Motor Wiring?

I have the PK296A2A SG7.2 with the G203V with the PDMX-134 and the PDMX-126. First off, my head is spinning on the half coil and unipolar wiring. Do you mean that to wired half coil is only to use 1/2 of the coil? I am not sure I fully understand this and how the motor would every function. Please forgive the lack of understanding, I am used to induction motor wiring where you parallel or series the windings but all the windings are always used. Oriental motors tech support on the other hand says to Unipolar wire that all six wires are supposed to be used, however, in the posts I have read a lot of them say that unipolar is half coil and only 4 of the 6 wires are used. My head hurts.....

Also since all the calculations are based on unipolar the stated current set resistor is to be 35.25 for 3A which leaves the option of 33.2k, 34.8k or 35.7k resistors. I just feel like I am missing something here. Sure I could just follow the rest of the setups posted and not fully understand but I am interested in fully understanding the wiring and why the discrepancy in the wiring information I am reading.

The best answer is this I found at CNCZone. It explains full and half coil but then what is unipolar?

Picture might explain a bit more
http://www.probotix.com/cnc_images/unipolar_ill_02.jpg

as I understand 8 wire motors have 4 separate winding and it is possible to use only two of them as a complete motor and have other two in some kind of "reserve". it gives less torque just as unipolar winding which got its name from unipolar way of controlling the motors.
Look for thread "selecting motors for mechmate" you might find some more info about it.

Danilo,

I have read most of them already and I am very familiar with the diagram. However my confusion came into when I called Oriental Motor to verify the unipolar wiring, the tech insisted that all 6 wires must be used and the center tap is the common and the other two are tied to the respectable phases (a or b) depending on which coil you are working on.

This was from Unipolar, Bipolar series, Bipolar parallel, choices for 2phase steppers thread. It has also been stated that due to the fact that the SG7.2 gearboxes cannot handle any more torque than a half coil configuration anyway, using full coil and the higher voltage needed just wastes more heat. The half coil configuration due to the much lower inductance needs a lot lower voltage. Hope this helps a bit. :)

Finally painting, ready for wiring

looking good (but you seem to have strange apparitions floating around your garage :) )

Alan,
I think that might be the censors. :)

Perhaps the "ghost" in the machine.

I was trying to keep the focus on the machine and not the junk I have cluttered around.

Pete,

Sorry to be so late to respond.

The current limiting resistor is not critical. You can use any standard value close to 35K. I normally use 30K or 33K.

Look at the chart posted by Danilo on May 15th. The Unipolar chart shows the wires for your motors. The A coil has three wires, Black, Yellow and Green. The B coil has three wires, Red, White and Blue. When wiring the motor half-coil, for the A coil, you can use the Black/Yellow wires OR you can use the Yellow/Green wires. For the B coil, you can use the Red/White wires OR you can use the White/Blue wires.

Bipolar means that each coil has two poles. The stepper driver has eight transistors that switch the polarity of the windings. That's why you only use four of the wires.

(Oriental Motor's unipolar drivers use an entirely different scheme. I have a bunch of them sized to drive the PK268 size motors. Because the drivers are different, you just have to trust Gecko that they are doing the best thing possible to drive your motors - they are. I've run some significant tests between the G203v and the Oriental Motor unipolar driver for the PK268-02A motor. The Gecko drives the motor much faster and much harder.)

Wiring the motors half-coil will give you 70% of the motor's torque. Because the motor has a gearbox, 70% is MORE torque than the gearbox can handle, so you get maximum speed, snappy performance and enough torque by wiring the motor half-coil.

Use a 35VDC to 45VDC power supply that has at least 20,000 uF of capacitance (30,000 would be better). The harder the motors are working, the more current they pull. The more current they pull, the more they drain the capacitors between charges. Having large capacitors means that the motors will always see almost the full voltage all the time. Smaller capacitors means that the voltage will droop, which will affect the top speed.

