That makes sense.
-
Scam Alert. Members are reminded to NOT send money to buy anything. Don't buy things remote and have it shipped - go get it yourself, pay in person, and take your equipment with you. Scammers have burned people on this forum. Urgency, secrecy, excuses, selling for friend, newish members, FUD, are RED FLAGS. A video conference call is not adequate assurance. Face to face interactions are required. Please report suspicions to the forum admins. Stay Safe - anyone can get scammed.
Weiss VM32 CNC Conversion
- Thread starter DavidR8
- Start date
I've asked about minimum number of slots. I have 60 on mine or 360 edges.
I 3D printed a disk for my Gingery to use with 3 slotted sensors. I had to paint the black PLA with metalic paint. The senso IR went right through the PLA. I tried different colours. No difference.
I'd post a picture but we're at YVR waiting for our flight.
For my mill I cut slots in 16g aluminum.
drewnabobber
Member
Thread milling has no requirement for any kind of synchronization between the spindle and the tool. It's just cutting metal with a specially shaped tool.
Yes, I'm not sure why I had it in my head that thread milling required spindle/A-axis sync. It's basically a helix into/out of the hole.
Sorry @DavidR8 - I am lost. I should just keep my nose out of it. But maybe a few stupid questions from the back of the class can bring more clarity to the teacher.
Are you suggesting that you will NEVER do regular threading? If so, then fine. You will only need to coordinate advance (z) with x & y.
But if you plan to do regular threading at some point in the future, the rate of advance is the thread pitch which depends on where the spindle is in its rotation.
Are you suggesting that you will NEVER do regular threading? If so, then fine. You will only need to coordinate advance (z) with x & y.
But if you plan to do regular threading at some point in the future, the rate of advance is the thread pitch which depends on where the spindle is in its rotation.
If by regular threading you mean chucking a tap in a collet , running the spindle at some low rpm and manually advancing the tap into the work then no because I won’t have the ability to manually control the Z axis advance at the rate dictated by the spindle speed and the pitch of the tap.
The Z axis is going to be controlled by a servo motor which is why to do rigid tapping an encoder is required on the spindle. The encoder tells the CNC controller exactly how much/far the spindle has rotated so the CNC controller can calculate how much to advance the Z axis.
I see. Well if you ever want to do rigid tapping, I think you will need a lot more sensing locations (pulses per rev) on the rotational encoder. But you might be able to increase the number of pulses by adding more sensors in a phased "or" array (any one generates a pulse) instead of adding more slots. They don't have to be located all around the perimeter. They can all be in one phased cluster. It's a handy way of improving resolution without changing the number of slots.
Yes that is what I’m learning.
It’s all a bit moot at this point though because I’ve discovered that the CNC controller I’m using doesn’t accept the g-code commands that are required for rigid tapping.
So thread milling is how I will be cutting threads.
I’m completely fine with this because it allows for more control over thread fit because I can specify how deep the thread is cut.
Another option for tapping on a cnc mill is a "tension-compression" head. this allows for a deviation to a degree between the commanded z travel and the actual amount the tap enters the work. I used one for the first time a few weeks ago, nerve wracking but exciting, and sure was faster than figuring out threadmillling and 20x faster than hand tapping. I'm still trying to figure out threadmilling, its easier if you buy a threadmill for the specific thread and pitch, but I'm trying to use an inexpensive metric threadmill (M6 x1) to cut an imperial thread (1/4-20). I think in theory it seems like it should work but I ran out of play time to keep tweaking it,
drewnabobber
Member
In the above vid I am using a 1.0mm pitch threadmill to cut a 1”-27 thread. Big advantage of threadmilling is that with a single thread cutter you can do many different and weird threads and hole sizes.
By tension-compression head do you mean a tapping head like a Tapmatic or something different?
Edit: found what you meant with a bit of googling
Well my eyecrometer needs to be sent out for calibration because when I tried to install the servo motor on the mill head the holes are out by about a 1/16".
Blast, I guess I couldn't get that lucky.
On the plus side, the wire guide, Dinkle terminal blocks and DIN rail have all arrived. Just need the enclosure now. Come on UPS!
Blast, I guess I couldn't get that lucky.
On the plus side, the wire guide, Dinkle terminal blocks and DIN rail have all arrived. Just need the enclosure now. Come on UPS!
slow-poke
Ultra Member
Comments from the peanut gallery.......
1) Try to keep your signal and motor power wiring physically separated.
2) Keep all wiring as short as possible. Turn those servo drivers 90 degrees so the power connectors are close to the supply, and the motor wires can exit the case to the RHS.
3) Avoid making wiring loops (they make great antennas)
4) Don't daisy chain power, especially if high current, wire each driver back to the source.
5) Twisted pairs for signals for example "step" are better than a bunch of single conductors.
6) The G3 drivers have differential outputs use them instead of single ended.
7) You might need a small fan at the top of the enclosure for cooling (probably not) but would be wise to have a little space if needed.
