Really cool to see you make improvements with the same machine the parts are meant for! Excited to see that z axis improvement!. Measured progress is great progress!
Thank you Matthew for your comment, yes this machine has built already several other routers for friends of mine. It is very satisfying to see the improvements for this one. The z axis stiffener plates are designed now and I am doing the CAM and raw material prep today.
Hey there, enjoyed that thanks. And like that you shared it even though it wasn't quite the result you hoped for. Looking forward to the z stiffening video :)
Thank you for your comment. So I have been measuring all day and scratched my head but I finally nailed it down. The whole gantry tips forward. The HiWin guide blocks are worn. And that is ok because there are miles and miles in them. I will see if I am able to service them.
Thank you! I ended up at Bal-Tec looking through the sizes. I took one apart yesterday and used a micrometer the diameter is 0.1285”. I think I will use + 0,0003” for the new balls
I think I have the same machine you do and this was going to be an upcoming project for mine as I noticed deflection as well. Good to see it made some improvements but I will definitely be looking at my linear rails as well.
Yes also start by upgrading the fixed bearings in the ball screws. They made a huge improvement on backlash. I have an older video about that … take a look
Nice video! I think that if you add another spindle mount to the top of the spindle you will see that there isn’t that much flex anymore. I did the same with my round spindle a couple of years ago. With the second spindle mount I was able to reduce chattering a lot. What also could help ist to add side plates to the z-axis plate to stiffen it up. Why did you cut out those triangles in the back plate? Aren’t those cutouts affecting the stiffness of the plate it self? Keep up the good work!✌️
All 3 points are spot on. Yes the spindle mount I will change next as I hate the looks of it and it provides for the highest amount of flex at the moment. In theory the triangular cut out of the plate does effect the rigidity in y direction however I measured this on a machine that I build for a friend and only had 1/100 mm difference full plate versus plate with the cutouts. Thanks for your comment !!! Oh forgot to ask you something how is the z stiffener mounted on the Sorotec ? I see screws from the side … not really intuitively right through the plate would add better holding power in y direction.
Thank you for that info. I have the front of the first one complete (machined) and the slot I planned to have 0.03mm play in it. The receiving plate is not machined so this might be interesting to get installed.
1 year further you know much more about the flexing points of your machine. To close the X axis profiles from behind is a good idea, but i personally wouldn't have made those design windows just in the middle of that part. Just the middle should be as stiff as possible and there must be the most material. I went the other way and built the Gantry out of a massive profile of 160x80 mm, filled with cement - but my machine is not working at the moment, so I have no measurements. I hope I can get values like yours :) The Z Axis is the most weak part - many guides Wagons, thickness of the plates - the side reinforcements make very very much sense. Maybe the spindle holder is another problem, too. Did you find out more on this? Kind greetings from Hamburg, Germany!
You you are correct if you know the Lexington direction and the equation then you will see that the windows I cut hurt more than they help by reducing weight. However overall I am happy how the machine performs and the accuracy is really good and surpasses my expectations. Most of my cuts are conservative and even the rapid moves are dialed down to improve the overall results … I have the time … a few seconds or minutes longer per part are not relevant. The span of the guide “wagons” is important so a design that sets them further apart will be good. I have been thinking to make my machine a bit wider (not really good for the gantry stiffness) and redesigning the X-axis. I probably would place one guide rail on the top surface. Thank you for your comment curious to see your results.
adding this plate is a good thing although measured improvement are not so impressive. when you load gantry like that it mostly twists rather than bend. when it twists both beams move closer to each other. and if you load spindle in direction parallel to gantry both beams will shift relatively. so binding them together into one structure preventing relative movements is a good thing. i does the same thing on my router ) design it with that backplate in mind but never install it and used like that for couple of years ) recently i thought it's time for some improvements and i bound both beams together (by some epoxy granite instead of aluminium plate which was planned). i also made the same measurements before and after and get about the same results with only 25-30% improvement in static load. but when i start to mill improvement was more prominent. most noticeable thins is significantly less chatter when spindle is in the middle of a gantry. P.S. and here is some thing on chatter. stiffening the gantry will have effect on more low frequency chatter which happened in low rpm range. but it will have little to no effect on high frequencies which all concentrated mostly in Z axis assembly (closer to spindle - less mass - higher resonant frequency)
Thank you for your comment. Funny how we sometimes use something for years because it seems to work just fine until we take another swing at it. I will do the Z-axis next and started the machining yesterday. I am not 100% satisfied with the x-gantry improvements and I do not expect much improvement from the z axis stiffener but like you mentioned they will add mass right at the spindle. Yes granite for the X- Axis beam would be epic. I think also having a granite plate as the machine bed would be fantastic. This would dampen all the vibration associated with the fixturing of the part.
