Grizzly G1006 Milling Machine CNC Conversion
27 Oct 2012 23:14 - 27 Oct 2012 23:37 #25893
by BruceLayne
Replied by BruceLayne on topic Re:Grizzly G1006 Milling Machine CNC Conversion
I've been slugging away in the background, with too many projects and not enough time devoted to any one of them.
The motors and motor drivers for my low cost and simple CNC upgrade of my milling machine didn't ship for three days. They shipped UPS ground, but they're crossing the country, so they won't be here for four more days. In theory, I could work from prints, but I have enough other stuff to do, so I'm not twiddling my thumbs. I prefer to wait until I have parts in hand to get too serious about designing the motor mounts, but I did some pre-imagineering. I was leaning toward mounting the motors directly to the shafts of the Acme screws (later to be ball screws), but after staring at it a bit, I'm leaning toward a timing belt driven system for a couple of reasons. Turning the handles, I think the Y axis will need some torque advantage, at least until the Acme screw is replaced with a ball screw, so I'll probably step down the speed by a factor of 2 or so, but the biggest reason for the belt drive system is to tuck the motors in closer to the milling machine. Having these ten pound motors hanging way out in the air would look goofy and it's an invitation to bending or breaking something in my messy shop. I think I'll mount the X motor low on the left front, under the table, the Y motor along the right side of the cast iron base, and the Z motor on the top right of the head casting. The castings don't have square sides, so some of these mounts will have some shallow angles. For the most part, the motors don't need to be perfectly aligned with the shafts they're driving with a belt drive system, but the motor mounts do need to be reasonably rigid.
I was planning to make the motor mounts from aluminum, (aluminium for Andy
) as is the standard practice, but this morning's wake-up epiphany is that I can probably make the motor mounts from plastic... er, I mean "an engineering polymer". The torque won't be that great given the inherent gearing in the Acme screws. The motor mounts only need to withstand forces comparable to those generated by turning the handles by hand. Plastic is cheaper than aluminum and even easier to machine. For anyone following along wanting to do something similar, or maybe make a CNC router, plastic can be machined using tools made for wood working. I may yet use aluminum, but I might have enough plastic in my scrap pile to machine the parts that mount to the milling machine, so when the motors arrive, I can make the parts that mount the motors to the brackets already mounted to the milling machine, order six pulleys and three belts, and I'm good to go.
Tooling
You don't need to hang around a machine shop very long to realize that it's not unusual to pay as much to tool-up a machine as you spent on the machine. In the case of a "cost effective" (to put it kindly) old import machine, the point of optimal usability will probably require more money spent on tooling than was spent on the machine. While waiting for motors and motor drives this week, I've been shopping for tooling. I've spent some time thinking about how I've used my milling machine manually for the last decade or so, and how I expect to use it after the low cost CNC upgrade.
The biggest problem I've had with the imported bench mill is the limited 5 inch quill travel and the round column that replaces the knee on a knee mill to make larger adjustments to the Z travel. Typically, I raise or lower the head on the column to find the sweet spot for whatever object I'm machining, set a zero reference point, and machine the part. All Z motion is then confined within that 5" of travel. That gets to be a big problem If I start off machining with a short end mill in an R8 collet, and then need to switch to a long drill bit in a long R8 keyless chuck. Invariably, I'm forced to raise the head on the round column, and that loses my XYZ zero, so I need to rezero everything. It's a major pain. The 5 inch Z travel problem will only get worse with CNC.
Limited Z Travel Fix #1 - Shorter Drill Bits
I had been using a full set of R8 collets in 1/64" increments to hold the jobber length drills, and that helped a lot compared to having the long drill bits protruding from a long keyless chuck. I decided to take that concept one step farther. I just bought a couple of sets of cobalt drills (89 drill bits) in the shorter screw machine length. Enco had them on sale (use-enco.com).
That'll help a lot more with the limited Z axis travel problem and it'll also result in stiffer drill bits that wander less. Maybe I'll pick up the 26 piece letter drill set in cobalt screw machine length when they're on sale!
