[mhackney]

I just picked up some small toggle clamps to try. I made a new fixture this weekend that has enough space around the stock to place the toggles. I'll try that first. KISS is the way to go.

cheers,
Michael

[Geof]

....I might even be able to design the vacuum fixture plate to include vacuum clamps for the sheet stock....cheers,
Michael

That is a good idea, a perimeter clamp held down by vacuum. Depending on the size of your machine you could get a large area under vacuum and get considerable clamping force. Also you could have dowel pins to locate the perimeter clamp so it could not slide sideways.

Also have dowel pins on your vacuum clamp washers to go through some of the holes in the parts. This way neither part nor clamp can slide or rotate.

I hope I can remember this idea it is worth stealing.

[mhackney]

Ray, the new motor arrived today and exhibits exactly the same problems with hesitation and direction reversal. So, I sat down and made up a chart to implement your test procedure. Here is what I did:

Firstly, I tested this configuration with 2 different drivers on 2 different ports on my BOB to eliminate a fault in driver/BOB/wiring. Results were the same on both sets. I placed a piece of red tape on the motor spindle as a flag so I could verify rotation direction.

So...

My wiring goes as follows:

from driver:
A+ is black
A- is green
B+ is red
B- is white

from motor (see spec sheet a page or 2 back in this thread)
phase 1 coil A+: blue
phase 1 coil A-: red
phase 2 coil A+: yellow
phase 2 coil A-: green

phase 1 coil B+: brown
phase 1 coil B-: black
phase 2 coil B+: orange
phase 2 coil B-: white

so, plugging those colors into your 3 configuration test:

Config 1:
driver A+ (black) to phase 1 coil A+ (blue), driver A- (green) to phase 1 coil A- (red)
driver B+ (red) to phase 2 coil A+ (yellow), driver B- (white) to phase 2 coil A- (green)

Config 2:
driver A+ (black) to phase 1 coil B+ (brown), driver A- (green) to phase 1 coil B- (black)
driver B+ (red) to phase 2 coil A+ (yellow), driver B- (white) to phase 2 coil A- (green)

Config 3:
driver A+ (black) to phase 1 coil B+ (brown), driver A- (green) to phase 1 coil B- (black)
driver B+ (red) to phase 2 coil B+ (orange), driver B- (white) to phase 2 coil B- (white)

and here are the results:
config 1: Motor does not work properly. It sporadically reverses direction if I jog-stop-jog-etc

config 2: Motor seems fine.

config 3: Motor does not work properly. It also sporadically reverses direction if I jog-stop-jog-etc

So, I am not sure how to diagnose this except that things are amiss! Help!

cheers,
Michael

[SCzEngrgGroup]

Michael,

I don't know exactly which motor you have, but assuming the color coding on Keling motors is standardized, you're not connecting the coils correctly. The results would seem to confirm that. Your first and third combinations appear to be connecting both coils of one phase, and no coil from the other phase.

I think the terminology is confusing as well, since you have A+/A- and B+/B- for both half-coils and driver connections. Let's leave A+/A-/B+/B- for the driver connections, and refer to the two half coils of each phase as coil1 and coil2.

Your phases are as follows:

Phase 1 Consists of:

Coil 1+=Blue, 1-=Red
Coil 2+=Yellow, 2-=Green

Phase 2 Consists of:

Coil 1+=Brown, 1-=Black
Coil 2+=Orange, 2-=White

Your normal bipolar parallel connection should be:

Driver A+ to Blue and Yellow
Driver A- to Red and Green

Driver B+ to Brown and Orange
Driver B- to Black and White

For this test, you want to try these combinations:

Driver A+ to Blue
Driver A- to Yellow
Driver B+ to Brown
Driver B- to Black

Driver A+ to Red
Driver A- to Green
Driver B+ to Brown
Driver B- to Black

Driver A+ to Red
Driver A- to Green
Driver B+ to Orange
Driver B- to White

Give that a try, and I think you'll get a much better result.

Regards,
Ray L.

