First movement... sortof

[Rance]

And No, this aint about Beethoven .

I've got my floppy disk drive stepper cobbled to the parallel port. Using TurboCNC, Y axis only driving the controller board directly. I boot up in DOS mode and run from there.

I seem to only be able to get the motor to turn one direction. I installed TurboCNC and am using ALL the default INI settings except for lpt port ($278 I believe). By pressing the up/down arrows, I can get the motor to move but in one direction only. I swapped the lines (step/dir) to make sure I did not have them reversed but then it only moves one step when I go from up to down (or down to up, I can't remember).

If I change EITHER of the pin configurations in the INI file to another pin, then the motor does nothing. I did this to confirm that I had the right two wires. I believe my test is conclusive on this.

Additionally, I can measure the voltage at one of the pins using my DVM and can see one of the wires go from 0v to 5v when I change direction (using Jog).

My inkling is that the config file needs some tweaking, that it's timing is off.

Tools at hand: DVM, electronics knowledge, very little money . I thought one of you veterans might be able to shed some light or point me to where I could find the answer. Thanks in advance.

Rance

[esmiller]

It has been some time since I hacked a floppy drive controller, but I *THINK* I recall that some of them had separate pins for CW & CCW movement.. That is Clockwise & Counter-Clockwise stepping...

Is there anybody who can confirm that i recall correctly on certain floppy controllers, or am I recalling something else and getting it confused with floppy controllers in my older age... ??

Eddie

[Rance]

Eddie,

I think its old age . The documentation I have shows Step / Dir. going into the drive controller. This could still be part of the problem though.

Edit: Eddie, actually cw/ccw is the same as the Dir pin. Thanks.

Rance

[abasir]

Check here:
http://www.epanorama.net/circuits/diskstepper.html

Maybe you need to extend the pulse width. The article said you need about 10ms, TurboCNC pulsewidth is specified in microseconds (us). Not sure whether you can put 10000us as pulse width in turbocnc

[Rance]

abasir,

That's what I've been working from. Thanks anyway.

I continued to make modifications to the INI file configuration. In checking things, I found out that it acted differently when I touched the DVM probe to the Step pin on the cable. Its times like these when I wish I had an O-Scope. I'm beginning to think that I'm having problems with the port not being strong enough to drive my floppy controller board. I believe the levels here are both 5v logic. The length of my cable is not more than 4' so I don't suspect that. Any other ideas to try? Thanks.

Rance

[abasir]

Tap +5V from the motherboard (disk power connector, etc).
Put a 470 ohm pull up on the step/dir lines.

[Rance]

abasir,

Good idea! Are the printer outputs open collectors?

Just tried that, no difference. Just for grins, I tried tying them to ground and no difference either. I'm guessing that I'll have to put a buffer (driver) in there to drive the lines. Yesterday I disconnected the printer port completely and just used hardwire inputs and the drive worked as expected, in both directions. If I hook up a driver in the line, would you have any suggestions as to which driver to use? Maybe I could just use some 6426 darlingtons w/pullups.

To answer your question the other day, TurboCNC only allows me to put a max of 999 for the pulse width. However, with a setting of just 10, I can load up the Sherline sample cnc file and run it and the motor will do stuttering turns. It should normally run each axis the full length of the settings. In addition, when it is running Sherline, if I touch the signals (either Step or Direction) it slows down a little.

Rance

[Rance]

Hooray, I have expected movement!

I had purchased a newer drive on the way home from work to replace the Tandon drive that I have been fiddling with. A quick test and I found out that this newer drive (a Teac drive) works! I can successfully Jog it back and forth, I can stop on a dime and give you nine cents change.

This concludes a BIG milestone for me on this small machine I am building. My next milestone will be to find out how much torque these small steppers will give me and if need be, is there anything I can do to squeze more torque out of them.

WoooHooo, we have movement!!! Last but not least, thanks for your help guys.

Rance

[basshead]

hi rance!

nice to see u got it running! i have tried all i can think of but i still don't get any movement on my floppy. can u please share the pinouts u used?

peace! (and CNC)

[Rance]

Bass,

From the http://www.epanorama.net/circuits/diskstepper.html page, Search for 'PC parallel port to stepper-motor interface' and the diagram is just below the text. And don't forget to connect the floppy to a power cable from a PC power supply.

Do a process of elimination, try another floppy (even one with the small steppers), and another computer (preferrably NOT a laptop). The computer would be my first suspect component. Not all parallel outputs are the same level. I believe that was what was causing me problems at first.

Until you get past this stage, don't even bother trying to disassemble the floppy, just work to get one that will work first.

Lastly, consider writing some quick code in QBasic to drive the printer port rather than trying TurboCNC or Mach software. Work to simplify the setup. Hope this helps.

Rance

[basshead]

cool man!

i forgot to mention that i am trying this on a modern, 3 1/2". maybe i'll just swing by the recycling station tomorrow and look for a 5 1/4" instead. tomi's page is great for info. i just built a kit 158 driver board which works great! but you know how it is, if you got something in your scrap pile, -"hey maybe i can use that too..." feeling. ;-)

thanks for the reply.

[Rance]

[i am trying this on a modern, 3 1/2". ]

The 3.5 will work just fine for testing. The pinout should be the same too. It's fun to see things move.