Limited Z Travel Fix #2 - Shorter Tool Holders
The R8 collets are as short as they get. They mount almost entirely inside the spindle. But the top of the R8 collet is long once its removed from the spindle, so I still needed some clearance underneath when dropping a tool out of the spindle. To use as much of the limited Z axis travel I have, I need to make all of the tools as close to the same length as possible. I decided to go whole hog and get some Tormach Tooling System ER-20 collet holders, so I ordered a set of ER-20 collets. I bought them first, because they probably have a long lead time from China.
www.ebay.com/itm/280801550138
I was finishing up a Tormach TTS order that included their TTS ER-20 collet holders, TTS set screw end mill holders, TTS Jacobs chucks, and their cool heavy duty TTS carbide insert SuperFly cutter, when I stumbled on a new YouTube video describing mostly TTS compatible ER-20 collet holders from China for under $16 each! Tormach gets $57 for their TTS ER-20 tool holders. You'd need to turn an annular ring on the OD of the collar of the Chinese version to use the $16 versions in the Tormach automated tool changer, and they don't have the taper cut on top of the 3/4" shaft. Both would be quick and easy features to add on a lathe... particularly if you were doing 10-20 at a time.
I just ordered ten of the Chinese ER-20 collet holders.
www.ebay.com/itm/251139490102
If I like them, I'll get ten more, order more ER-20 collets, and call it done. If I can get ER-20 collet holders for $16, I don't see much point in getting more expensive set screw end mill holders. I'll probably get one of the Jacobs chucks, rather than messing with finding the right collets if I'm swapping a lot of drill bits, but I see all production tooling as using the low cost ER-20 collet system. I'll buy a lot of individual ER-20 collets in the sizes I need for the fixed tooling... mostly 3/8" and 1/2".
It'll be very nice to have a tool table with known Z axis offsets for a bunch of tools that I use a lot. I'm looking forward to some of these CNC benefits.
A Poor Man's Automated Tool Changer?
I'm also thinking of a way to devote the outlying areas of the 27" X travel on my milling machine that I seldom need to make tool caddies that substitute for an automated tool changer. It's an ambitious long term project that's still in the pipe dream stage. The tools aren't indexed, and that could be a problem. My preliminary solution is to have a decent resolution spindle encoder with an index pulse to index the spindle before the G code goes to the tool caddy position, lowers into place and releases the tool, raises up, goes to the next tool location, lowers onto the tool and grabs it. Teaser: I'm thinking of using magnets to maintain/realign tool index alignment in my homemade tool caddies. I might have room for 8 to 16 tools, depending on how much I'm willing to restrict my work space. The nut I haven't cracked is how to keep chips and other debris off the tops of the tools when they're resting in their tool caddies. I have an idea on that too, but it's still percolating. I'll need a lot more I/O for this level of machine complexity, but it should theoretically be within grasp, implemented incrementally. One thing at a time, though.
I think I'll work on my laser project some more.
The motors and motor drivers for my low cost and simple CNC upgrade of my milling machine didn't ship for three days. They shipped UPS ground, but they're crossing the country, so they won't be here for four more days. In theory, I could work from prints, but I have enough other stuff to do, so I'm not twiddling my thumbs. I prefer to wait until I have parts in hand to get too serious about designing the motor mounts, but I did some pre-imagineering. I was leaning toward mounting the motors directly to the shafts of the Acme screws (later to be ball screws), but after staring at it a bit, I'm leaning toward a timing belt driven system for a couple of reasons. Turning the handles, I think the Y axis will need some torque advantage, at least until the Acme screw is replaced with a ball screw, so I'll probably step down the speed by a factor of 2 or so, but the biggest reason for the belt drive system is to tuck the motors in closer to the milling machine. Having these ten pound motors hanging way out in the air would look goofy and it's an invitation to bending or breaking something in my messy shop. I think I'll mount the X motor low on the left front, under the table, the Y motor along the right side of the cast iron base, and the Z motor on the top right of the head casting. The castings don't have square sides, so some of these mounts will have some shallow angles. For the most part, the motors don't need to be perfectly aligned with the shafts they're driving with a belt drive system, but the motor mounts do need to be reasonably rigid.
I was planning to make the motor mounts from aluminum, (aluminium for Andy
) as is the standard practice, but this morning's wake-up epiphany is that I can probably make the motor mounts from plastic... er, I mean "an engineering polymer". The torque won't be that great given the inherent gearing in the Acme screws. The motor mounts only need to withstand forces comparable to those generated by turning the handles by hand. Plastic is cheaper than aluminum and even easier to machine. For anyone following along wanting to do something similar, or maybe make a CNC router, plastic can be machined using tools made for wood working. I may yet use aluminum, but I might have enough plastic in my scrap pile to machine the parts that mount to the milling machine, so when the motors arrive, I can make the parts that mount the motors to the brackets already mounted to the milling machine, order six pulleys and three belts, and I'm good to go.Tooling
You don't need to hang around a machine shop very long to realize that it's not unusual to pay as much to tool-up a machine as you spent on the machine. In the case of a "cost effective" (to put it kindly) old import machine, the point of optimal usability will probably require more money spent on tooling than was spent on the machine. While waiting for motors and motor drives this week, I've been shopping for tooling. I've spent some time thinking about how I've used my milling machine manually for the last decade or so, and how I expect to use it after the low cost CNC upgrade.
