AC Servo Motors and FogBuster Mist Coolant (DIY CNC Mill Upgrades 1)

hey what's up guys it's dr d flow and i can't believe it's been almost a year since i converted this manual mill into a cnc one now if you saw that video you'll know that this was a relatively bare bones conversion with me simply swapping out the lead screws for ball screws and the handles for stepper motors nevertheless i was able to tackle some really awesome projects namely the cyber scooter which consisted of over 13 unique parts machined on this mill however there are some quality control issues that i want to rectify because the end goal is to produce some unique products on this mill to possibly sell to you to do this we're going to be upgrading the mill in this video and in some future videos to not only improve part quality but also the speed at which i can produce parts dr d flow now it may be obvious what the first upgrade is with the stepper motors sitting on the table and yes we're upgrading to dmm ac servo motors now it may be kind of sacrilegious for me to install servo motors on the mill because my logo is a stepper motor but i'm really excited to walk you guys through all the benefits of servo motors but to reap those benefits we need to address other inaccuracies that are present on the mill the biggest culprit are these aluminum flex couplers now these join the ball screw to the stepper motor and they're fine for lightweight machines like the plasma cutter and the 3d printer but with a heavy duty machine like the mill taking a deep cut i actually noticed the couple winding and unwinding as the mill switched directions it's really not good when you can see backlash now the solution is an easy one albeit a little bit expensive i opted for these circle grade disc couplers which shouldn't have that problem so super excited about these as well now a more complicated solution to possible backlash that is present in the mill is to upgrade to these double nut ball screws now these nuts will tension against each other and minimize backlash in the axes but we're actually not going to install these today because i have to take apart the whole mill and the plan is to do another upgrade alongside of that so get subscribed for that but i am also very excited about these double nut ball screws but let's get back to the upgrades that we're doing today so this is my old vice it's obviously broken now and one of the problems with it was that the movable jaw actually went up and down like a roller coaster as you clamp down your part now this made your my part sit not flat against the parallels and this was especially troublesome when i was flipping my parts in between operations so my girlfriend bought me a curt vice and you know what they say there's curt and there's all other vice manufacturers the jaw moves like butter i'm so excited i opted for the magnetic jaw so that i no longer have to make sure that the parallels are following along as i open and close the vise sweet great quality of life and accuracy upgrade the last addition i'm going to make today is adding a coolant system hinch manufacturing reached out and said hey david you got do you want to check out our new line of mini fog busters and briefly what this system does is it sputters coolant right at the effector zone so where you're cutting the material and it keeps the end mill cool for a longer lifespan and keeps the park cool for you know just better cut quality and with the compressed air you get some chip evacuation too this system is perfect for me because as you know i don't have an enclosure so i went over and grabbed a servo motor and a stepper motor i want to highlight some differences before i start talking about the electronics stepper motors as you may know are open loop systems the microcontroller tells the stepper driver to move the motor does it move no one will ever know we just trust that it does and you know these are really reliable so they do move the servo motor operates in a closed loop it has a component that's known as an encoder that reports back the position of the motor microcontroller tells the driver to move the driver tells the shaft to move the encoder confirms that the shaft actually moved and reports it back to the driver if the shaft doesn't move in the case of a crash or large resistance the encoder reports back in air and the microcontroller relays and the driver relays that to the microcontroller so your machine will stop moving this closed loop system is one of the major benefits of the servo motor now i should mention that stepper motors also exist in closed loop variants so you know you don't have to shout all the cash for a servo motor to get this closed loop feedback so we're going to continue to talk about some more of the benefits of the servo but let's talk about these drivers because clearly there are some differences so this is my old enclosure really briefly it's just a dc power supply that supplies power to the stepper drivers now if we open up my new enclosure there is no dc power supply that's because the ac server drives take ac power i am supplying single phase but they will also take three phase and i want to talk about a couple different components that are present here as well these silver boxes are line filters and they're going to remove emi electromagnetic interference that could be present in the power coming in you don't want these drives misstepping speaking of you know misstepping and erratic behavior this box has contactors now contactors are electronically controlled switches that will connect the power allowing the drives to power up and these contactors are controlled by an e-stop circuit so if i press the e-stop it will physically cut power to the drives and this is a critical safety feature that was missing on my old enclosure i had a couple of scary crashes and all i had was that digital e-stop on my pendant here i can immediately cut power to the drives physically now once you wire the drives for power you just connect the motors and the encoders and then really getting data from the mesa 7i 76e which is the microcontroller i use because i use linuxcnc is simple you just need to send a step in direction to each drive as well as receive an alarm state so that if the motor stalls