If you are concerned, add a 2-3 amp fuse. It will protect the motors.
Do not use the series configuration. Use parallel. Moving torque is more useful than holding torque.
Printable View
Thanks for the advice. I could find any fuse at home so took a shot and tried to assemble averything. Nothing worked and since I never had a working system I'm not sure of what may be wrong.
This is my setup:
Desktop:
Intel e2140
2 GB RAM
160GB SATA
PCI controller with a Parallel port LCS-6019 configured as EPP+ECP in BIOS
Steppers and Driver:
3 Axis TB6560AHQ without any mod yet
3 NEMA 23 3.1Nm Steppers ( http://cnc4you.co.uk/resources/60BYGH301B.PDF )
12v 20A Power supply
I'm using these stepper motors as Bipolar series which according to the documentation will draw 2.1A and because of these I have configured the DIP switches like this:
http://i.imgur.com/hZPPRyKl.jpg
1 AND 5 OFF and all of the others ON. 1OFF | 2ON is supposed to give me 50% of current = 1.5A. This is not the 2.1A of the documentation but I was afraid of using the 75% which is equivalent to 2.25A, 1.15A over the limit.
I know the power supply is weak, but I'm still waiting for my 24v power supply to arrive. In the meantime this should be enough just to see the steppers moving :-(
The first thing I did was to download the linuxcnc live CD and run it. The configurations I downloaded from the linuxcnc wiki ( LinuxCNC Documentation Wiki: TB6560 ). I then connected the driver to the PC WITHOUT any stepper motor just to see if I could see the access leds blinking while moving the axis.
The first thing it happens is that three leds lights up as soon as I connect the parallel cable to the PC and turn the power supply on. If I connect the driver without the power power supply and just the parallel cable the same three lights turn on dimly. When I connect the power supply they get very bright.
Without the PS on:
http://i.imgur.com/UREftdpl.jpg
With the PS on:
http://i.imgur.com/vD9ozjVl.jpg
I loaded a G code sample and started the operation and nothing happened :|
The multimeter confirmed that no voltage was geting to the A+-/B+- of any pin axis. Because I wasn't sure that the parallel port was correctly detected and configure under linux (even though I had the /dev/lp0 and /dev/lp1 available).
Then I installed Windows XP Home 32bits and MACH3. The LPT port is correctly detected but I had the same problems: No activity leds blinking :-(
I then decided to take the risk and connect a stepper motor. The Z and Y axis are dead, but the X axis shows 6 volts between A+ and A- without any command sent from the MACH3. The stepper does some noise when connected and it looks like static noise.
The computer, parallel controller, steppers, driver and PSU are all brand new so I have no idea on how to troubleshoot this system.
Is there any way of testing all of the elements separately?
I'm not really familiar with the board, but be sure you assert the enable pin, or there will be no motion.
You may have fried your driver chips. You should never power up the board without the motors connected.
Hi,
After all that I've read here I thought that the only thing I shouldn't do is to connect the steppers to the driver if the driver is already powered on.
Anyway I've found out that the problem was due to a damaged 74hc14 and a problem with the parallel port.
I was using a PCI parallel port that for some reason cannot have the legacy I/O addresses assigned. I tried the DOS configuration tool nmdosin.exe, but the settings keep beeing reseted after rebooting the computer.
I'm going to try to find another PCI Parallel controller or switch computer.
The jumpers you show will probably not work for the enable lines. On a 3-axis board they need to go straight across rather than diagonal. see circuit on second page of thread.
You can leave the optos in for the enable lines if you like. This will make the mod simpler.
Else it looks good. the 10k resistor removal will lower your noise. I suggest trying each mod one at a time.
Is there any modification that can help smooth out higher feedrates? I've jumpered the step optos, removed resistors, and replaced timing caps. All these helped. Haven't tried jumpering all optos.
I get a jittering from my steppers at higher speeds requardless of microsteps, or current setting percentage. This creates a vertical pattern in my pieces, not relative to part geometry, also has jitter pattern in straight X or Y passes. Less pronounced at lower speeds, very faint pattern at 5mm/s.
1.8 degree motors, almost seems like the motor higher torques or 'snaps' into magnetic alignment as it passes each tooth of the rotor.
Running constant velocity Mach3, going to test LinuxCNC tomorrow.
Hello Neon22,
Thanks for your response.
I've managed to get the PCI parallel port to work. It was a driver related problem and now everything works, but I think it is extremely slow!
I tried to use a jumper wire between pin 2 and 4 of each optos without success (nothing happens). I tried to search the schematics to remove the optos on 2nd page as you said but couldn't find anything.
I could find though, another post that is sugesting the same thing I did:
This time I'm just going to jumper those three optocouplers and see what happens.
Anyway, I have another issues ...
I bought a 24v 15A but while it doesn't arrive I'm using the 12v rail from an ATX PSU to test the steppers. Also, since I cannot use the steppers in bipolar parallel mode because it would draw 4.2A each I'm using them in bipolar series mode.
But the tb6560 settings don't allow me to go near the 2.1A. 25, 50, 75 and 100% are 0.75A, 1.5A, 2.25A and 3A respectively (I'm using the 3A as MAX current to calculate the percentages, not the 3.5A peak, is this correct?) so I'm using the 1.5A setting which is far below the required 2.1A? Could this be the problem?
I've seen people that replaced a resistor from the dip switches to use a right amount of current for their steppers, could someone do the same thing for me?
Just one more question: If I used the 2.25A setting would this kill the steppers?
Thanks!