Losing X driver intermittently

[cnc_works]

I hope I can be forgiven for this being a little long, but I'm sure most of you can empathize with the frustration of trying to track down a subtle problem that affects production. I have a Shopbot PRT96, about 2002 vintage. Upgraded to an Ascension box just before ShopBot's board hit the market and to a spindle a couple months later. For the most part the combination has been great with one major exception.

On a very unpredictable basis one of my X steppers just stops, the car just binds up and I begin losing steps. It is has always happened on a fairly fast long traverse in the X direction, usually over 200ipm, whether rapid or cut. I can tell the stepper has lost its drive because after it stops and I hit the spacebar to pause, I can move the dead end of the car freely while the other end is still locked by the active stepper. Once I recycle the control box, everything is back to normal until the next event. After it happened a couple of times on my offside X stepper, I replaced the Gecko driver and the problem seemed to go away. A few months later, it began happening with my near side X stepper. Then my work went to mostly smaller 3D work, during which the Bot ran faultlessly for six to eight hours at a time, in both and either the X and Y directions. Yesterday, I began to flatten my table and the offside X stepper began stopping again during the long X cut at about 220ipm. I never had this problem until I changed to the Ascension box and spindle, but, to be fair, I never ran my PRT over 200ipm when it was still on the SB controller.

It can happen whether I have my vacuum pickup on or off, though the wire and 2 1/2" hose for the vacuum are nowhere near any stepper cables. I have walked along the X axis and wiggled and pulled the connections on the X steppers while it was running with no affect. Both X and Y axis wires run in e-track guides with the power and stepper control wires running parallel. I'm running a 2ghz P4 with 512mb RAM and Mach3. I can rapid it back and fourth on the X axis at 400-500ipm and can't make it happen. Right now, the only common denominator I have is the fairly fast long traverse in the X direction.

So there it is, any ideas would be greatly appreciated. I am posting this in the Mach forum as well since my Bot is a bit of a hybred.

Donn

[srwtlc]

Donn,

Does this happen right at the start of the jog or move? Could you have your acceleration value set too high? Does your carriage move freely without any binding when the motors are dropped away from the rack.

Have you been using Mach3 at a different kernel speed and then it somehow got set back to the default? This one caught me a number of times after upgrading to the latest version. I ran it at 35000hz and it would get set back to 25000hz and I would all of a sudden loose steps big time.

Sounds like you've checked the cables, but check them again or give the wagos a new bite on the wire.

Could the spindle be causing it?

Just some thoughts to get the elimination process started.

[garyb]

Don put your hand on the gecko drives, is the one x driver warmer than the others? if so check the blue connectors, they slide up, look for over heat marks on the pins, if driver is hot check the resistor is good else you may have a driver shooting craps $134 for a 202

[richards]

Donn,
You wrote that the only cure is to power down the Ascension box and then power up. That makes me think that the G202 went into auto-shutdown to protect itself. Here is a quote from the G202 manual:

"When a short-circuit condition is encountered, the G202 goes into protective shutdown by turning off the winding outputs. The G202 outputs stay in protective shutdown until the power supply is turned off and discharged, then turned on again. Should this happen, please check the motor wiring and the cable going to the motor for errors."

Carefully check all wiring to see if anything has worked itself loose or is any wire strands are sticking outside of the connectors. Assuming that everything is still connected properly, with the power off, check the resistance of the A-coil and the B-coil on the offending motor and the A-coil and the B-coil on a motor that is working properly. The readings should be within 10%. Assuming that the motors check out, finally check the temperature of the G202. With stock motors, you're not pulling much current, but with hot weather, your heatsink may not be pulling the heat away from the G202 drivers.

The 200-ips jog/feed speed seems moderate to me.

[daryl_alberson]

Don, I have a 2004 PRT upgraded to the 4G and have experienced the exact same problem. I talked to Shopbot and they said the driver was getting too hot and shutting off. They advised me to put the control box on its side and leave the side panel off and put an external fan for circulation. It has worked so far but I have not ran the machine very hard since.

[cnc_works]

OK, now I have egg on my face for not doing more accurate troubleshooting before my last post.