the mesa board can detect that that's about it we're going to go ahead and install those servo motors and those new couplers on the mill the first thing we need to do is swap out the aluminum helical couplers for the servo grade disc couplers now the nice part about these couplers is that they don't use a set screw to hold onto the shaft instead they clamp down so they won't mar up the ball screw shaft both the y and x axes will use couplers the z-axis uses a pulley and timing belt just going to reinstall these mounting plates the nice part about these servo motors is that they have the exact same form factor as the stepper motor so they're going to drop right into my existing motor mount plates as you can see here both the stepper motor and servo motor have the same mounting holes all right what's the y-axis the x-axis uses the exact same motor while i'm installing this motor i want to talk about another benefit of servo motors and that is that they don't suffer from resonance now stepper motors while they're very powerful at low speeds they also vibrate now this vibration comes from something known as ringing where when it takes a step it slightly overshoots the desired position and it gets drawn back now at certain rpms this ringing can become amplified and not only will the motor vibrate but the whole mill will vibrate here you can see when i'm probing this part at low speeds that there's a lot of vibration coming from the motors servo motors on the other hand are basically vibration free and they move super quietly the other benefit of the the servo motor is that they're efficient unlike stepper motors they don't draw current independent of load but they match the amount of power they need to how much weight they're moving the z-axis servo motor which will sit up here is not a drop in replacement for the old nemo 42 stepper motor so here is the old z-axis stepper motor and you can see with the new servo motor the shaft is longer now unfortunately with the current riser plates that i had on the mill the shaft would actually run into the column of the mill so bruce nelson over at heavy metal cnc machined me plates that are about a centimeter taller so make sure if you go with the servo motor and the heavy metal cnc conversion kit that you go with the taller riser plates there's one thing you have to be wary about when converting from servo to stepper motors when it comes to the z-axis and that is the tendency of the spindle head to want to crash into the table right now there's no motor but these screws are pushing into the gives preventing the head from falling but usually when i cut power to the mill the detent torque of the stepper motor it's very hard to move the shaft even when it's not receiving power prevents the ball screw from spinning freely so whenever the mill loses power the natural torque of this motor prevents the head from falling now the shaft on a servo motor without any power actually spins uh quite easily so my concern with installing a motor like this on the z-axis would be that when you lose power you know the mill is going to crash into the table now the solution for this is to use a servo motor with a spring set brake and basically these are the same motors but you can see there's a an extra module here at the end of the motor and what's in here are springs that are forcing a plate that's pressing into the rotor which prevents it from spinning whenever the motor loses power now when power is supplied to the brake an electromagnet becomes active which pulls that plate away from the rotor allowing the shaft to spin freely so for z-axis especially when there's no counterweight to prevent the head from crashing into the table you need to use a servo motor with a brake okay so next we need to install the pulley on the top of the z-axis okay time to put the motor on top as you can see that pulley is a nice height i'm going to go ahead and attach the belt okay now i'm going to grab my ladder so i mounted this new electrical cabinet using the same points as my old one obviously this cabinet's much bigger it's now way closer to the mill don't worry in my up next upgrade which i'm still keeping a secret i'm going to rearrange the positioning of the mill electrical cabinet pretty much everything so this is just going gonna be for testing purposes for now i know i sped through most of the wiring if you need more details check out the diagrams that are located on my website so now we just need to plug everything in so the brake is connected in the back of the z-axis motor and then if we come around to the cabinet the way this works is it's going to connect down here into this little aviation connector and whenever the z-axis drive which is right here turns on it'll flip this relay sending power to the brake and that will disengage the brake e-stop button plugs in down there and as i mentioned before it actually runs through the 24 volt power supply in through these contactors so once i plug in the machine the only thing that turns on right now is the mesa board but if i release the e-stop you're going to hear the contactor switch and about half a second later you're going to hear the brake release on the motor and that other click was the motor on the z-axis so like i said this controls everything and i'm going to mount it right here all right we're moving right along i went ahead and gave the table a little stoning just to remove any high spots before applying some lithium grease to protect it from rust i then slapped down my brand new shiny curt vice and i'm very excited to make parts but before we can do that there's still some software things we need to handle we need to take a couple extra steps to get these servo motors operating properly again to compare them to stepper motors which use discrete burst of current to actually move the rotor servo motors use continuous current supplied by the server drives also known as amplifiers now the amount and duration of this current is not only dependent on the distance left to travel but is also dependent on parameters that are generated during a process known as tuning the tuning process which we will perform in a second helps the drivers understand what kind of mass is attached to each motor clearly a heavy object like this