Simond Posted December 4, 2016 Share Posted December 4, 2016 Hi Thinking of replacing the stepper controller in my laser - please see the CO2 thread at http://www.rmweb.co.uk/community/index.php?/topic/110058-co2-laser-the-learning-curve/?p=2524596 Comments very welcome! Thanks Simon Link to post Share on other sites More sharing options...
Simond Posted December 5, 2016 Author Share Posted December 5, 2016 Thanks for this, it's pretty clear that the motors are either missing steps, or the controller board is somehow losing its position. Curiously it only happens in x, which is lighter and therefore less inertia. I'm not sure how either would occur in a system that was properly set up. There are no apparent tight spots or interference, so unless the motor is actually failing, I'd be surprised if they were the cause, although it might be a quick (and cheap) fix. And, if a motor is failing, it will have to be fixed anyway. And if one is dying, the other probably will too The other problem, that of the Newlydraw, which is locked by a USB dongle, and is a very poor piece of software, and the dodgy USB link to the machine, and the lack of power control through software all conspire to make me think that a replacement controller might well be a good investment. And if I can crack the setup and operation of a controller board, a further £200 will sort out numeric control on a mini-miller too! No rush, not going to happen this side of New Year anyway! Best Simon Link to post Share on other sites More sharing options...
Simond Posted December 5, 2016 Author Share Posted December 5, 2016 thanks Newlydraw is.... not very user friendly. It was supplied as standard by all the Chinese laser suppliers a few years back. I do not believe the controller board will work with other s/w so... I suppose I could try it with Visicut, as it's free. best Simon Link to post Share on other sites More sharing options...
raymw Posted December 5, 2016 Share Posted December 5, 2016 If the stepper motors are apparently gaining steps - e.g. sw says go 20mm, laser moves 20.5mm, then it is possibly a form of 'backlash' in the mechanism. As the steppers are running as open loop system, the software has no idea as where the positions are, it simply sends move 100steps, or so, and hopes the mech. does as told. If the movement depends on toothed belts, then check the adjustment/tension. Toothed belts are not so good with steppers. USB should be OK, but only if decent cables and connections and proper ground connections. I expect, it will be better to scrap the machine, and start over. It will have been built down to a price, and it seems that not much of it is any good now, and you don't sound too happy, nor seem to have any idea as to where to begin detailed trouble shooting, or have the tools to do so. I can see you spending money and time needlessly on replacing components, and then still ending up with something you are unhappy with. I think you mentioned somewhere about contacting HTC - if you bought it from them, then get them to fix it. But, I don't think you bought it from them. Link to post Share on other sites More sharing options...
Simond Posted December 5, 2016 Author Share Posted December 5, 2016 Ray, It's clear that the system is open loop, and that is one of the key issues, this machine does not have any form of zero calibration. I'm not at all sure there's anything inherently wrong with steppers and toothed belts, practically every industrial plotter, laser, and a good many other devices use this combination. Equally, I doubt it is a backlash issue, the whole image is shifted, accurately, but not predictably, when multiple passes are made. The belt tension is within the specified range. USB has known issues with interrupts, and with microsoft updating things when you don't want them to. I have managed to avoid the latter by isolating the rather old laptop on which the ND software runs. It's not connected to the outside world except by the pen drives on which I transfer drawings. I'm not sufficiently knowledgeable about the inner workings of computer interrupts to sort anything out in that area, but I don't have an issue with it afaik. In terms of detailed troubleshooting, I have an oscilloscope, and access to just about any measuring kit I might need. It's a question of priority. Do I want to spend days trying to understand why a bit of kit doesn't work reliably (and intermittent failures are the bane of any problem resolution approach) or £130 on upgrading a £350 second hand machine to modern standards? The latter sounds like very good value to me, as a replacement from HPC would be £2150 if they had any in stock. And I'd rather spend my spare time modelling. The machine itself, mechanically, seems reasonably well made, is solid and functions quite smoothly. The laser and optics all work as they're supposed to. Replacement seems unnecessarily expensive, and isn't in the budget. As I said, it was bought second hand. HPC were rather disinterested when I called them. They couldn't/wouldn't sell me a replacement controller, and were not hugely helpful in terms of advice. I won't be racing to buy there, in the unlikely event that I ever find myself in the market for a new machine. The Smoothieboard offers LAN connectivity, compatibility with (free & pay) software other than NewlyDraw, and zero point detection on all axes. It also offers direct software control of laser power, so potentially improves the capability of the machine considerably. What I'm hoping to find is someone who has tried it, and confirm that it delivers what it offers. Best Simon Link to post Share on other sites More sharing options...