[mhackney]

Thanks Ray. So, here is the motor spec sheet again:


S0, looking at the bipolar parallel connection, it shows:

blue & yellow to A+
red & green to A-
brown & orange to B+
black & white to B-

it appears that the red and yellow color labels are swapped on the drawing. That means I had both motors wired improperly. I am going to go change those now and see what happens. Be back in a few...

thanks!
Michael

[mhackney]

Well, I used the corrected wiring and it still has a stutter/stall problem. I only tried 1 of the 2 steppers (the one I received today). I will do the individual tests next.

regards,
Michael

[SCzEngrgGroup]

Thanks Ray. So, here is the motor spec sheet again:


S0, looking at the bipolar parallel connection, it shows:

blue & yellow to A+
red & green to A-
brown & orange to B+
black & white to B-

it appears that the red and yellow color labels are swapped on the drawing. That means I had both motors wired improperly. I am going to go change those now and see what happens. Be back in a few...

thanks!
Michael

Michael,

I'm sorry, my post is wrong, their drawing is right. No matter how many times you read your own words, you see what you want, rather than what's there.... I have corrected my previous post.

The two coils on the left of the Keling drawing comprise one phase, the two along the bottom comprise the other. Any valid driver connection must have either ONE or BOTH of the coils on the left connected to one phase of the driver, and ONE or BOTH of the coils on the bottom connected to the other phase of the driver. When connecting both coils of any phase, you must always respect the polarities, exactly as you would when connecting multiple batteries. So, if connecting them parallel, always connect the two + wires to each other, and the two - wires to each other. If connecting them series, then make sure you have the + of one connected to the - of the other.

How you connect each phase to the driver is not really important. You can pick either motor phase and connect it to either driver phase, in either polarity, then just connect up the remaining motor phase to the remaining driver phase in either polarity. So, you can connect the controller A+ to the phase 1+ wire, and the controller A- wire to the motor phase1 - wire. Or, you can connect the controller A+ wire to the phase1 - wire, and the controller A- wire to the motor phase1 + wire. Do the same with the controller B wires. The worst thing that will happen is the motor will turn the opposite direction to what you want, in which case, simply reverse ANY pair of controller phase wires - swap connections to A+ and A-, or B+ and B-. The motor will then spin the other direction.

Regards,
Ray L.

[mhackney]

So, as you pointed out (of course I was testing at the time!) the Keling color coding is correct. The test with the reversed red/yellow wired was horrible on all 3 tests. So, I started from scratch. I removed both motors and had them sitting on the bench in front of me, wire nuts, and a deep breath. I went through and wired the original motor up EXACTLY the same way I did on Sunday when I discovered the problem. Lo, it worked fine. What? So I switched the wiring over to the new motor and lo, it worked find too! I have checked the wiring against photos I took earlier and they are the same. I timed the motor watching the Z DRO in Mach to see what its jog rate was expected and it was right on 150 IPM that I configured it. Is it possible that I had some motor tuning issue and was overdriving the motor on my earlier tests? I didn't think so since I was testing at 30 IPM in the tuning dialog earlier. I have no idea what's going on but both motors are wired as per the data sheet and both seem to perform as expected.

Odd this life sometimes,
Michael

[SCzEngrgGroup]

So, as you pointed out (of course I was testing at the time!) the Keling color coding is correct. The test with the reversed red/yellow wired was horrible on all 3 tests. So, I started from scratch. I removed both motors and had them sitting on the bench in front of me, wire nuts, and a deep breath. I went through and wired the original motor up EXACTLY the same way I did on Sunday when I discovered the problem. Lo, it worked fine. What? So I switched the wiring over to the new motor and lo, it worked find too! I have checked the wiring against photos I took earlier and they are the same. I timed the motor watching the Z DRO in Mach to see what its jog rate was expected and it was right on 150 IPM that I configured it. Is it possible that I had some motor tuning issue and was overdriving the motor on my earlier tests? I didn't think so since I was testing at 30 IPM in the tuning dialog earlier. I have no idea what's going on but both motors are wired as per the data sheet and both seem to perform as expected.

Odd this life sometimes,
Michael

Michael,

It's just the Gods of CNC laughing at you.... They do that a lot.... :-)

Regards,
Ray L.

[mhackney]

Yes, I guess I was overdo! Thanks for the help. I'll sleep on it and hook everything up on the mill tomorrow.

Cheers,
Michael

[Titaniumboy]

Michael,

Great thread on your G0704 CNC conversion. I've bought the Hoss plans and hope to start my conversion in the Fall.

I'm glad you've ironed out the problem you were having with the Z-axis motor/driver. I hate it when a persistent problem just goes away with no explanation.

Regarding G0704 mounting, did you ever consider using the Grizzly G8683Z Mini Mobile Base which is shown in the G0704 owners manual?

How is the Harbor Freight tool cabinet working out as the base for your G0602 10x22 lathe?

