PDA

View Full Version : CNC router occasionally makes jarring noise and home jumps when it does this. Help!!



itsmillertime
12-01-2014, 11:49 PM
Hi Guys I was hoping that you could help me track down the problem behind why my CNC 6040 router now and then (but pretty much most jobs) it will happily be machining away. When at some point the machine will make a louder 'jarring' kind of noise and it appears to be along the Y axis by what it does to my work pieces (the longest axis which the gantry moves across). The machine doesn't appear to stop but it looks as though it jumps by about 10mm. I haven't been able to physically see this by eye as everything is always too quick and it's not spectacular enough. But I can see it jumps on my work.

You can see from the photos attached what it is doing.

The machine appears to be operating just fine until this happens i.e circles are all correct as are square shapes.

I know that they have lots of quality issues, but I would be really grateful to you guys for advice on how to diagnose this.

Upon starting up the CNC router and running mach3 on my desktop all 3 stepper motors lock as they should and the router happily runs/jogs along all axis. I can turn the stepper knobs by hand freely and cannot feel any rough parts of travel in the ballthreads.

Please help me diagnose this with simple steps. Is it more likely that it's wiring or a stepper motor or driver?

Just guessing possible Y axis wiring issue. But would love to see what you all think.

As you can see this error has made it into all 3 jobs below. Looking forward to hearing what you guys think.

Cheers andy

112671126811269

EddyCurrent
13-01-2014, 12:05 AM
On the second photo it looks like it jumped then came back into line again, did it do that itself or did you intervene and correct it ?

Is it direct drive or via toothed belt ?

Jonathan
13-01-2014, 12:29 AM
I think in the second photo it jumped twice by different amounts, as it's not gone perfectly back into line.

It could just be the Y-axis motor is very occasionally stalling. I'd try running this code (http://www.mycncuk.com/forums/gcode/5843-random-g-code-testing-motor-tuning.html) to test it more thoroughly and report back.

Washout
13-01-2014, 12:32 AM
There are way better guys on here than me to advise, but here goes....
.
What you are describing sounds like a motor stall on the y-axis, which is likely due to 2 causes:
.
1. You have a mechanical problem, such as the rails, ball or lead screw(s) or bearings being out of true and causing binding i.e. a resistance the stepper motor is having difficulty overcoming or;
.
2. You have an electronics issue where data or pulses are getting lost and the steppers are not moving when they are supposed to be.
.
For 1. I would see if you can first disengage the ballscrew/leadscrew and see if the gantry can be moved across its full range of travel with only a little pressure by hand - my own gantry weighs a fair amount but I can push it easily with a finger. Or failing that see if there is any adjustment possible on the ball/leadscrew alignment and redo that - not sure how your ballscrews/motors etc. are attached, so can't really advise on the method of doing this.
.
For 2. Only advise I can give at this stage is if your wiring is not using shielded cable, then replace that with shielded CY cable and ground the shield at the control box end only and to the same ground point all the other earths on your machine should be going to, which should in turn be wired to mains earth. Make sure the ends of the shielding at the stepper motors are isolated i.e. not contacting anything. Other than that you are looking at driver, control board/BoB replacements, but I can't advise on diagnosing your existing components. Also I have seen some stories of inadequate grounding within the control box itself, so check that out.
.
Oh also if the machine came with a cheapo parallel cable for connection to your PC then try another one. I had a problem with my machine getting spurious e-stops that turned out to be a shoddy parallel cable. Replaced it with a chunky shielded one and that problem went away.
.
There'll likely be other guys that can advise further, but having tripped over most of the pitfalls in building reliability into my own machine that's where I would start.
.
Hope that helps.

JAZZCNC
13-01-2014, 12:57 AM
If it's a sudden jump straight after the Jarring noise and it's just on the Long axis then I suspect it's the ballscrew end bearings floating.?
Grab the gantry firmly at both sides and try to move back n forth. If you have end float you'll feel it clunck.

If it's losing position over time then I'd suspect a lose coupling first, after that then I'd look at the wiring. These cheap machines come with crap wire on the motors which gets brittle and breaks internally but still works and causes all sorts of strange things like this. Same with the connectors there rubbish and need replacing.

From you saying it works fine then does this then I don't think you will have it over tuned or anything so wouldn't go messing in that direction just yet.

With the noise your saying it's making then 90% chance it will be mechanical 10% wiring or electrical.

GEOFFREY
13-01-2014, 01:20 AM
It would appear to me that the problem occurs during rapid moves and not whilst cutting. if this is the case you may be able to work around the problem by changing G0 to G1 as a short term fix until you are able to resolve the problem. G.

itsmillertime
13-01-2014, 08:18 AM
Thanks guys loads of ideas to look at here! eddy its a direct drive stepper motor without any drive belts. As for the photo and it correcting itself. I think it did the top text line in a different order I.e do a few letters and go do something else then come back. I haven't corrected anything in any of the pics.. hope that helps?

I think it is most likely a mechanical issue as is suggested. so I'm going to take a few videos while its cutting to see if I can narrow it down further.

