วีดีโอ

I think its in the neighborhood of $3500 for the setup to stay with your steppers. I would stay with the steppers in your case because you do very fine milling at low speeds. I think AVID is giving a refund for the current controller box you have as well. You can call them and then can give you an exact quote.

I don't know about the relative hardness of the rack vs. pinion. However, the pinion is pretty small, so if the hardness is about the same, I would expect that it would wear faster than the rack typcially since the wear on the rack should be pretty spread out normally. However, I don't have any data on that, and havn't hear from AVID about that either. They recently changed the pinion assembly so you can replace the pinion itself w/o getting a new belt pully and bearing, which saves some money. I think you could do an experiment with backlash to find out if its the rack or pinion. If you have a section of the axis travel that you more heavily use, that should show more backlash in that area if the rack is worn. However, if its the pinion, I would expect it to be bad across the whole travel. You would just measure backlash across the whole travel of the axis and then plot it to see.

On mine, it does show up as reversed in the wizard too. If i change it, then it definitely jogs backwards.

@@JNWoodworks That is very interesting. In your very first video with the steppers that wizard screen clearly showed no reverse. Things must have changed for the servo motor profile. I configured my Clearpaths three years ago using Mach4 and a custom SmoothStepper electronics configuration. Now I am trying to make the change to Acorn and having to do a lot of detective work since Avid still has not published the details of their configuration.

@@Sehast Ah, well, that was the one with steppers. I have the servos now. I just checked in the wizard, and the servos and steppers have the opposite direction settings for all 3 axis' motors. I never noticed that before.

cool, glad it was helpful.

Hi Ron, I'm looking forward to seeing some videos of what you do with your XT-50 on those awesome pool cues you make!

I didn't measure it, but the input pipe is 3/4" and I maintained that size, and then on the plastic bucket I have a lot of holes. I don't expect that there is any significant increase in restriction, especially compared to the stock silencer which had fewer and smaller intake holes in it.

I think you'll like it. One is still going strong.

That's a big bolt. I'm in Italy right now so I can't go look at it, but I'm 99% sure that it's a standard right hand thread so should loosen counter clockwise.

I'm waiting for some really good software engineer to come along and write me a new UI for it. If only I knew someone like that.....😂

@ Do you think it’s possible to mount both the ESS and Centroid boards and switch between them? Maybe with a few wire swaps or nice connectors in between? I’ll get you that software it’s my resolution ;)

@@PaperTools With the servos like I have the wiring is too much of a changout. However, If you keep your steppers, then its not too bad. The cables in that case are a 1:1 match, so if you labled all your sensor and motor cables well, you coule probably do a swap in about 5 minutes.

@@JNWoodworks Thanks I might give that approach a shot!

That is easy to do. It's just constant power for flat etching. You would just do the opposite of this.th-cam.com/video/sjGmWgwnIJs/w-d-xo.htmlsi=AQ4YuaAyAsbAcRTu. and etch out the background around the letters. I would start with 60% power, 50ipm, and a crosshatch pattern with 0.010" stepover to start with, and the work up or down from there depending on how deep you want it.

@@JNWoodworks I am just curious as to how small of letters we can get from the XT8. I will definitely try your settings tho.. Thanks

​@@scottchausee2342 The main limitation with raised letters is they become fragile at small size

Possibly. I think the main parts that would wear over time are the small ring that the steel balls are held in, and the lid sheet. Everything else really shouldn't wear. I don't know if that ring would be better machined out of Al or not.

Awesome, have fun with your new machine. Half the fun of these is putting them together and figuring out all the things you can do with them .

Ya,mine is still working smoothly. I don't know of any recommended lubrication. I just spray some WD40 or whatever else I have available up in there once in a while. Probably a light machine oil would be better, but I don't have that in a spray and I'm too lazy to take the chuck off just to lube it. Mine is still very smooth and grips like crazy just with hand tightening. I like "Vactra Oil No. 2" for machine stuff like this, but you have to drip it in so it requires pulling the chuck off unless you have a way to spray it up into the chuck.

Thank you very much and Merry Christmas to you too.😊

Yes, you can still calibrate linearity and backlash. I have done backlash on mine, and the values are added in the Utility Menu->Advanced config->Mach.->Motor. In there are the settings for Lash comp. I have not yet done linearity calibration on mine since converting (but I plan to soon). In that same menu is "Motor revs/in". Mine is set to 1.01859 by default from AVID fpr X and Y. It is the setting that you would fine tune to adjust the linearity. There is probably a formula listed on Centroid's website or forum for figuring out much to adjust it for a given amount of error, but I found with Mach4 the easiest thing is if you measure your error, make a change to the setting and measure the error, now you know how much the number affects the linearity and its easy then to figure out what the new number should be. Maybe I'll do a quick video on it when I get around to adjusting mine.

One thing that I should mention is that with the EX controller you cannot do sperate linearity and backlash adjustments for the two Y motors. Since they share the same axis output (the slave is just a copied and inverted version of the step signal) they have to use the same parameters. You can measure the error at each end of the gantry, and then just use a midpoint between the two if they are different.