Above all, don't sweat the electronics. A CNC machine is not a constant motion machine. The speeds vary. The load varies. More important, the density of wood varies and the sharpness of your tooling changes - while you are cutting. You'll probably find that setting your speed and depth to 50% or 75% of maximum will give you better over-all throughput than trying to run at 100%.

I learned that lesson the hard way with my Shopbot PRT-Alpha. I drove it at maximum speed and maximum depth, and learned that the "Alpha" feature worked because I had divots in the work when an axis bogged down and then recovered. When I slowed things down, by 25%, the machine never went into "Alpha" mode. The divots disappeared and I realized that I had wasted thousands of dollars buying a machine with the "Alpha" feature when the Shopbot software did not compensate for the speed variations of the motors.

Mike,
I think I have read every article on the web but your was the easiest to understand. I understood the wiring but did not understand the stepper and how it works. It is nice to fully understand every component. I have been reading about Mach3 and understanding the interface with the 126 breakout board. I am adding a second parallel port to handle the additional inputs. Have done a lot of reading to find one that works with Mach3. Been busy building a fort for the boys so I have not much time to post.

Some of my previous posts I had questions but never told the solutions that I used. I will be trying to post pictures of what I have done and progress.

I used C7 x 9.8 channel and the machine is all bolt together. The C channel had a U profile to it and I needed to level out the X rails. I mounted the x rails with M12 bolts for extra holding power to compensate for the bowing in the x rail from machining so much of the 2.5" side down to mount the v rail. This was only needed on the X and not the Y. The first pic shows the side view with the M5x0.8 set screw visible in front of the mounting bolt. The second is a top view. The third is a view the the v rail mounted the the angle iron, the pic is deceiving but the rail mounts on the rail on the vertical edge and does not set on any part of the radius.

nice job

Pete,
What was your reasoning in using the angle iron instead of the Aluminum channel for the prefab Vrail?

Heath, mostly price and availability. The aluminum would have been easier by far but the cost and lead time here was too high and too long.

Pete,
That is a nice solution to the "toe in" challenge that happens with some of the C channel available lately.

The "squaring" of the leg of that structural shape has been done a number of ways over the years.
Some of the solutions used that I can remember seeing here:
- Building the end gates of the assembly with adjustable tie-rod type adjustment to allow toeing in or out of the C channel to the top face gets in plane.
- using BONDO, (car body filler) on the top face, and sanded square
- shims used in the same place you put set screws to get the angle to set square.

Nice work.

Wiring and Mach setup almost complete
 

Have some minor wiring changes to do. Need to change the 8mm proximity home/limit switch for the Z axis to a 12mm. Seem to have some problem with the 8mm switch picking up interference when it is grounded (barrel of the proximity switch touching the steel). It does not do it with the 12mm switches though. All the steel is grounded by a connection wire and all wires are shielded. It is odd. I am using 24VDC pnp switches with the PDMX-126 board. One switch each for each axis for home/limit. I know most of the posts I read that it is not needed but I like the idea of having Mach3 take care of the racking by calling it at the start of a program if needed. Had a bit of trouble setting up Mach3 to see my other parallel port. Finally got it then had to find out why the motors were not all working. Ends up I did not have the pins set correct. Next it took me a while to get the A axis to run reverse. Finally found that as well. The only thing I need yet is to make the mount for the touch off probe used as a tool offset. That will be next week. Will start looking a dust collection next, I was looking at a JET DC-1200VX-CK1 mostly because of the 2 micron rating. Was looking at a thien separator build to use in conjunction.

Pete,
I had the same problem with the Z axis proximity on machine #28. The issue is that it touches the steel and is grounded on that machine. For some odd reason, the z axis proximity to the spindle can pick up the EMF in that area and cause issues with the home circuit.

Things to do / check that have fixed this issue in the past:
- use a doubled sided grommet (plasitc - non conductive) to isolate the z axis proximity from the frame
- change the debounce # in mach3 config to a higher number 100, 1000+
- (in some cases) add a isolation relay to the proximity circuit

Good luck

Sean,
I just changed the prox to a 12mm. I don't know why but the 12mm works fine but neither of the 8mm would work correctly.