table will require a lot more power to move and more time to accelerate and deaccelerate than something that is a lot lighter the tuning process will output parameters that will help the drivers understand how to move each load without overshooting the final position by too much or oscillating back and forth between moving too much and too little now to tune the motors you want everything to be as it will be when you're actually running the mill which is why i went ahead and put the vise on already the tuning process is actually quite easy with the dmm servo motors i have this usb cable that i'll plug into one drive at a time and the other end will go into my pc which i will run the tuning software so first we'll start off with the x-axis drive which is this one on the bottom we'll use this special tuning cable plugs right into the drive and the other end plugs into my pc open up the dmm application we will connect to the drive next we're going to go to auto tuning it's very important that there's enough room for the drives to move at least five rotations we'll select our motor [Music] for the x and y axes we're going to leave these settings as default so we're going to use the standard load inertia mismatch our mechanism is rigid because it's directly coupled to a ball screw and i'm going to go ahead and turn the torque filter on standard just to make sure we don't have any resonance [Music] and then after that we just start the tuning process so you can see in the beginning here it's doing a low frequency ramp not moving much [Music] high frequency ramp a little bit faster i can already tell these motors are going to be so much more quiet incredibly smooth too and that's it the tuning process for the x-axis motor is complete and there were three important parameters that were a result of that process and they are the main gain the speed gain and the integration gain now the main and speed gain are going to increase with loads of higher inertia so moving really heavy masses the integration gain which is a metric for servo stiffness where a higher integration gain would be best for high acceleration rapid movement performance would be decreased with heavier loads we're going to repeat this process for the y-axis i'm going to do it off-camera because it's basically going to be the same thing oops i almost forgot we need to switch this servo motor back to pulse direction mode because that is the command that the controller board is going to send to the driver and that finishes up the y-axis the parameters are very very similar and finally we'll switch the tuning cable to the z-axis drive now the z-axis is a little bit different than the other two axes because the servo motor does not drive the ball screw directly but instead there's a timing belt in between the two now what that means is when we go to tune it first off we have to select this big boy motor this 1.3 kilowatt with break we will keep the inertia mismatched as standard the load mechanism selection it's not going to be rigid but it's going to be a belt mechanism belts are less responsive so this is going to have a direct impact on that integration gain we saw before i'm going to change this to standard we are going to start the auto tuning process a little nervous because that motor is so big the results of the z-axis tuning process are as expected this axis has the highest inertia because it has to counteract gravity when you pull the spindle head upward so we have higher main gain and speed gain values and again the integration gain is going to be won not only because of the high inertia but also because this is a belt driven mechanism i hope that quick introduction demystified the process of tuning servo motors for you as someone that came from stepper motors i always wondered how complex the tuning process would be however with the cnc mill where the linear motion mechanics are relatively simple the auto tuning software was perfectly adequate i clicked a couple buttons and 10 minutes later all the motors were tuned now if you have some tips and tricks for getting servo motors to operate absolutely perfectly then please let us know in the comments down below because us beginners would really appreciate it now it's time to make sure that the drives communicate properly with the controller but because they take a step and direction signal i only had to make a couple changes to configuration file namely making sure that the controller looks out for possible fault signals from the drives that configuration file will be on my website i'm going to go ahead and get linuxcnc running and our first operational test will be to make sure that all the axes can home there goes the z-axis looking good there x-axis y-axis that was a simple test and you know more problems could arise down the line but it looks like everything's operating as it should let's switch gears and talk about coolant now obviously this is an open air mill and i can't run flood coolant but even if this mill was in an enclosure that doesn't mean the enclosure would be set up to recapture the flood coolant but i can tell you i'm tired of spraying at the part and the end mill with this spray bottle to keep everything cool so how can i automate this process without having copious amounts of liquid well minimum quantity lubrication systems exist like the fog buster that dispense just enough coolant to keep everything cool now we're going to talk in a bit why this thing is called the fog buster but its operating principle is relatively simple coolant sits in the reservoir and pressurized air forces it out the nozzle which is aimed directly at the tip of the end mill now because this system requires compressed air i'm going to have to plum some pneumatic tubing from my compressor over to the mill so let's head over to my bench to talk about that setup so here's the setup for the compressed air this pneumatic tube is coming from my compressor which will hook up to the inlet of this motor guard filter now in my affordable plasma cutter video i talk about how dirty and wet air is coming out of a compressor and even i was surprised how often i had to switch out the filters that are present within this unit if you haven't seen that video i would go ahead and check that out