I looked at the drivers as a source of the problem because the immediate symptoms seemed to be that of my earlier driver problem. This time though, I assumed instead of actually seeing the driver light on the controller go out, that it had done so. Further troubleshooting showed that the driver was continuing to hold without recycling the box. I unplugged the steppers and slowly pulled and pushed the car and noticed a kind of sporadic growling accompanied by a hitch in the movement on the outside problematic stepper. Not all the time and not even at the same spots.

I pulled the steppers and looked at the spur gear and below is what I found. Would wear like this actually cause the gear to hang in the rack and cause lost steps?

When I upgraded to the Ascension a year ago it was recommended to me to go to the 20 tooth spur gears which I did, but now I wonder if that was a good idea seeing the wear after just a year and probably less than an average of 10 or 20 hours a week of use. Are they going to wear faster because of less teeth engaged at any one time? Would 25 or 30 tooth spur gears be better? What about general smoothness in relation to the number of teeth?

My rack seemed to show a little wear, but nothing at all like the spur gear. How often do you guys lube your rack? And do you clean it every time?

Anyway, you guys, I appreciate the previous feedback and learn from it even if it was based on my faulty info.


3729

By the way, I'm pretty computer and graphics literate, but the time it took me to get this image down to 50k was ridiculous, even using the resizers and hints from the forum. My vote is a modest increase to 100k for forum photos.

Donn

[cnc_works]

Well, I didn't remember in time to edit the last post within the 30 minute timeframe, so here is a PS question. Any comments on why all of the wear seems on the stepper side of the spur gear? And that goes for both X steppers which showed similar wear patterns, though the nearside X wasn't binding up to cause lost steps.

Donn

Donn

[Brady Watson]

Donn,

Q) Are they going to wear faster because of less teeth engaged at any one time?

A) Yes. If you consider that you are putting down all of your stepper's power through the pinions, that power must be divided up by the number of teeth engaged in the rack at any one time. For example, let's say you are running a 300 Oz stepper & 20 tooth pinion. Let's also say that you have only ONE tooth FULLY engaged in the rack at any position of revolution. That means that 300 Oz" of torque is excerted on just ONE tooth. The smaller the gear, the more tendency for it to 'cog' because so much force is being applied to it with so few teeth engaged in the rack.

Q) Would 25 or 30 tooth spur gears be better?

A) It depends. If you have sufficient gear reduction on your motors (EG - 3.6 or 7.2:1 ratio), then yes. Changing to a higher tooth count will increase top speed & mechanical smoothness with a small reduction in step resolution. If you are running non-geared motors, a 25 or 30T pinion will render your CNC useless (as it applies to 1:1 PRT & PRT Alphas) in most cases. The steps will simply be too coarse. The mechanical smoothness gained from switching to a 25 or 30T pinion is well worth it if you have a geared motor. There are more teeth engaged in the rack, and therefore less torque per tooth, since it gets divided up. If you have 3 teeth engaged in the rack and are running a 300 Oz" motor, then you effectively have a max of 100 Oz" of torque being applied per tooth. It is easy to see in this example, the reduction in wear.

Q) What about general smoothness in relation to the number of teeth?

A) The more teeth you have engaged in the rack, the smoother motion & transfer of power will be. Imagine a bicycle chain going around two 50-tooth sprockets compared to going around two 7 tooth sprockets. The chain will not even stay on the 7-tooth setup.

Q)I'm pretty computer and graphics literate, but the time it took me to get this image down to 50k was ridiculous...

A) If you are graphics literate, than you no doubt use Adobe Photoshop Elements or full version. Just pull in the file you want to resize, goto File->Save for Web, resize and set quality. It will tell you what the size will be depending on the changes you've made.

Q) How often do you grease the pinions/rack?

A) Motors should be removed and pinions examined for wear every 1 to 4 months depending on the use of your tool. They are the only real consumable on the ShopBot. Pinions do NOT last forever...If you have problems with chatter or smoothness & haven't check or changed your pinions in more than 6 months of use, I guarantee that your pinions are shot. Despite what you may think, you cannot adequately inspect pinions looking upside-down at a dark/greasy rack/pinion. You have to clean off the pinion anyway...pull the motor. It takes 2 minutes with a 9/16" wrench on most tools. (Not talking about the Z.) Visual inspection of each rack should be conducted after ANY 'dirty' job where debris as worked it's way into the rack by static electricity or air turbulence. Teeth should be raked clean with a brass brush & fresh thin coating of grease applied. Either the black lithium grease supplied with your tool or white lithium spray grease should be applied as needed. If you fail to grease the rack & pinions, you will wear out the pinions in a matter of days or weeks. I have done this personally and learned to keep an eye on adequate lubrication for the drive system. Do yourself a favor & wipe off the excess grease from the front/flat area of the rack. The idea is to lube the teeth...we're not slathering on BBQ sauce on a side of beef here people...

