Communication issue? Grind noise, then position loss

[Brett1970]

So, i've had this gremlin for 2+ years. My machine will be running, i'll hear a grinding noise, and i've lost X, Y, or Z position, and my sheet is ruined. Sometimes it happens once every 20 sheets, sometimes every 3 sheets. The problem is definately NOT physical, i've had every stepper off, gears are tight, teeth are good, all the V rails and teeth are good. Verified all the grounds, no issue there.

I am running a laptop, no internet, no other software. Automatic updates, etc every non essential service is turned off, i followed the guide on the forums i found to do this. Using the USB hub that came with my shopbot. In speedtest i get 74.9 I am cutting at 4 inches a second, and my jog speed is 6 inches a second (i think, whatever the factory setting is) I have no idea what to do. My machine had two other issues since day one, it's more or less been a lemon since i bought it new. It leaves a nub on every single part, and my proximity switches never worked right since day one.. it could be out in the middle of the table cutting away and it will say "stop hit" constantly, so it's disabled.

I thought 74.9 would be fine. I have a prs standard 48X96 with a 2.2 hp yaskawa spindle. Has anyone tried a desktop with USB 3.0 with a USB 3.0 hub, and what kind of speeds are you getting? seems to me a serial port would have been a better way to go than USB.

[gc3]

lemon......don't think so

you need to trouble shoot your machine...laptop issues?

a old VW saying, "be kind to your ass for it bears you"

[bruce_taylor]

Your speed seems a bit fast for a standard, what are you cutting? I think your loosing steps because your pushing to hard. Just my humble opinion. I'm sure some more knowledgeable guys will have some opinions. I've never heard a shopbot called a lemon before.

[adrianm]

Have you contacted ShopBot support directly over this? If you've had the machine from new it would seem to be the thing to do.

A USB 3 port defaults to USB 2 speeds when connected to a USB 2 device so it won't make any difference. A serial port is far, far slower than USB 1 let alone 2. With the speed test ratings you're getting it's not the transfer speed that will be causing your issues I wouldn't think.

Much more likely to be a build up of static or you're just pushing it too hard I have an Alpha rather than a standard but there's quite a few things I can't cut at 4ips. What are you trying to cut, pass depths, etc?

Given the issues you've had with the proximity switches my money would be on static or some other problem with interference on the electrical side.

[Brett1970]

oh beleive me, i've talked to the guys at shopbot plenty, and the proximities and the nub it leaves on the part have them baffled. Nothing fixes it. I cut 3/4 birch, and also 3/4 melamine. The melamine i cut at .405 per pass with a compression bit 1/4" and i cut the ply at .375 per pass with a double flute straight bit. My system is grounded, starting at the breaker box in the house. From there, it goes to the back of the control box, to the nearest machine rail. From that rail, one wire runs up the machine wiring to the X car, then up the chain wire guard, and directly to the side of the spindle. Back at the first rail, it runs to the second rail, and on to the dust collector base. then, it goes to stranded wire/no cover, and it goes inside the dust collection skirt, and attaches to the dust collection boot.

The entire system is properly grounded. All my cords to everything are 3 prong, so it's a total mystery. Frank at shopbot is telling me it's a communication issue. 74.9 is the best i can get with a laptop. I'm hoping a desktop with usb 3.0 and a powered usb 3.0 hub gets the numbers up into the 80% range.

[garyc]

David...
I have tested a number of computers with native USB 3.0. In fact I have 2, one with XP Pro-32 and one with Win7-64 on my test bench here. Both machines have USB 2.0 & 3.0 ports. I have seen nothing that would cause me to believe that a USB 3.0 system as you describe it would increase the comm rate to a ShopBot.

[richards]

I think that Bruce identified your problem. Try cutting at 2.5" or 3" per second and see what happens.

A stepper motor loses torque as its speed increases. If you are running the motors too close to the fall-off part of the torque curve, a very slight difference in materials can overload the motor.

A "grinding" noise indicates that a motor is loosing steps. Slowing down the speed puts the motor on a different part of its torque curve which may allow it to handle those slight differences in the material.

If you go to Oriental Motor's website and look at the torque curves for your motor (or a similar motor), you will see that a stepper motor has maximum torque at low speed and progressively less torque at higher speeds. As the speed continues to increase, the motor will reach the "knee" portion of the torque curve. The knee portion is where a minor increase in speed causes a major decrease in torque.