@ Jonathan I think you are correct about it having jumped different amounts, it looks like that to me too. I will run the code tonight when I get in from work. thank you!

@ washout and jazz cheers guys I will look into your suggestions after I can nail this this down on video.

and Geoffrey I shall indeed try changing those settings as it does look like its only doing this whilst moving fast and not actually cutting. cheers guys I will report back. hope I haven't missed anyone out trying to type this from my mobile!

cheers andy

richie00boy
13-01-2014, 09:25 PM
Check your pulse train speed as I had a sort of similar issue and found I was running too fast. If you want to post up your axis settings I'll check them against mine. I've got a 3020, but I think the settings should be similar?

itsmillertime
13-01-2014, 11:38 PM
Thanks Richie I will post up some screenshots of my settings tomorrow it could be that the stepper motor is running to fast in one direction and trying to reverse too quickly (if this makes sense?) I have managed to capture it doing this on video twice and have attached a youtube video. You will be able to hear the 'jarring/grinding noise' briefly. Sorry for the crappy fastening down I did the first clip without the spindle running. The second part of the video is just to highlight that it's doing the same thing with the spindle running too.

I am a beginner to all of this and appreciate all of the help you guys have offered. I tried rocking the gantry back and forth and left and right and I was unable to find anything loose as such.. no real obvious play. I will look into this more tomorrow night.


http://youtu.be/hqf3sHZVPjE

Any more thoughts welcome? does this rule anything out or more clearly point to any of the above suggestions?

Thanks again Andy

JAZZCNC
13-01-2014, 11:58 PM
100% the noise is motors stalling from being over tuned and trying to cut too fast.

Looking at the speed you where cutting at I think your being too ambitous on the Feed n speeds for this little machine and the motors are running out of torque and thats why your getting stalling motors.

Drop the feed rates down and try again and I think you'll find it works fine.

You could play with motor tuning which would help but I think your just pushing the little thing too hard.!! . . .. Try lowering the Accel first if you adjust motor tuning has that will have the biggest affect and would cause this kind of stalling/missed steps.

itsmillertime
14-01-2014, 12:22 AM
phew! I'm pleased to hear you say that jazz! Is just me being far too ambitious with it. I will have another play tomorrow night and get those speeds down. what kind of speed would be a conservative setting? also this may sound stupid. I can change the feed speed whilst cutting in vectric whilst setting up my toolpath. Is there a separate option in maxh to adjust the rapid movements in between cuts? or is this dictated purely by feed settings? guessing there's something in mach3 for this?

Thanks again and appreciate your opinion jazz. was beginning to feel open wallet surgery about to take place ;-) .

thank you

JAZZCNC
14-01-2014, 12:36 AM
Rapid speed is the Velocity setting in Motor tuning.
If you watch the video it stalls while cutting not rapiding between moves so I think if you tweak the Accel first and lower the feed while cutting then you'll be ok.
Basically it's stalling under the cutting load which isn't much at this depth but your feed rate is so fast your at the limits of the motors torque so it doesn't take much to make stall.

Just drop the Accel by 10% and try again.

What feed rate was it set at while cutting.?

Edit: If you want to test and see the affects of lowering feedrate without altering anything, machine or G-code, then just lower the feed rate over ride to 75% and try it again.!

Clive S
14-01-2014, 10:40 AM
phew! I'm pleased to hear you say that jazz! Is just me being far too ambitious with it. I will have another play tomorrow night and get those speeds down. what kind of speed would be a conservative setting? also this may sound stupid. I can change the feed speed whilst cutting in vectric whilst setting up my toolpath. Is there a separate option in maxh to adjust the rapid movements in between cuts? or is this dictated purely by feed settings? guessing there's something in mach3 for this?


Just for the sake of clarity Velocity and Accel are two different settings in Mach they are in config\motor tuning. Not to be confused with each other Jazz is referring to the Accel setting that might be a bit too high. Good luck I am sure you will get this sorted. ..Clive

kingcreaky
15-01-2014, 09:28 AM
The way i think of it is, the Acceleration and Velocity settings in Mach act like your speed limiter, they over-ride whatever settings the CAM software specifies in the job.

Its also critical (although I cannot explain why) to ensure the Acceleration and Velocity settings are the same for both your Y axis and X axis. (Z can be different) Otherwise (I found) it effects 90deg corners. Where (I think) the machine is still decelerating (moving) when the next move starts

Personally (others will disagree, I set the accelleration very low (often 100), and the velocity to a comfortable speed for the machine as Im far more interested in quality, than speed) I also find having acceleration set so low helps when jogging your machine manually into a start position and also makes a nice smooth start and stop to each move.

As mentioned above, Mach is your speed limit. So if you have set your velocity to 6000mm/min in mach, and your tool to 8000mm/min in vectric (or your cam software) mach wont let the machine exceed 6000.

The machine jogs (ie with the arrow keys, at the maximum speed stated in mach)
The machine moves (between cuts, at whatever speed the Gcode outputted by your cam software) -- but limited by mach.

The above is just my opinion.

