Somewhere on my z axis I must have a loose Dir wire.

diycnc-discuss
gplus
(George Allen) #1

Somewhere on my z axis I must have a loose Dir wire. When I try jogging my Z axis the motor cannot decide which way to go. Sometimes up, sometimes down. Doesn’t that sound like the issue?

(Paul Frederick) #2

Check your motor lead wires. A loose one of those can do the same thing. Don’t ask me how I know this.

(George Allen) #3

Yes. I just re wired the z axis and it’s working now. Thanks

(Paul Frederick) #4

@George_Allen so was it a motor lead? What you described sure sounded like when a motor lead got loose on me. Though then I was sure it was something else myself. At the time I just knew all of those leads were good. They weren’t and me knowing they were didn’t improve matters any either. But yeah one loose coil wire on a stepper motor can make it act possessed. It’ll run choppy, and randomly the wrong way. I was convinced I had some major control signal issue myself. Or even a faulty motor drive. Nope was just a loose coil wire.

(George Allen) #5

@Paul_Frederick Exactly. However, after I got the motors and everything working I ran a job and made a stupid mistake. I ran the bit too deep into the workpiece and the spindle got hung up. There was some smoke, it took too longer than it should have to turn off my spindle. and now I’m headed to the shop and see if there’s damage to the spindle, motor or electronics. What a bummer.

(Paul Frederick) #6

@George_Allen what you described did sound exactly like what a loose motor wire acts like. When it happened to me I thought it was something wrong with the logic too. It really does look like something more complicated than what a single loose wire can do. But now you’ve seen it so you’ll know what it is in the future. When I run jobs on my machine I’m right there babysitting it so if anything happens I shut it down real fast. I also simulate jobs before I run them for real. That way if there’s anything really weird I hopefully catch it in sim before for real. Because I’m green when it comes to making G Code myself.

(George Allen) #7

@Paul_Frederick After looking at things, my spindle seems okay, and the motors all power up. But for some reason my Smooth Stepper stopped communicating with the computer. It says it’s no longer on IP 10.9.9.9 but on 10.9.9.2 and I cannot get it back. The configurator won’t run b/c it says there’s a “bootP” that’s being returned. It says it’s in configuration mode and to remove a jumper, but there’s no jumper on those pins.

(Paul Frederick) #8

@George_Allen a buggy Mach setup? Say it isn’t so! http://linuxcnc.org/

(George Allen) #9

@Paul_Frederick It was my WiFi adapter. It kicked my ESS off its IP address. All good now

(George Allen) #10

I think I’m gonna get gecko drives. Whenever feeds get just above normal rates my y -axis messes up.

(Paul Frederick) #11

@George_Allen what kind of a power supply are you using? Gecko drives are good but they can’t fix inadequate power. Have you checked the current draw of your drives you’re using now? Over current actually makes stepper motors weaker. It is a strange thing. But if stepper motors had feet they’d be tripping over them with too much current applied. How you test phase current is a bit of a trick with microstepping drives. Power your machine down and disconnect one phase and put a current meter (ammeter - amp meter) in series with the drive coil. Now power up and set your jog speed really really slow. Now tap jog and read the meter every time you stop. Look for the highest reading you can see. That’s the current you’re running at. Cycle through the whole step sequence a few times to make sure you’ve found the peak. Compare the value you’ve found with the rated current your motors are supposed to operate at. If what you’re seeing is higher reduce the current setting of your drive. If it is lower increase it. The closer you can get to your motor’s rated current the better. Do the same thing for each axis on your machine. Proper set current will net you the best performance and smoothest running. It can make a big difference depending on how far off you may be initially. Gecko drives use fixed resistors for current sense. Which would be a bit of a pain to setup. My TB6600 drives have potentiometers on them which makes it easy. Optimized run voltage is a bit more involved to setup. It plays a huge factor in torque at speed though. You have to realize that stepper motors are inductors and as such are subject to back Electro Motive Force (EMF). Which means when you stop supplying current to a stepper motor the magnetic field in it collapses and sends a voltage spike back into the circuitry. Which can put you into an over voltage situation. Read you can let the magic smoke out of your motor drives. But again the closer you can get to your drives maximum voltage without exceeding it the better for torque at speed. That’s due to another phenomena associated with inductance called reluctance. Which is an inductors characteristic to reject changes in current flowing through it. Precisely what we want to do to step motors. We overcome this reluctance using higher voltage potentials. But drives can only handle so much voltage before they blow up. So it is tricky. Often drives are rated absolute maximum with no headroom for back EMF. So your supply value cannot be at the drive’s maximum voltage handling value to allow for the return spikes. Still as high as you can get away with going is best. You’d be surprised what even a couple more volts can do for stepper motors at speed. I know I was. But let’s say your drive handles 32V and you’re running at 24V. Going up to 28V will net you noticeably more torque. It can easily be the difference between it doesn’t work at some speed and it works fine. Adding some bulk capacitors to your supply rail might help you out too. Just putting caps at each of your drives voltage input pins could do something for you. Something is the 1,500-2,200 uF range. Though Mariss does make nice drives. He rates his drives at operating voltage not max. The last time I chatted with him he told me his parts can handle 300V peak. He has to deal with a lot of idiots using his products. Poor guy. But yeah if you get Geckos you’re going to need a new PSU to take full advantage of them. They can really run at 80V input. They should run at that for full performance. So there’s that hidden cost to consider in your decision.

