[Madclicker]

Got some more done today. Laminated top, bottom and sides. Top, bottom and 2 sides in p-lam. The sides that show I did in re-sawn red oak for the aesthetics. First pic is the bottom after gluing the oak panels, laminating 3 sides, and trimming edges and bolt access holes. 2nd is after laminating the top and trimming edges and t-track grooves. Also added red t-track to see how purdy it is gonna be......VERY!

This one piece is definitely the heavy lifting for the project.

My double secret oil blend really makes the red oak pop.

[ViperTX]

Gotta commend you all.....you make some working machines from low-cost materials! CONGRATULATIONS! Just a fine example of ingenuity!

[Madclicker]

I've purchased quite a bit of materials. I'm trying to get a handle on what this first project will end up costing. I'm posting my running total as it stands so far:

3/4" MDF $42
1/4" MDF $24
VT Laminate $44
NEMA 42 1125 oz/in stepper $25
NEMA 34 450 oz/in stepper $43
Red Oak $16
T-Track $70
UHMW PE $20
2" EMT $16
3/4" flange bearings $17
ACME rods and nuts $38
4 Heavy duty castors $17
3/4" MDF $27
2 X 1 X 1/8 AL channel $10
abec-7 skate & R8 ZZ bearings $44
*10A 4 axis drive $135* (contemplating...not purchased yet)
-----------------------------------
Total $453

[2muchstuff]

Where did you find a 1125 oz/in stepper for $25. That was a steal. Also what kind of drive is it for $135, are there more.

[Madclicker]

Actually I got 3 of these 4.2" motors for $73 delivered off ebay. They were called "pulled from working mill" and very grungy. I'll clean them up, paint them and keep my fingers crossed. The shafts turn smoothly and I can feel the detents.

This is the link to the drive site:

http://dtllc.com/

p38nut uses this drive and swears by it and the guy that makes and sells them. My major reservation is that from the pics it looks like there is no optical isolation onboard. I think that if I'm careful with the wiring before I flip the switch this will not be an issue. Also, it's a unipolar drive, but I can live with that because I have these gorilla motors. The feature I like best about the drive is that it supplys 10A and my 4.2's can use 9.1A of that.

I think I'm going back to my original plan of using 2 of the 4.2's on the X, 1 on the Y and use the 3.4 on the Z. New cost sheet:

3/4" MDF $69
1/4" MDF $24
VT Laminate $44
3 NEMA 42 1125 oz/in steppers $73
NEMA 34 450 oz/in stepper $43
Red Oak $16
T-Track $70
UHMW PE $20
2" EMT $16
2 5/8" flange bearings
5/8" ACME rod and nuts
6 3/4" flange bearings $49
3/4" ACME rods and nuts $38
4 Heavy duty castors $17
2 X 1 X 1/8 AL channel $10
abec-7 skate & R8 ZZ bearings $44
*10A 4 axis drive $135* (contemplating...not purchased yet)
-----------------------------------
Total $668 (includes drive price)

I originally wanted this project to come in under $600....oh, well. Maybe $850 or less.

[2muchstuff]

I'll say this again, you stole those steppers. What luck. I checked out the drive site and I like the 10 and 15 amp ratings but I need something that is good up to about 60 volts in a bipolar chopper. It keeps looking more and more like Gecko's all the time. I know what you mean about trying to keep costs down. I must be my own government, cost over-runs are running rampid.

[jeffs555]

Using that drive may not be as economical as you think. I dont know what voltage rating your steppers have, or what power supply voltage you plan to use, but be aware that using those drives you are going to need at least 8 massive power resistors to handle 4 motors at 9 amps. For example, if your motors are rated at 3 volts and 9 amps, and you use a 12 volt power supply, the resistors will have to handle 72 watts each, probably 100watt resistors to be safe. These will cost between $7 to $12 each. In the winter you may save some money, because it will probably heat your whole shop.
Also, you will need a massive power supply. You will need 18 amps per motor, or nearly 80 amps for 4 motors.