The biggest problem I've had with the imported bench mill is the limited 5 inch quill travel and the round column that replaces the knee on a knee mill to make larger adjustments to the Z travel. Typically, I raise or lower the head on the column to find the sweet spot for whatever object I'm machining, set a zero reference point, and machine the part. All Z motion is then confined within that 5" of travel. That gets to be a big problem If I start off machining with a short end mill in an R8 collet, and then need to switch to a long drill bit in a long R8 keyless chuck. Invariably, I'm forced to raise the head on the round column, and that loses my XYZ zero, so I need to rezero everything. It's a major pain. The 5 inch Z travel problem will only get worse with CNC.
Limited Z Travel Fix #1 - Shorter Drill Bits
I had been using a full set of R8 collets in 1/64" increments to hold the jobber length drills, and that helped a lot compared to having the long drill bits protruding from a long keyless chuck. I decided to take that concept one step farther. I just bought a couple of sets of cobalt drills (89 drill bits) in the shorter screw machine length. Enco had them on sale (use-enco.com).
337-2807 29PC 1/16-1/2X64 135D HERTEL COB SCR MCH DR SET $118.95
337-2814 60PC NO.1-60 135D HERTEL COB SCR MCH DR SET $105.95Limited Z Travel Fix #2 - Shorter Tool Holders
The R8 collets are as short as they get. They mount almost entirely inside the spindle. But the top of the R8 collet is long once its removed from the spindle, so I still needed some clearance underneath when dropping a tool out of the spindle. To use as much of the limited Z axis travel I have, I need to make all of the tools as close to the same length as possible. I decided to go whole hog and get some Tormach Tooling System ER-20 collet holders, so I ordered a set of ER-20 collets. I bought them first, because they probably have a long lead time from China.
www.ebay.com/itm/280801550138
I was finishing up a Tormach TTS order that included their TTS ER-20 collet holders, TTS set screw end mill holders, TTS Jacobs chucks, and their cool heavy duty TTS carbide insert SuperFly cutter, when I stumbled on a new YouTube video describing mostly TTS compatible ER-20 collet holders from China for under $16 each! Tormach gets $57 for their TTS ER-20 tool holders. You'd need to turn an annular ring on the OD of the collar of the Chinese version to use the $16 versions in the Tormach automated tool changer, and they don't have the taper cut on top of the 3/4" shaft. Both would be quick and easy features to add on a lathe... particularly if you were doing 10-20 at a time.
I just ordered ten of the Chinese ER-20 collet holders.
www.ebay.com/itm/251139490102
If I like them, I'll get ten more, order more ER-20 collets, and call it done. If I can get ER-20 collet holders for $16, I don't see much point in getting more expensive set screw end mill holders. I'll probably get one of the Jacobs chucks, rather than messing with finding the right collets if I'm swapping a lot of drill bits, but I see all production tooling as using the low cost ER-20 collet system. I'll buy a lot of individual ER-20 collets in the sizes I need for the fixed tooling... mostly 3/8" and 1/2".
It'll be very nice to have a tool table with known Z axis offsets for a bunch of tools that I use a lot. I'm looking forward to some of these CNC benefits.
A Poor Man's Automated Tool Changer?
I'm also thinking of a way to devote the outlying areas of the 27" X travel on my milling machine that I seldom need to make tool caddies that substitute for an automated tool changer. It's an ambitious long term project that's still in the pipe dream stage. The tools aren't indexed, and that could be a problem. My preliminary solution is to have a decent resolution spindle encoder with an index pulse to index the spindle before the G code goes to the tool caddy position, lowers into place and releases the tool, raises up, goes to the next tool location, lowers onto the tool and grabs it. Teaser: I'm thinking of using magnets to maintain/realign tool index alignment in my homemade tool caddies. I might have room for 8 to 16 tools, depending on how much I'm willing to restrict my work space. The nut I haven't cracked is how to keep chips and other debris off the tops of the tools when they're resting in their tool caddies. I have an idea on that too, but it's still percolating. I'll need a lot more I/O for this level of machine complexity, but it should theoretically be within grasp, implemented incrementally. One thing at a time, though.