@@JNWoodworks It also looks like you could use the Acorn controller screw mapping and correction capability to greatly improve accuracy in the XY plane. Building the table of correction values for every half inch of travel would be quite a job but it appears that CNC12 would execute the corrections seamlessly. It should work on rack and pinion systems just as well as ball screws.

@@Sehast Yep, Centroid has a LOT of adjustment capabilities. You can also adjust the acceleration that you want to use when the system is changing direction to take up the backlash. However, all of that stuff is IMO a little unnecessar for a large cnc router since the flex and deflection in the system are much larger than the backlash and linearity issues across the rack.....but its there if people want to go to the nth degree :-)

I wouldn't really recommend it for 3/4" plywood, unless you are going to do two sided cutting ( see my video on that). Dense plywood like Baltic birch is very hard to get through beyond 1/2", and software plywood will tend to burn in the lower part.

Thank you. Yes, for engraving organic materials the beam width is very dependent on the power level and feedrate. It is less so for engraving metal and cutting organics because you need such a high power level for that that you are running relatively slow and high power. However, for engraving things like wood, paper, leather, or even doing things like the Norton White Tile method on ceramics, you will get a significant variation in pixels line width so you always have to do a little characterization to figure out what step over you want for fills, or what you want to call you DPI number when setting up a picture raster. Opt has to specify the beam size in a consistent way, so they have some fancy tools to measure beam profiles so they spec the beam size to be something like the area that contains 90% of the total power (or something like that). This gives you a means to compare each laser. However, you can make a significantly smaller line than their spec based on how easy the material burns, and your feed and power level. The beam profile of the laser probably affects this as well. A long or short focal length laser will probably have a different shape to the power distribution.

Hi, I'm glad you found the video helpful. There are a lot less things to set in CNC12 for laser than my last controller. I am currently setting the PWM frequency at around 5000hz. The XT-8 will actually run much faster than that, but it has a limited turn off time that has the effect of limiting the min on pulse of the laser. So short story is that if you want to have a very low power level for test fire, then you want to run it at a lower frequency like 5000. I am currently trying to decide if I need to use 100 or 1000 as the max power value (I have a few experiments to do yet to understand all the nuances of that wrt to applications like Vectric and LaserGrbl, and test firing resolution). If you send me an email at jnwdwks@gmail.com, I can send you a screenshot of the laser wizard window, and answer any other questions you might have. As far as the VCP, I'm not sure what you mean by non-cnc setups, did you mean non-AVID setups? Most of their custom VCP is just to make a cleaner screen for controlling a CNC router/laser setup vs. the packed one that CNC12 has typically had for mills and stuff. Basically they removed a lot of the stuff that wasn't needed on a gantry style machine, and I think did a great job of making the icons look a little easier to read. I think Centroid may be incorportating some of that in their latest router builds of CNC12.

I'm working on that 😀

@ awesome!!

Subscribed.

Yes, those are magnets that I used for stops. Most of the time I don't bother with stops for laser work because the object doesn't move. The main exception is paper and cardboard because the laser fan and air nozzle tend to blow them around. In this video, I just threw the magnets on there because they were handy and I needed to make sure they didn't move even a few thousandths. On the lens cover question, Opt lasers don't have lens covers, the main lens is exposed inside the air nozzle. As long as you always use the nozzle and don't forget to turn on the nozzle air pressure, or get a spider inside the nozzle (that actually happened to me once) the lens will stay clean. I don't do production work (I'm just a hobbyist), but I have run a couple of laser picture rasters that took 7-8 hours and the lens wasn't dirty. I run about 5 lpm of air for light engraving, and 10-15 lpm for heavy engraving and cutting. The outside of the nozzle can end up very dirty, but nothing seems to get into the lens area.

They can use the same power supplies that they have for steppers this way, and electrical safety ratings are easier to deal with under 50v. The power difference only matters is it's not enough, and these move things along just fine.

Thank you!

The dogs are these DCT Woodworking Plastic Bench... www.amazon.com/dp/B076MGP4F6?ref=ppx_pop_mob_ap_share And the clamps are these: Kreg KBCIC In-Line Clamp www.amazon.com/dp/B0774WBGPX?ref=ppx_pop_mob_ap_share The dogs are cheap and hold well. The clamps are not cheap, but they really hold up over time and clamp very well. I haven't had any failures except when I hit them with an end mill 😅. No side clamp holds things down flat to the table, for that I use hold down clamps and usually some oak sticks so I can keep the clamps out of the way.

Agreed, the collets need to be clean. Its the same issue with the standard ATCs, some people use an air jet like you suggest, some just clean them often.

@@JNWoodworks yeah nothing is totally no work you have to do some lol

Could you elaborate more? I'm not sure what you mean by "do the sensors in the software".

Well thats not good. It sounds like you've run your's more than I've run mine so far. I've been mostly doing a bunch of experiments on vibration and laser accuracy. I have not seen any issues like that with mine, the only time I ran the Z axis down was my fault when I got a little careless with the pendant. I guess that could be a motor problem (internal encoder issue), or noise on the step signal from the controller to the motor. Weird. I'm sorry I don't have any good suggestions but maybe to reconnect the Z motor's signal cable at both ends just to make sure it has a good connection.