After you have changed your pinions, the correct number of turns for PRTs & PRTAlphas' turnbuckle/spring is 3.5 turns from the point you just start to feel resistance from the spring.

-B

[cnc_works]

Grady, your willingness to spend the time for such a complete answer is greatly appreciated.

My PRT96 came with 25 tooth spur gears, so that is what I'm going back to. The Ascension box w/geckos supposedly gave me the higher resolution, so would 30 tooth gears be even better? I'm more concerned with smoothness than speed.

Donn

[Brady Watson]

25 or 30T will be fine on your machine. 30T for increased pinion life, mechanical smoothing & speed. 25T for higher resolution, greater pinion life than 20T. If you go with the 25T, then you can just use the same unit values as the 4G on your non-ShopBot box. Keep the 20T on the Z axis.

Although I full embrace the Sanford and Son philosophy, there is no 'G' in Brady.

-B

[cnc_works]

My own spur gears must be worn, 'cause I know I programmed a "B".

Now, on my Bot, is it normal for new spur gears to make a kind of occasional light "rattling" noise until they get broken in?

Donn

[Towersonline]

Looking at your photos tells me there is something else going on. The wear pattern seems to be on the left side of the gear. There is mushrooming in the in the bottom of the gear between the teeth. The metal is being squeezed out the side of the gear. I wonder if perhaps the gear is not perfectly aligned with the rack. I would also check the spring tension, it may be to tight. I have seen badly worn pinions but none worn quite like these. Brady my have some additional information he could add.

[cnc_works]

I definitely noticed that as well, Bill. I keep them tensioned to the classic 3.5 turns. My mind is having a hard time stretching around a way to do any re-alignment. The alignment all seems to be pretty well factory set.

One thing that I haven't seen covered is how tight the stepper mounting bolt should be. I tighten mine to the point that I feel the spring could just move it enough to keep the gear in contact with the rack.

One thing this whole event has pointed out to me is to watch the gear wear and to lubricate more often. I'm trying the motorcycle chain oil that Brady recommends now.

Donn

[conceptmachine]

Donn,
I recently replaced my pinions and noticed the same wear as you have.I agree on the rack posibly being mis aligned,what i plan on doing is to move the X or Y car which ever one you will do first to the max travel then loosen the bolts on the rack just enough to move it then slowly jog the car to the opposite end of the table stopping the pinion near each bolt and lightly snugging each bolt.After you have gone through the travel then go back and tighten each bolt.This hasn't been tried yet but i can't think of any other way to align the pinion to the rack since there is nothing square to reference from.
shawn

[conceptmachine]

Edit previous post:
You may want to back off of the 3.5 turns a little on the spring tensioner that may force the pinion a little too much while aligning the rack(may not be the case but just a thought)
shawn

[Brady Watson]

Donn et al,
The lube that you want is made by Cyclo as discussed in this (http://www.talkshopbot.com/forum/messages/28/21510.html?1181041856) thread. Make sure you shake it well. It comes out like whipped butter.

As far as tighness on the pivot bolt, you want it somewhat tight to prevent torsional movement of the mount. If you have the older angle iron motor mounts, you should see if you can get the 4G PRT mounts from ShopBot. They completely mount the motor and have bends in them to eliminate torsion. Make the pivot as tight as you can while still being able to swing the motor into postion & ensure that you are fully engaging the pinion to the rack.

Pinions with that much wear could be attributed to many factors. The number one factor is that they were simply run beyond their operational life, with or without adequate lubrication. They may also have not been full engaged in the entire width of the rack, or not squarely seated, as Bill suggests. You'll want to check that the new pinions are engaging squarely to the rack AND completely in the rack before you lube it. If it appears that the pinion is running on an angle to the face of the rack, you'll want to figure out what is going on & shim in between the motor if necessary. It is a tricky thing to figure out being that you need a very small machinist square or an extra pair of hands and a way to check pinion to rack distance from the front of the rack teeth and the pinion to the back of the rack teeth (checking parallel of square portion of pinion to square portion of rack).