to learn more about that but once the air leaves the motor guard filter it comes to this manifold now right now i only need one air connection for the fog buster but in the future when i'm going to add a power drawbar i will need more connections to the compressed air so i decided to go ahead and incorporate a manifold now this dewalt manifold has two unregulated outputs and two regulated outputs controlled by this knob my air compressor is most often used to run my plasma cutter and so it's usually set at 90 to 120 psi and i don't want to expose the fog bus to that high of air pressure because it really only needs 10 to 20 psi so i will set the regulator to about 30 psi that regulated air will come down to the electric solenoid which when it receives 24 volts it'll actually open allowing air to flow through the system this is how the microcontroller will turn on the fogbuster at the beginning of an operation and turn it off at the end of an operation let's take a closer look at this circuit so here i have a breadboard that's being supplied 24 volts of power it's got some components on it that are interfacing with this relay now this relay may look funny but it's really just uh tall and skinny and able to sit on a din rail that's the only thing unique about it and every other aspect is just a normal relay this yellow button symbolizes one of the numerous outputs from the microcontroller when that output is active or when the button is pressed the relay will switch now the connection to the solenoid is hooked up to the normally open terminal so that it only receives 24 volts when the button is pressed or the output's active so if we hook up the solenoid press the button you can hear the solenoid actuating so if you're watching this video you're probably already aware of a simple relay circuit but i want to draw your attention to this diode now this is a flyback diode that's connected across the positive and negative inputs for the relay when the relay is switched off the electromagnet will experience a voltage spike as that magnetic field is converted back to current now this voltage spike could traverse back to the output of the microcontroller possibly damaging it but with the flyback diode it will divert that current away from the microcontroller and back to the relay where that current will go around in a loop until it dissipates through resistance it's always good practice to have a flyback diode when connecting a relay to a microcontroller so this time everything's pressurized so let's hit the button awesome i'm going to repeat this electronics in the cabinet and then we're going to carry everything over i went ahead and mounted the aluminum extrusion to two studs in the wall i may run more aluminum extrusion in the future because i just have so many different ways to connect to it i have a lot of different brackets left over from past projects such as the plasma cutter or my 3d printer builds now it's definitely getting crowded over in this corner i probably need to get the belt sander and grinder out of here but you know small garage problems to mount the fog buster reservoir i drilled and tapped two holes in the right side of the column on the right side it's easier to access the reservoir i can unscrew reload screw it back in pretty easily and i don't have to worry about mixing electronics in water however the quill handlebars are kind of in the way too i'm really close to just getting rid of the quill function on this mill it's nice to be able to do some manual drilling every now and then but i've been doing less and less of it so for now i'll probably keep the handles off two tubes go from the reservoir to the nozzle one tube carries compressed air and the other one carries coolant inside of the nozzle is where that patented mixing technology comes into play where the coolant is mixed with the compressed air but it's not atomized we will talk about why the process of atomizing which creates an aerosol is bad to do with coolant especially in a small shop like this in a couple of minutes i want to finish talking about the setup of the fog buster first so as supplied this tubing was a little bit too long and i ended up cutting about two and a half feet off of it now here's a little tip for you if you ever have trouble removing plastic hose from the barb run a lighter underneath it for a couple of seconds and then you can pull the tube off so i filled the reservoir up with duracut 7000b it's a lubricant that's meant for these micro or minimum lubrication systems now this is the mini fogbuster line and only take 16 ounces or about half a liter worth of lubricant the fogbuster comes in a couple different variants with there being a half gallon a gallon and i think a fogbuster with a 2 gallon reservoir but the nice part about having a smaller reservoir is that it'll be more responsive to turning on and off the air pressure because there's less volume that needs to be pressurized most of my cutting operations are pretty quick so i think that this is going to be more than enough especially when you see just how little coolant actually comes out when this thing's turned on so over here we have my pro basic linux gui with there being a button called mist now that button is connected to the output pin which is connected to the solenoid so when i click that button it will trigger the relay which turns on the solenoid now the amount of coolant is controlled by this needle valve you can really only see it when it accumulates on the paper towel so i don't have a high speed camera but maybe you can still discern some of that coolant coming out of the nozzle let me go ahead and turn it off so we can talk about atomizing now atomizing is the process of creating an aerosol and aerosol is just liquid that's suspended in the air and that's bad because if it's in the air then it's not on your part and that's just a great way to waste coolant especially when you've got flow rates that are this high also if the coolant's in the air and you're operating the mill you're most likely breathing in that coolant especially in a small shop where when i put a little wd-40 on the part when i'm cutting and the smoke literally rolls out of my garage when i open up my garage door you don't want to be atomizing