Frage Drift

We have eventually got our CNC working and need some advice. We are cutting circular arc 1.45mm thick through 120 degrees - imagine a cake slice with the edge peeled off. Despite checking our drawings for accuracy the pieces are tapered - from 1.45 to 1.25. Also the taper is not gradual - for about 2/3 of the length it is narrowing only fractionally...

We have positioned the cut at the edge of the table and in the centre and the amount of taper is the same. Is this a 'backlash' issue? A fellow ukulele maker who supplied me these also has the same problem.

Pete

Last I read, you threw your toys out of the pram big style and were offering to sell your Stepcraft at a knockdown price, now you're back asking advice...

My advice is that it's not adjusted properly but you provide no details of what you've done during your assembly and set up efforts so we can only speculate. Have you trammed the XZ axis so that it is level with the machine bed for example?

Doug, what a Philistine you are See St Luke 15:7

I say unto you, that likewise joy shall be in heaven over one sinner that repenteth, more than over ninety and nine just persons, which need no repentance.

Pete, be patient and check everything is parallel or perpendicular, and all you brass rollers have no slack but are not excessively tightened either. Also the endless screw or whatever their name in English is move freely and do not jam at extreme positions.

Now that I have ironed all the bugs out (sez I, but Murphy's Law is insidious) it will be a long time before I get rid of it

Can't resist having my two pennyworth here!!

If the taper to your part appears only over part of the toolpath, I'd bet a pound to a penny that it is a backlash problem.

You'll need to set up a simple engraving test pattern like this :-



X and Y amplitudes are set to 1mm. Measure the actual amplitudes (strong magnifying glass needed!!) and adjust backlash in X and Y axis setup pages. The compensation acceleration needs to be set as well - I think its set to 50% of the normal acceleration value. Don't forget to enable backlash compensation!!


Peter

OK, I'll bite. I've heard of whiplash, but thought backlash was when a newbie forum member asks a stupid question (just kidding). So, to be lazy and not search the web, would someone explain backlash, and why I should be concerned. I can almost bet on the ones who will respond, and my sincere thanks to you guys or gals.

PeterG, I had no idea you were a Crypto-Nazi

Well disguised eh!!!

To answer questions and ask some...

Yes I was about to sell and may still do so. However, I paid Rory £360 to come and help and had confirmed that certain parts were not right but that we had put the machine together OK.

We eventually got it into commission this week and over the past few days, it has slowly begun to create ovals instead of circles, uneven and tapered lines. Also, after 200mm from the zero position there is a difference in the sound the gantry makes as it advances along the y axis. This is announced by a click and then a change from a quiet rasping noise to smooth running.

We've managed all the programming issues and stuff. We just cannot seem to get consistency. For instance - we cut 5 identical pieces and thought that the machine could carry on. On the sixth piece it moved it's zero position 20mm so all the pockets and holes were milled off. Nothing had changed, there was no power outage or anything like it...

We are dispirited...

do you use mach3?

We use the step craft recommended UCCNC software.

Pete Howlett schrieb: To answer questions and ask some...

Yes I was about to sell and may still do so. However, I paid Rory £360 to come and help and had confirmed that certain parts were not right but that we had put the machine together OK.

We eventually got it into commission this week and over the past few days, it has slowly begun to create ovals instead of circles, uneven and tapered lines. Also, after 200mm from the zero position there is a difference in the sound the gantry makes as it advances along the y axis. This is announced by a click and then a change from a quiet rasping noise to smooth running.

We've managed all the programming issues and stuff. We just cannot seem to get consistency. For instance - we cut 5 identical pieces and thought that the machine could carry on. On the sixth piece it moved it's zero position 20mm so all the pockets and holes were milled off. Nothing had changed, there was no power outage or anything like it...

We are dispirited...

And your conclusion is to sell a faulty machine on to somebody else?

Oh well, mine is working well and I'm on my way to creating a Brian May Red Special replica.

Doug

Pete Howlett schrieb: We use the step craft recommended UCCNC software.

Please share your cut parameters.

Depth, feed speed, tool diameter, what spindle you use.

Also, have you tried "Exact mode"?

Due to aproximations, constant mode won't cut true circles, but exact mode will.

I got no idea if it is the machine or me Doug. Rory's memorable remark when he came to help us out was, "It's a tinkerer's machine..." This did not give me a lot of confidence and I am sure that someone with more intelligence and patience than me can do it justice. With the time and effort we have put into it I could have bought purfling and rosettes for my last 180 instruments!

The attached files show the nature of the problem. IS this software or hardware? You tell me...

Thanks everyone for your help. We gave it one last chance today and after sending images to Rory confirmed what we suspected - the x axis was doing something peculiar. So following your advice we started to do a physical check and discovered the brass coupling between the motor and lead screw was loose. Access to this meant stripping the gantry so we drilled an access hole, tightened the grub screw and hey presto, concentric circles.