Aluminium extrusions are very prone to torsion and bending forces. I had the same problem with my diy cnc with the x axis built with a 30x150mm aluminium extrusion. The deflection you see at the spindle sticking out is due to torsion stiffness, or the lack there of. measuring directly on the rail will show you a small deflection but your conclusion is incorrect, it's deflection that get amplified the longer the distance is from the x axis, like the spindle sticking out under it. The best shape to counter torsion is a round tube, but since it's hard to mount a square tube is the best runner up. your alu plate is inadequate to counter torsion flex. After mounting a square tube of 150x150mm with 3mm thickness steel tube the deflection was reduced down to a quarter it was before. I used a spindle with a square extrusion running the entire length of the spindle stiffening up the z plate too, but adding a box structure or square tube like structure will also improve stiffness. Steel is 3x stiffer compared to aluminium and about 2x heavier, it saves space to use steel instead of aluminium. I used shims to level out the steel tube sine I didn't have any means to make it straight. It worked out for me. It can now machine mild steel as easily as aluminium.
Nice Job Josha, however the “amplification” can easily be calculated over the length of arm from lower rai to the spindle. It still will not make the 0,14mm we see in the video. What actually happens is that the entire gantry tips forward because the Y-Axis HiWin guide rails are worn out. I ordered new balls and will recall them hopefully this week using more precise balls and a different diameter to increase the preload. This will bring the deflection back down. However yes a closed profile is always more torque resistent and from the material standpoint only the E-modulus goes into the formula and yes steel is way higher than Aluminium. Let’s see how far I can work that deflection down
Me too. Been machining all day the Z Stiffener plates but honestly I don’t think they will make a lot of difference. The z plate is already rather thick. Anyways I will keep going if you keep watching and giving me feedback !!
@@JBWorx Absolutely will keep watching, and commenting when I've got something to share :) It's a lot of work, but I'm sure it'll pay off in the long run. I just finished my diy fogbuster today, I'm going to do some aluminium machining tomorrow to see how it performs.
The plate is adding stiffness in Z direction, but almost no improvement in torsion stiffness. For torsion stiffness You need round or square pipe in X direction. But, because of leverage most forces are coming when force is applied in Y, as you tested. First improvement would be to put two plates in XY plane right under the rails. Look the wood cnc from life latitudes as very good solution. Second improvement would be to add side reinforcement to the Z plate, but you seem to do it in next video. Aluminum profiles as the most people use them are actually X profile. Relatively stiff at side forces, but complete disaster for torsion.
It is quite an improvement over the individual profiles not being connect at all. So “almost zero” like you say I disagree with but your general approach is correct. I remember that I measured both top and bottom rail movement simultaneously with a load applied to the tool in the Spindel and got very little deflection Of the rails for torsional movement. So yes that worked great. However the Z stiffener that you mentioned are a total waste for my machine. There is such a small improvement that I think they are not worth having. Changing the gantry to a closed box design is something I wanted to do for a long time. Well one day ! Thank you for your comment.
I found some plate material in my shop that is almost 1” thick so my thinking right now is that I will keep the ugly bottom piece for now and just add a top bracket.
Not easy to do but there is a sound tool that you can use as an app with your phone that will measure the frequency during machining. Also have a look at my Video for the second spindle mount there I measured acceleration.
Really cool to see you make improvements with the same machine the parts are meant for! Excited to see that z axis improvement!.
Measured progress is great progress!
Thank you Matthew for your comment, yes this machine has built already several other routers for friends of mine. It is very satisfying to see the improvements for this one. The z axis stiffener plates are designed now and I am doing the CAM and raw material prep today.
Wow congrats your homemade CNC Machine
Thank you Rany !
Hey there, enjoyed that thanks. And like that you shared it even though it wasn't quite the result you hoped for. Looking forward to the z stiffening video :)
Thank you for your comment. So I have been measuring all day and scratched my head but I finally nailed it down. The whole gantry tips forward. The HiWin guide blocks are worn. And that is ok because there are miles and miles in them. I will see if I am able to service them.
@@JBWorx Nice detective work! I would be interested in the servicing process... guess you might want some slightly bigger balls to refill? Good luck!