It probably wouldn't be that much more, and might even be cheaper to get some Geko drives, which won't need power resistors, and will also have microstepping to help eliminate resonance. They will only handle 7 amps, but running bipolar series, would require only 4.5 amps to get the same torque as 9 amps unipolar. Also, using chopper drives, the power supply can be much smaller and cheaper. A resistor drive needs the power supply to handle the full motor current, but with a chopper drive and a higher voltage supply, the average current required will be much less than the rated motor current.

[2muchstuff]

Resistors are the reason for going with the chopper type of drive. I just can't see spending all that money on resistors and transformer to watch it all go up in heat. I want to build a power supply/driver not a glorified over priced shop heater.

The motors that I have are 34 frame, 2.3 volt, 5.5 amp, 450 oz/in, 4wire. I would have swore that they were 690 oz/in motors but I guess not. It had been a while since I last looked at them. Due to the size of the router that this thing has grown to be, I will more than likely have to get bigger motors. I'm guessing that both the x and y motors will have to be alike because of the circular interpolation that I will be doing. Unless software will compansate for that which I don't know. I haven't even thought about software yet, I'm still working on the mechanicals.

[Madclicker]

Jeff,

9 x 9 = 72?.... -10

I always try to approach projects from a systems (I'm a systems engineer) standpoint. Some things I have to use (the motors) and some things I have choices on. I think there are two very real design mistakes you made in your argument against this drive:

1. You seem to have arbitrarily chosen 12V PS's?

2. You have completely ignored the FET SD resistance.

Let's forget the FET resistance for now and go with your original premise. If I were to choose or build a 5V PS I would need to drop 2V across a power resistor. 18 watts. BTW, I found a 5V 150A PS today for $70.

BUT, the FET resistance matters (many blisters confirm). FET's typically have an 'on' resistance of 1 to 2 ohms in their working envelope. The resistance varies somewhat linearly with temperature. Of course, temperature increases with current. Taking the best case (1 ohm), your 12V PS would drop 9V across the FET and no additional ballast would be needed. Worst case, at 2 or so ohms, the 12V PS wouldn't power the motor to it's capacity. You would want more voltage!

I don't know what FET's are used on this driver, but Ill sure find out before I make my decision.

[mdreitzusa]

very nice looking machine.looks good enough to use as living room furniture.
as a systems engineer the controls shound be a wiz.cann't wait to see that conrtol panal that you mentioned.
keep up the good work.

[p38nut]

You are right, I really like this board for my application. I bought the 6 2 ohm 100 watt resistors for $6 bucks each from www.mouser.com. I also used 2 ATX computer power supplies, purchased from the local computer store, at $29 each. The total is 44 amps at 12v available. My steppers were 4.6 amps at 1.2v. Using the resistors has something to do with getting the best response and torque from the steppers. The on board enable relay and Mach2 shut down the PS's at the end of program. Some of this is technically beyond me. However, I do know one thing, it works, and keeps my garage warm at the same time.
Madclicker, if you have any questions, call Dave direct. He's easy to talk to, and knowledgeable about his product and the cnc craft.
Thanks. Mike

[Madclicker]

md,

Here is a link to the site that has a slick board that can be programmed to provide KB shortcut sequences with single button pushes.

I still need to find arcade style button switches.

[Madclicker]

gosh, missed the link:

http://www.ultimarc.com/

[ger21]

I still need to find arcade style button switches.

Ultimarc sells them. Check here too. http://www.happcontrols.com/pushbuttons/pushbuttons.htm

[spalm]

Mr Clicker, this is not my area of expertise and you seem confident of your conclusions on FET drivers, but…

I have not designed a motor controller or used many high power FETs, but have designed many switching power regulators and have never come across a FET with 1 or 2 ohms of resistance. Take a common example of the IRLZ44 MOSFET that I see in many hobby controllers. It has a RDS(on) resistance of .04 ohms. Even if it is normalized for temp, at 150C it is still only .07 ohms, and that is hot!

Now, I live in a world were efficiency is king, and heat is a bad word. So I would naturally tend towards a constant current PWM type solution.

IMHO,
Steve

[mdreitzusa]

thanks for the link
it could be fun to see how far this control can go.with the option of buttons,trackballs,joysticks and opical switches all on the same machine.have you looked at
http://www.futurlec.com/DevelopmentBoards.shtml
they have development board and control boards designed as modules.if you wanted to make an embedded system.

