The answer is NO. Mach 3 will not work with xPRO.

 

Page 84

 

Really if you want something reliable, and smoother running go for a Mach 3 board. I recommend a uc100 or uc300 motion controller, octocoupled 100khz Bob with descent mosfet microstepping drivers. In fact I'd like to see a video of an xpro actually working properly and not sounding rough with nema 23s.

 

So I reinstalled GRBL, re-flashed XPro firmware, reset all settings and created a new cut file. No glory, low speeds are definitely not something the Xpro is going to perform smoothly. I would say if your going to run at anything below 1000mm/min then you should get a more reliable controller. Wish this conversation would have happened before I went with the XPro, would have saved me a lot of aggravation and discouragement. The XPro really shouldn't be billed as a CNC controller without also cautioning buyers of it's serious limitations.
My options now seem to be a higher rpm spindle or a new controller...ugh.

 

Probably worth going for a new board, Well depending on what you plan on cutting in the future, really for wood you need a setup that can go down to at least 8000 rpm.

Having had tried a couple of grbl controllers, I can say neither performed anywhere near as smooth as a mach 3 board, The setup will generally cost a bit more as you have to buy mach 3 but you get what you pay for and with quality mach 3 parts you know your getting a lot more reliable system at the end of the day and better quality cuts.

For the hardware you can pickup a bob for around £35, a motion controller for £85 and your looking at £35 each for decent Hexfet Power mosfet drivers. Giving you a descent pc data buffer and pc opto-coupled protection, something you dont get with grbl controllers meaning that all pc data is buffered to send a smooth response to the bob giving you smooth stepping, true current from the mosfet drivers and in an unlikely failure you can swap out the broken part without forking out $129 a time as you would with the xpro.

 

Really? I thought I could use UGCS, GRBL, Mach3, or ChiliPeppr...

 

I guess I'm still a noob and don't really know what I'm talking about. I got a lot of learning to do...

 

Hey Jonny, could you share a bit more of your expertise...

Can you explain a bit why ? I thought wood usually requires high RPM

Could you recommend some brands / products

Does it make sense to keep the low cost GRBL (with its free control SW) but use HexFet Power Mosfet drivers. I think it could be an intermediate step towards a more professional setup, recycling the Mosfet Drivers later when upgrading to Mach3

I'm not unhappy with the xPRO. I understand it has limitations, but I had no CNC experience at all when I started building this OX. I consider the 129$ an investment in learning. (and I learned a lot!) I try to learn with cheap parts, then hit the limitations and have a good reason to upgrade.

have a nice day!

 

I'm no expert but think you will find its all down to chip size, I have some wood specific compression spirals that need to be run at 8-9000 rpm @ 1200mm/min in order to achieve the manufacturers recommended chip size. This is more important with composite woods, mdf especially, any faster and i find the mdf burns and gums up the bit. Though you could speed up the feedrate if your capable of faster but usually you may have reduced feed rates for plunge and inner corners etc this is where your bit if run too fast may gum up. In an ideal situation you would have the spindle rpm controlled by your software and drop the rpm during these moves and setup dwell stops in order to keep the chip size the same, but if you dont have that luxury then really you need a find a happy medium. As a rule my wood bits go from 8-16000 rpm, metal bits from 12-19000 and you only need go faster for engraving and pcb bits.

It does change from bit to bit but you will find some plastics need low rpms also.

Certainly changing to hexfet drivers would make a huge improvement. Though it may not solve it completely, one could only try it and find out.

 

Something else i have just become aware of that was previously unknown to me is that unless you know your bit is completely center balanced with no run out you can actually knacker the bearings and kill your spindle running too fast.

 

We've all been there. I learn something new every day.

 

I have another questions guys...once I have everything wired up and ready to go, how do I install the CNC xPRO (software wise) on my PC? Is there a driver or wizard I need to run/install? How do I get my PC to notice the xPRO? I think I'm going to use Universal GCode Sender since I can't use Mach3. Any advice is appreciated. Thanks.

 

And just to confirm...it's NOT possible to run Mach3 with the xPRO...correct?

 

No the xpro won't work with Mach 3, in fact it's debatable whether the xpro works for cnc at all.

 

So what are some alternatives? How about the Geck G540? Is it a contoller and drivers all in one? Is it a direct replacement for the xPRO? And can I use it with Mach3? If I use it, do I need any other supporting hardware?

 

These things are either grbl or Mach 3 never both.

 

Does anyone know if the Mach3 / GRBL differentiation apply to whether someone is running Mac OSX or Windows?

What, also, would be the differences or comparison in the effective functionality, performance, etc. of the Grbl vs. Mach3, in people's experience or opinion?

 

I have to correct myself from my previous post, I reversed the situation. Series circuits split voltage and parallel circuits split current. The reason is that with only one path (series) all the current must flow equally through all loads and split the voltage, equally if the resistance is equal or proportunately if it is are not. In a parallel circuit there is more than one path so the current takes the path of least resistance, equal in each leg if the loads are equal or proportunate if they are not and the voltage is equal on each leg. Sorry if I caused confusion...

 

I dont think mach3 is mac compatatble. I have run both and found mach 3 to be alot smoother.

 

if you use a motion controller such as a uc100 you can use a lower spec pc, its a good idea to use one regardless of the spec imo, reduces chance of lag causing issues and have seen these improve smoothness even with a fast laptop.

 

As a side note, I am looking at something like the small format Ox, and notice there has been some discussion of difficulties and limitations with the xPRO V2 all-in-one controller.

Are there some workable suggestions that could be made for a person to use the xPRO more effectively within it's limitations?

For example, I have seen a video where someone mentioned seeing improvement by adding heatsinks to the drivers. Would it be more advisable for someone to source lower amp draw Nema23's or consider using stronger Nema17's throughout? Or does it seem like that would take the capability of the build down too far unless someone knew that they have very limited use with softer materials, etc.?

I know, generally, increasing voltage where possible can or will reduce amperage. Is that something possible that someone could do to make the xPRO or similar more effective regarding the amp output limitations?

 

Think you will find xpros have huge value resitors on them already which suggests they have already been optimised in this manner, giving these thier rated 24v. Without them I'd expect the chip would fry and I have heard these as standard Cooking at 27v.

 

Thanks for the feedback.

It looks like this gShield V4 is basically the combo used in the build video for the Ox. I saw it might have similar concerns, but do you, or anyone, know how that worked out, or in comparison to the xPRO?

Also, how about the TinyG as far as similar concerns, that I understand it may have similar amp ratings, if I recall, ?

Do any of these allow something like easy driver upgrade or piggy-backing that you know of, (if I'm not just imagining too far 'outside the box') ?

 

Think the question you need to ask here is " is anyone getting good results with grbl and if so what hardware are you using?"

 

Got it (at least somewhat).

I will probably go with something like a TinyG to start, as this is my first build and upgrading to Mach3 later would mainly require only the controller system components and leave the rest of the build components usable, while the first controller (TinyG, etc.) might also be able to be put to some other use.

And, I've seen at least some level of acceptable 2.5D work, etc. done with other builds using the same or similar controller setups, such as ShapeOko 2, etc. That will also give me a chance to become more familiar and proficient with everything else involved, in the case that expanding the build dimensions and/or upgrading to Mach3 is something I would then know better that I would want to do.

Thanks again for your input.