I think I'll work on my laser project some more.
Please Log in or Create an account to join the conversation.
30 Oct 2012 04:16 - 30 Oct 2012 04:56 #26019
by BruceLayne
Replied by BruceLayne on topic Re:Grizzly G1006 Milling Machine CNC Conversion
Some Serious Turd Polishing Today
I received the vurra vurra nice Edge Technology Double Vise Stop on Friday, and I mounted it today.
That vise stop is crying out for some levers that can be operated by hand to lock the four adjusting screws (two big SHCS for side arm angle adjustment in the back, and two small set screws on the bottom front under the sliding bracket for adjusting the length of the stop rods).
Alas, my Chinese vise uses M12 cap screws where the vise stop mounts on the back, and Edge Technology assumes their customers have fancy US made vises, so they supply the fractional inch fasteners. The socket head cap screws also stick out a bit too much, and I already have very limited Y travel on my milling machine, so I decided to save the few mm and ordered M12X30 button head cap screws which have 4mm tall heads. The BHCS touches the base of the milling machine column at the same point as the lip on the bottom of the vise casting when the vise is all the way back, so I didn't lose any of the precious Y travel by adding the vise stop. Here's a picture from the trial fitting a few days ago with the vise's M12 SHCS, temporarily on loan from its previous duty attaching the rear jaw.
Dang! The sweet Edge Technology vise stop is making my cost effective Chinese machining vise look shabby. I see a $500 vise purchase at Glacern in my future.
This is addictive! Now I see why people call them vices instead of vises.
In addition to the M12 screws from McMaster-Carr (along with some Acme rod and nuts for my laser build) I also received some new spindle belts that I bought on eBay. I had read online that the Power Twist belts greatly reduce the spindle vibration, but I never had too much of that anyway. I've heard similar claims made for table saw blade vibration reduction.
Here are the old V belts for comparison. They were original equipment.
Honestly, I'm not sure that I can tell much difference. The spindle may run a bit smoother than before, but it's human nature to think it's better after spending some time and money on it. It's a good thing I'll be putting a three phase motor and VFD on there to adjust the spindle speed, because I always HATED moving the belts on the Tower Of Hanoi pulley puzzle, but the new Power Twist Plus Link V Belt is much more difficult to move on the pulleys, so I hate it even more. I was already machining too slow many times because it's slower overall on small jobs to spend time moving belts to be able to machine faster. I'm looking forward to "set it and forget it" where these spindle belts are concerned. I wish I could have eliminated on pulley and one belt, but the column is in the way and I need the intermediate pulley to avoid hitting the column.
More Stepper Motor Musings
I thought some more about this CNC retrofit over the weekend, and watched a lot of YouTube videos. I saw some similar machines with Acme screws direct driven by 527 ounce inch stepper motors, so I may not need any gear reduction for added torque when using my 1600 ounce inch stepper motors. As John pointed out, this is the holding torque spec, and the operating torque drops off quickly with step frequency, but the 527 vs. 1600 is an apples to apples comparison of holding torque ratings, even if the actual torque at speed is much less.
Part of the reason I went big on the stepper motors is because I wanted a lot of holding torque. When manually machining, I'll move the axis where I want it and if I'm going to do much machining there, I'll lock the axis in place with the brakes (screws that stop the ways from moving). With CNC, I won't have that ability, so I wanted motors that could hold the axis in place against some fairly aggressive machining. This is a 2 HP milling machine that's probably overpowered for its mass. The new three phase spindle motor is 1.5 HP, which is still not too shabby for a small bench top machine. If I'm hogging out some steel, I don't want the stepper motor slipping. That's probably not a realistic concern because the motors would be back driven through a high gear ratio with the ten turn per inch Acme screws, but I wanted to err on the safe side. I didn't want a quick and dirty system that worked, but when I upgraded to ball screws I'd lose the friction of the Acme threaded rod and might have stepper motors slipping when back driven.