Jonathan
15-01-2014, 09:35 AM
You're noticing the variable corner rounding effect because the acceleration is too low. You should be aiming for around 1000mm/s^2 for the acceleration - certainly at least 500... otherwise in some circumstances your parts wont be as accurate.

Neale
15-01-2014, 09:42 AM
Isn't there also a gcode parameter which affects accuracy at corners? You can either specify stop/start for maximum accuracy but lose feed rate at corners, or give a maximum error value so that Mach or LinuxCNC can approximate the corner to maintain cutting speed. Coupled with a max acceleration figure, it can all start getting a bit complicated.

kingcreaky
15-01-2014, 10:02 AM
thanks for this boys....although since I have set the two the same this no longer happens... not to cloud the issue... its still good practise to ensure the acceleration and velocity values for both the X & Y are the same right or wrong?

JAZZCNC
15-01-2014, 10:35 AM
Matt just going to make some things a bit clearer for sake of others not picking on you here mate but it's not clear and your Soooooo wrong on some it. .:hysterical:


The way i think of it is, the Acceleration and Velocity settings in Mach act like your speed limiter, they over-ride whatever settings the CAM software specifies in the job.

As mentioned above, Mach is your speed limit. So if you have set your velocity to 6000mm/min in mach, and your tool to 8000mm/min in vectric (or your cam software) mach wont let the machine exceed 6000.

The machine jogs (ie with the arrow keys, at the maximum speed stated in mach)
The machine moves (between cuts, at whatever speed the Gcode outputted by your cam software) -- but limited by mach.

This is true but I'll make it clearer because it does apply here.

In the G-code this will only apply to the G1 moves which have F### feed rate parameter but all G0 moves will be at the Velocity setting in motor tuning. So if your over tuned on velocity then you will still lose position if G0 moves are in the g-code even if you lower the feedrate in Cam software.!



Its also critical (although I cannot explain why) to ensure the Acceleration and Velocity settings are the same for both your Y axis and X axis. (Z can be different) Otherwise (I found) it effects 90deg corners. Where (I think) the machine is still decelerating (moving) when the next move starts

Personally (others will disagree, I set the accelleration very low (often 100), and the velocity to a comfortable speed for the machine as Im far more interested in quality, than speed) I also find having acceleration set so low helps when jogging your machine manually into a start position and also makes a nice smooth start and stop to each move.

This is the bit your Sooooo wrong about matt... .Lol

It's not critical and perfectly fine to have differant setting for each axis, Mach will only go fast has the slowest axis anyway when moving 2 axis at same time. Try it.!! . . Double the Speed of Y axis velocity (if machine can do it) and home the machine and jog away good distance and Type G1 X0 Y0 F ( Feed value abobe X axis velocity or same has Y axis velocity) in the MDI box.

Like Jonathan pointed out most if not all of trouble is coming from having the Accel far far to low.
Molasses moves faster than that so MAch's gets tired of waitng and Knobs off to the pub for a quicky but has few too many so when it gets back it's forgot where it was upto and thinks bollocks to it i'll just carry on from here and that's why your getting square circles.!!. . . . Mach pissed.!! . .Lol . . . . Ok well maybe being bit silly but the point is your messed up on tuning.!

You really want to be within 500-800 just for normal use and 1000 and above this if doing mostly 3D type work.
Just remember the golden rule that you can't have High Accel and High Velocity you need a balance and you won't go wrong.

kingcreaky
15-01-2014, 10:41 AM
removes hat. Thanks for taking the time to clarify.

itsmillertime
15-01-2014, 11:13 AM
wow, thanks guys for all the info. I'm at work at the moment but late last night I had a quick look at the settings and my velocity looks way too high. I have found someone elses motor settings for the 6040 machine and Mine is at least 25‰ greater! no wonder why I'm having issues. I'm going to get in tonight change the settings and report back. Thanks to you all for your input. will be good to see this fixed lol. got my fingers crossed for tonight. I will screenshot the settings for anyone else with a similar machine should this quandry happen for others.

I'm sure I won't be the first ;-) cheers guys

Jonathan
15-01-2014, 12:47 PM
In LinuxCNC you can specify a tolerance on the G64 command, so the controller sacrifices constant speed to maintain the toolpath within the specified tolerance. e.g. put 'G64 P0.05' at the begining of the G-code and the path will not deviate by more than the P from the true value.


It's not critical and perfectly fine to have differant setting for each axis, Mach will only go fast has the slowest axis anyway when moving 2 axis at same time. Try it.!! . . Double the Speed of Y axis velocity (if machine can do it) and home the machine and jog away good distance and Type G1 X0 Y0 F ( Feed value abobe X axis velocity or same has Y axis velocity) in the MDI box.

You've raised a mildly interesting point there, which I think is worth expanding to show why what you've observed occurs. The velocity of each axis cannot exceed the values in the motor tuning, however the speed of the machine (i.e. the feedrate) can, unless the machine is only moving in one axis. This occurs because when two axes are moving at given velocities, the resultant feedrate is the vector sum of these velocities. One way to visualise it is as a right angled triangle, where (for example) the velocities of X and Y are the sides adjacent to the right angle - the length of the hypotenuse will equal the feedrate.