We are going to persevere and I suspect there will be many more trials. HI have to say that when it works it's fantastic. So thanks again. I'll try and post some images of the inlay work we are doing..

Hi Pete.

I'm glad you're on your way and that the fix was simple. In a way that can be a huge relief but also truly maddening. I chose to apply a small quantity of cyanoacrylate superglue to the grub screw on my x axis after it vibrated loose during early trial runs. Thankfully the problem was obvious. I was tempted to do the same with the Z axis grub screw but it has held firm so far - surprising give the amount of vibration when drilling at low Z feed rates I have experienced. I also changed the grub screws to stainless steel items and flattened off the pointed end in the hope of gaining a better contact in the lead screw recess.

I am looking forward to seeing the results of your inlay work. I am fascinated by this - what a great way to put a CNC machine to use. Get a quality result on an intricate design and take the manual labour out of it that can be put to better use.

My next effort on the luthiery front will likely be a 3D cut of a long tenon guitar neck from (initially) Khaya hardwood. Rory is pessimistic about this working, but my growing confidence with the Stepcraft leads me to believe that it is not only possible, but a good result requiring minimal hand finishing will be realistically obtainable. My final cut will be on a vintage Cuban mahogany fireplace surround which cost me £150 in total, so I will not attempt it until I have a high degree of confidence that the Stepcraft can execute the job. I don't care if it takes five test cuts on medium quality tonewood before I go for it - failure is not an option here.

Apologies for getting shirty with you during your difficulties; I'm afraid that I have become increasingly annoyed and disillusioned with all the insight and advice I have offered to others on guitar forums who then take it for granted, bugger off and ignore you or simply use the particular forum selfishly then don't contribute anything in return.

Doug

Don't worry Doug - it's comforting to know you had the same problem... I have guitar size neck blanks in Honduran Mahogany plus some made up blanks for sale. I also have a neck carving DVD - takes me about 40 minutes to fully hand carve a neck. This is about how much time it would take to CNC it. Got to Taylor's 'factory Friday' series and you will see that they DON'T use CNC router set-up for neck carving per-se. They use an incredible turret style shaper - much faster. When I visited Hawaii recently, Kamaka had a geat CNC set-up which was 5 axis working on the principle of a lathe to present the work at the various angles to the tools.

Best of luck with it - it will be a pretty much academic exercise with an expensive long reach bull-nose cutter...

Pete Howlett schrieb: Kamaka had a geat CNC set-up which was 5 axis working on the principle of a lathe to present the work at the various angles to the tools.

Best of luck with it - it will be a pretty much academic exercise with an expensive long reach bull-nose cutter...

Interesting. My original plan was to use the fourth axis rotary table to rotate the blank and thus avoid the need for an expensive long reach ball nose cutter However, MeshCAM didn't seem to handle this processing too well (it only indexes anyway) and because it isn't smart enough to know what it has already cut on a previous index you have to mess around with varying the depths of cut manually.

In the end, it seemed easier to just do a two sided cut. I did indeed waste a lot of money on a 100 mm long, 38 mm flute length 4 flute ball nose cutter then discovered during some smoky trial runs on Khaya offcuts that I should be using two flute cutters for timber anyway and the ball nose cutters are not for roughing.

My latest draft plan involves a two sided cut with a 5 mm 2 flute end mill for roughing on the top (fretboard side) which has a 5 mm rounded bottom truss rod channel then finish with a 5 mm 2 flute ball nose cutter. Maximum cut depth 15 mm.

For the bottom side, roughing with a 6 mm 2 flute end mill then finish with a long series 6 mm 2 flute ball nose cutter to a maximum depth of 24 mm. To reach the full depth (38 mm) on the central rounded section, I think I will just remove the uncut material at the sides of the blank then run a second toolpath with a machining region restricted to this (i.e. avoiding the tenon end and headstock). I think this will work out o.k.

It's all a bit of a faff really but the Brian May neck does not have a uniform C or U profile along it; it evolves towards the tenon whereupon it blends into the rectangular section. Then at the headstock end, there is an enigmatic shaped bump as the fat neck transitions into the lozenge shaped headstock which Gibson Les Paul enthusiasts might refer to as a volute This would be difficult for somebody with no fine woodworking hand-eye co-ordination experience like me.

Perhaps after I have knife and forked my way through one CNC cut, light bulbs will illuminate and I will find more efficient ways of doing it.

Doug

Our inlay below.

Machine not 100% accurate again tho- the left leg of Snowdon is narrower than the right and we now have slightly oblate holes and a 'step' but identifying the problem is proving much harder this time. We only ran it for a few hours too...

Holy cow, your exotic woods leave me jumping mad! Over here if I want to buy a piece of anything they'll only sell me the entire log!!!

Can any UKer or thereabouts provide sources for fine woods in 'modelling quantities' and who'll be willing to sell overseas? Online sales preferably? I'd like to try my hand cutters at doll furniture.