 and the torque is multiplied by 40. On the gantry itself, you have two motors each rated at 2.39 Nm. That implies there is about 160 Nm continuous on tap before the rack comes into play. I'd give it a try with the existing Yaskawas before considering going up in size.


From what I've seen, we use the Aux Keys on the jog panel for the 5th and higher axes. The jog panels have keys for four axes in the "Axis Motion Controls" area and up to twelve keys that can be assigned in the "Auxiliary Controls" area. The machine I use for a lot of my project testing has the B axis jog controls on AUX 1 and AUX 2.

In your case, I'd assign the vertical axis that will need to be manually positioned the least to the Aux Keys.

According to the ISO standard 841, your secondary axis parallel to Z should be W (or Z2) and the tertiary should be R (or Z3). This post has a great breakdown of that with helpful diagrams: viewtopic.php?t=3648. However, CNC12 does not support R or numbered axis labels so your best bet is to choose one of the remaining available from [A, B, C, U, V]. I'd choose U or V because A, B, and C should be reserved for rotary axes.

[HAL9000]

I'm not an expert on this but you have a 40:1 reduction from the motor to the rack pinion, correct? So, the speed is divided by 40 and the torque is multiplied by 40. On the gantry itself, you have two motors each rated at 2.39 Nm. That implies there is about 160 Nm continuous on tap before the rack comes into play. I'd give it a try with the existing Yaskawas before considering going up in size.


From what I've seen, we use the Aux Keys on the jog panel for the 5th and higher axes. The jog panels have keys for four axes in the "Axis Motion Controls" area and up to twelve keys that can be assigned in the "Auxiliary Controls" area. The machine I use for a lot of my project testing has the B axis jog controls on AUX 1 and AUX 2.

In your case, I'd assign the vertical axis that will need to be manually positioned the least to the Aux Keys.

According to the ISO standard 841, your secondary axis parallel to Z should be W (or Z2) and the tertiary should be R (or Z3). This post has a great breakdown of that with helpful diagrams: viewtopic.php?t=3648. However, CNC12 does not support R or numbered axis labels so your best bet is to choose one of the remaining available from [A, B, C, U, V]. I'd choose U or V because A, B, and C should be reserved for rotary axes.

Thanks for your useful comments centroid467. The torque numbers make sense so we'll give it a go. There's nothing to lose at this point.
I'll update this post as things progress.