Example 1:
Suppose you set your maximum velocity settings for X and Y both to 1000mm/min, home the machine (so it's at 0,0) then enter 'G0 X100 Y100'. Mach will move both X and Y as fast as it can (as it's a G0 move), so the feedrate you actually get is \sqrt{100^2+100^2}\approx 1414mm/min, assuming the machine can accelerate to that feedrate within the specified distance.

The practical upshot of this is you can specify a feedrate higher than the velocity setting, and get that feedrate in some circumstances.

Example 2:
Same as example 1, except we enter 'G1 X100 Y200 F1100'.
The required speed is 1100mm/min, so the following must hold: 1100=\sqrt{X^2+Y^2} (equ. 1), where X and Y are the X and Y axis velocities.
The Y axis must travel twice as fast as X to go in the correct direction (as 200/100=2), so we have \frac{Y}{X}=2 (equ. 2)
We also have the condition X,Y<1000,due to the motor tuning settings.
From (2), Y=2X, so substitute that into 1:
1100=\sqrt{X^2+(2X)^2}
1100=\sqrt{5X^2}
X=\sqrt{\frac{1100^2}{5}}\approx 492mm/min
Whence,
Y=2*X=2*492\approx 984mm/min

(You can also solve this graphically quite easily - which may be more intuitive.)

So the X and Y velocities do not exceed 1000mm/min, yet the machine can in this example cut at 1100mm/min.

In reality, it's probably not a good idea to use this 'feedrate bonus' as it makes the feedrate a function of the direction in which the machine is travelling, so unless you're cutting a single straight line the feedrate you actually get will vary (continuously if an arc move).

JAZZCNC
16-01-2014, 02:01 AM
From (2), Y=2X, so substitute that into 1:
1100=\sqrt{X^2+(2X)^2}
1100=\sqrt{5X^2}
X=\sqrt{\frac{1100^2}{5}}\approx 492mm/min
Whence,
Y=2*X=2*492\approx 984mm/min

I see you like your new toy. . . . Lol . . .(Shame Nobody got F'#$ing clue what your saying. .:hysterical:)

irving2008
16-01-2014, 09:42 AM
Some of us do :) but it's true J is playing with it. It's a nice new toy tho if I could make it work for me..

Clive S
16-01-2014, 10:19 AM
I see you like your new toy. . . . Lol . . .(Shame Nobody got F'#$ing clue what your saying. .:hysterical:)

Dean: its a recipe for a sponge cake anybody can see that:yahoo: ..Clive

Neale
16-01-2014, 11:13 AM
Dare I ask for a bit more clarification? Dangerous ground, here, perhaps! But here goes...

What I'm not quite sure about here is the difference between G0 and G1. So, a few examples to see if I understand. In all cases, we start at (0,0).

G0 X100 Y100
This is a rapid move at max permitted machine speed. If max X and Y speeds are 1000mm/min, then Mach (or LinuxCNC, I presume) will move each axis at max permitted speed which actually moves the cutter at 1414mm/min across the bed (hypotenuse of right-angled triangle).

G1 X100 Y100 F1000
Mach/LCNC calculates the speed needed for each axis to achieve an actual cutting speed of 1000mm/min along the cutting path, which translates to a speed along each axis of 707mm/min, well within max permitted limits. So you could write G1 X100 Y100 F1400 and actually cut at that speed.

What Jonathan has said is that (using these figures) the max cutting speed along either axis is 1000mm/min, but in principle you could ask for higher cutting speeds and if you are cutting along a diagonal, you might achieve them. What is also relevant, maybe, is that it doesn't matter if you have different max speeds along each axis; Mach/LCNC will do the sums needed to make sure you don't exceed either max although this may mean that rapid moves will achieve different speeds over the bed depending on direction.

That all seems pretty straightforward, but it's the difference between G0 and G1 that I'm not quite clear about. G0 takes max speeds for each axis which you have defined for the machine, and will use them for G0 rapid moves and may even achieve speeds over the bed above the max for an axis if you happen to move along a diagonal. G1 needs a speed defined (either by setting it once in the gcode and using it as a default or by setting it in each G1 command) and will use this to calculate actual cutting speed whether along an axis or at some angle (or even around an arc), as long as max speed per axis is not exceeded. I think this means that if I stick a G0 move in at the top of my gcode file with a ridiculously fast F setting, then LinuxCNC will always give me the fastest rapids that preset max X and Y settings permit, but I should make sure that the F setting for g1 cuts is the actual cutting speed that I want. I use LinuxCNC, by the way, so maybe there is also a slight difference of interpretation compared to Mach 3.

As far as the original question is concerned, all this means is that you can look at max speed and accel along each axis independently in order to avoid missing steps and other nastiness, and actual speeds over the bed are just an accident of geometry and don't affect design and tuning parameters. If you are missing steps along X, say, during rapid feeds you might choose to reduce accel or max speed on that axis but you don't necessarily have to change Y if that is working fine with its current settings. However, if you get skipping during cuts, then it's a bit more complicated and you probably need to reduce the F cutting speed parameter just to reduce load as you can't tune this by axis.

And my apologies, guys - I was desperately trying to find an excuse for using nested integral signs and the odd greek symbol from the LaTeX library but just couldn't find a way to fit them in...

Jonathan
16-01-2014, 11:52 AM
So you could write G1 X100 Y1000 F1400 and actually cut at that speed.

I think you meant Y100 or X1000 there.

Otherwise, nicely summarised.


I think this means that if I stick a G0 move in at the top of my gcode file with a ridiculously fast F setting, then LinuxCNC will always give me the fastest rapids that preset max X and Y settings permit, but I should make sure that the F setting for g1 cuts is the actual cutting speed that I want.

The 'F' command does not affect G0 moves. By definition, G0 moves the machine as fast as permitted by the motor tuning, unless you've overridden it using the feedrate adjustment, in which case the speed would be lower.


As far as the original question is concerned, all this means is that you can look at max speed and accel along each axis independently in order to avoid missing steps and other nastiness, and actual speeds over the bed are just an accident of geometry and don't affect design and tuning parameters. If you are missing steps along X, say, during rapid feeds you might choose to reduce accel or max speed on that axis but you don't necessarily have to change Y if that is working fine with its current settings. However, if you get skipping during cuts, then it's a bit more complicated and you probably need to reduce the F cutting speed parameter just to reduce load as you can't tune this by axis.

Yes, that's correct.


And my apologies, guys - I was desperately trying to find an excuse for using nested integral signs and the odd greek symbol from the LaTeX library but just couldn't find a way to fit them in...

Here's an easy way to get some different symbols in... derive a formula to find the direction (angle) for which the machine can travel at the highest feedrate, for given X and Y rapid speeds.
It's just occurred to me that you could actually do something useful with that formula - potentially surface your bed faster by using a raster pass set to that angle.

Neale
16-01-2014, 12:10 PM
Ooh, an excuse for some calculus! That'll get 'em rolling in the aisles...

Thanks for pointing out typo, Jonathan - correction made.

On the F setting and G0, though - my comment was based on my experience with LinuxCNC, and I wonder if I have misremembered or LCNC works a little differently to Mach 3 (which seems from casual observation on this forum to be much more popular). If I type G0 commands into LCNC to move the machine "manually", my recollection is that it bitches if it has not seen a previous F setting in a G0 context. I tend to have to type something like "G0 X100 Y100 F1200". Once I've done that, I don't need the F parameter for subsequent G0 commands. In other words, it doesn't seem to default to the max machine speeds. The way you describe it makes much more sense and I wish it worked like that on my system but it doesn't seem to. I'll have to experiment - I might have missed something subtle.

Jonathan
16-01-2014, 12:19 PM
If I type G0 commands into LCNC to move the machine "manually", my recollection is that it bitches if it has not seen a previous F setting in a G0 context.

I don't remember... by all means check.
According to the definition here, it shouldn't be a problem:
G Codes (http://linuxcnc.org/docs/html/gcode/gcode.html#sec:G0)

Also notice that they mention what we just discussed - 'The maximum rapid traverse rate can be higher than the individual axes MAX_VELOCITY setting during a coordinated move.'

GEOFFREY
16-01-2014, 07:32 PM
I see you like your new toy. . . . Lol . . .(Shame Nobody got F'#$ing clue what your saying. .:hysterical:)

Hey Jazz, I was under the impression that you were into latex!!! G.

JAZZCNC
16-01-2014, 09:07 PM
Hey Jazz, I was under the impression that you were into latex!!! G.

Nah but I do have a pallet of Latex I can sell you.???. . . .Gloves. :hysterical: . . .( Also Nitrile and Vinyl if that's your fetish. . Lol)

itsmillertime
17-01-2014, 08:08 PM
Nah but I do have a pallet of Latex I can sell you.???. . . .Gloves. :hysterical: . . .( Also Nitrile and Vinyl if that's your fetish. . Lol)

OMG lol What did I just walk back into? :whistle::beer:

I wanted to update you on my progress with getting this jumping/ noise issue out of the way. Well... I figured that you guys Jazz I think mentioned it first that I was expecting too much and the speeds were too high with the rapid and cutting settings.

Well, Lo and behold after trying lots of different tuning combinations. I discovered setting the speeds too low also made for some horrible sounds. But I came across a another post with some supplier recommended settings. That were indeed a lot slower than what I had been gave.

112971129811299

These settings were actually found to be the best for my 6040 machine. And they are

steps velocity accel' step dir
X 320 2000 200 10 5
Y 320 2000 200 10 5
Z 320 800 150 10 5


My old settings that made the occasional noise and stalled the motors were

steps velocity accel' step dir
X 320 3000 200 3 1
Y 320 3000 200 3 1
Z 320 800 200 3 1


It sounds loads better now and doesn't fly about making my table shake slightly with the speed of movement of the gantry.

I have re-cut a few test pieces and haven't had the horrible sound or any missed steps.... yet ;-) Never say never eh!


I just want to say a really big thanks to all you guys for all of your invaluable advice I really do appreciate it and would have been stuck without all your help. cheers guys

Andy :-)

itsmillertime
18-01-2014, 07:33 PM
Well, Spoke to soon! Although the motor tuning appears to be better and it certainly still sounds better. I have had the same problem come back. The stalling and the engraving being out of line by about 10mm as though the motor stalled again (complete with grinding noise!)

I turned the power off and swapped over the X and Y axis connectors at the 'blue' controller breakout box. Looking to see if the fault would swap over the the X axis (My fault has always shown on the Y axis so far)

Lo and behold I rotate my workpiece by 90 degrees and power up and run the machine again. The error appears again exactly the same (but without any really noticeable grinding on the machines X axis. Apart from the engraved text appearing backwards the error is in the same place on the workpiece. This I think tells me that the problem is in my Y axis driver board? Plus my Z axis (at approx the same time as the initial Y axis error reappeared seemed to plunge deeper by about 2mm after/ same time as the Y axis stalled. Inside the breakout box the driver boards are separate.

Am I right in thinking that it can only be a crappy control box (mixed with maybe crappy unscreened -and not star earthed either..)

I'm now looking at replacing the existing chinese cable in the 6040 router with screened cable is this the best thing to do and any links to the best 4 core stuff to buy for the job please?

Also I am going to need to replace the break out box. What is the best and easiest stepper controller solution in the UK. Balancing cost and quality. I would stick with the stepper motors I have for now but probably need to upgrade this crappy control box.

Would really like to hear your suggestions guys.. :-)

Thanks again Andy

JAZZCNC
18-01-2014, 07:58 PM
Calm down Mr mannering don't go scraping things just yet.!! . . . Still some other things it could be and if you have separate drives there's half a chance it's just a setting or pulse issue.!

Looking at your motor tuning with 320 steps Per setting and presuming a 5mm pitch screw means the drives must or should be set to 1600 micro steps.
This is relatively high for these machines so It could just be your system can't handle that many pulses and your getting missed steps.?

So first we need to confirm this so can you give us a good quality picture of the drives and there wires and Dip switches if any.
Try and find out ther PSU voltage.
Next what is the Kernal speed set at.? You'll find it in ports & pins.

Show us more before spending any money.

itsmillertime
18-01-2014, 08:51 PM
Lol Cheers Jazz ;-)

Just got frustrated with trying to get it running consistently! I have just gone out and taken the case off and had a look inside.

It looks like the switches are set for 1600microsteps. From the PSU I can see on the label (DC output 24v and current 12.5A) outside case says drive voltage 24v DC drive current 3A.

The kernel speed is set at the lowest setting 25khz? I have attached some pics of the controller.113071130811309113101131111312113131131 4

Jonathan
18-01-2014, 08:58 PM
The kernel speed is set at the lowest setting 25khz? I have attached some pics of the controller.

320 step/mm * 2000 mm/min = 640,000 step/min
640,000/60=10.67kHz
10.67kHz is less than 25kHz ... so that shouldn't be a problem.

itsmillertime
18-01-2014, 09:08 PM
Thanks Jonathon for showing the workings of that, makes it a load easier to understand the reasons behind the settings. cheers andy

JAZZCNC
18-01-2014, 10:30 PM
320 step/mm * 2000 mm/min = 640,000 step/min
640,000/60=10.67kHz
10.67kHz is less than 25kHz ... so that shouldn't be a problem.

Jonathan I was more concerned that the Kernal speed was set too high which can cause strange happenings like this.!!. . . Long shot but it happens.

Andy.

First are you 100% sure there's no mechanical issues.? No binding etc. It won't take much stickage at these speeds with such low powered motors/drives to rob torque.

If not then next try cutting the same part but in air without spindle running. This will eliminate VFD/Spindle interference.

After that then carefully inspect wires at the motor connectors and at the BOB/drives for any lose connections.

The drives are better than some of these machines but still cheap low quality drives. That said they should do this job thou with just 24v you can't expect too much.

Again I'm clutching straws here but you could try setting the micro stepping to 400 on drives and see if it changes anything. Just do the problem Axis first to see if it alters any thing.
You'll need to go into Motor tuning and set the Steps Per to 80 and retune the motors. You'll see can get much higher velocity on that axis now but I wouldn't anyhigher than you have it now.

Failing any of this working and machine being mechanicly sound then scrap it all and get this MX3660 - 3-Axis DSP Based Digital Stepper Drive Max 60 VDC / 6.0A (http://www.zappautomation.co.uk/en/digital-stepper-drivers/979-mx3660-3-axis-dsp-based-digital-stepper-drive-max-60-vdc-60a.html) along with 48V PSU.
It will transform the machine and I'm pretty sure it will fit neatly into your existing box with only slight modification. It comes complete with Digital drives, built in BOB and Spindle speed control.
Rated upto 60V 6A and will easily run motors at 50-55Vdc with a unregulated PSU so if you want to upgrade motors to say 3nm at a later date then you can. . . . .Or take it with you to another bigger machine.!!

george uk
18-01-2014, 11:17 PM
Hi

My friend had a similar problem a few years ago, after a lot of messing about, we found a few important but simple things helped out.