Titaniumboy

[mhackney]

Thanks Titaniumboy. I have mobile bases on my Delta Unisaw and 14" vert metal bandsaw. I really don't need the mobility with the mill or lathe. I would be concerned with the G0704 since the base is narrow with respect to the table width. The HF tool cabinet has been great and the height has even been a positive. Nice and stable and the drawers are great for holding all the tooling. I think I'm going to bolt my stock mill stand to the floor and get a narrow tool cabinet to sit to its right as a workstation with storage. I'll probably take the wheels off to make it less wobbly since I will use it like a desk. I've considered putting the mill on a tool cart too and still might. Maybe even the same wide HF that the lathe is on. My lathe and mill are back to back so that would work. I love having the storage.

Cheers,
Michael

[mhackney]

Just a quick update. Got my control box finished off this weekend and buttoned up. Also did the 3 bolt mod on the head. Wow, was that a PIA! I used a corded 1/2" hand drill with lots of torque. Clamped the head assembly in a big workbench vise. Once I got up to the larger drills (> 3/8") the bits really bit in and almost ripped the drill out of my hands. I stepped up in 1/64" increments and just used brute force. Lubricant helped a little.

I also finished off machinechick's head tramming aids. Went on to do a rough tune on the motors and started tramming. Should be done with that tonight. I was able to "cut air" and the movement looked and sounded great.

cheers,
Michael

[photomankc]

Michael,

Too Late now, but I had better success in hand drilling with only a pilot hole for the web and then going at it with the big bit. Much less grabby and less likely to go screwing into it. The break-through is what gets me every time though. I either rub things to death trying to ease up or go smashing through into stuff behind. I hate hand drilling through holes in heavy metal.

[mhackney]

I've been working on the home switches and wanted to use the same Optek OPB830 optical switches that Hoss used and designed the really nice aluminum enclosure to protect them. But, being a frugal Yankee, decided to build a circuit that basically hooks the opticals up so they mimic 3 NC mechanical switched wired in series. The output only requires 1 pin on your BOB and Mach just like the mechanical equivalent. (FYI, when Mach homes it does each access individually so there is no advantage to using a separate port for each axis).

While I was at it (and because the integrated circuits I used had unused components) I added in support for a forth switch that could be used for an ATC or maybe even a tach. The test here just shows 3 switches. Here's the prototype:



and a short video explaining how it works:

[ame=http://www.youtube.com/watch?v=cIpO-9sHauA]Optical Home Switch Ganging For Mach 3 CNN - YouTube[/ame]

So now that I know it works properly, I am going to layout a PCB and mill it on my X2. I will post the layout and parts list once it is cleaned up and ready to go.

cheers,
Michael

[Turbo442]

So now that I know it works properly, I am going to layout a PCB and mill it on my X2. I will post the layout and parts list once it is cleaned up and ready to go.

cheers,
Michael

Thanks Michael,

I am looking forward to your post.

Matt

[photomankc]

Very nice Michael, I see we think along the same lines. I've been contemplating something similar.

[mhackney]

Ok, first a little background:

My objective was to use Optek OPB830 optical switches as home switches wired to a single pin on the BOB and in Mach 3. This conserves pins since Mach3 homes each access in sequence (there is no advantage to having each home switch wired to its own pin). I also wanted to mimic the behavior of a mechanical NC switch circuit, i.e. there is normally a +5 volt on the line which drops to 0 volt when the switch is triggered. If used for combination home/limit this will trigger a limit if a wire to one of the switches breaks for instance (a little extra safety measure). Since there were extra unused components in the ICs used, I was able to add a 4th optical switch that I can either use for an ATC home switch or maybe as a spindle tach. It is highlighted in yellow in the circuit diagram, simply omit V04, R7, R8 and connector A.



I plan to put this circuit in a small box mounted on my mill's column with short leads to each of the optical switches. A 4 conductor cable carries power, ground and the signals to my control box. I've been using these 4 pin connectors from Radio Shack and will likely use them on the connections to the switches and to the control box. (you need the mating plug too)

At this point, I've wired up the circuit on an experimenter's board and tested the logic. It works great. Next I will build it on a protoboard and hook it up. I haven't drawn electronics in decades so please excuse me if the drawing sucks! Feedback welcome though.

cheers,
Michael

[H500]

I think the symbol for the opto is wrong. It shows a triac instead of a transistor.

[mhackney]

Thanks, no doubt! There were no 4 lead opt switches so I just used a symbol that looked close. I probably should just have left them completely off. The OPB830 has 4 leads color coded that correspond to the 1, 2, 3, 4 positions.

cheers,
Michael