I know which video you are talking about. Ya, they use the backlash as a means to adjust that right now. I'm currently characterizing it vs. feedrate. These clearpaths are stepper replacements, so they are closed loop within the motor only, as far as the controller is concerned its a stepper motor. The other thing is that this is really a timing lead/lag between the PWM control and the motor control and its hard to build that in to be dead on without needing a little calibrations. The ESS card with Mach4 is the same way.

Hi Scott, I am actually working on some of that very laser stuff today. I'm using both Vectric and LaserGrbl. I don't know much about Lightburn's customization capabilities for Gcode output. LaserGrbl writes out Gcode that is Centroid compatible, and it lets you write your own header, footer, and multipass mid code sections, so it is easy to get all those "special" gcodes and macros added it. I suspect Lightburn is the same. A lot of the recent new router additions to the stock CNC12 came from collaboration with AVID. A lot of AVID's customization that is on top of that are just simple macros, so you can do that stuff yourself easily. For example, I like the stuff AVID has done with the laser, but its a little too integrated for me because I do a lot more experimentation rather than production, so I want it a little more stand alone and not tied to the spindle. It will be easy to make the changes I want for myself.

Hi Tyler, You are hitting on the one thing I think falls short on CNC12 (at least compared to mach3/4). They do have a graphing screen, but the current one is only offline. However, in CNC12 5.20 they did make it so you can jog around and it will show where the origins are, workpiece boundaries, and the tool is relative to the toolpath. So you can find out if all the zero settings and the toolpath fit on a weird shapped workpiece. However, I don't know if the live version of graphing is also updated with that info. This version should be coming to AVID soon. Centroid needs to do some improvement on the CNC12 screen. It is bascially set up to be optimal for a smaller LCD touchpanel that you'd see on a large CNC mill. However, everyone I know on a CNC router uses a bigger monitor, and the screen real estate is not used very well for that situation. I also had the line counters and percent complete statistics on Mach4, and I would like to see the toopath graph along with all the other info. Rigth now their graph drops on top of everything else when its turned on. They have added a lot of improvements lately for CNC router use, and I think they are planning on more so I'm hopefull that some of the things we're used to having on display all the time will be coming.

@@JNWoodworks Thank you for the reply! I wanna switch over so bad!

I don't think I'd have a concern with that. The test I did with the very long, dull bit in Mesquite was probably worse than aluminum, especially since you'd typically be using a smaller diameter mill for aluminum. You'd want to make sure that the collets were not getting chips dropped in them when doing tool exchages, that would be my biggest concern.

Hi Stephanie, Thanks for all the good questions. First, 0.23" plywood (especially baltic birch) is kinda the upper limit of cutting capability for that laser. Plywood has a lot of glue and that makes it harder to cut than plain wood as well. If I remember right, the 15W will cut quarter inch stuff at maybe around 15 inches per minute, but will probably take 2-3 passes. In the above video I'm using a 45W laser which handles quarter inch wood like butter. Your laser has less diodes and a different lens. You can try the procedure I outlined in the video with the 15W laser. The 45W laser I was using has a longer focal length lens than your 15W, so you'l have a shallower depth that you want to set the focus at for cutting deep. I never characterized my 15W laser when I had it with the above method, but I'm going to guess a good starting point would be to set the focus to a depth of about 0.070" or so below the surface. Make sure you've established your best height for the smallest focus on the surface first (I've gone over that in some other videos), and then you know that number, just set your Z zero 0.070" lower than that. As far as not hurting your laser/maintanence. #1 thing is to keep the lenses clean. If they get dirty, performance will drop off rapidly. At some point the dirt can get hot enough to damage the lens as well. If you always operate it with the nozzle on and good airflow (~5-10 lpm for etching, 10-15 lpm for cutting) you lens should stay clean. The airflow keeps soot and stuff from getting in the lens. Its good to inspect the lens once in a whle though just to be sure - I had one laser where a spider got in there and built a little web in the nozzle once. I looked at the lens and it was clean and then inside the nozzle and wow. The other thing is you can damage a laser with very reflective material like aluminum. If you are etching painted aluminum, you don't need to run at full power, so its better to only use maybe 20% or so. I did a quite a bit of marking stainless steel with my 15W laser and didn't damage it, so thats probably ok. The other good thing (for you as well), is to keep the smoke away from the fan so it doesn't blow through and make the fan and cooling fins dirty.

I was thinking the same thing. I now have the servos installed, and they are definitely smoother, but the Acorn controller cleaned up the vibration so much on the laser there is no improvement to be had there. So now I'm looking at optimizing the opposite way where I keep the vibration low, but push the speed up until the vibration starts to increase again. It would be great to get back some runtime performance.

You're welcome Steve.

I haven't looked in detail about the controller only, but I don't think the upgrade is that much. The price you are quoting I think might be for the whole EX control kit which would include motors and wiring. I can't answer if it's worth it for anyone else, it is for me.