1. Check your parallel cable, and both connections, ( someone on here already mentioned it ),
2. Try not to cross any of the cables for the axis, especially with the spindle cable and keep them away from estop cables. but try rout them away from each other, or at least insulate them a bit.

The version of Mach your using, did it come with the machine

3. Some of the Chinese suppliers send out hacked versions of mach ( can not remember the build number ), and its full of bugs,

2 years ago, when dealing with a problem on my friends machine, the hacked version of mach was buggy on the axis drives and spindle speed setup for slow spindle rates.

edit. just looked at your wireing again, you have 3 axis power cables running very close to your other wires, If your Y axis driver is the one on the right side, i would route them power cables elsware, or at least dont let them get close.

itsmillertime
18-01-2014, 11:24 PM
Cheers Jazz,

You have been pretty thorough with your suggestions and I do appreciate you taking time out to give your advice. I will go over the machine tomorrow. take off the suspect axis stepper motor and move the gantry by hand to see if it is binding / any resistance present.

As for possible interference from the spindle? I did think of this before. So I've ran g code with the spindle powered off/and in the air and its still stalled. Its very random when it happens.. but it happens frequently enough to prevent me working on anything expensive or intricate. Possibly happens once on every other job. But its enough to ruin whatever I'm doing!

I will physically swap over the x and y driver boards and see if the fault does indeed transpose to the x axis as I suspect. I can barely live with the one problem. But with the Z axis losing its zero. Its a step too far!! boom tish ;-)

Going to also swap out my parallel desktop cable for another one I have and change/ ensure my desktop is in standard PC mode not acpi..

Then I will try the changing steps as you suggest and if that doesn't work after inspecting all cables etc?

I will replace/upgrade all stepper cables and shield the vfd cables. Earthing all of the screens.

Still not working? I will be delighted in swapping out the controller with the one you suggest.

Having these issues are a great way to get to know my way about the machine so I view it as a learning exercise in getting to understand how it all works.

Will let you know how it goes tomorrow. wish me luck!

thanks again for all the help

andy

itsmillertime
18-01-2014, 11:32 PM
thanks George for your reply too. only noticed your reply after answering jazz so we just cross posted!

Yes the Mach that I have was supplied on a CD and I'm sure its been unlocked as it never asked me for a key. not sure of the version I will have to look that up tomorrow. Going to get a legit version of Mach but need to try and get this machine running first( if possible!) as this could really break my budget the way this is going!

thanks for the cable suggestions I'm going to give that a shot tomorrow.

will let you know if I find anything!

thanks andy

george uk
19-01-2014, 12:02 AM
looking again at your pictures, i would definably suggested routing the wires away from each other, especially the power cables away from the data cables. I would also check that anything supposed to be earthed to the case, is firm

I think mach3 comes in a trial version from there site, ( limited to 500 lines i think ) If your only running at 25kh, the hacked version might not affect you that much, in fact, it was only the spindle controll that we found buggy. Because of all the other things you have tried, its probably interference with the wires, or the driver board ( cheap to replace ) .

Mach3/4 well worth buying a licence for, the support and help you get on there is like the help you get on here, saves you hours of reading and headscratching

JAZZCNC
19-01-2014, 01:01 AM
looking again at your pictures, i would definably suggested routing the wires away from each other, especially the power cables away from the data cables. I would also check that anything supposed to be earthed to the case, is firm

It won't hurt doing this but I would be very surprised if this is anything to do with it has he's only running 24vdc. Only the 240V mains voltage will be worth routing signal wires away from has 24Vdc won't produce much noise.
When it's noise it's often a very slow process of missed steps or the complete opposite and it's tripping drives or E-stops etc. If the VFD isn't affecting it then wires being a little close to each other won't be.!

The Parallel Cable is always worth swapping out for another when having trouble and it will probably been me who you've seen suggest before George has i've seen it many times. Thou I don't think it's that in this case other wise would have siggested it.

Andy george is correct about the Hacked version of Mach3 and now knowing this I wouldn't do anything else untill you have un-installed it and replaced with the genuine and currant lock down version.

To be honest I'm still thinking it's a Mechanical issue with binding that's robbing the low amount of torque that's left at the higher feed rates. These motors are 4 wire motors so can run out of torque very quickly when rpm rises and with only 24V then there won't be very much torque to start with.!

itsmillertime
19-01-2014, 11:58 PM
Ok guys,

I spent a few hours this afternoon stripping down bits and taking the bed off and removing the stepper motors from the router trying to look for any mechanical binding etc along the Y - the gantry driving axis.

It doesn't sound particularly smooth but I dont have much to compare it to. The gantry moves quite freely by turning the leadscrew by hand. I can freely push the gantry across the length of the machine and back with a gentle concerted push by hand. It's definitely less resistive in the middle section and approx the last 4" at either end the resistance increases by about an extra 15% and the noise from the ballscrew is slightly more pronounced and it's just a little more 'jerky' towards the last 4-6 inches of travel.