The stepper motor drivers have automatic current reduction when idle. It's usually 50% current after 200 ms with no motion on that axis. I think these drivers might drop to 40% current when holding at idle. If they were being back driven and were slipping, I could always turn off the idle current reduction to keep the full 1600 ounce inches of holding torque. I wish there was a way to greatly increase the dwell time before the current reduction kicked in. I'd like to make it two minutes. Most machining I do might move an axis into position and then machine the part by moving the other axis for a little while, but it wouldn't be two minutes. Think of cutting a rectangular pocket. It's a lot longer than 200 ms, but it's not two minutes. After being idle for two minutes, the milling machine has probably finished the job and is truly idle. It'd be nice not to heat the motors when truly idle. Again, this is probably much ado about nothing. The almost instantaneous idle current reduction will probably be fine. If not, I'll turn off the current reduction and waste a little bit more electricity heating the motors.
The other reason that I got the big 1600 ounce inch motors was to have some reasonably fast rapid travels. Watching a CNC machine moving slower than I could manually turn the handles using my Armstrong drive didn't sound like much fun, and this is a fairly heavy little machine by bench top standards. It should scoot along with these motors. I wouldn't care much about the slow speed if it was unattended operation, but I suspect a lot of my uses for this CNC milling machine will be automating very simple jobs that run too fast to leave unattended, but slow enough that I don't want to wait on them. The Acme screws have low mileage. I plan on trying to wear them out in a hurry under CNC power, before swapping them for ball screws.
More Tools!
I spent a lot of online time this weekend looking at tools, CNC tool tables where the tool parameters are stored, tool holders, tool height measurements and offsets, etc.
I have a set of 12 imported ER-20 collets inbound, ten of the Chinese version of pseudo TTS ER-20 tool holders coming, as well as 89 Hertel cobalt screw machine length drill bits (fractional and numbered sets). I also have a $300 tap order that's pretty much ready to go (tip of the hat to John for his good info about CNC tapping in his previous link in this thread). I'm nowhere near ready to start machine tapping, but a girl can dream, can't she?
I also have a Tormach order mostly ready to go. I'm definitely getting the SuperFly 3" carbide insert cutter, largely based on their video showing surface finish and material removal rates at 1.5 HP. I loved the low cost approach of using a single carbide insert. Changing one insert is cheap and easy. On smaller machines, the material removal rate is largely determined by spindle power instead of RPM or number of cutters. The single cutter also helps to ensure a good surface finish when fly cutting. I hope to use this as a fly cutter and a facing mill cutter. I need to pop my keyless chuck off its R8 adapter and measure it. I think it's a #2 Jacobs chuck taper. I'll order the corresponding TTS adapter so I can have a keyless chuck for seldom used drill bits, which I seem to use very often! The various clearance drills and tap drills will have their own dedicated tool holders.
While watching lots of videos, I saw John Grimsmo and Hoss were both using a <$50 Craftsman powder coating gun they bought on eBay. It looked a bit flimsy, but it seemed to do a great job without needing an air compressor, so I bought one on eBay for $43. I had bought some black powder coating powder from Eastland a long time ago when ordering some stainless bluing solution, but never had a chance to try it. I think I'll powder coat the motor mounting brackets for my milling machine and several other little incidental parts I make for the shop. The matte black will also hide some of my crappy minor tooling marks. It's good to have some pride in your work. Maybe my CNC conversions won't have unfinished rough cut edges on the aluminum plates, with "ALCOA" clearly visible on the flat unmachined sides. Not that there's anything wrong with that. I don't want to offend anyone. 
More Tool Changer Musings
Speaking of turd polishing... I also spent some dream time this past weekend obsessing some more about a future automated tool changer, of sorts. It'd be a covered tool crib on one or both ends of the table. I could make them removable, so I could remove one or both and still have the machining space if needed, but most of my jobs don't use much of the milling machine's 27" X travel, so I could set the tool crib in place over the end of the table and tighten a couple of pointy set screws on the front side of the table to precisely return to zero each time so the taught positions of the tool cribs didn't change (in absolute machine coordinates). I could get 3-8 tools per side. This Turd Polishing Extreme upgrade is way down the road, if ever, but I wanted to consider it now. If it's no extra cost or time to accommodate that future possibility, then why not?