[jeffs555]

Madclicker,
I was only trying to be helpful. I am an electrical engineer, and have designed motor controllers, and Steve is right about the on resistance of modern N-channel mosfets. For a board like that to handle 15 amps, I would guess the on resistance is closer to 0.01 ohm. At 15 amps, a mosfet with 1 ohm of on resistance would be dissipating 225 watts, and the whole board would be dissipating 1.8 kW, and would be a molten lump.

The main reason for using a higher voltage power supply is to allow the steppers to achieve reasonable rpms. If you look at the torque curves for most stepper motors, they are all specified with supply voltages of 24v or 36v or higher. Still don't know what model your motors are, or what voltage they are rated, but running a 3 volt motor with a 5 volt suppy, the torque at more than a couple hundred rpm would be very limited. Even a 12 volt supply is usually marginal. Your motors may be so overrated for your application that you can get away with it, but you will be spending a lot on the drives and power supply, so you might want to check the specs and do some calculations to make sure they will work for you.

Jeff

Ok, so I am getting old and can't even multiply 9x9 without a calculator. So 81 watts is even worse.

[Madclicker]

I appreciate any help I can get. I'm in absolutely uncharted waters here. If I question things that are said it doesn't mean I think what you are saying is wrong. I've done bunches with DC servo's, but this is my first exposure to steppers. I would use servo's, but they are so expensive...even used. Which brings me to why I'm looking at this drive board as opposed to buying 4 Gecko's. If I can use this board I would save $350 or so.

I threw away my data books a long time ago and googled "power fet data" or something like that to get the numbers i quoted. The guy that makes these boards got back to me and told me the source-drain R is about .1 OHMS. That would give a drop of .91V. My big motors are rated at 2.1V, so I see I would need to drop about 2V across a ballast if I used a 5V supply. 18W. Which brings me to my next question:

Why would using a larger supply voltage increase rpm's when what you're really doing is just dropping the extra V across a resistor and making waste heat? I'm an electical engineer also and might understand a technical answer.

[jeffs555]

Steve,
The main thing that limits the stepping speed is the inductance of the windings. Each coil inside the motor is equivalent to an inductor in series with a resistor. Current through an inductor cannot change instantaneously. When you apply a step pulse to the driver, and it switches the current to the coils, it takes a finite amount of time for the current to change. The rate at which the current will change is equal to the applied voltage divided by the inductance, so the higher the applied voltage, the faster the current reaches the rated current. For example, if the coil inductance was 1mH which would be typical, and you applied 2 volts to the coil, it would take at least 4.5 mS to reach 9 amps. For a 200 step motor, 4.5mS is only around 67rpm.


With a chopper type drive, the full supply voltage is applied to the coil until the current reaches the set value, and then the voltage is switched on and off to maintain the set current. Since the full supply voltage is being applied to the motor, it is easy to see that the higher the supply voltage, the faster the motor will reach the desired current.

On a unipolar drive with series resistors, when you switch the voltage to a coil to step the motor, the current initially starts out at zero and ramps up. Since the current is zero, the voltage drop across the series resistors is zero, and the coil initially sees the full supply voltage. As the current ramps up, the voltage drop across the series resistor increases, so the coil sees the applied voltage decrease as the current ramps up, but for most of the ramp up period the voltage across the coil is much higer than the rated voltage, so the current does ramp up faster.

It is generally accepted that for good performance, the supply voltage should be around 10 or more times the rated voltage of the motor.

Jeff

PS I would question the cooling requirements for a drive that claimed to handle 15amps, and used .1 ohm mosfets. At 15amps, that would be 22.5 watts for one mosfet, which is an awful lot for one small device to dissipate. It would take some major cooling to keep from burning it up.

[Madclicker]

Jeff,

Thanks a lot! That clears it up completely. I was sure you were talking about the dynamics of the motor, but I don't know why it wasn't obvious to me sooner. Been a long time since my circuits courses. What did they call that? "unit step response"? (chair)

Damn, I was sure hoping I had found a cheaper solution. I can't dump that much heat into my shop. Not even in the winter. I have 2 window AC's going out there right now.