Thank you! I ended up at Bal-Tec looking through the sizes. I took one apart yesterday and used a micrometer the diameter is 0.1285”. I think I will use + 0,0003” for the new balls
I think I have the same machine you do and this was going to be an upcoming project for mine as I noticed deflection as well. Good to see it made some improvements but I will definitely be looking at my linear rails as well.
Yes also start by upgrading the fixed bearings in the ball screws. They made a huge improvement on backlash. I have an older video about that … take a look
Nice video! I think that if you add another spindle mount to the top of the spindle you will see that there isn’t that much flex anymore. I did the same with my round spindle a couple of years ago. With the second spindle mount I was able to reduce chattering a lot. What also could help ist to add side plates to the z-axis plate to stiffen it up.
Why did you cut out those triangles in the back plate? Aren’t those cutouts affecting the stiffness of the plate it self?
Keep up the good work!✌️
All 3 points are spot on. Yes the spindle mount I will change next as I hate the looks of it and it provides for the highest amount of flex at the moment. In theory the triangular cut out of the plate does effect the rigidity in y direction however I measured this on a machine that I build for a friend and only had 1/100 mm difference full plate versus plate with the cutouts. Thanks for your comment !!! Oh forgot to ask you something how is the z stiffener mounted on the Sorotec ? I see screws from the side … not really intuitively right through the plate would add better holding power in y direction.
how far has everyone between mounts?
@@JBWorx My Sorotec AluLine stiffeners has very tight tolerances. It took me light hammer taps to get them off.🔨
Thank you for that info. I have the front of the first one complete (machined) and the slot I planned to have 0.03mm play in it. The receiving plate is not machined so this might be interesting to get installed.
Interesting experiment. Make sense that the spindle holder is a weak point.
I think that the surface finish is better on most parts since I installed it.
1 year further you know much more about the flexing points of your machine. To close the X axis profiles from behind is a good idea, but i personally wouldn't have made those design windows just in the middle of that part. Just the middle should be as stiff as possible and there must be the most material. I went the other way and built the Gantry out of a massive profile of 160x80 mm, filled with cement - but my machine is not working at the moment, so I have no measurements. I hope I can get values like yours :)
The Z Axis is the most weak part - many guides Wagons, thickness of the plates - the side reinforcements make very very much sense. Maybe the spindle holder is another problem, too. Did you find out more on this? Kind greetings from Hamburg, Germany!
You you are correct if you know the Lexington direction and the equation then you will see that the windows I cut hurt more than they help by reducing weight. However overall I am happy how the machine performs and the accuracy is really good and surpasses my expectations. Most of my cuts are conservative and even the rapid moves are dialed down to improve the overall results … I have the time … a few seconds or minutes longer per part are not relevant. The span of the guide “wagons” is important so a design that sets them further apart will be good. I have been thinking to make my machine a bit wider (not really good for the gantry stiffness) and redesigning the X-axis. I probably would place one guide rail on the top surface. Thank you for your comment curious to see your results.
@JBWorxStudio may be some video about your speed and feed when cutting aluminium / plastic, when slotting or drilling
Thank you David for the suggestion. I have in fact a 3 part series come up stating with the beginner in mind on cutting wood all the way to steel.
@@JBWorx nice
Great video
Good that you got something out of it.
adding this plate is a good thing although measured improvement are not so impressive. when you load gantry like that it mostly twists rather than bend.
when it twists both beams move closer to each other. and if you load spindle in direction parallel to gantry both beams will shift relatively.
so binding them together into one structure preventing relative movements is a good thing.
i does the same thing on my router ) design it with that backplate in mind but never install it and used like that for couple of years )
recently i thought it's time for some improvements and i bound both beams together (by some epoxy granite instead of aluminium plate which was planned).
i also made the same measurements before and after and get about the same results with only 25-30% improvement in static load.
but when i start to mill improvement was more prominent. most noticeable thins is significantly less chatter when spindle is in the middle of a gantry.
P.S. and here is some thing on chatter. stiffening the gantry will have effect on more low frequency chatter which happened in low rpm range.
but it will have little to no effect on high frequencies which all concentrated mostly in Z axis assembly (closer to spindle - less mass - higher resonant frequency)
Thank you for your comment. Funny how we sometimes use something for years because it seems to work just fine until we take another swing at it. I will do the Z-axis next and started the machining yesterday. I am not 100% satisfied with the x-gantry improvements and I do not expect much improvement from the z axis stiffener but like you mentioned they will add mass right at the spindle. Yes granite for the X- Axis beam would be epic. I think also having a granite plate as the machine bed would be fantastic. This would dampen all the vibration associated with the fixturing of the part.
maybe the second clamp at the top of the spindle?
