Further Adventures with the Sieg KX3 & KX1

[Ketan SwaliParticipant @ketanswali79440]

Posted by mick on 01/02/2022 17:06:41:

First off thanks for everyone's input. I have to confess that most of the electronic stuff is way outside my comfort zone. I did say earlier on that the stepper rotated when disconnected from the machine, I tried it again this morning and I can't stop it rotating with hand pressure which would indicate that there's torque there. So with the slide moving freely with only light hand pressure without the ball screw connected, but appreciably stiffer when it is connected, although it will move with only finger and thumb pressure on the ball screw square, but the jog won't move it, could it be this torque adjuster brain cell that Adam was talking about a few posts back?

Apologies, I failed to properly understand what Adam said.

If you were unable to stop the motor by your hand, then issue maybe mechanical.

Between the motor and the ball screw is the Oldham coupling with an acetal disc in-between the couplers. Here is how the assembly works: link

is the disc worn?. Technical details of the disc can be found here: link

Ketan at ARC

[Michael GilliganParticipant @michaelgilligan61133]

At the risk of stating the obvious …

Oldham couplings are excellent when the input and output shafts are parallel but not concentric

… they struggle when there is angular mis-alignment.

MichaelG.

[Martin ConnellyParticipant @martinconnelly55370]

I suppose the Oldham coupling alignment could be checked by just loosening the motor on its mount and seeing if that eases any binding.

Martin C

[Michael GilliganParticipant @michaelgilligan61133]

[Ketan SwaliParticipant @ketanswali79440]

Posted by Michael Gilligan on 01/02/2022 18:04:21:

At the risk of stating the obvious …

Oldham couplings are excellent when the input and output shafts are parallel but not concentric

… they struggle when there is angular mis-alignment.

MichaelG.

Agreed Michael.

• At present, I am only presuming that an Oldham coupling is fitted.
• if an Oldham coupling was fitted, then the probability of angular mis-alignment during assembly installation were/are low.
• It has lasted over ten years without issue. An event which could cause angular mis-alignment would have been some kind of hard shock, which could bend motor shaft/ball screw shaft, or more likely break/damage the acetal disc.
• the reason why HUCO Oldham coupling were requested and supplied by ARC to the factory at the time was that these were preferred by us over flexi coupling which the factory used on all other machines. In original tests, we found that a mis-alignment issue caused the end of the ball screw to snap due to the force/strain placed on it with flexi coupling. In addition, the movement got strained, leading to inaccurate cuts being made, for example when cutting a circle. The extra strain would cause a defect in the cut… failing to cut a proper circle in certain cases where flexi coupling was used. This has been seen on several hobby CNC machines, including a well known ‘American’ hobby CNC machine, presented as problem on a negative YT video, without anyone really realising that strain put on a ball screw by a flexi coupling was the real cause of an inaccurate circle cut.

At present I am only trying to help Mick to check and deal with obvious mechanical issues which I can remember about.

Mick says that he is unable to stop the motor with his hand, so this suggests that there is more than enough torque coming off the motor…. much more than hand turning the ball screw. For this reason, I suggested to look at the acetal disc.

Ketan at ARC

[JasonBModerator @jasonb]

Is it possible that while adjusting the gibs the coupling was undone and has gone back together with too small a gap between the faces and they are tight on the acetal disc once the motor is screwed to it's mounting.?

Similar problem if the end of the ball screw is too close to the end of the motor shaft.

Edited By JasonB on 01/02/2022 19:44:27

[Michael GilliganParticipant @michaelgilligan61133]

Posted by Ketan Swali on 01/02/2022 19:00:10:

Posted by Michael Gilligan on 01/02/2022 18:04:21:

At the risk of stating the obvious …

Oldham couplings are excellent when the input and output shafts are parallel but not concentric

… they struggle when there is angular mis-alignment.

MichaelG.

Agreed Michael.

• At present, I am only presuming that an Oldham coupling is fitted.
• if an Oldham coupling was fitted, then the probability of angular mis-alignment during assembly installation were/are low.
• It has lasted over ten years without issue. […]

At present I am only trying to help Mick to check and deal with obvious mechanical issues which I can remember about.

[…]

.

My post was intended to be entirely non-confrontational, Ketan

… Just a comment for Mick’s benefit
… apologies if it appeared otherwise

If he has dismantled things, there is always the risk that they have not gone back together exactly as originally built.

That’s why I supported Martin’s suggestion

MichaelG.

[Ketan SwaliParticipant @ketanswali79440]

Hey Michael,

I didn’t take your post to be confrontational at all.

I know you are trying to help. I was just playing through the possibility of angular mis-alignment.

Ketan at ARC

[mickParticipant @mick65121]

I've had a day off to do some gardening but tomorrow I'll check the coupling as its been removed a few times now and I assumed that all the time the drivers on the motor and screw hadn't been altered the plastic drive plate would have gone back in the same alignment.

