Entertaining, as always. I was going to suggest, now the machine is back together, to cut another plate - so you never need to use the "one-time" plywood plate again. But I guess the only reason to make another one would be if it needed modifying - and you probably can't know what needs modifying yet... so I'll keep that suggestion to myself 🙂 Thanks for the videos.
When I originally cut all the plates for the Z axis I was stupid and didn’t square the gantry very well. I’d probably redo all the plates at some time in the future. Possibly change the tram plate slightly so I can box in the sides of the spindle to make it slightly more rigid.
Yeah I gotta sort out all the exposed rails and ball screws. It’s becoming a problem with all the chips getting in there. I have some stainless steel sheet left over that i’ll used to make some wall for the Y axis. Still trying to figure out the best solution on covering the X axis rails top/bottom and ballscrew. Z axis I can wall off the sides and figure something out for the bottom rails.
Hello Brayan, thank you a lot for putting up these videos and sharing your journey. I would like to know if the laser level and webcam system can be used for surface flatness assessment as I intend (hope) to replicate for that purpose. in the affirmative, what would the limits for that system to asses surface flatness to microns ? can it replace other techniques like the one using autocollimator
It can be used for surface flatness but you’ll want to put in the effort in calibrating the lens to the laser. I’ve been using a Powell lens and to the best of my knowledge it’s the best lens to convert a laser dot to a line. The alignment of the lens to the laser is absolutely crucial to be able to use it for measuring the flatness of a 2d plane. We want the beam to be in the shape of a straight horizontal line. This can be calibrated by projecting a dot parallel to a wall surface. Place 3 points equal distance from each other along that line for the webcam. You’d want some adjuster to raise and lower the platform the webcam sits on so you can get each spot dialed in to zero out on the laser. After you know you have 3 points in space that are straight to each other, you have a reference you can use to adjust the angle of the powell lens. This can be done by projecting the laser line perpendicular to the 3 points aimed at the middle one and just the pitch axis of the Powell lens till all 3 points zero back out. When you can zero out on all 3 points perpendicular to the original reference point, you’ll have a straight line generator that you can use for mapping out a surface with. I’ll have to make more videos and one showing this whole process in the future. Here’s the results of probing a very small object with 10,000 points ruclips.net/video/xf5xHpGlwvQ/видео.html You can also do a similar approach by sampling nothing (air) on the CNC to map out how accurate your XY axis is. If there is any twist in a dual Y setup such as my CNC or if there's any warping of high and low spots on the rail surface.
@@BryanHoward thank you very much indeed for your reply. I will definitely investigate more into it and also read all information you already made available. It would be a huge help if you'd find time to investigate more into this flatness testing application and more videos from you regarding that would be highly appreciated. Also will share any feedback regarding adopting this method in our project.
hey man been watching your channel for a while, also working on my own cnc too. what i was wondering was how low of an RPM the spindle can be operated at. you did try some agressive drilling at 6k rpm, but how about milling/drilling steel? can it handle 3000rpm os we can do some light cuts on steel? thank you in advance
yeah the spindle isn't designed for it. You can get down to low rpms depending on how much load is going onto the spindle. This is about as low as I've gone in steel @ 3k rpm: ruclips.net/video/guddMZIETKs/видео.html and going in the opposite direction here's 24k rpm in steel: ruclips.net/video/8v-yXavlFhU/видео.html when it comes to cutting steel on these gantry cnc machines, they need to be made as ridged as possible if your using insert tooling like this. I usually gravitate to using insert tooling in steel due to them lasting far longer than solid carbide tooling. cheaper too.
@@BryanHowardthank you for your reply. Yea i have seen both of them and the “fly cutter” is actually pretty scary. Maybe having 2 inserts at 6k rpm is the way to go
The pump is one of those cheap pond pumps. 120V. No more than a $10-$14 pump. They can be found on Aliexpress all over, search "submersible pond pump ". Amazon if you want to spend 3x the price for quicker shipping. I have it connected to the VFD through an internal relay on it. That way when the spindle turns on, the pond pump turns on as well. It’s been in the green coolant for probably 2 years now without any issues running.
Entertaining, as always. I was going to suggest, now the machine is back together, to cut another plate - so you never need to use the "one-time" plywood plate again. But I guess the only reason to make another one would be if it needed modifying - and you probably can't know what needs modifying yet... so I'll keep that suggestion to myself 🙂 Thanks for the videos.