RMweb Premium nick_bastable Posted December 5, 2016 RMweb Premium Share Posted December 5, 2016 Ray, It's clear that the system is open loop, and that is one of the key issues, this machine does not have any form of zero calibration. I'm not at all sure there's anything inherently wrong with steppers and toothed belts, practically every industrial plotter, laser, and a good many other devices use this combination. Equally, I doubt it is a backlash issue, the whole image is shifted, accurately, but not predictably, when multiple passes are made. The belt tension is within the specified range. USB has known issues with interrupts, and with microsoft updating things when you don't want them to. I have managed to avoid the latter by isolating the rather old laptop on which the ND software runs. It's not connected to the outside world except by the pen drives on which I transfer drawings. I'm not sufficiently knowledgeable about the inner workings of computer interrupts to sort anything out in that area, but I don't have an issue with it afaik. In terms of detailed troubleshooting, I have an oscilloscope, and access to just about any measuring kit I might need. It's a question of priority. Do I want to spend days trying to understand why a bit of kit doesn't work reliably (and intermittent failures are the bane of any problem resolution approach) or £130 on upgrading a £350 second hand machine to modern standards? The latter sounds like very good value to me, as a replacement from HPC would be £2150 if they had any in stock. And I'd rather spend my spare time modelling. The machine itself, mechanically, seems reasonably well made, is solid and functions quite smoothly. The laser and optics all work as they're supposed to. Replacement seems unnecessarily expensive, and isn't in the budget. As I said, it was bought second hand. HPC were rather disinterested when I called them. They couldn't/wouldn't sell me a replacement controller, and were not hugely helpful in terms of advice. I won't be racing to buy there, in the unlikely event that I ever find myself in the market for a new machine. The Smoothieboard offers LAN connectivity, compatibility with (free & pay) software other than NewlyDraw, and zero point detection on all axes. It also offers direct software control of laser power, so potentially improves the capability of the machine considerably. What I'm hoping to find is someone who has tried it, and confirm that it delivers what it offers. Best Simon are you sure there is no float in any bearings ? Nick Link to post Share on other sites More sharing options...
RMweb Gold SHMD Posted December 5, 2016 RMweb Gold Share Posted December 5, 2016 Is the supply voltage to Driver board correct? (Do you know what it should be?) Any drop in supply volts will have a dramatic effect on the Stepper's ability to maintain position at speed. Can you specify/try slower speeds - even if you have to turn the LASER off. Does the error only occur in one direction? Set up a run (say) 50mm out and then 50mm back and repeat many times. Which direction does the error head towards or does it cancel/average out? Kev. Link to post Share on other sites More sharing options...
Simond Posted December 5, 2016 Author Share Posted December 5, 2016 Nick, No, I'm not sure... but it passes the "gentle shake" test. I'd have to dig more to ascertain the looseness, or lack thereof. Kev, No, no idea*, I agree I can turn down the speed, but if I don't have the laser on, I won't know whether it has mis-stepped. I could turn it down low. It seems to be one direction only, the X-axis, moving to the right. As I said earlier, it doesn't appear to be binding, and it is the lighter, longer, axis. It does not appear to average out. I noticed the issue a couple of times early on, but it was particularly noticable when I tried per Giles' suggestion to cut by etching the same shape over and over again. Not exactly sure how many passes, maybe 12, 15? * I guess I could put a meter on the psu and look for droops in the supply volts when it's running. You'd hope that it would be stable within quite close limits. Thanks for all the suggestions. Not much going to happen during the next few weekends with Reading & stuff & Christmas & so on. Best Simon Link to post Share on other sites More sharing options...
Simond Posted December 6, 2016 Author Share Posted December 6, 2016 Ah, NASA and the Russians again Thanks Tovarisch!!! Link to post Share on other sites More sharing options...
RMweb Gold unravelled Posted December 6, 2016 RMweb Gold Share Posted December 6, 2016 Were you talking to hpc directly, or on their user site? I haven't logged on there for a few years for years, but the user site seems to have a lot of experience and advice to share. At least one of the contributors seems to work for hpc, others may well be into the sort of hacking you are looking at. As a 3020 user, I'll be interested in your findings. Having software power control would be a great advance. Hth Dave Link to post Share on other sites More sharing options...
Simond Posted December 6, 2016 Author Share Posted December 6, 2016 Dave I phoned them I will have another look at their user site Cheers Simon Link to post Share on other sites More sharing options...
Recommended Posts
Archived
This topic is now archived and is closed to further replies.