Frage Back to zero?

I'm sorry if this has been addressed before but if it was I couldn't find it.

I needed to make a hole grid on a 9mm plywood to serve as a baseplate for my S600, in order to hold a clamping system like this guy did.



This consisted of a 8x12 grid of 6mm and 12mm diameter holes, half of each. Also I needed to cut 9mm holes to a depth of 5mm concentric to the 6mm holes to house the clamp nuts. I used a 3mm flat end tool to mill everything out, feed 240mm/min, layers 1.5mm deep.

It all went well and nicely, with what I think were slow feed and gentle plunge in order to avoid problems.

First task was the 6mm holes, then another pass for the concentric 9mm holes, and finally the 12mm. Trouble was that the first batch was ok, but the concentric holes where not really so, and the 12mm ones were out of alignment.

At the end I sent the spindle back to the job 0,0,0 but the Y-axis was out by circa +5mm. X and Z axes were ok. Y-axis is well lubed and runs well from o to 590 with no funny noises. Both pillars are aligned.

Any comments please?

I can only think of obvious and generally unhelpful comments like:

"the machine has lost steps".

If the axes are indeed working well unloaded, an axis has met resistance so it's down to whether your toolpath and feeds/speeds parameters are sensible or not.

I.e. are you trying to cut too much material away in one of the passes?

Doug

Losing steps as in loose clutch slipping back? That's a problem as I had already made all tasks slow in order to avoid loads. Tomorrow I'll try running a test track from a o point slightly off from Home to the end limits and back to 0 to see how it recentres in no-load.

This was one of the tests that some 20 years ago I used to run on A0 pen plotters before we bought them

I would say the Y-axis belt is not adjusted properly. I tighten mine very strong.

Try also milling with the end-mill some mm above surface and see what will happen.

DocBrown schrieb: I would say the Y-axis belt is not adjusted properly. I tighten mine very strong.

Try also milling with the end-mill some mm above surface and see what will happen.

Agreed. The Stepcraft U.S. assembly video put me in fear of having too tight a belt which was a mistake. Another area where Stepcraft leaves the user to guess what a suitable tension is.

I then experienced big issues a couple of weeks ago until I put a decent amount of tension on the belt. Problem solved.

I have commented before but I was horrified to see on another of the Stepcraft U.S. setup videos which showed their XZ module stuck fast in the track extrusion so that it would only move when jerked fairly strongly! My gut feeling was that was nonsense so I tuned mine so that the assembly would run freely under gravity when the extrusion was tilted.

Doug

I would say the experience with mechanics is different for each user. That means for example "tighten a screw firmly" is very different if there is no reference or experience.
I've done modelmaking on ships, planes, helicopters, cars and railway-models for more than 35 years and that gave me a "feeling" what is correct and what not.

Maybe one step in the correct direction could be listing a torque value for each relevant screw. Don't know if that helps...
A second one could be a list of FAQ's and answers:
- I'm loosing steps, what to do..
- Circles are not round, what to do...
- my X-axis stuck...
- ...

Hartmut,

I think your idea is an excellent one - perhaps we need an independant "Wiki-step", whereby questions can be posted and relevant answers recorded. I know there is a CNC-wiki, unfortunately only in German, which could act as a model for "Wiki-step". Additional information such as videos and build pictures etc. etc. could be added a well. Your torque settings could be incorporated too.

Unfortunately someone would have to "police" such a site to ensure that only qualified answers were accepted, and to remove irrelevant or incorrect material and chit-chat. Then of course there would be a funding issue - I personally have no idea of the costs involved in maintaining such a website.

Any bright ideas on how to start such project?

Peter

Hi Peter,

my ideas are more adressed to Stepcraft themself. I'm not interested to do their quality assurance ...
So I like to help here from time to time but not with permanent commitment.

The idea with the torque settings can be simply implemented with no further effort from Stepcraft if done once. The wiki idea indeed needs permanent support and for sure during the start-phase it will be a lot of work.

The guy who`s hosting the German wiki (Speedo) could do that also in English if he likes...but he always ask for support...

Thank you for your input Gentlemen As a matter of fact the belt is taut enough though I cannot say how taut in BTUs/second but I know I'm not having it jumping teeth between both branches of the Y-axis.

I did a couple of test runs of my previous long toolpaths on no-load, and still I get differences between Zero at start and at end on Y-axis, the others being spot-on. To me it looks like I have to tune the entire Y running gear again, rollers and screw nut. This screw nut is very very sensitive as it is too easy to misalign and distort the Y-screws.

Hi Miguel

How annoying that the Y-axis needs tuning again - its a real pain.

When I tuned mine, I made sure that the moving rollers were all slackened off when I adjusted the end plates. That way I could be sure they were not the cause of any tightness. It's important to make sure the two gantry sides are equidistant from the end plates too (i.e. the X-axis is at 90 degrees to the Y-axis).

Having completed that adjustment, the rollers were then tightened individually so that they could just not be turned by hand (finger). I carefully degreased everything before doing this so I could feel the increase in resistance to turning as they were screwed down. Of course it was all lubricated again once I was happy with the adjustment.

I know Stepcraft suggest a 1/4 turn - IMHO that is a nonsense, it leads to grossly excessive forces and is likely to cause surface damage to the aluminium extrusions and the rollers themselves.

As a kinematic design that part of the system is very poor - in fact the rollers should cylindrical and not dished to match the extrusion. At least that would remove 4 unnecessary contacts and eliminate any twisting effect on the gantry sides.

Regards,

Peter

Thanks for the advice Peter, that's what I do in general terms, what I'm missing is the grease cleaning bit but I'll try that next time. That Y screw nut adjustment is a pain, it has to be checked at several points along its run. And next time I do that seriously I'll think of inserting a few retention washers while I'm at it.