1) Try to keep your signal and motor power wiring physically separated.
2) Keep all wiring as short as possible. Turn those servo drivers 90 degrees so the power connectors are close to the supply, and the motor wires can exit the case to the RHS.
3) Avoid making wiring loops (they make great antennas)
4) Don't daisy chain power, especially if high current, wire each driver back to the source.
5) Twisted pairs for signals for example "step" are better than a bunch of single conductors.
6) The G3 drivers have differential outputs use them instead of single ended.
7) You might need a small fan at the top of the enclosure for cooling (probably not) but would be wise to have a little space if needed.
Last edited:
Good call on rotating the drives. I was thinking that would be a better orientation.
One design consideration is that the enclosure has a opening in the bottom that is probably 4" x 12" which is where I'd like all the cabling to enter/exit.
All of the servo cables are pretty stout, thicker than a 14/2 extension cord and there are eight of them that have to snake their way in. Two have to go to the drive in the upper left corner.
The cables all have non-removable plugs molded onto them so they are going to be trickly to manage.
We'll all do our best to throw peanuts after the wiring is complete.
Actually excellent suggestions from @slow-poke. Gives you something to chew on.
@jcdammeyer do you leave your servo drives powered on all the time or are they switched?
slow-poke
Ultra Member
My attempts at tidy control panel wiring never seem to turn out very well. Partially because the emphasis is on function instead of aesthetics. The electrician types that wire panels for a living really do have a talent for making them look good.
A few images from my mill panel that might help....
Notice the yellow & red power wiring to the T6 drivers, it might be messy, however it's kept separate from the control wiring above and is as short as possible and there is a separate feed from the source to the driver no daisy chaining. I added a ferrite bead for good measure. In hindsight I should have run the control wiring to the right instead of the left to avoid the power wiring altogether, as it is it crosses at 90 degrees so fairly benign but would be better the other way. Also note the control wiring is all twisted pair, tip of the day use your cordless drill to twist the pairs.
The T6 servo connectors are quite large, however by temporarily removing the shells (T6 end) and drilling out the center protrusion of the cable clamps (note round holes in center four cable clamps) I was able to get the encoder cables through standard cable clamps.
I need to share the DRO pickup data because I have a separate DRO and also need to get the DRO signals back to Linux. So I made a simple DB9 splitter PCB that mounts to the bottom of the cabinet, and uses small JST connectors to get the data up and into the cabinet to the MESA board. Not having to make ten DB9 cutouts and then parallel the wiring was well worth the $10 cost of the splitter board, and was a great excuse to use the Greenlee punch for three more holes. BTW Greenlee does have DB9 punches, but Mach also has a wizard for electrical shapes.
For the cable clamp holes, I have a set of Greenlee punches and they make for nice clean holes. If you don't have Greenlee punches a step drill is a bit messier but will get the job done. Any excuse to purchase new tools is a good excuse so If you do go in search of Greenlee punches just be aware that the advertised size is not the actual hole size, it's the conduit size. These Greenlee punches are plentiful on Kijiji I think I paid $25 for a complete set, the old ones with the bearings are the better ones if you get to choose. Apparently the Amazon clone specials are complete and total junk from what I have read.
A few images from my mill panel that might help....
Notice the yellow & red power wiring to the T6 drivers, it might be messy, however it's kept separate from the control wiring above and is as short as possible and there is a separate feed from the source to the driver no daisy chaining. I added a ferrite bead for good measure. In hindsight I should have run the control wiring to the right instead of the left to avoid the power wiring altogether, as it is it crosses at 90 degrees so fairly benign but would be better the other way. Also note the control wiring is all twisted pair, tip of the day use your cordless drill to twist the pairs.
The T6 servo connectors are quite large, however by temporarily removing the shells (T6 end) and drilling out the center protrusion of the cable clamps (note round holes in center four cable clamps) I was able to get the encoder cables through standard cable clamps.
I need to share the DRO pickup data because I have a separate DRO and also need to get the DRO signals back to Linux. So I made a simple DB9 splitter PCB that mounts to the bottom of the cabinet, and uses small JST connectors to get the data up and into the cabinet to the MESA board. Not having to make ten DB9 cutouts and then parallel the wiring was well worth the $10 cost of the splitter board, and was a great excuse to use the Greenlee punch for three more holes. BTW Greenlee does have DB9 punches, but Mach also has a wizard for electrical shapes.
For the cable clamp holes, I have a set of Greenlee punches and they make for nice clean holes. If you don't have Greenlee punches a step drill is a bit messier but will get the job done. Any excuse to purchase new tools is a good excuse so If you do go in search of Greenlee punches just be aware that the advertised size is not the actual hole size, it's the conduit size. These Greenlee punches are plentiful on Kijiji I think I paid $25 for a complete set, the old ones with the bearings are the better ones if you get to choose. Apparently the Amazon clone specials are complete and total junk from what I have read.
Last edited: