Milling Machine Design Request

[Ahmad_k]

Now my machine works great but my motors cannot rotate higher than 195RPM (650Hz step frequency). With a 5mm ball-screw mechanism i got 38.3 IPM

And that a low speed. Someone have managed to spin the motor at 600RPM without any problem and he looks now for 1200RPM with 70VDC (Bipolar-Parallel connection). After searching the WEB i found that these Hybrid Stepper Motors can speed up to 3200RPM without load. And the company confirm that we can rotate it at 1200RPM without any problem but we need a good driver with closed loop system.

So i started a new design on my stepper driver, i will try everything until i get a very good stepper motor driver. This design is a modified version from my last driver but i have added a new method to drive the MOSFET (Used in 5KVA modified sine wave UPS) and redesigned the PCB.

So i need from experts in High power PCB design to see it before trying the driver (It is not tested yet, two connection still not connected and these are sens1 and sens2). My project is done with Altium Designer 9.

[MoarK]

I have used this before (for unipolar steppers):
http://www.mondotronics.com/PDFs/3-303_Stepper_v14.pdf
with IRLZ44 hexfets. I love the logic fets
but I can not tell you how fast it will clock in your system.

[Malcolm]

 Hello Ahmad_k

Stepper Motor Driver.rar (671.26 KB - downloaded 1 times.)
How to open this rar.files?
Showing-- unkown  application

[MoarK]

http://www.rarsoft.com/download.htm

[Ahmad_k]

I have used this before (for unipolar steppers):
http://www.mondotronics.com/PDFs/3-303_Stepper_v14.pdf
with IRLZ44 hexfets. I love the logic fets
but I can not tell you how fast it will clock in your system.

This design does not have current limiting feature and that is the whole problem for me. Thanks anyway.

[MoarK]

Oh.
Is it not acceptable to adjust the supply voltage to suit the coil resistance?

[Ahmad_k]

Of course not. Higher Speed require higher voltage. If you read the previous replies, you will find a working stepper motor driver that can deliver up to 10A with current limiting, but it works on low voltage < 6v.

[MoarK]

Is it not current that produces the field?
What does a higher voltage do except push more current? ( and make much more noise)
If you must limit the current, is the voltage not excessive?
Perhaps there is something I do not know.

[oldvan]

Follow ELI the ICE man to understand why higher voltage with current limiting is beneficial to quickly charge and discharge an inductive load.

Stepper motor performance is strongly dependent on the drive circuit. Torque curves may be extended to greater speeds if the stator poles can be reversed more quickly, the limiting factor being the winding inductance. To overcome the inductance and switch the windings quickly, one must increase the drive voltage. This leads further to the necessity of limiting the current that these high voltages may otherwise induce.

http://en.wikipedia.org/wiki/Stepper_motor

[Ahmad_k]

For 23H2100-30-8A Nema23 motor :

Max RPM = (0.191 * V)/(Ls * Is)

Is should not exceed 3A as specified in datasheet. Ls is constant = 2.5mH . So the only way is to use higher voltage supply

On PWM drive the max voltage is 20x rated voltage.

Rated voltage = Coil resistance * Rated Current = 1.4 x 3 = 4.2v

Max voltage = 4.2 x 20 = 84 Volts

On my design i use 5v

RPM = (0.191x5)/(0.0025x3) = 127 . Mine spins on 195 without any problem

Please guys do not forget this : http://www.sonsivri.com/forum/index.php?topic=27112.msg103996#msg103996

[MoarK]

Thank you. I understand the need if you are pushing for highest speeds.
I found this:

The graph is for a motor similar to yours but not specific for yours.
It suggests a maximum practical speed to expect for that series of motor may be determined by load as well.
Is it possible that the motor is simply not big enough for what it's doing?  Because 60rps is 1200 steps/sec, so it seems you are not near top of the range for it at 600.
My desktop mill, for example, will do 30 in/min on a 20 tpi screw & 200 step/rev 200 oz.in. nema 23 motor. I calculate that to be 2000 steps/sec.
http://www.taigtools.com/cmill.html

[Ichan]

I am not an expert in high power pcb.

I would prefer to exchange the position of heatsink with other components on the left of the mosfet.

Where is the motor power came from?

-ichan

[Ahmad_k]

the power will be connected directly to the center of each coil.

I think 3cm is sufficient for the heatsink.

[Ichan]

I think 3cm is sufficient for the heatsink.

Not about heatsink space, but puting the sense resistor to the right (where now occupied by heatsink) will make it easir... I think.

-ichan

[Ahmad_k]

Yes that's right , now i figure it. Any recommendation about GND track ? Tomorrow I'm gonna try it.

[Ichan]

Any recommendation about GND track ?

From the center of the motor connector.

-ichan

[Parmin]

I designed drives for some movie FILM cameras (A-Tone, Bolex, Eclair etc) commercially for cinematographers
  the type of camera that utilize stepper motors to pull the film, spool, and shutter mechanism at HIGH speed (up to 200 FPS)
At normal cinema firming speed of 25frame per second (hence 25 rev/second),
  I need the stepper to go at 1500RPM within 2 seconds from the time the trigger is pressed.
  To go at 200 FPS require a mixture of HIGH torque, timing belt and pulley ratios and a very rigid drive mechanism.
 
What I learn from this? To achieve the 1000's RPM you need to have a decent acceleration algorithm.
I am sure if you fix the direction signal, and add a pot controlled analog step signal (like an LM555) that you could control BY HAND
 you can get very near to the motors top speed..
Human can "sense" acceleration and adjust accordingly, whereas most machine would need an algorithm or read from a table.

I am not an expert in High Power PCB, but IMHO the current board design is adequate
 and any changes from that would not improve performance more than 5% - 10%  if any at all.
The basis of my opinion is the fact that it is about on par among quite a few commercial stepper drivers that I have used and serviced.



Ahmad, I read all your reply but did not see the solution of your NOISE problem?

[Ahmad_k]

I didn't find any solution for the noise, i live with it. And for higher speed, Mach3 have an acceleration feature in motor tuning option. For me 600 RPM is more than enough because i'm using 5mm ballscrew mechanism and that gives 120IPM verru good speed for wood working and drilling

[Ichan]

Hi Ahmad,

I modified a bit your layout, this way there is enough area to connect the sense pins.

I would also prefer to use multiple 1 ohm 1206 SMD resistor in parallel as sense resistor.

-ichan

[Ahmad_k]

But there will be no space for heatsink. I have made a PCB for testing and tomorrow i will get the IRF540. I hope that this driver will make a big improvement for the machine

[ktek]

try these

http://www.cncitalia.net/forum/
www.cncproject.net/Forum


end
I have bought a good kit here

http://www.cnc-creostamp.com/
 

[waicon]

china open-source CNC design. here ( i hope google transilator works correctly )
http://www.ourdev.cn/bbs/bbs_content.jsp?bbs_sn=3355500&bbs_page_no=1&bbs_id=9999

circuit; stepper; powersupply design, pdf download link
http://www.ourdev.cn/bbs/bbs_content.jsp?bbs_sn=3334237&bbs_page_no=1&bbs_id=9999

stepper driver using A3977, PLCC package.
http://www.ourdev.cn/bbs/bbs_content.jsp?bbs_sn=3274271&bbs_page_no=1&bbs_id=1027
(in my country A3977 is cheaper than L297)

[glitch_inc]

I know you would rather do your own design but there are a bunch of designs available on this site for both Uni and BiPolar. Maybe you can find an answer in one of their designs.

EDIT: Link removed has it trys to install virus on my computer and has been repoerted by others

[Wizpic]

glitch_inc I hope this was a genuine mistake and not an attempt to try and to spread virus.

Please explain your actions before I take action

[myheadhurts]

Some time ago Marius from Gekko did a tutorial on how one of his microstepping drives worked.
Its quite interesting..

http://rapidshare.com/files/370379490/5_microstep_tutorial.pdf

Rob