I have been having a lot of fun playing with the CNC machine now that I have it running. Note I said playing with the machine. I haven't settled into any projects built with CNC as I am still deciding on what software I will be using. The one that really has my attention right now is Kelly Ware. I am exploiting their free download. The other package in which I am interested is DeskCNC. Both these programs will drive external serial input controller boards that are supposed to vastly improve the pulse train timing from Windows based CNC control programs (theirs).
Of course I am having a wonderful time watching my machine carve things (by the numbers) without such fancy accessories. So it is NOT a required addition. But this is a hobby. I am the only one making the rules around here. (That's a pleasure in itself!}
The Piker board is working faultlessly. That board and its attached "shop heater" will probably run forever. Except now I want to play with a little more technology. The trade-off will be considerably more expense. Considerable only in relation to the original low cost of assembling the Piker. I am easily going to equal the cost of my total initial project investment just to incorporate these new features.
The next step up in driving stepper motors is to use much higher voltage. About 42 volts is my new target. Right now I am using about 14 volts with dropping resistors to limit current to the steppers. A chopper hits the motor with the full 42 volts (approximately). Because the motor windings are an inductance, the current lags the voltage spike. The chopper senses the rise in current and "chops" off the voltage at a determined current set point. This "chopping" settles into its own frequency and can be heard in the motors (harmlessly). The high voltage is what gives the stepper its extra "push". (Remember the basic Ohms laws... Voltage is push, current is flow. The magic in a chopper is because of the "off" cycle, the AVERAGE power (voltage x amperage) (over time) to the motor is not exceeded.
That is a very brief introduction on how steppers work. Do some searches on the Internet and you, like me, will find all you ever wanted to know about the process. The great thing today are the manufactures who produce packaged I.C.'s that perform all of the magic. The board I bought contains four of these modules. Very few other external components are required.
There are some limits on how far you can (and should) go with open loop stepper motors. The best CNC systems are closed loop. That means there is an external signal that feeds back to the controller to tell the controller exactly how much movement has occurred. This works for steppers, servos, D.C. motors... almost any other drive out there.
I am staying with open loop for awhile. It is less expensive and as long as you don't stall the motor (loose steps), can be very accurate. If I build a larger CNC machine someday, I will surely change to closed loop. I have noticed all the low cost CNC ads in the magazines are mostly open loop stepper controllers on very small mini-mills. The bigger the machine gets and the more dependable it has to be, the costs greatly increase.
As can be seen, I have done a lot of homework on stepper motors. I did the same on controllers. I decided, I just wanted the features of the chopper. The best deal I found for a controller is the one I chose. It is a kit, so I needed some skill at assembling components on a printed circuit board. I have been doing that since PCBs (printed circuit boards) were invented. I think I am OK!
Here is the link to HobbyCNC. I purchased the 4AUPC driver board kit from these folks. The cost is about $100.00. I didn't consider the three channel kit as some day I might want to use the extra channel. The really good news is that they now provide a complete package (kit) with power supply (less transformer and case) and stepper motors!
Here are all the parts laid out prior to construction. There is not a lot of components so the assembly went quickly. The solder pads for most components are very small and some leads are close together. |
The PC board is absolutely first class. It is double sided with plated through holes. The board is completely solder masked and all component locations are clearly marked. They don't get any better than this. |
In about a half day the board is completely built. There is a recommended voltage check before installing the controller modules, and my work passed with flying colors! No solder bridges on this board. Not that solder bridges are impossible, but with the excellent solder masking I would almost have to try to build one. |
Heat sinks for the controller modules are not provided with the kit. The plans show a simple 1.5 x 2 x 0.125 inch aluminum plate that can be made for each module. I will be running my stepper motors at about 1/3 of the controller's capacity. However, I will be using it's full rated voltage. Sometime in the future I may want to push the controller harder with bigger stepper motors.
My hobby is machining,. A flat plate heat sink is way too simple. Besides, I have some aluminum bar stock on hand that is 1.25 x 0.250 and about 10 inches long. Ah Ha!
The first time I have used my two flute 3/4 inch end mill. |
All I have to do is thin this end down to 0.125. This is the surface that will contact the body of the control module. |
Four of them is pretty quick work to this point. |
The ribs have been milled and I am drilling #43 holes for taping to #4-40 screws. Who needs nuts? |
All four heat sinks complete, polished and washed. |
Now those are MACHINIST's heat sinks! No wimpy flat plates for me... |
Back side. See Pa... no nuts! I could have put some short ribs back here too. Umm... could they then be called spare ribs? |
Finished and ready for the next step. That will be building the 42 V.D.C. power supply. The 30 volt A.C. transformer is on order. I'll explain the voltage difference when I get there. |
I am using the case that previously contained my old 12 volt 30 Amp power supply. Here I have already fly cut the fan opening and just finished milling the opening for the 110 volt power cord connect. The power cord socket inserts here and the socket provides bypass filtering on the power line. The little hole to the left was where the original power cord entered the case. It is the perfect size for a new fuse holder. |
|
I have been thinking that this size (300 VA) power transformer is way overkill for the chopper CNC controller I need with my current (pun intended) steppers. It is the power that does the work. The voltage and current are relative. The small steppers I have are rated 1 amp and 5 ohms. E=RI (Ohms Law) so 5 volts will provide that current. Change that to power P=EI and we have 5 watts. So I hit the stepper motor with high voltage to give it a "kick" but cut off the pulse at a specific current level so I don't burn up the windings with too much power.
40 volts DC with 5 ohms load (I=E/R) is 8 amps and that is (P=EI) 320 watts! POOF! That is only thinking of one motor. So I know that load isn't there. That brings me back to my power supply sizing thoughts. Current is lagging voltage (only at the leading edge of the pulse) because of the inductance (thank goodness) or this thing would never work. I know A/C motor currents and inductance don't follow simple ohm's law values (so don't write me) but with D/C motors it should get me close.