Talking privately to someone else from here the impression I got was that the saying "You pay peanuts, you get monkeys" is valid but the fact remains that this is quite a substantial peanut both in capital terms and staggeringly so in time (equals money!) terms due to assembly and maintenance. Especially for me who's trying to start a business up where this CNC voodoo is paramount and here I am wasting my precious time and leaving behind other tasks equally important.

Again, I am convinced that the theoretical design of the Clockwork Orange is quite good for the spec, but production and commercially wise not quite so. I find it infuriating that having received a 4th-axis assembly it was completely jammed and I had to "open the lid, shake, and put the lid back in place" for it to work at all! Don't those people simply test their stuff before shipping out?

Oh, and give thanks for one certain representative who shall remain nameless, if it weren't for him many people me included would be left in very deep lurch.

Miguel,

I think the whole of the English speaking Stepcraft community owes a vote of thanks to the nameless one!!!!

I feel for you having to waste your precious time fettling your machine when it should be earning its keep.

Good luck with your next "Battle of the Y's"!!

Peter

I like peanuts. Well, peanut butter anyway. I reckon Stepcraft is more Nutella, or perhaps peanut butter with palm oil (which I dislike) rather than peanut oil.

Meanwhile back at the ranch, I received the promised small industrial quality spindle from Rory today. It looks great but I have no power supply cable or time to test it until at least Tuesday next week.

Here is a short video of me jog testing all three axes with it fitted. It weighs 1.5 kg and stickout is 30 mm below the base plate.



Doug

I got the impression from StoneyCNC that my troubles were to be expected because of the entry level of the 'hobby' machine. I build ukulele fora living and my entry level instrument costs £425... a little bit different to the near £2000 I paid for my 600 which after 2 days of assembly and reassembly twice I still cannot get to work. My customers get a replacement if they are dissatisfied regardless of them spending £425 or £2500... I am very frustrated that the support always defers to advice to watch the assembly videos with the implication that you haven't done it right so try again. I am to say the least disillusioned and disappointed with the machine and supplier.

Hang on there, Pete. As you can read in this and other threads I myself had immense trouble in getting the machine to work acceptably but I am getting there step by step. Of course that's not the best way for anybody to do business, much less when the customer is - like you and me - a commercial concern who needs to get finished products out of the door.

My thinking on this is that I have invested too much money and precious time to throw the bath water with the baby still in it, I'm not going to waste all my investment by going out after a different mousetrap which may not turn out better after all, especially in this el-cheapo bracket of the CNC router market. So I'm sticking with what I have and do the best I can. If I have to insert multiple calibration points in my toolpaths then that's what I'll do.

If you and your colleague can be as stubborn and punctilious as I had to be, you'll get rewarded with a surprisingly good machine in the end (but keep careful watch all the adjustments so it won't jam on you!)

In my experience there is no cause to blame StoneyCNC, he's not the source of your troubles, the problem lies up-stream from him. Quite the opposite, as I said elsewhere he's the one who saved me from an expensive failure. I just wish he'd be more like St. Patrick and expel the snakes out of Stepcraft

Doug schrieb: Meanwhile back at the ranch, I received the promised small industrial quality spindle from Rory today. It looks great but I have no power supply cable or time to test it until at least Tuesday next week.

Doug

Holy Shit, Doug! Rory has dragged you into the Dark Side of the Kraft

There is some good advice for you there Pete. I punted just over £3,000 for my "Steppi" with all the bells and whistles at the time except an automatic tool changer and several times have angrily regretted it, thought about selling it and a guitar and upgrading to a £5,000 commercial quality machine. However, I have persisted because I have mellowed a lot with age and I now believe that I understand its capabilities and inherent flaws well enough to appreciate that it WILL do what I require of it once the system is tuned and optimised.

Stick with it. It's easy for me to say I suppose because this is a long term hobby for me, not a commercial enterprise.

I'm not just saying this to be sycophantic to the main player here but Rory has been fabulous, just fabulous. Look at that amazing quality spindle he has offered to let me evaluate because of the repeat issues I had with the Stepcraft system. That's above and beyond normal customer service. He really cares and that's rare in my experience. Normally, people just want to sell you stuff and the after sales care is offered grudgingly.

Doug

Picking up on the thread where I'd left it, I decided to repeat that long test I had done before, the one where that had finish with a difference of several mm in Y-axis. First I gave the Y assembly a good check, and there was an difference of less than 1mm between the gantry pillars. This was annoying because that's one of the things I check often but there it was, so I exposed the belt which was quite taut but I removed it anyway so I could adjust both screws independently.

After enough head scratching the point was that the grubscrew on the farther assembly had come loose a couple of mil and the sprocket had a lot of slack! How did this come about I don't know, and I can't believe some Stepcraft agent came into my shop to sabotage my sprocket, so I'll assume it was me who failed to adequately screw the thing in during assembly.

After putting everything back into its proper place, I ran the aforesaid test, and I'm happy to say that 0-0-0 at start and at finish were satisfactorily coincident!

So one up for Stepcraft!

MMorao schrieb: ...and I can't believe some Stepcraft agent came into my shop to sabotage my sprocket, so I'll assume it was me who failed to adequately screw the thing in during assembly.

Don't beat yourself up, we had similar issues with slipping pulleys. (Parts would be uniform and consistent but the aspect ratio x:y would be out as the y pulley grub screws slid up to opposite edges of their recesses. Erm, probably.)

This is why:-
(a) Splines on shafts = Very Good
(b) Woodruff Keys = Good
(c) M4 Grub Screw on circular shaft = Bad
(d) M3 Grub Screw on circular shaft with pulley made from cottage cheese = Utterly pathetic, and shouldn't have made it past prototype.

Just my thoughts...

Owen S.