It'd use the Tormach Tooling System tool holders. Raise the quill to Z max, slowly rotate the tool to the proper indexed position for loading and unloading, rapid move in XY, slide the tool crib lid to open it (the table moved under the tool, so the lid on the tool crib opens into the same footprint that was occupied by the table, so no collision surprises in my cluttered shop), descend to the unloading height of the tool crib, make a slow sideways move so the U shaped tool crib cutout engages the slot in the TTS tool holder, and magnets added to the TTS tool holder align/realign the tool holder with magnets in the tool crib. Activate the power draw bar to release the TTS tool from the TTS tool collet, and raise the spindle .100". Check to see if that tool crib's limit switch is detecting the tool is in the crib, or did the tool hang in the TTS tool collet and raise the spring loaded tool crib holder? If the tool drop was successful, raise to full height.
Picking up the next tool is mostly the reverse of the tool drop. That's the crazy plan, anyway.
The tool changes would probably be hard coded in the G code, rather than trying to use the automated tool change G code. At least, I think that's possible. I'd do more research and mock it up before building anything.
I received the vurra vurra nice Edge Technology Double Vise Stop on Friday, and I mounted it today.
That vise stop is crying out for some levers that can be operated by hand to lock the four adjusting screws (two big SHCS for side arm angle adjustment in the back, and two small set screws on the bottom front under the sliding bracket for adjusting the length of the stop rods).
Alas, my Chinese vise uses M12 cap screws where the vise stop mounts on the back, and Edge Technology assumes their customers have fancy US made vises, so they supply the fractional inch fasteners. The socket head cap screws also stick out a bit too much, and I already have very limited Y travel on my milling machine, so I decided to save the few mm and ordered M12X30 button head cap screws which have 4mm tall heads. The BHCS touches the base of the milling machine column at the same point as the lip on the bottom of the vise casting when the vise is all the way back, so I didn't lose any of the precious Y travel by adding the vise stop. Here's a picture from the trial fitting a few days ago with the vise's M12 SHCS, temporarily on loan from its previous duty attaching the rear jaw.
Dang! The sweet Edge Technology vise stop is making my cost effective Chinese machining vise look shabby. I see a $500 vise purchase at Glacern in my future.
This is addictive! Now I see why people call them vices instead of vises.In addition to the M12 screws from McMaster-Carr (along with some Acme rod and nuts for my laser build) I also received some new spindle belts that I bought on eBay. I had read online that the Power Twist belts greatly reduce the spindle vibration, but I never had too much of that anyway. I've heard similar claims made for table saw blade vibration reduction.
Here are the old V belts for comparison. They were original equipment.
Honestly, I'm not sure that I can tell much difference. The spindle may run a bit smoother than before, but it's human nature to think it's better after spending some time and money on it. It's a good thing I'll be putting a three phase motor and VFD on there to adjust the spindle speed, because I always HATED moving the belts on the Tower Of Hanoi pulley puzzle, but the new Power Twist Plus Link V Belt is much more difficult to move on the pulleys, so I hate it even more. I was already machining too slow many times because it's slower overall on small jobs to spend time moving belts to be able to machine faster. I'm looking forward to "set it and forget it" where these spindle belts are concerned. I wish I could have eliminated on pulley and one belt, but the column is in the way and I need the intermediate pulley to avoid hitting the column.
More Stepper Motor Musings
I thought some more about this CNC retrofit over the weekend, and watched a lot of YouTube videos. I saw some similar machines with Acme screws direct driven by 527 ounce inch stepper motors, so I may not need any gear reduction for added torque when using my 1600 ounce inch stepper motors. As John pointed out, this is the holding torque spec, and the operating torque drops off quickly with step frequency, but the 527 vs. 1600 is an apples to apples comparison of holding torque ratings, even if the actual torque at speed is much less.
Part of the reason I went big on the stepper motors is because I wanted a lot of holding torque. When manually machining, I'll move the axis where I want it and if I'm going to do much machining there, I'll lock the axis in place with the brakes (screws that stop the ways from moving). With CNC, I won't have that ability, so I wanted motors that could hold the axis in place against some fairly aggressive machining. This is a 2 HP milling machine that's probably overpowered for its mass. The new three phase spindle motor is 1.5 HP, which is still not too shabby for a small bench top machine. If I'm hogging out some steel, I don't want the stepper motor slipping. That's probably not a realistic concern because the motors would be back driven through a high gear ratio with the ten turn per inch Acme screws, but I wanted to err on the safe side. I didn't want a quick and dirty system that worked, but when I upgraded to ball screws I'd lose the friction of the Acme threaded rod and might have stepper motors slipping when back driven.