I can see where I don't really need 100 VA (watts) transformer power per motor. I shouldn't push much more than 5 watts (average) into each phase. Even if I was loading 10 watts (average) per motor the transformer is still 10X oversize! The capacitor provides the reserve voltage kick The transformer supplies the current. Lots of room here for future growth!
Woo-hoo! It's great fun thinking (engineering) this stuff. Even better when I get to take measurement to see what is actually going on. Hobbies are great because I can get involved in every detail (if I choose). I don't accept everything I see or read as correct. The time I spend learning is my own enjoyment...
Hey! Just did something I can only call a "learning experience". I "blew up" my brand new controller! Yeah, I let the magic smoke out. Actually, there was a good bit of flames as well. Yes, Fire!
I had everything wired up and I thought, triple checked. I had thrown the big switch and everything seemed fine at first. Everything was cool, literally. As it turns out, the wiring was fine
I went around the corner (see pictures elsewhere of my shop layout) to punch some buttons on the computer. Hmmm… just some weird groans out of the stepper motors. Gee, they ran OK with the old controller. Let's see here. (More button pushing, trying to "go" in any direction.)
Hmmm. A new sound! I take a quick look around the corner. Holy Smokes! Again, I make a literal statement. (I never mince words.) I immediately started looking for the weenies and marshmallows. I had a good fire going IN my new controller. 46 volts DC with 300+ watts available doesn't fool around when it wants to get out.
When it was all over, I had lost two control modules, two sense resistors and, I found out later, a stepper motor winding. Well, I wanted a bigger Z axis motor anyway…
One problem was that the step and direction bits were reversed in the software. I was sending solid "ons" rather than "steps" to the controller. From this I first assumed the more I had held down the "move" key the more heat went into the motor and controller. Then POOF! After correcting the software and suffering intensive mental anguish rechecking EVERYTHING, I was able to run the two remaining steppers (on the still good half of the controller) with no problem.
However, not being easily satisfied, I did some futher investigating. I tried to duplicate the event by deliberately reversing the the step and direction bits again. This time no burn or heating problems! So now it seems there must have been other contributing factors and not just the reversed control bits. I did swap motor ports once. I may have made that swap before the capacitor bled down. There is a very strong caution not to do that sort of thing in the controller (chopper) instructions. The other opinion is I had a bad motor that just couldn't take the higher voltage.
The board is now repaired and all is working as intended. The lesson is, "Sometimes I only see what I expect to see" (when checking the software configuration). I am too conditioned that software on a PC can do no harm. Although it now looks like the port configuration was NOT prime the reason for failure, anytime software is controlling shop machinery and high current or potential (voltage), it is a whole different ball game. There is no excuse for a wrong set up. I should know better. My profession is software control of major buildings and large energy consuming devices. Homer (Simpson) has a word for it, "Doh!"
This is the inside of the burned stepper. The coil at the top of the picture has some melted varnish on the windings. Take a look at those nice ball bearings on both ends of the rotor. I discovered lots of magnetic force here. The rotor "clamped " itself to the side of the motor case as I set up for this photo. |
I love the color in these photos. The full size shots are outstanding. The parallel cable connection is only temporary at this point. There is a surprise coming on how I am going to next run this system. However, running as shown here, I seem to be able to get close to 40-50 IPM on a 16 TPI lead screw! That's after the ramp up. The motors and controller heat sinks are all only very slightly warm. What a great difference! |
Click on the picture above to enlarge. I built the capacitor clamp from scratch. The resistor on the capacitor is required. It is five watt and is consuming about two watts. It gets a bit warm, not hot. It provides some surge protection and a little regulation but is really needed to bleed down that massive capacitor. With no load the RC time constant is about 5 minutes to loose 66% charge. Never plug and unplug loads into a "hot" power supply like this. It will definitely blow the controller. I have seen enough smoke, thank you. |
Even home brew equipment should be properly labeled. I thought about insulating these connections (good practice) but I want to be able to make another connection here. The board visible at the upper edge of the picture is a leftover form the previous power supply. One LED for each of the four Lambda power supplies. |
This shows the four new output plugs. The fuse holder on the left is the primary for the transformer. |
This is the case with the perforated top cover on. The fan pressurizes the case so the air will be leaving through this grill. |
The power socket contains a line filter and is a very good case ground connection. The fuse holder on the right is the fuse for the DC circuit. I made use of the previous power cord inlet hole. I insulated every connection very well in this corner because of the proximity of the input and output voltages inside the case. Low voltage wiring is 14 GA. THHN rated 600 volt insulation. Power supplies of this class (although simple) must be treated with great respect. If someone is not comfortable with this kind of construction, it is best to leave the work to someone else who is knowledgable. A power supply like this should never be assembled or operated without the proper enclosure and safety measures. |
This is the sample board I ran with my new software (DeskCNC). I Chose this software because it has a serial port interface controller that the software uses to generate all the stepper timing pulses. Let me tell you. There is a world of difference in step flow when running Windows software directly from the parallel port or through the external serial interface. I have recorded a 6.5 minute video of my CNC machine running with my new control setup. The demo is running the program that carved the tree in the above picture. I am not running the Dremmel in the video as I wanted to let folks hear what the steppers sound like when doing an intricate pattern as in the picture. The Dremmel is MUCH louder than the steppers! |
| HIGH BANDWITH .MOV 36.6 Meg |
HIGH BANDWITH .WMF 51 Meg |
LOW BANDWIDTH .WMF 18.8 Meg |
They are all are the same video. Pick one of the above and enjoy!