[mickParticipant @mick65121]

With the gib adjusted to allow free travel with minimum hand pressure the ball screw is re-connected, which requires more effort to move the slide to both limits, reconnect stepper motor and its a bit stiffer again while the table won't move in either direction on jog, but I can still move the table with finger and thumb pressure on the ball screw square, which must suggest that the coupling isn't impeding the rotation of the ball screw. slacken gib by releasing tension on the right hand adjusting screw and increasing on the left, this is now pushing the tapered end of the gib back, but the slide now has sideways play and will only jog a short distance in the + quadrant but no movement back to the – quadrant. with the gib completely removed there's free jog in the + plus quadrant but again no jog movement in the -. Either jogging with the gib removed in the – quadrant is somehow introducing a slight twist to the slide which is making the rear dovetails jam or there's a electronic issue with the jog in the minus. As all this started when I removed the gib to clean the slideways the only situation I can come up with is that after thirteen years and over 2,800 hours of work there is a degree of wear to the central areas of the dovetails, which, when the gib is reinstalled and adjusted to give free movement and no sideways play is bearing on an slightly shallow area of wear and causing the gib to jam against the slightly higher areas where there is less wear, but even so I don't see why this should affect the jog function as I would have thought there would be sufficient torque to overcome this. I'm probably missing something quite fundamental but my options are rapidly running out as I think I've done everything mechanically I can think of that on a manual machine would have proved effective.

[Adam StevensonParticipant @adamstevenson91624]

Mick, is the motor (eletrics) on when you are trying to move by pushing the bed?

[Martin ConnellyParticipant @martinconnelly55370]

Is the gib strip tapered and moving when the bed moves? It could wedge the bed if this is the case.

Martin C

[JasonBModerator @jasonb]

As Martin says a tapered gib that is not restrained at either end will wedge tight when the table moves one may and loosen off when moved the other.

If you want to set the gib fit loose you need to slacken the screw at the thick end and ALSO tighten the one at the thin end to push it up against the other screw so the gib can't move

[mickParticipant @mick65121]

When the gib is adjusted by means of moving the two adjusting screws in which ever direction is required to allow free travel by light hand pressure to both limits and with no side way movement, the gib is held captive by the two adjusting screws at either end and therefore cannot move as the slide passes. The gib can only be fitted in one direction, that is inserted from the right hand side. The power to the stepper is off when the I move the slide by hand, as this allows the motor to freely rotate. After too many years that I care to remember I'm reasonably confident that I can recognise and repair most mechanical problems that manifest themselves on most conventional machine tools, its the addition of the electronic dimension that clouds the issue for me. Which doesn't explain why the slide will move freely when disconnected from the ball screw, but jams and won't jog when the ball screw and stepper are connected.

[mickParticipant @mick65121]

If anyone else experiences the problems I've been trying to get to the bottom of I would suggest the first thing to check might be the coupling clamping screw on the ball screw. Once tight full jog is restored and might also explain the program stalling when cutting in the direction of X- . But it doesn't explain why I still have the problem of the spindle over shooting absolute zero in the Y axis when go to zero is clicked.

[John HaineParticipant @johnhaine32865]

Mick, good to hear that you have found and fixed one problem.

As to the other, are you saying that if you select machine coords and click "ref all home", then the Y axis hits the end stop without seeing the limit switch?

If so, have you tried jogging it well away from zero, using the MDI to put on a long, slow move towards zero, and checking that if you manually operate the limit switch it stops the machine? If not that would perhaps indicate a wiring fault, OR that something has corrupted the ports&pins configuration that assigns the limit switch.

If I recall correctly, Mach 3 stores a backup copy of its configuration in an XML file every time it closes down, in a folder within the main Mach 3 folder, possibly within a further sub-folder called something like "XML Backup". The folder gets filled up with files each of which has a date tag.

If you look in that folder, can you find a file dating back to when you can remember the machine working correctly? Don't do anything with it yet, but if it is a config corruption problem this could give you a way to restore it to a working state.

[mickParticipant @mick65121]

Hi. John.

Perhaps I haven't explained myself very well. At the end of a program or cycle I return the spindle to the work piece zero from where all the dimensions are programmed. The x axis returns to this position while the spindle over shoots by about 10mm in the Y- quadrant.

[JasonBModerator @jasonb]

Mick is this a Zero you have set for the particular item eg one corner or the ctr.

Or is it the home zero which is done by a proximity switch?

[mickParticipant @mick65121]

Hi. Jason. Yes the zero position is always a point on the component. This is one of the conditions I was experiencing that prompted me to start this process. However today it didn't happen once, but a new condition appeared on the Z axis. The first feed depth move is always 2mm deeper than programmed, i.e. Z-0.5 is Z-2.5. No matter how many times the offset was re-established I had the same situation. The only way round it was to program the first two depth moves as + values. I can now use the machine as the problem on the X axis is resolved, but I'm just waiting for the next control problem to rear its head!

[JasonBModerator @jasonb]

I got a cry for HELP a couple of weeks ago, it would seem that Graham Corry's loose piece that produces the name on his Retlas engine's base casting had been lost and could I help with a replacement.

After a bit of playing about with Alibre and then F360 I had about 100,000 lines of G-code on a USB stick which I promptly stuck in the KX3 together with a piece of 1/4" brass and after 103mins this is what I ended up with.

I think I will put it on a chain so he can wear it around his neck which may also help prevent this one getting lost too

[mickParticipant @mick65121]

If anyone's interested the problem was a 2.5 value in the gauge block window. The question still has to be how it got there as it needs to be manually entered and as its a feature of the control that I've never seen the point of and have never used in the thirteen years I've operated the machine.

[John HaineParticipant @johnhaine32865]

Important for auto tool zeroing.

[mickParticipant @mick65121]

We all have our own ways of doing things, for the past thirty plus years I've used a piece of paper positioned beneath the cutter face, when it just starts to grip that's the established TLO, which works for me.

[JasonBModerator @jasonb]

I gave the part I made in my previous post a quick bead blast and was happy that all the toolmarks have gone, goes to show that they are more visual than physical and that the tool is going up and down at a consistent height.

[James Mitchell 3Participant @jamesmitchell3]

Great work! Nice to see what these little machines are capable of!

[Author]

Posts