The stepper motor drivers have automatic current reduction when idle. It's usually 50% current after 200 ms with no motion on that axis. I think these drivers might drop to 40% current when holding at idle. If they were being back driven and were slipping, I could always turn off the idle current reduction to keep the full 1600 ounce inches of holding torque. I wish there was a way to greatly increase the dwell time before the current reduction kicked in. I'd like to make it two minutes. Most machining I do might move an axis into position and then machine the part by moving the other axis for a little while, but it wouldn't be two minutes. Think of cutting a rectangular pocket. It's a lot longer than 200 ms, but it's not two minutes. After being idle for two minutes, the milling machine has probably finished the job and is truly idle. It'd be nice not to heat the motors when truly idle. Again, this is probably much ado about nothing. The almost instantaneous idle current reduction will probably be fine. If not, I'll turn off the current reduction and waste a little bit more electricity heating the motors.
The other reason that I got the big 1600 ounce inch motors was to have some reasonably fast rapid travels. Watching a CNC machine moving slower than I could manually turn the handles using my Armstrong drive didn't sound like much fun, and this is a fairly heavy little machine by bench top standards. It should scoot along with these motors. I wouldn't care much about the slow speed if it was unattended operation, but I suspect a lot of my uses for this CNC milling machine will be automating very simple jobs that run too fast to leave unattended, but slow enough that I don't want to wait on them. The Acme screws have low mileage. I plan on trying to wear them out in a hurry under CNC power, before swapping them for ball screws.
More Tools!
I spent a lot of online time this weekend looking at tools, CNC tool tables where the tool parameters are stored, tool holders, tool height measurements and offsets, etc.
I have a set of 12 imported ER-20 collets inbound, ten of the Chinese version of pseudo TTS ER-20 tool holders coming, as well as 89 Hertel cobalt screw machine length drill bits (fractional and numbered sets). I also have a $300 tap order that's pretty much ready to go (tip of the hat to John for his good info about CNC tapping in his previous link in this thread). I'm nowhere near ready to start machine tapping, but a girl can dream, can't she?
I also have a Tormach order mostly ready to go. I'm definitely getting the SuperFly 3" carbide insert cutter, largely based on their video showing surface finish and material removal rates at 1.5 HP. I loved the low cost approach of using a single carbide insert. Changing one insert is cheap and easy. On smaller machines, the material removal rate is largely determined by spindle power instead of RPM or number of cutters. The single cutter also helps to ensure a good surface finish when fly cutting. I hope to use this as a fly cutter and a facing mill cutter. I need to pop my keyless chuck off its R8 adapter and measure it. I think it's a #2 Jacobs chuck taper. I'll order the corresponding TTS adapter so I can have a keyless chuck for seldom used drill bits, which I seem to use very often! The various clearance drills and tap drills will have their own dedicated tool holders.
While watching lots of videos, I saw John Grimsmo and Hoss were both using a <$50 Craftsman powder coating gun they bought on eBay. It looked a bit flimsy, but it seemed to do a great job without needing an air compressor, so I bought one on eBay for $43. I had bought some black powder coating powder from Eastland a long time ago when ordering some stainless bluing solution, but never had a chance to try it. I think I'll powder coat the motor mounting brackets for my milling machine and several other little incidental parts I make for the shop. The matte black will also hide some of my crappy minor tooling marks.
It's good to have some pride in your work. Maybe my CNC conversions won't have unfinished rough cut edges on the aluminum plates, with "ALCOA" clearly visible on the flat unmachined sides. Not that there's anything wrong with that. I don't want to offend anyone. More Tool Changer Musings
Speaking of turd polishing... I also spent some dream time this past weekend obsessing some more about a future automated tool changer, of sorts. It'd be a covered tool crib on one or both ends of the table. I could make them removable, so I could remove one or both and still have the machining space if needed, but most of my jobs don't use much of the milling machine's 27" X travel, so I could set the tool crib in place over the end of the table and tighten a couple of pointy set screws on the front side of the table to precisely return to zero each time so the taught positions of the tool cribs didn't change (in absolute machine coordinates). I could get 3-8 tools per side. This Turd Polishing Extreme upgrade is way down the road, if ever, but I wanted to consider it now. If it's no extra cost or time to accommodate that future possibility, then why not?