When I originally cut all the plates for the Z axis I was stupid and didn’t square the gantry very well. I’d probably redo all the plates at some time in the future. Possibly change the tram plate slightly so I can box in the sides of the spindle to make it slightly more rigid.
How are you planning to meassure the distance in the ball bar?
Wanna make some way covers? Bellows, they can be made diy with tape
Yeah I gotta sort out all the exposed rails and ball screws. It’s becoming a problem with all the chips getting in there. I have some stainless steel sheet left over that i’ll used to make some wall for the Y axis. Still trying to figure out the best solution on covering the X axis rails top/bottom and ballscrew. Z axis I can wall off the sides and figure something out for the bottom rails.
Hello Brayan, thank you a lot for putting up these videos and sharing your journey. I would like to know if the laser level and webcam system can be used for surface flatness assessment as I intend (hope) to replicate for that purpose. in the affirmative, what would the limits for that system to asses surface flatness to microns ? can it replace other techniques like the one using autocollimator
It can be used for surface flatness but you’ll want to put in the effort in calibrating the lens to the laser. I’ve been using a Powell lens and to the best of my knowledge it’s the best lens to convert a laser dot to a line.
The alignment of the lens to the laser is absolutely crucial to be able to use it for measuring the flatness of a 2d plane. We want the beam to be in the shape of a straight horizontal line.
This can be calibrated by projecting a dot parallel to a wall surface. Place 3 points equal distance from each other along that line for the webcam. You’d want some adjuster to raise and lower the platform the webcam sits on so you can get each spot dialed in to zero out on the laser.
After you know you have 3 points in space that are straight to each other, you have a reference you can use to adjust the angle of the powell lens. This can be done by projecting the laser line perpendicular to the 3 points aimed at the middle one and just the pitch axis of the Powell lens till all 3 points zero back out.
When you can zero out on all 3 points perpendicular to the original reference point, you’ll have a straight line generator that you can use for mapping out a surface with.
I’ll have to make more videos and one showing this whole process in the future.
Here’s the results of probing a very small object with 10,000 points ruclips.net/video/xf5xHpGlwvQ/видео.html
You can also do a similar approach by sampling nothing (air) on the CNC to map out how accurate your XY axis is. If there is any twist in a dual Y setup such as my CNC or if there's any warping of high and low spots on the rail surface.
@@BryanHoward thank you very much indeed for your reply. I will definitely investigate more into it and also read all information you already made available. It would be a huge help if you'd find time to investigate more into this flatness testing application and more videos from you regarding that would be highly appreciated. Also will share any feedback regarding adopting this method in our project.
hey man been watching your channel for a while, also working on my own cnc too. what i was wondering was how low of an RPM the spindle can be operated at. you did try some agressive drilling at 6k rpm, but how about milling/drilling steel? can it handle 3000rpm os we can do some light cuts on steel? thank you in advance
yeah the spindle isn't designed for it. You can get down to low rpms depending on how much load is going onto the spindle.
This is about as low as I've gone in steel @ 3k rpm: ruclips.net/video/guddMZIETKs/видео.html
and going in the opposite direction here's 24k rpm in steel: ruclips.net/video/8v-yXavlFhU/видео.html
when it comes to cutting steel on these gantry cnc machines, they need to be made as ridged as possible if your using insert tooling like this. I usually gravitate to using insert tooling in steel due to them lasting far longer than solid carbide tooling. cheaper too.
@@BryanHowardthank you for your reply. Yea i have seen both of them and the “fly cutter” is actually pretty scary. Maybe having 2 inserts at 6k rpm is the way to go
@@TheRealBleach1000 I don't think my machine is ridged enough to handle it at 6k. Probably not rated for that rpm either.
4:11 name of the coolant pump? I tried googling the model number but nothing comes up.
The pump is one of those cheap pond pumps. 120V. No more than a $10-$14 pump. They can be found on Aliexpress all over, search "submersible pond pump ". Amazon if you want to spend 3x the price for quicker shipping.
I have it connected to the VFD through an internal relay on it. That way when the spindle turns on, the pond pump turns on as well. It’s been in the green coolant for probably 2 years now without any issues running.
@@BryanHoward thanks for responding, i will check them out. 🤝