(George Allen) #12

@Paul_Frederick okay, I’ll check before I buy. Thanks for all the info. I haven’t read it all yet.

(George Allen) #13

@Paul_Frederick well, before I decide on new drivers, I’m getting a new dust collection system in. I’m not crazy about the space it’s going to take up, but for the cost it should be good. It’s the Harbor Freight 2HP dust collector. It has gotten good reviews, mainly due to the cost: $209, plus I can use the 20% off coupon. The rating says it can move about 1550 cfm (I think) of air, which is only a bit below the $320+ shop fox. But it’s going to make my limited space even more cramped. Yesterday, I reinforced the beams underneath the table with 2 additional beams of extruded aluminum and today I “dadoed” (if that’s a verb) 2 dados into my MDF spoil board to house some t-tracks I bought to clamp down my workpiece to the table. I also had to tighten some screws and, I put thread lock on them to try to keep them in place. I need to recheck how level the surface is, but I’m definitely thankful that no real damage occurred during the last crash, and I was able to reconnect my controller to the computer. I really need some kind of enclosure for my electronics and a more accessible e-stop. When I get everything in optimal working order, I’d like to see how well the machine can make faux columns and carvings to go around panels. I have a collection of router bits that I’ve been practicing with and I may try doing some Wainscoting. I never had heard the term until about two or three weeks ago. Interesting what you run across on Pinterest.

(Paul Frederick) #14

@George_Allen I ran across some of my own pictures on Pinterest the other day. I didn’t even upload them there either. Apparently some of the sliding grinder tool rests I’ve made are pretty popular on the net though.

(George Allen) #15

@Paul_Frederick Just finished reading your full message. Yes, if I get the gecko drives I’ll likely need an unregulated toroidal PSU, but I’m going to check the current draw on the motors today. I’ve also got to check to see if I have the controller settings in the software consistent with the drivers. Currently, I’m using those TB6600 drivers for 3 of the 4 motors. I don’t have a potentiometer for the drivers, just those DIP switches. But, I’ve got to figure out why it cannot move faster than it is. Right now, even at the speed it is running, the motors are not smooth. There sounds like a grinding or stuttering when it’s running. It doesn’t make that sound when I’m jogging the axis though. Also, I’m wary of using any small endmill or bit on my large machine. I tried using a 1/8” ball nose yesterday and it snapped right off. I use that same bit on my 3020 in the same type of wood and I’ve never had any problems. I will attempt to upload a video of the running problem.

(Paul Frederick) #16

@George_Allen the way you describe it it sure sounds like over current to me. Are your motors running hot? Like really hot. That is another symptom of over current. Stepper motors are supposed to get hot, burning hot is another thing. But if you’re jogging OK and it only happens when you execute code it could be a software issue. Mach’s trajectory planner might just be awful. I don’t know. Or your Smooth Stepper might be losing something in translation? That’s going to be hard to troubleshoot. Past my pay grade. You might try lowering your acceleration and seeing if it smooths out some? Acceleration is the ramp up speed. High accel is nice, but it can be jerky too.

(George Allen) #17

When I tested the current, it was only drawing up to.98 A at its peak. I did adjust my Z-axis motor a little, which seemed to make a little difference. Because the amps are so low, I may need an unregulated PSU to get the power up. I will check the PSUs to see if they have potentiometers to adjust. But no, the motors don’t seem to be running hot. I’m also going to see if I may need to adjust the buffer settings on Mach, if the planning programming might be the issue.

(Paul Frederick) #18

@George_Allen an amp sounds pretty low to me. That was the highest reading you could get? You saw it fluctuate as you stepped and stopped right? Each microstep is a bit different current. One is going to be the max current. You have to manage to land on that one max current microstep to get a good reading. Though going around a few times should give you a fair idea of what current range you’re running.

Now you have to pin down where the deficiency lies. Is it the PSU or the drive? You can load test your PSU and see if it is capable of delivering the current you need. Light bulbs can make good loads. A plain old incandescent light bulb can burn up quite a few Watts. I’ve used light bulbs in the past to load test PSUs. Once you’ve cleared the PSU we can see what we can do with your drives to spiff them up a bit.

You have TB6600 chip drives right? If yes there may be some hope. I have V 1.2 TB6600s myself They look like this http://diyprojects.eu/wp-content/uploads/2016/02/BL-TB6600-v1.2-green-pcb-open-case-front-side.jpg
Well that’s a picture of one with the heatsink removed so you can see all of the components on the board. The blue box is the trimmer pot for adjusting drive current.

(George Allen) #19

@Paul_Frederick I think I may have gotten the issue mostly corrected. I think I had the acceleration too high and it was causing jerks

(George Allen) #20

@Paul_Frederick It seemed to be working pretty well yesterday. It may still be a little off; I snapped 3, 1/8” endmills yesterday. I definitely had the acceleration too high, and I may still need to tweak it a bit. I may try to get down there today, though it’s really cold.