Sometimes there are ways to get more speed and more torque out of a motor, but it may require different stepper drivers, different wiring of the motors, a different power supply and sometimes a gearbox or belt-drive. None of those options are particularly easy; however, identifying the problem by reducing the cutting speed is easy.

[ken_rychlik]

It could be as simple as an incorrect bit for the job. If you are cutting full depth in one pass, slowing down like Bruce said will help.

For the proxy's not working, it's probably not grounded well.

There are so many things that will cause this and thus the many different suggestions. It may be a machine problem, but most likely not.

[khaos]

About the false proxy warnings. Two things come to my mind assuming your ground system is perfect as described. Line noise: This could be from a power line running parallel to your motor or proxy lines. Magnetism from the power line creates electricity in the other lines that could be read as a positive proxy hit.

The second issue could be a spike or drop in line voltage. This could be a surge amperage(current) from an aggressive ramp value. Or some appliance in the shop like a compressor, grinder, or saw starting. Or a surge from the grid. I have my bot on a quality UPS from APC (http://www.apc.com).

Another piece of data is that resistance increases with heat. So if ANY of your power lines get warm with use longer use means more resistance in the wire which equals a voltage drop at the device and often noticeable current increase which raises temps more in a cycle.

The SB tech boys are really good at what they do but electricity is a fickle beast.

HTH,

[bruce_taylor]

I gave up on proximity switches long ago really not a big deal for me I cut 3/4 plywood alot at 3 passes .375 bit at 2.5ips 11000 rpm onsrud 60-121 or 60-123 compression, This is on my alpha, I know I could push it a little harder but it sits idle a fair amount and I don't like to push my tools to hard. I would slow down and see what happens.

[jerry_stanek]

I had the same problem with the proximity switches triggering until I move the switch ground to a different ground post. Haven't had a problem now for a year.

[Brett1970]

i don't run anything else in the shop when the bot is going. The bot is on it's own breaker, as is my dust collector, and my vacuum hold down. Any power lines going to the bot itself, are not zip tied to the motor wires, i kept them seperate to eliminate cross talk. I don't ramp in to any cuts, and i cut my melamine and ply in 2 passes. I followed the rule of depth of cut max = 1.5 bit diameter. I'm over just a touch on the melamine, at .405 but that's just to make sure i am cutting all the way through it.

I've had it cut 15 sheets with no issue, other times it's only 2 - 3 sheets before it acts up.

[fredtoo]

"My system is grounded, starting at the breaker box in the house."

How far is that? Are you maybe looking at a ground loop or GPR (ground potential rise). Try moving that ground over to one of your local outlets or maybe disconnect it temporarily and run a test cut. Too may grounds can, if at different potentials, can cause BIG problems.

[garyc]

David...
Try cutting a few sheets with your speeds set at half of what you now cut at. When all is good your cut speeds are near the max for the machine. Add some dust, wear and tear and a partially dull bit and max speeds must be reduced.

Have you dropped out the motors and cleaned the racks lately? have you done a free wheel motion test of all axes? How about pinion replacement? Many that cut sheetgoods fairly steady forget to dissassemble the z axis motion parts to clean and lube. Out of sight is not out of mind when cutting aggressively.

[ron_moorehead]

A couple of days ago, I got a switch proxy warning (input 4) and the shopbot shut down. I checked all my wiring and took apart the emergency stop switch, and found all the wires tight. The wire were tight at the connector on the circuit board. I was still having the same problem. I put jumpers on the input for the emergency stop thinking I my have a bad cable, still had the problem. I finally figure out that I was running the Z into it's end stops and stalling out the motor. You get the same fault warning as a proxy switch. Repositioned my work so I would not run into the end stop and everything worked fine. I have a PRT Alpha and I only run it at about 2 ips. I also heard the grinding noise just before she would stop.

[Brett1970]

I dropped my cut speeds from 4 inches a second to 3.5 i ran off 20 sheets with zero issues. now today, it's doing the same damn thing, ruined 2 sheets in a row. First time was on the Z axis as it was cutting. Wasn't as severe as before. The last time was a pronounced grind. This time, it's more like it hit a speed bump, just a quick THUNK sound, first sheet i lost position. it did it right before it went to plunge, and it lost position on the Z by 1/4". the next time it did it, it was on the X axis while cutting, heard a thunk, and lost position by 1/8th inch.