Yes, I think that will be an improvement. Best would be to capture the spindle at the location of the bearings.
Aluminium extrusions are very prone to torsion and bending forces. I had the same problem with my diy cnc with the x axis built with a 30x150mm aluminium extrusion. The deflection you see at the spindle sticking out is due to torsion stiffness, or the lack there of. measuring directly on the rail will show you a small deflection but your conclusion is incorrect, it's deflection that get amplified the longer the distance is from the x axis, like the spindle sticking out under it.
The best shape to counter torsion is a round tube, but since it's hard to mount a square tube is the best runner up. your alu plate is inadequate to counter torsion flex. After mounting a square tube of 150x150mm with 3mm thickness steel tube the deflection was reduced down to a quarter it was before.
I used a spindle with a square extrusion running the entire length of the spindle stiffening up the z plate too, but adding a box structure or square tube like structure will also improve stiffness. Steel is 3x stiffer compared to aluminium and about 2x heavier, it saves space to use steel instead of aluminium. I used shims to level out the steel tube sine I didn't have any means to make it straight. It worked out for me. It can now machine mild steel as easily as aluminium.
Nice Job Josha, however the “amplification” can easily be calculated over the length of arm from lower rai to the spindle. It still will not make the 0,14mm we see in the video. What actually happens is that the entire gantry tips forward because the Y-Axis HiWin guide rails are worn out. I ordered new balls and will recall them hopefully this week using more precise balls and a different diameter to increase the preload. This will bring the deflection back down. However yes a closed profile is always more torque resistent and from the material standpoint only the E-modulus goes into the formula and yes steel is way higher than Aluminium. Let’s see how far I can work that deflection down
@@JBWorx ah right, worn out rails will do that too. Looking forward which accounts for the highest part of the deflection!
Me too. Been machining all day the Z Stiffener plates but honestly I don’t think they will make a lot of difference. The z plate is already rather thick. Anyways I will keep going if you keep watching and giving me feedback !!
@@JBWorx Absolutely will keep watching, and commenting when I've got something to share :) It's a lot of work, but I'm sure it'll pay off in the long run.
I just finished my diy fogbuster today, I'm going to do some aluminium machining tomorrow to see how it performs.
The plate is adding stiffness in Z direction, but almost no improvement in torsion stiffness. For torsion stiffness You need round or square pipe in X direction. But, because of leverage most forces are coming when force is applied in Y, as you tested. First improvement would be to put two plates in XY plane right under the rails. Look the wood cnc from life latitudes as very good solution. Second improvement would be to add side reinforcement to the Z plate, but you seem to do it in next video.
Aluminum profiles as the most people use them are actually X profile. Relatively stiff at side forces, but complete disaster for torsion.
It is quite an improvement over the individual profiles not being connect at all. So “almost zero” like you say I disagree with but your general approach is correct. I remember that I measured both top and bottom rail movement simultaneously with a load applied to the tool in the Spindel and got very little deflection Of the rails for torsional movement. So yes that worked great. However the Z stiffener that you mentioned are a total waste for my machine. There is such a small improvement that I think they are not worth having. Changing the gantry to a closed box design is something I wanted to do for a long time. Well one day ! Thank you for your comment.
mount should be fairly fine, but add those stiffeners to the Z axis plate, or box that spindle inn :) Then you will see improvement for sure :)
Yes stiffener design is on its way as we speak. Also a new Spindel mount
@@JBWorx i designed and made my own spindle mount, stiff as i need on my machine, enought to do steel on it ;)
That is very nice. I probably will need a new spindle for steel though.
Can i ask what do you mean by “box that spindle inn”?
I think he is saying to build a frame all around the z plate not just left and right.
I split the spindle clamp in half and separated them by about 100mm
I looked at it today several times and thought of the very same solution … it’s a good idea nothing new to make and no material to buy !!
@@JBWorx i lost a hole but it wasn't a hole that was doing anything beneficial
I found some plate material in my shop that is almost 1” thick so my thinking right now is that I will keep the ugly bottom piece for now and just add a top bracket.
@@JBWorx always good to do something more complicated lol
😀😀
I really thought you would have gained more from that upgrade
Yes me too. Well it’s on there now
@@JBWorx I wonder if it helped buy absorbing some vibrations? How would one measure that?
Not easy to do but there is a sound tool that you can use as an app with your phone that will measure the frequency during machining. Also have a look at my Video for the second spindle mount there I measured acceleration.