It'd use the Tormach Tooling System tool holders. Raise the quill to Z max, slowly rotate the tool to the proper indexed position for loading and unloading, rapid move in XY, slide the tool crib lid to open it (the table moved under the tool, so the lid on the tool crib opens into the same footprint that was occupied by the table, so no collision surprises in my cluttered shop), descend to the unloading height of the tool crib, make a slow sideways move so the U shaped tool crib cutout engages the slot in the TTS tool holder, and magnets added to the TTS tool holder align/realign the tool holder with magnets in the tool crib. Activate the power draw bar to release the TTS tool from the TTS tool collet, and raise the spindle .100". Check to see if that tool crib's limit switch is detecting the tool is in the crib, or did the tool hang in the TTS tool collet and raise the spring loaded tool crib holder? If the tool drop was successful, raise to full height.
Picking up the next tool is mostly the reverse of the tool drop. That's the crazy plan, anyway.
The tool changes would probably be hard coded in the G code, rather than trying to use the automated tool change G code. At least, I think that's possible. I'd do more research and mock it up before building anything.
Please Log in or Create an account to join the conversation.
30 Oct 2012 11:54 #26045
by BruceLayne
The stepper motor kit I bought ships with three 60V at 5.8A switching power supplies, one per axis. I figure the manufacturer probably knew what they were doing when they bundled it.
I took a look at my well made Polish knockoff of an Albrecht keyless chuck. I wanted to tap it out of the R8 to Jacobs taper adapter so I could buy a TTS Jacobs taper adapter and use my nice keyless chuck, but I remember that it fell out of the R8 adapter, even though I thoroughly degreased the mating parts and wasn't using it to hold an endmill or any other tool that generates lateral loads. After that, I epoxied the two parts together. I guess I'll sell it on eBay as a one-piece unit, along with a lot of other nice R8 tooling I bought and used little or not at all before deciding to use TTS style tools in my milling machine after its CNC retrofit.
I'm still thinking about indicating tools. After some more research, if I decide I need anything else, it'll either go in an ER-20 collet in a TTS style collet holder, or I'll buy a TTS style tool from Tormach in a future order.
Replied by BruceLayne on topic Re:Grizzly G1006 Milling Machine CNC Conversion
I'm only using one side on the vise stop. I removed the other arm to have better access, as I do some jobs where I hold a fixture in the vise and push a spinning spindle mounted part into the fixture tooling. I need access to the left of the vise to load and unload parts. I keep the other vise stop arm handy, in case I want to machine two parts in the vise at the same time. I never do this now, but I expect to do this after it's a CNC milling machine, so I went with the double.I like my Edge vise stop but I got the other one that can move more and you can just swap the arm from side to side.
I understand about inductive reactance. I are an electrical engineer.Something to think about on stepper drives the higher voltage the more "power" and "acceleration" you can get from a stepper. A 24v system can be rather slow to watch compared to a 80v drive.
The stepper motor kit I bought ships with three 60V at 5.8A switching power supplies, one per axis. I figure the manufacturer probably knew what they were doing when they bundled it.
32428 - TTS SuperFly Cutter Kit $148.50
Insert 32653 Quantity Options: 32653-1 - Single Carbide Inserts for Aluminum, Non-Ferrous, Plastic, Cast Iron (1)
Insert 32654 Quantity Options: 32654-1 - Single Carbide Inserts for General Purpose Steel (1)
30106 - 22mm Wrench for TTS-ER20 $7.70
30151 - 30mm Wrench for TTS-ER20 Nut $7.70
32087 - TTS R8 Adaptor Collet for Tormach ATC $23.75
32730 - Screw - TTS SuperFly Cutter $3.01
30149 - Keyless Drill Chuck - 5/8" - JT6 $47.85
31828 - TTS-JT6 $19.11
Total $257.62I took a look at my well made Polish knockoff of an Albrecht keyless chuck. I wanted to tap it out of the R8 to Jacobs taper adapter so I could buy a TTS Jacobs taper adapter and use my nice keyless chuck, but I remember that it fell out of the R8 adapter, even though I thoroughly degreased the mating parts and wasn't using it to hold an endmill or any other tool that generates lateral loads. After that, I epoxied the two parts together. I guess I'll sell it on eBay as a one-piece unit, along with a lot of other nice R8 tooling I bought and used little or not at all before deciding to use TTS style tools in my milling machine after its CNC retrofit.
I'm still thinking about indicating tools. After some more research, if I decide I need anything else, it'll either go in an ER-20 collet in a TTS style collet holder, or I'll buy a TTS style tool from Tormach in a future order.
Please Log in or Create an account to join the conversation.
Moderators: cncbasher
Time to create page: 0.116 seconds