I am positive i have no ground loop in my grounding system. The length of my ground from box to machine is no more than 16 feet. much shorter than if i hit the ground in the nearest outlet by alot.

I am so frustrated and sick of losing money on ruined materials i am at my wit's end. if i slow this thing down any further, i might as well get out the hand router.

[bruce_taylor]

David,
I can relate to the frustration of ruined material and time. How many sheets out of the bit? What are you using for a bit. 3.5 ips still sounds a bit fast. You are loosing steps which implies to me that you may be pushing a bit hard still. I know you don't want to here that but give a little more info on rpm cutter wear etc and what you changed after the first sheet was ruined. I use a reference point to zero to incase of loss of position or loss of power so I can re zero my position if I have an issue that might help. What is the recommended feed and speed from the dealer for your bit selection. Is cut quality acceptable? I would really like to help you resolve the issue. Anything else you can tell us about the variables would help .

[garyc]

David...
In most of the cases that I have seen or heard a quick "thunk" as you describe, the cause was static. Do you have your dust collection hose grounded? Is there a bare wire run inside the dust hose? Is it grounded on the end away from the machine? Is the dust foot alligator clip attached to it?

[Brett1970]

i have it all grounded per the guide i found on here. I think i've covered my grounding before but i will go over it again in case i haven't. ground starts at the breaker box, goes to control box, then to Aluminum X rail, from here it splits off - one leg goes through my wiring looms, and is attached to the X car, then through my wire chain, and is attached to the side of the plate the spindle is attached to. The other leg from the X rail, goes over and connects to the other X rail. From there, it goes to the base of my dust collector. Then from the base, it goes to the stranded bare wire that goes through the dust collector hose, to the dust skirt. I also have a wire through the plastic on the dust hose, and it's connected to the ground back at the base of the dust collector too.

Using a 1/4 double flute full carbide bit, cutting birch plywood. cut quality is fine, i have maybe 10 sheets on that bit. i usually get around 25 sheets before i swap it. Cutting in .375 deep passes. Jog speed of machine is whatever factory setting is. Pc is optimized to bare bones, no internet, defragged, etc. The other stuff i was cutting was black melamine, using a 1/4 compression bit, same speeds, but .4 max depth of cut per pass, and had no issues. I switch to ply and it all goes to **** : /

[Brett1970]

i adjusted the motors, cleaned all the tracks, even the Z greased everything, etc. Still get the grinding noise, and have to hit the stop before it ruins a sheet. Then start over, cut again... sigh. won't be much longer before i just sell it to fix the problem.

[bleeth]

David:
It could be that there is a service kicking in on your computer resulting in a sudden loss of comm and hence lost steps. The grinding noise is likely the stepper losing steps. You mentioned you were using your laptop, and if this computer is used for other things and has not been "cleaned" then it could be as simple as another program looking for an update or some other scheduled task like an antivirus scan or automatic de-frag.

[Brett1970]

defragged and disk cleanup regularly. I followed the guide on here, for cleaning it up, turning off unneccessary services, etc. It does not go online (disabled) and i use it for nothing except the shopbot exclusively. I am still baffled. On the speed test, i get almost 75% and i've dropped my speed to 3.5" a second and using new bits. What is the absolute BEST setup to get maximum USB speed?

[danhamm]

Static discharge, probably on the z-axis, from the spindle.. get a ohm meter..and resistance test every thing on it "using" a direct cable or ground wire to a referenced grounding lug on your service..

I know youve done it.. but your clues are quite obvious...copper and aluminum will coat with time and become resistive..then become a capacitor...then discharge... causing a floating reference...if your going to get rid of your headache...lol.. I'll trade you...

[adrianm]

defragged and disk cleanup regularly. I followed the guide on here, for cleaning it up, turning off unneccessary services, etc. It does not go online (disabled) and i use it for nothing except the shopbot exclusively. I am still baffled. On the speed test, i get almost 75% and i've dropped my speed to 3.5" a second and using new bits. What is the absolute BEST setup to get maximum USB speed?
75% should be fine anyway. I never get over 73% and mine is fine.

Do you have all the power saving features of the laptop disabled? Even on mains power I've come across laptops that turn off the hard disk after a few minutes, reduce CPU power and turn off USB power etc.

[bleeth]

I use an APC Hub and a serial/USB convertor rom Radio Shack. I get 84%.
I don't think it is that though.
Dan has a good point-How long since you grounded it? Have the ends or connectors built up "crud"? (You may not even be able to see it) Think how the cables on your car battery get.
Was the air humidity higher or lower on the "good" day? This won't cause a problem, but it could be an indication of where the problem is. Please don't say you have the issue when there is an electrical storm!!
You can also be having the bad luck of someone nearby cranking up a heavy welder or something and so getting surges.

[jimmya]

Cutting at 4 inches a second, I think is to fast for the standard machine. If I get over 3 inches a minute I start losing steps. Just for the heck of it slow it down and see if you still lose steps. Just a thought.

Jimmy

[Brett1970]

well, it just did it AGAIN!

The grounding was done earlier this year, and problem seemed to go away. I don't know why, i don't have near as many problems cutting melamine than i do ply. I was cutting 3.5" a second, 12,500 rpm (spindle) and a new 1/4 double flute full carbide bit. Just turned on the laptop and machine, first sheet i cut. This isn't anything complicated i'm cutting..

http://i68.photobucket.com/albums/i24/dptwiz/Starcastle.jpg


I had the pockets and inside cutouts as 1 file, i ran that. 2nd file was to cut out the parts, outside the line. I'm doing .375 depth passes. It cut the square on the left, the two squares towards the right side of the sheet, then it went to cut the square 2nd from the left.. it made 1 full pass, plunged, went about 5 inches, and i hear "THUNK"

totally at a loss here.

[danhamm]

What do "you" think that clunk was..? my guess is its lost communication and because it lost communication with no ramping it went clunk..!!

I wouldnt be putting anymore plywood in that machine until I had licked my problem...its not going to disappear on its own.

Get and ohm meter and a long piece of grounding wire heavy stuff 1/4 to 3/8 in. hook one end to your wire goin to your ground rods or casing ground.. and check every piece on the machine...

[srwtlc]

Just out of curiosity, is that screenshot from the computer that you run the bot with? It appears that you have a lot of programs resident in your sys tray (by the clock) that could cause issues depending upon what they are for and doing (monitoring).

[Brett1970]

nooo design computer is what i'm using now, i do all my design work on the PC in the house, and use a flash drive to take files to the laptop in the garage. I don't leave the flash drive in the pc either lol i download, remove hardware, and pull the drive.

Personally, i think it's a communication issue. I have a friend bringing me a new desktop next weekend to replace the laptop. Also bringing me a new powered usb 2.0 hub. As i said 74 - 75% and 3.5 inches a second.

Ryan from shopbot suggested i cut melamine at 4" a second at 12,500 and it worked great for a while, then i started having the thunk issue. I have ran a ton of melamine with hardly an issue at 3.5

[Brady Watson]

David,
Please check your dust hose ground one more time. The symptoms you list point at ungrounded high voltage static. You cannot use the same ground as the control box, as it may cause a ground loop issue. Below is the suggested 'tried & true' method of correctly grounding your dust collection system. Take NOTHING for granted...

First, verify ground between electrical system ground and your dust collector chassis. Put one probe of meter in ground in electrical outlet & other probe against a metal bolt, screw or lug on the DC chassis - preferably the exact location you plan on grounding the end of the bare copper wire. The spiral bound wire encapsulated in the hose is not suitable for grounding. It is there for physical support only. The ground must be a bare copper wire without exception. Doorbell wire stripped of insulation works perfectly.

Run the bare copper wire through the dust foot, connecting to the alligator clip, then through the entire length of hose, then go through the hose and tie off at the place where you checked continuity. You can tie a rag to the wire, fire up the DC and throttle it until it get close to the DC, before turning it off. Then test continuity between the copper wire and your electrical ground.

The high voltage generated by dust collection can, will & does overpower low voltage DC switches and sensors on the machine, that operate at 5vdc (USB, input switches) and 24vdc (prox switches). If you don't give the voltage a 'path of least resistance' via bare copper wire inside the hose that acts like an antenna to wick up the static - then, it will find it's own path, which will either be the X or Y Prox switch wires (long antenna) or USB cable (long antenna)...Prox switches that get 'stepped on' via high voltage result in false 'Limit Switch Hit' and 'Emergency Stop Hit' messages, interruptions in the communication stream or 'thuds & bumps' in movement are caused by the USB 5v cable getting 'stepped on'. If you have good Com speeds, and computer is good...you've got static issues.

For the heck of it...check it again. AND - also check that you have continuity between the dust foot clip and the dust collector. Wires CAN break!

I've seen it happen so many times that people think they are grounded, only to find they are not because of a broken wire, a corroded contact or other thing that doesn't work properly.

-B

[steve_g]

I recently discovered that the cable guy broke my clamp where my electric panel attaches to my ground rod outside... he just left it!

Steve

[Brett1970]

i could swear the grounding guide i found on here DOES tell you to go to the dust collector from the leg of your shopbot. Should i ground my dust collector directly the earth ground outside? It's only about 6 - 8 feet tops from my dust collector to my ground lug. I don't think this would cause a loop, unless it did throught the electrical ground coming from the plug to power the the dust collector motor. I removed my aligator clip and replaced it with an eyelet, and attached it to the dust boot at one of it's rivets. No chance of my wire being sucked through the dust collector.

[garyc]

David...
Attaching the ground wire to the dust foot mounting rivets will allow static to dischage directly to the machine. Brady gives very clear instructions above. It will be in your best interests to follow them to the letter.

[Brett1970]

noooo lol i guess i shouldn't have said rivets. It's actually a screw, and i'm talking about the plastic dust boot itself. i have my bare strand wire attached to the base of the dust collector, through the dust hose, and i put an eyelet on the end, and it's attached to one of the screws on the dust boot, where it goes from the formed top, to the slatted sides. It's plastic to plastic right there, no metal involved.

Here's my thought. I'm going to cut the ground going from the machine frame to the dust collector, and go from the dust collector right to the earth ground in my breaker box. This shouldn't cause a ground loop that i can think of, since it will terminating at the plastic dust boot.

Here's a diagram of how it's grounded

http://i68.photobucket.com/albums/i24/dptwiz/shopbot.jpg

[michael_schwartz]

I recently discovered that the cable guy broke my clamp where my electric panel attaches to my ground rod outside... he just left it!

Steve

somebody did that to the second grounding rod for my shop.... And of course the wire, had been loosely wrapped around the rod, having never been clamped to begin with. I don't even want to know who installed it. It is still laying on the ground, since it is not my responsibility to pay to have it redone properly.

[garyc]

David...
Delete your wire from (G) to (F). Verify as per Brady's instructions above.

[Brett1970]

Alright, i'm going to go get a spool of wire tomorrow, so i can eliminate the ground from table to dust collector, and go dust collector to breaker box.

[garyc]

David...
Assuming that the DC has a 3 wire cord, and you verify as per Brady's instructions, it is already grounded to the electical box. Adding the wire to the machine creates a ground loop.

[ken_rychlik]

I had one from C to F and also from B to the case of the PC. It was tagged to the x and y gantries as well along the way. Also the dust collector base was grounded to it's electrical supply ground just to be sure. My DC hose has the wire embeded in it and I didn't use an internal wire like most gurus on here will suggest. I tried the internal wire, but it kept getting clogged up with wood fragments.
Not sure mine was perfect, but it worked.

[Brett1970]

i'll double check that the dust collector is 3 prong. pretty sure it is, i have it and the vac hold down on 2 seperate extension cords, and they are 3 prong.

[ken_rychlik]

Not just a 3 prong cord, but tie a wire from the body to the bare ground wire. Don't depend on the DC to be grounded properly to begin with.

[Brett1970]

i'll do that. it's in a bad spot, between my woodrack (that doesn't move) and the wall. shouldn't hurt to get into into it and tie in to the ground coming in from the plug and go right to the stranded wire.

[michael_schwartz]

not to deviate from the topic of the thread, but since there is discussion of ground loops, I do have a question.

