[TroyO]

I'm moving, but having some trouble setting max current on my setup. Using the formula " Io = Vref * 1.667 " since i want 4A I can simply set my voltage (In "Hold" mode) to 2.4V, in theory. 2.4v / 1.667 = 4.0008 amps... close enough, LOL.

Two issues arise, though.... I can't adjust to 2.4v vref, I can only get 1.9v and the drivers error out as soon as they are told to step when I try that. I'm not sure about the following line though in the adjustment directions on the Massmind website "Carefully adjust the current setting to 30% of the motors rated capacity."

I assumed (and we all know the trouble that causes) that the formula, by using the * 1/3 factor was already taking care of the "30% of rated capacity" calculation. Am I maybe supposed to apply that as well?

(IE, 4A =2.4fullvref /3 = .8finalvref when in "idle"?)

Not that it will really matter until I get them hooked up to something anyway, but I am just fine tuning/setting up the motors as best I can in "air-cut" mode.

[James Newton]

Not sure why you are getting such a low voltage at vref... can you check the resistors installed for R4 (2k7) and R12 (470)? And your logic supply is +5V? What motor voltage are you running?

Honestly Troy, I don't set it by the calculation: I always just adjust it to the lowest setting, fire up, start the motors spinning, and then crank it up until they move reliably. There isn't any point in running them at the peak current if less will do. Then I let it run for a while to make sure nothing is overheating. A little drop of scented oil on the motor and heatsink is a good trick to monitor temperature over a long run: If you start smelling it, something is toasty.

[TroyO]

LOL, scented oil is classic.... great idea.

I'm not sure why either.... especially on all 3.

5V is coming from the PC supply and tests within a few hundreths of a volt.

I think I have the right resistors, I measured them and figured they were close enough?

I'll be danged if the color codes make sense to me though. (At my best I have trouble reading those even with a magnifying glass.)

[James Newton]

Yeah, those resistors look right. Can you check the label text on the potentiometer? It should say M502 on the top. If that's right then... I'm really confused. Can you post a shot of one of the drivers from right on top so I can see all the resistors and wiring?

More than that, do the motors turn ok when you reduce the vref?

Oh, and what motor voltage?

[TroyO]

I set my motor voltage to 42v. I can take a pic but I may have to bring in my "real" camera for it to be any use to you. (Blackberry camera sucks, LOL). Confirmed that the pots are labeled M502. The steppers seem to be working just fine, although I don't have a physical load to put on them at the moment they seem strong enough to do the job. (At .8V Vref). Just to verify, I measure from the GND testpoint to the Vref testpoint. I'm at a loss..... it seems OK anyway but I don't like stuff going on that doesn't add up, LOL.

[john-100]

Hi TroyO

after looking at the boards circuit diagram and the IC data sheet

with no step/clock (pin21) pulses being received by the stepper IC ,
the IC pin 25 should be switched to ground
this connects R13 a 2K2 resistor in parallel with VR1 the 5K potentiometer and R 12 ( 470 ohms )

the extra current that now flows through R4 (2K7) the maximum V ref will be 1.84 V

as soon as the IC receives step pulses pin 25 should go open circuit
and the maximum V ref will increase to 3.35 V

John

[TroyO]

John, that is a great analysis and jives precisely with the observed behavior! I ended up at about 1.6v in "Idle" which climbed to about 2.4V when in motion. Mystery solved.

[James Newton]

So I just heard back from Luc (who designed the driver) and he agrees with John's analysis. I think our documentation just needs to be updated in order to be a bit more clear. I'm working with Luc on that now.

[TroyO]

Heh, nothing like life to slow you down huh?

So, almost 2 years later LOL.....

https://www.youtube.com/watch?v=JeURJxPkzCc

That's using 3 of the Massmind stepper kits running a G0704, I could run at 180 IPM but have it detuned to 150 (Good question... is that enough headroom or should I maybe drop to 120?) The PC and all electronics are still in a single full size ATX case, no signs of heat problems.

Spindle control is a C41 from CNC4PC, using the stock G0704 controller.
ArizonaVideo's conversion kit.
X,Y Axis KL23H276-30-8A
Z Axis KL23H2100-50-4B

I'm not sure oak is the best of test materials, but it was handy. That 1" circle was as close as I could measure, so within .0005 if I trust my calipers (which I don't particularly, LOL.) Same fit on and off axis too so it's not egg shaped.
(Bwahaha..... I tried the shake reduction tool in Youtube and it made it like the bit was a wet noodle, bending and twisting all over.... I undid it but you may notice it looks like a mill from the twilight zone until they update it.)

[James Newton]

Nice! So glad to see it running!

I'd probably do something stupid like running it at 180 and then every time it skipped a step, I'd back it off 10 until the worked. But then I like to destroy parts and cause problems.

I'm curious what made you decide on our drivers over the KL electronics offered by Grizzly? Not that I'm complaining of course... LOL. But honestly, they are cost competitive (given the time required to solder up the kit) and appear to be pretty good drivers... I'm fishing here: Did you hear bad things? Are their other users not getting good speeds? Just like to support open source electronics?

[TroyO]

I definitely love open source stuff. I wasn't particularly aware of Grizzlys kit. I had looked at some other options (all in one kits, packages, cheapie eBay stuff) and mostly the 6064 looked like it would be just barely behind in power and specs but had the advantage of being a modular design plus I was looking to dive in to electronics more at the time as well.

I also got to work at my own pace and buy a piece here and there as finances allowed rather than trying to scramble to come up with big ticket money all at once. I feel like it would be easy to toss a driver and swap in a new one if something goes wrong. Especially since I know this setup inside and out.

This was pretty much a budget build, not sure what the Grizzly kit sells for but it would have to be pretty inexpensive. I'll have to add it up someday. My greatest fear is someday my wife will sell it for what I told her I paid for it...:.

[TroyO]

I'm grossly ballparking here, but I get about $1500 as my budget for conversion...

Power Supply $45
Drivers $100
Breakout $20
Steppers $180
PC Free
LinuxCNC Free
Ballscrew/Mounts kit $650
Gear drive (Metal gear swap) $40
DC Motor PWM Controller $60
Wires and misc. crap for the rest of it.

Add in $1200 for the machine and a couple hundred hours hand scraping and hand fitting.... ;-P

There's some stuff I bought and didn't use, but they will just get recycled in to the next project. (4th axis? Hrrrrmm....) and if I were to do it over I'd use my $8 Vactrol pwm to motor control for the DC motor, it worked almost as well, LOL. I'd also probably scratch build a power supply, the switching supply wasn't necessary. I softened the blow of my one really big ticket item (the kit) by selling off $300 worth of ballscrews I wasn't going to need so I managed it all in budget friendly chunks.

[James Newton]

That really isn't bad for a quality mid sized result. Just don't ask the wife if she agrees! ,o)