I checked the travel and the resistance of the x axis and this is noticeably smoother to push but I think thats because it's lighter and better supported on the 2 bars and less to travel. It does make a similar noise to the Y axis it's just less 'clanky' and smoother.

I moved the gantry right up to the ends and looked for gaps/ being square with the frame. At one end of travel (near the stepper motor) it's pretty much bang on. But at the other end of of the router I measured the gap at approx 2mm in one corner with the other corner at 0mm flush with the gantry. 2mm is quite a noticeable gap across the end. I have attached a photo.

So what's the best way to work out if its true? Could this be causing some resistance? I guess measure across rails at one end and do the same at the other? I haven't removed the lead screw or anything yet as I didn't want to really disturb anything else until I asked here!!

Is there any easy way of doing this, maybe loosen up the rails and run the gantry from the one end to the other? Then tighten up the rails? I checked all the allen screws and all are tight.

I may have found a possible cause of the motor stalling. One of the grub screws on the Y axis coupler didnt feel very tight (possibly loose) and maybe causing the coupler to slip when the load got too much? Not 100% certain that is it. But it's the very same axis that has had the problems.

I guess what I'm asking is, is the noise my machine makes when moved by hand (pushed) normal? Should I expect any resistance at extremes of travel. And whats the best way of making sure its square/ true and not a bent ballscrew.

Where should I go from here please guys?

While I have the bed up and I'm looking for biding/mechanical issues I want to rule them out I guess. Also should there be any oil of grease on the ballscrew? No nipple or any obvious lube in place.


http://youtu.be/5mxulmQ8Ivo

Sorry that the video is 3 mins long but it shows the freedom of movement and sound of my y and x axis being moved by hand.

Below is a photo of the gantry with the gap approx 0mm/ flush at one side and 2mm at the other. I took measurements to show the difference in the gap.1132711328

11326

Thanks again for all your advice. Just hoping to rule out any big mechanical issues before going for drivers etc..!

andy

JAZZCNC
20-01-2014, 12:56 AM
Andy wouldn't worry about the 2mm gap it's just the frame thats out of square which doesn't mean a lot. More important that the x axis gantry is square to the Y axis rails and the y axis rails are parallel to each other.

The Y axis doesn't look to be binding too much but it does look and sound very dry so get some grease in the ballnut or at least bit of light lube oil on the screws.! . . . If the coupling was loose then it's 90% chance you've found your problem.

One thing I didn't like the sound of was the rattling when you grabbed the end of the ballscrew. Check if the ballscrew is floating back n forth by grabbing the end, there should be no movement.

If this is ok then I'd just grease and lube everything and put back together. If the ballscrews don't have a small flat on the end then file one on to give the grub screw a flat surface to grip on then Loctite the coupling grub screws.

richie00boy
20-01-2014, 10:54 PM
I had a bit of an issue with getting tighter at one end. I loosened the stepper bolts and coupling screws and bearing block and found the tightest spot then jiggled all the above to try and get it into the best position to bind the least. Seems to have done the job. I also had a slightly less tight coupling screw - wonder if it's a symptom of binding?

itsmillertime
13-02-2014, 11:10 PM
richie - thanks mate for your reply. It could be something like that causing a slight bind but I was unable to replicate any real significant difference in 'feel' during pushing the axis' by hand.

Sooooo, what did I do next? Well after getting really fed up with it all. I decided to take the advice of you guys especially jazz and his recommendations and I went shopping...

I bought MX3660 - 3-Axis DSP Based Digital Stepper Drive Max 60 VDC / 6.0A


SPS487. 48V unregulated power supply

3x SY60STH86-3008B Nema 23 Stepper motors

I also bought the screened cable recommended in another thread and rewired all 3 motors back to the control box. I’ve earthed all cables at the box end and replaced all components. It all fits nicely in my blue box too.

After all of this I have fired her up and she is running perfectly. No more stalling, missed steps, or dodgy inconsistent cuts!!!

Just want to say thanks to all you guys and to Jazz for his kit recommendations. So much smoother, quieter and working well! Now I finally have a machine I am happy to work with at last.

cheers guys. :-)

Got a few questions to ask about my setup to perfect it. But will ask them later. just wanted to let you guys know how it turned out!

thanks andy

JAZZCNC
13-02-2014, 11:41 PM
Great news and happy your sorted. . :yahoo:

Can't beat good electronics they make a world of differance and don't workout that much more expensive in the grand scheme of things when you consider you only spend the money wounce and they work perfectly so No spending weeks pulling hair.!!

itsmillertime
13-02-2014, 11:49 PM
lol ain't that the truth. Sat watching nervously while it did its first test cut after being rebuilt. Phew that was nerve wracking!

Glad I listened to the advice regards new electronics. The bigger motors and the electronics make a HUGE difference with the way its running now.Looking forward to actually making things rather than fixing them lol. Thanks for all the time you guys took for your advice.

andy