More Pics
First picture is the machine itself
Second is the X stop modified to accommodate proximity switch
Third is the slaved axis stop to accommodate proximity switch
Fourth is the Y axis stop to accommodate proximity switch
Fifth is the electrical panel
Sixth is the inside of the the electrical panel (still needs straightened up)
Seventh is the PDMX breakout board and motherboard for the Gecko 203V

Still trying to work out the Velocity and Acceration for the X/A axis. Acts as if it is losing steps when I try to tune them. Works fine with V at 120 to 220 but will start missing above 220. These are inch settings. Have worked out test program. Still working on Fine tuning and setup for work surface, offsets from home, etc.

I have found that it does not matter what the speed there still is a little jitter to the axis x a and y. Z is fine. It is almost like it is fighting resistance somewhere. I have tried loosing tension on the motors, adding more. The power supply reads 35.8 at standstill and 35.4 and max speed with all the axis moving at once. I do not think it is a power problem. The resistor for the 203 is a 33K. I have the steps per in at (3055.775) and have measured the distance moved and it is precise. I don't know if it is maybe just the new gears and racks that need to just wear in or not. It sounds like the steppers are pulsing and once in a while it has an nice solid hum to them. However I can not feel it on the shaft or gear mechanically but can hear it audibly. Do I have to adjust the pot on the Gecko Drive or something else to smooth it out? Any help is appreciated.

Pete,
A few things to check:
- did you lubricate the racks and pinion?
- Loose tension on the motors, is better than tight when binding happens
- double check for parallel alignment on the x1 and x2 rails - these can cause a "trap" bind on the v-rail. The quick way to see this happen is slowly move the gantry by hand with motors disengaged and have a helper see if the gantry "rides up the v-groove" If so, this could create pinion tension where the misalignment occurs.
- check to see if rack is parallel to v-rail above - a misaligned rail-to-rack arrangement can cause bind

As far as Gecko tuning -
The G203 manual shows how to do this.
On a disengaged motor, set your motor RPM to that in the manual and rotate the potentiometer on the drive until you have a smooth hum. You should easily hear it in a quiet shop, if you have a helper - you can feel it with a hand on the motor while trimming.

It sure sounds like some thing in the electric domain but, what I have found using Mach2010 screenset is that with so large toolpath display any less capable computer will choke on it and set processor usage to 100%, it was my case until I upgraded to the new one.

Sean,
The racks are lubricated, I tried loosing the tension on the motors already. I will check again for parallelism. Occasionally you will hear the hum that I am used to hearing with a stepper. I am familiar with start up problems with machines however they are usually something I am a lot more familiar with.
Danilo,
The computer has a 3g cpu with 2g of ram. It is only a clean install. The only thing I can possible attribute to this is when you do a driver test on mach it starts out as pulsing too fast but after takeover it states System Stable and then System Excellent. Pulsing too fast does not return. However I do not like it stating it.

This is definitely a computer problem, I started by disabling my antivirus, then windows firewall, then read so more posts and disabled my wireless adapter. The jitter mostly went away. as happy as this may seem the problem did return. I might be searching for a different computer. I was using dell xeon workstation 3GHz with 2 gig ram. I might disable all the programs in msconfig startup but there are not many. I know there are a lot of processes running but don't know which to turn off.

get into BIOS and look for speedstep, its an automatic feature to lower the CPU frequency, and disable all power saving features. that might do the trick.
Also check the mode of LPT port. there are 3 (ECP EPP etc..) but I cant remember which one would be correct :)

Danilo,
Which mode am I looking for, my head hurts right now from all the troubleshooting. I know I read something about the this before. I know legacy support is not selected.

Will check later today. I will keep you posted.

http://www.machsupport.com/forum/index.php?topic=3060.0

I also checked and its ECP+ EPP set on mine, so try that if its not already set.