Keep an eye on the new pinions & check them often until you can verify if you have an alignment problem or not.

-B

[Brady Watson]

Shawn,
The easiest way to check if the rack is parallel to the rails is to butt a tape or tail of a digital caliper against the front facing edge of the rack and take samples down the rail. You mention unbolting all the way down the rack...not everyone has bolts all the way down the rack & I believe all of the racks have VHB tape between them and the angle iron. Before opening up a can of worms, 1st measure parallel via the front face of the rack and the bottom edge of the rails. I doubt they are off that much, and if they are off a little here or there, the machine is designed to compensate for these small fluctuations. If your rack has VHB on it, you're going to have to pull the whole thing down to adjust it even a little bit...Be sure you've done all you can to diagnose if there is in fact a problem before you go pulling the rack down.

-B

[Towersonline]

Don, Here is something else to look at. Is there a large diameter flat washer between the motor mounting bracket and the X-car mounting point? If its not there that could be the root of your problem. In servicing these machines I have found them missing or installed someplace else. I little lubrication on the faces of the washer won't hurt.

[cnc_works]

Thanks for the input, Bill. Yup, they are there and I lubed them with the same stuff I'm putting on the racks.

Right now, I'm still doing test cuts and dealing with some tuning in relation to Mach's constant velocity features. My only real frustration with Mach!

Donn

[Brady Watson]

"My only real frustration (is) with Mach!"- Imagine that!


You know you can always just switch back to SB3, right? You can buy a bare 4G board and transfer your power supply & existing Geckos over to it for less than you think. I had my PRT running via Mach2 & then Mach3. I didn't notice ANY improvement over SB3 in terms of cut quality, speed and smoothness caompared to a 4G setup...and power supply, Geckos, motors etc were identical between the two setups. Yes there is a massive difference between the original 3G PRT and a system running Geckos regardless of the software used to control it.

Come on in...the water's fine!


-B

[cnc_works]

Was that a purposeful mis-read, Brady?

Other than the CV, I really like Mach and I do have hopes that I will find the sweet spot in the settings. And further, I remember that similar features in SB were pretty obtuse and occasionally didn't behave in a consistent manner for me. Of course I've used G code on a smaller machine I have for ten years so I was G code oriented well before I had a ShopBOT. But it maybe also that I like the internal controls possible with Mach just for the same reasons I use a PC over a MAC. A little more friendly to get into the guts and tinker, not that each won't do the job just fine.

I was interested to know about the 4G motherboard, though, as I have had some feelings about not being completely under the SB umbrella anymore. But then, isn't there a licensing issue for the control software that would add more expense?

Donn

[conceptmachine]

Donn,
I have the ascension/mach setup as well and i am very pleased with it.I feel the same as you,seem's like if you don't use all of the "original" equipment or upgrades offered by SB then you are the red headed step child.I personaly get a kick out of all of the reason's people use to downgrade non SB item's.In my opinion use whatever driver,control,etc...that you are happy with,what's good for some may not be for others.For me i'll never go back to the SB software not that anything is wrong with it i just like the layout and being able to customize the mach sofftware to suit your own need's and mach has never let me down in the middle of a cut either
.
shawn

[Brady Watson]

Like anything else, whatever works for YOU is the right tool (or control software) for the job. Does it really matter? In most cases...No. There's no need for anyone to feel like a 2nd rate citizen if they are not running SB3...it just makes it a little harder to get answers when you don't have the same configuration as most of the other people running ShopBots. For some who are more astute than others, this is not an issue. I just wanted to let people know that they can switch back to SB3 if they wanted to without investing too much money. If you want to know the cost, call SB.

Question: Just out of curiosity, what features does Mach have that SB3 doesn't? Citing reliability is not a feature...both Mach & SB3 have locked up on me in the middle of a file until things were configured properly for each system. Let's be fair & honest here...

Donn,
I actually did mis-read your post! It makes sense today....why it didn't yesterday is beyond me.

-B