coolant the way that the fog buster works is it creates droplets that are small enough to fly at the park but big enough that they don't become suspended in the air we got the servo motors installed the coolant's ready to go it's finally time to make some parts and i've got a great first project so because the side of the spindle heads kind of textured magnets tend to slide down them pretty easily so i'm going to make a permanent mount for the nozzle so that when i position it it doesn't fall away typically when i make small mounting brackets i will use a thicker than required stock material cut out the entire side profile in one operation and then flip over the part and cut off the excess material that was held in the vise now this is by far the easiest way to make a part but it doesn't tell you much about the accuracy and repeatability of the mill's linear motion and work holding so i'm going to make this part the hard way by cutting out the top half of the part in one op and then flipping it over and machining the back half the parting line that will be present on the side of the part will be an indication of how well everything is aligned now for this type of operation it's not recommended to use a vice even a curt vice because the slightest amount of rotation in the vice will be compounded two-fold when you flip the part over also the crushing force of the vice can distort the part especially as you remove material from the stock usually you want to use a special mounting plate that has dowels which register the features machined in the first operation but that would be extremely overkill for this part so we're going to make do with the vise um the first thing i want to draw your attention to is how quiet the motors are during the probing cycle they are not vibrating the casting like the stepper motors did after facing the stock i'm using a slotting operation to create two slots for z-axis adjustment of the nozzle i am using conservative speeds when it comes to the cutting operations because these servo motors are at least twice as powerful as the stepper motors that they replaced and i would prefer not to figure out how much force is required to stop the motors from turning with a lot of these adaptive toolpaths the mill spends a lot of time retracting the end mill and moving it to the next position if you're new to cnc motion these non-cutting movements are typically called with a g0 command which results in a high velocity movement known as a rapid even if you use low cutting speeds because your mill lacks rigidity or the spindle can't output enough rpm you can speed up machining times by having quicker rapids and with servo motors you can easily wrap it at 400 inch per minute or 10 000 millimeters per minute on a mill with dovetail ways technically these dmm motors have a max rpm of 3000 when supplied 120 volts and 5 000 when supplied 240 volts but those speeds would require linear rails a larger diameter lead screw to prevent whipping and very careful motion planning because a crash would be disastrous after this part i will increase the rapids to about 200 inch per minute or 5 000 millimeters per minute the fog buster has been great for this part the compressed air is strong enough that it clears the chips out of the way of the cutter without blowing them clear across my garage with the coolant adjustment i've come to realize that less is more just having the needle valve slightly open to allow a small portion of coolant to mix with the compressed air has been more than enough for keeping the end mill cool low coolant usage also means that even this mini fog buster reservoir will last a really long time now i haven't gone through a whole reservoir yet but i'm thinking about 10 to 15 hours of cutting time at my current coolant usage the center hole will be tapped for a 10 32 bolt [Music] [Music] [Music] i realized after the fact that i forgot to clean up the perimeter of the part with a 2d contour but even without it this part has a respectable finish that parting line is barely noticeable just slightly protruding on the far edge which i would think would be from that slight rotation in the vise to finish up this part i 3d printed some inserts for those two big holes in the bracket the inserts will hold magnets which in turn will hold two hex keys one key for adjusting the placement of the fog buster nozzle which is accomplished through loosening and tightening a series of set screws and another smaller key for adjusting the concentricity on my probe which i do from time to time i just need to mount the bracket to the mill and that was a quick and easy project before ending this video i'm going to quickly make one more part to showcase higher rapids that i have set in the configuration file for the mill this part requires a lot of drilling so there will be plenty of g0 moves the x and y axes are set to 200 inch per minute and the z-axis is set to 100 inch per minute at first i had the z-axis set to 200 per minute but the end mill came down so fast at the beginning of a cut that i hit the emergency e-stop button a couple times thinking that the mill was going to crash i'm much more comfortable at this lower speed for the z-axis clearly the rapids are much faster now and you can appreciate how faster rapids equals quicker parts this part is a fixture plate and i will use it to image large samples at work there is a grid of holes which are being spot drilled as i speak that will act as tie-down points unfortunately i'm going to have to tap all those holes by hand this time around in a future video i will be adding rigid tapping capabilities to this machine so definitely be on the lookout for that video as i make upgrades to the mill i immediately post them to my instagram story so if you want real time dr d flow content follow me over on the gram i want to thank dmm and hinch manufacturing for partnering with the dr d flow channel and making this video possible if you're in the market for server motors or minimum quantity lubricant systems then just know that both of these companies make great products [Applause] [Music] you

Channel: Dr. D-Flo

Views: 48,080

Rating: 4.910543 out of 5

Keywords:

Id: bPM9XwzF2LY

Length: 40min 36sec (2436 seconds)

Published: Sun Jan 17 2021