Since I have the UPS for my control PC plugged into a completely different circuit than my control box with about 10 feet of extension cord, am I creating a potential ground loop, since the PC is tied into the control box via USB? After replacing my control PC I haven't had any problems, but I want to avoid future issues if I can.

[Brett1970]

ok, i have cut the wire from the machine leg to the dust collector. I have verified with my meter there is no continuity between the bare stranded wire and the machine, and no continuity from the dust collector to machine. I do have continuity on the bare stranded wire at the cutting head, to the ground prong on the dust collector cord.

Ran a test sheet, still getting the "THUNK" so still at a loss.

[Brett1970]

i am starting to think it's a pc issue. It's extemely frustrating to say the least. I'm switching it out this weekend.

[garyc]

David...
Before you replace the computer, try running the same file with all the dust collection components removed from the machine.

[Brett1970]

Ok, so i talked with tech support, and made some changes.. took the ground off the control box, and off the spindle motor plate. Ran a test, cutting 50 circles all over the table in the air, with everything running, vac hold down, dust collection, etc. at 6" a second, no thunk, no loss of position, so i ran off a sheet of parts, no issues. Today, i load a sheet of melamine to cut out 1 large part, just an outline. 6 feet into it, proximity switch goes off, i'm no where near one. I resume, 6 inches later THUNK it's doing it again. Use C3 to run it back to home. Then i do C2 to zero the bit.. i get a limit switch hit before the Z is 1/2 way to the plate, and resume, Zeroing failed. Try to run C2 again, limit switch hit, i can't even zero my machine now.

[jerry_stanek]

That is exactly what mine was doing. I had to move the proximity ground wire to a different post in my control box.

[Brett1970]

i'll go try that , pretty sure i moved it once early on. This THUNK i get it is only for a split second, and it keeps going.

[paul_z]

David,

Are you using the ZZero clip and the ZZero plate? If not, try using the clip.

Paul Z

[ken_rychlik]

Is your z zero plate by chance touching ground while you are running? It needs to be isolated from ground so it doesn't give a false signal.

[Brett1970]

ok, my zero plate and clip are isolated from the machine and eachother. The double circle light is not lit. Still getting proximity switch hit on the Y axis, roughly 9 - 13 inches into the Y axis. getting ready to move the Ground, i am positive i've moved it before.

EDIT- maybe i haven't moved it, both grounds are where they go according to the book, along with a white wire. I moved the two ground wires to the ground on the two blue blocks that have no other wires going to them. going to go test now.

[Brett1970]

ok, i did notice i where i moved the ground, one is on the input section ground, and one on the output section ground. Shouldn't matter in the box, a ground is a ground in there. Worked fine for C3 command to set X0 Y0 i ran C2 and used a 1" block of wood to set 0 above my sheet so i wouldn't actually be cutting.

Got the THUNK and it continued on after about 3 minutes of simulated cutting (vacuum on, spindle on, dust collector on.) about 1 minute later, Limit switch hit, no where near one again. Also went through the PC some more. Here's the processes i have running:

taskmgr.exe
explorer.exe
svchost.exe
svchost.exe
svchost.exe
lsass.exe
services.exe
winlogon.exe
crss.exe
smss.exe
system
system idle process

I have verified the above with two IT guys, said the above are essential services.

No internet, no wireless, no other software than control software, no screensaver, no power saver, Usb hub with only shopbot on it. No mouse, no nothing else.

still baffled.

just for the hell of it, i'll go try the same thing with no vacuum and no dust collection. I'll even unplug them to make sure they are out of the loop.

[Brett1970]

ok, same test, breaker to spindle off, vacuum hold down off, and unplugged, dust collector off and unplugged. Got the Thunk and it continued on, then got the proximity switch hit again, no where near it.

[ken_rychlik]

you could unplug the sensors one at a time and run air test also. May have a bad one or a bad wire.

[Brett1970]

It's done it in X Y and Z. Wouldn't explain the thunk. I can cut circles in the air, (50 of them, all over in different places), no thunk no stop hit. How bizarre is that?

[Brett1970]

ok, Big thanks to Frank and Chris @ shopbot. I am now running the new beta control software, and seems to have corrected the THUNK and also the false proximity switch hits. Ran 4 sheets with no issue. It also pockets small circles a little slower, but it isn't jerky acting as before, I got to say the beta software is extremely cool!