I did that thing.
At 1000mm there is still the tiniest bit of flex in the beam when I pull on it, but all my carving flex issues went away. I also saw a huge increase in quality by switching to a screw drive. Originally I had some problems but once I properly tightened the anti-backlash screw the X axis became rock solid.
I would never use heat on a stepper motor as the heat will transfer down the shaft into the stator and windings. taking a good chance of ruining the motor.
As with any pulley and shaft that is press fitted always secure the pulley at the area closest to the shaft and press the shaft that way all the force is transferred at the press fit. this is how we remove all the bearing and pulley without damage in the shop.
As much hassle it is to remove the pulley I would just consider buying a new stepper motor. Or possibly redoing the other pulleys and belts so that they all match the new one. The larger pulley is probably there to support a stronger belt.
I would be really (!) surprised if there is enough heat travelling off the pulley, down the shaft of the stepper and into the interior overheating the magnets/melting the coil shielding.
Most steppers can take 80degC (175degF) og exterior/case temperature without issue, and if its is 80degC on the outside it is much hotter inside. I have heated pulleys of several types of motors (including steppers) before and the motor case have not taken any heat at all (hand held during the heat-up)
do you think the heat difference is from where the heat is generated. when you run the motor heat is generated on the outer coils and not in the stator. when heat is applied to the shaft it is traveling down the shaft and into the stator, of course you can do whatever you want to your equipment, I would never do this to mine and never recommend you do it to yours without the understanding of the damage you could do. same as I would never place pressure through a bearing to a shaft to remove (it will more likely than not damage the bearings) and why would I want to take the chance when it is not needed. I just wanted to express that damage could occur. but I always remember that everyone’s luck runs out eventually.
We may disagree on this, which is fine by me, but I am certain enough that the risk of motor damage is so slim, almost non-existent and therefore quite comfortable suggesting that heating up the pulley with a blow torch is safe for the motor.
With a torch you won’t need to heat up the pulley more than a minute, enough to make the pulley expand and let the shaft go a little easier. It will not be enough for the heat transfer quick enough, in a magnitude large enough to cause problems further down the motor. With lots of margin.
But I´ll leave it at that, I still stand by my opinion that heat can aid the pulley removal - with little/no risk of motor damage.
sounds great to me, have at it. not my stuff you be working on so no loss on this end. totally ignore all previous post as I know little to nothing at all and I only speak from experience.
Well, ill stick to not heating my stepper motor for now, ill just try hammers, and grinders and what not, if I become desperate ill use heat, the worst that can happen is I buy a new stepper motor. I have to wait on this because I’m working all week, so ill do it on my day off.
I took a quick gander at what you’ve done, and I really like it. I’m probably going to do what you have, except the beam is an actual C shaped, and my carriage will be on the inside of the beam. I’m also thinking of getting 2 beams, and linking the carriages, onto the Z-Axis, which I believe would eliminate any flex completely. Also I will redesign the X-Axis plates to extend down to the lower Y Rails to add extra stability. Ill draw some cad files soon, and show everyone what I mean.
Phil, Phil, Phil, Phil…
Read the rest of the sentence: “Beyond that you need more modification.” Also, I should state very clearly, that there are those on this forum who have been here WAY longer than I. They helped me get to where I got. I will not take credit for other’s effort. All of these mods have their genesis long before we got here. FWIW. Except the steel plate mod. :). Congratulations on your .03 per pass.
Ive started brainstorming some ideas for the new X-Axis, I’m thinking of extending the X-Gantry plates to include the lower makerslide, to add rigidity. I’m thinking of using two C-Beams, with one “Linear Actuator” if I am correct. The X-Carriage plate is extended to both these beams and is held together with a set of wheels lining the inside of the beams. I do see problems with this, and I’m thinking of having T-Slot nuts line the inside of the rails so that the carriage slides along the inside instead of rolling inside the C-Beam. I haven’t had time to make a proper drawing so I just did a 5 min sketch in autocad. I am open to any ideas, or a better design.
Sorry, I didn’t take the best photos, both the X and Z are a c-beam kit. The acme screw is inside the channel of the “C”, which is on the opposite side of the router to help keep dust out. I never thought to take pictures from the “back” of the machine 
Thanks, I really do see the resemblance between your machine, and that R7 one. Hopefully I can finish the cad files today, and see how everything plays out.
Oh boy a lot has changed since the past 90 days or so, and I am getting very close to achieving my original goals!! Right before I left for a months trip to the Philippines I completed this monstrosity.
Now it was all fine an dandy but I had to leave the next day, so I came back to test it a couple weeks back and I have some good and bad results.
The bad news is my lead screw, even with an anti backlash nut has 0.04" of backlash (Now I’m not too sure how this is measured, but I took the difference between the distance I wanted the machine to move and the result), and there is a spot on the lower C-Beam where something grinds into it.
The good news is that I got between 0.005 and 0.01" of deflection (Not bad for a CNC router, and besides finishing passes are meant for that stuff). I was able to mill aluminum nicely at a feedrate of 30 IPM, with a 0.5MM stepover, and a 5mm DOC.
Now in order to curb the backlash issue, I bought a nice ball screw, and with that I have to redesign the X-Carriage to clear the issue with grinding into the C-Beam, and to mount the ballscrew. So in conclusion, it works but it still needs more work.
Any suggestions as to why the grinding happens in a certain spot (It happens in the middle of the beam), and why I’m getting 0.04" of backlash?
Backlash is the travel lost during transition from one direction to the other.
Under/oversized parts is usually a factor of back lash, runout, tool deflection, bit aize (actual vs stated) and system rigidity.
What is the diameter of your end mill? Number of flutes?
The lead screw does fit with the anti backlash nut, and is very snug like it should be, I do have a little bit of deflection so that may contribute to that issue just a bit. I’m using a destiny viper 1/4" 3 flute endmill with the black coating (forgot what its called, I don’t have the bits with me right now).
I’m wondering if the weight of the carriage has an issue with the engagement of the screw. The carriage fully loaded is quite heavy and its being driven by a small 8mm lead screw. I’m sure the ball screw will add rigidity as well as curb the backlash issue
ACME T8 screw and Delfin block should be very tight and capable enough for this task (I use ACME throughout on my CNC) But ball screw will be a better performer for sure, a conversion I am currently taking myself.
I used a 3mm 3F bit before which worked quite decent, used it down to 9mm per pass with a 0.6mm stepover. Feed was 600mm/min (22ipm or so) at 10k rpm and cut quality was a little shabby.
Have since then started to use 1F bit (3mm) which seem to thrive a lot better 
I modded my gantry to make it stiffer and now I can take 1mm depth per pass, full width at 22k RPM and 550mm/min.
Strive for rigidity when you want to mill alu.
What did you base your DoC/WoC/Feed upon?
(Disclaimer - diy-CNC based on MDF, not X-Carve)
Blockquote[quote=“HaldorLonningdal, post:42, topic:42228”]
What did you base your DoC/WoC/Feed upon?
[/quote]
My feeds were based on tool deflection, and chip rate, mostly chip rate. I ball parked the range from other peoples CNC’s with similar performance. I just got the aluminum to make the new carriage so we’ll see how that goes
![20171126_221226[1]](http://discuss-assets.s3.amazonaws.com/original/3X/c/2/c290f6d5028a7e197d7443ef2151ab04872f830f.jpg)
