[jrz126]

crongrats on the no smoke

maybe try switching the encoder channels?
If the encoder and motor are connected backwards, the controller could spin the motor the wrong way, and results in an error.

[Chris D]

Thanks! Will try that in the morning.

I also will double check the step and direction outputs from Mach with a known controller to verify that too. I tried connecting my scope to the outputs and it appears as though that is working correctly - but I am new to scopes and new to servo drives, so I could be dead wrong too!

Chris

[ulihuber]

Chris,

sorry, you got your controllers before the US-version was made.

Hence you have to reference the 'older' manual UHU-Servo_Controller_300_en.pdf, which is in english too, but uses german words for the instructions.

Controllers sent to people who asked in english language after 20th of december are US-Version.

Both manuals can be found on the webpages www.uhu-servo.de.

Your problem with 'running away' of the motor is a good sign. You just have to exchange channel A/B of the encoder or reverse connection of the motor.

Uli

[Chris D]

Chris,

sorry, you got your controllers before the US-version was made.

Hence you have to reference the 'older' manual UHU-Servo_Controller_300_en.pdf, which is in english too, but uses german words for the instructions.

Controllers sent to people who asked in english language after 20th of december are US-Version.

Both manuals can be found on the webpages www.uhu-servo.de.

Your problem with 'running away' of the motor is a good sign. You just have to exchange channel A/B of the encoder or reverse connection of the motor.

Uli

HOt dog! That should not be a problem using the german command set - heck I am not sure I understand the Engish commands yet!

I will swap the motor leads this morning and start playing with the servo again.

Will report any successes or failures later this morning.

Thanks Uli

Chris

[Chris D]

Getting much closer now!

I now have it so the motor holds position without error. If I try to rotate the motor shaft, it resists and I can see the graph going up and down very slightly.

Using the terminal, I can make the motor rotate a little bit and that shows up on the graph.

The graph shows a quick rise up - then a flat area - then a quick line down, then it appears to jump up and down a lot. I am guessing this is the motor trying to hunt for position to keep it there? Sorry, I am not sure how to interpret the graph as I have never tuned a servo. Are there some graphs showing good moves and bad movments?

I also cannot get the motor to move via the CNC software. I checked the interface using a stepper driver I had handy. Using that stepper driver, I can make a stepper motor move nicely - so I know the "configuration" is correct in the CNC software - in other words, I am sending data to the right pins for step and direction.

I am assuming that the motor can be jogged from the CNC control while I am doing all the configurations and such with the terminal program (which is running on a sepperate computer).

Also, no excessive heating anywhere - the motor is barely warm, the mosfets are barely warm and the voltage regulator is a bit warm - so, all looks good there.

Thanks again guys.

BTW, I have a hunch this servo driver is going to catch on pretty big here in the U.S. I will gladly work with some of the other folks here to create documentation and instructions -etc. to help out people like myself who are new to servos. I don't know enough about the technical stuff, but I figure if a couple of work together we can help make it easier for the "next guy".

Chris

[Chris D]

Need a little more help - electronics guidence...

I used the HCPL-2631 OPTO and placed that on the OPTO3 location.

From what I can see on the layout, it looks as though there are two 1k2 resistors controlling the current for the input on the OPTO. My calcs (Keeping in mind I don't know what I am doing) say I should be using a 200 ohm resistor there.

Can someone clarify this for me ?

Also, the reason for this is that I put the scope on the input side of the OPTO and it is triggering, however, on the output side there is nothing going on. I think this is why I can't control the servo with my CNC control.

Chris

[vroemm]

BTW, I have a hunch this servo driver is going to catch on pretty big here in the U.S. I will gladly work with some of the other folks here to create documentation and instructions -etc. to help out people like myself who are new to servos. I don't know enough about the technical stuff, but I figure if a couple of work together we can help make it easier for the "next guy".

Good idea.
There are lots of details which would be usefull to put in a documentation of some sort.

Any thoughts on what form to use ?
Web site, text file, wiki ?
I have been playing a little bit with a UHU servo controller wiki http://gsst.wikispaces.com/

There is a new version of the UHU servo controller in the planning.
Then there has to be a clear doc about the current version, which would then become the old version.
And a doc about the new uhu version.
How can this best be organized ?
I do not know how best to do this in the above wiki.

[Chris D]

Good idea.
There are lots of details which would be usefull to put in a documentation of some sort.

Any thoughts on what form to use ?
Web site, text file, wiki ?
I have been playing a little bit with a UHU servo controller wiki http://gsst.wikispaces.com/

There is a new version of the UHU servo controller in the planning.
Then there has to be a clear doc about the current version, which would then become the old version.
And a doc about the new uhu version.
How can this best be organized ?
I do not know how best to do this in the above wiki.

I am involved with a couple of WIKI's right now and frankly I don't like them - not even sure why.

One suggestion is this. If one of us started a Yahoo group, everyone could post documentation in it. Everyone in the world could have access to it. I suppose for documentation, we could start out with a file format the everyone can edit such as plain text, HTML, or even MS word. Once it built to the point of being useful, we could publish it in to PDF format so that the final "document" is somewhat controlled and consistant.

Seeing as I got my first board up and running, I plan on making the other 2 next weekend if time allows. I will try to shoot some photos of the process to add to the document.

I also noticed the number of views on this thread - it is extremely high which leads me to believe that there are A LOT OF folks that are willing to give this a go - perhaps the documentation will take them onto the next level!

Chris

[ulihuber]

Chris,

you are right, for the 2631 it is more advisable to have a lower value. I'd try 560 Ohm and maybe go down to 200 if necessairy.
1.8 k were for the HCPL2231 which has 1.5 mA high level input current.

You are really close, aren't you ? :-))
Uli

[Chris D]

Chris,

you are right, for the 2631 it is more advisable to have a lower value. I'd try 560 Ohm and maybe go down to 200 if necessairy.
1.8 k were for the HCPL2231 which has 1.5 mA high level input current.

You are really close, aren't you ? :-))
Uli

Oh I am so close I can almost feel it :-))))))

I just dropped down to 360 ohm and that didn't help. I am going for 200 ohms now to see what happens!

Chris

[Chris D]

Hmm, I just can't seem to get the opto to trigger on the far side. I have dropped the resistor values down to 220 ohm (I don't have a 200 ohm).

I am wondering what could be wrong in this area.

Any suggestions on what to try next or what to look for?

When I ordered the optos, I ordered three. I did try a different one and the results are the same.

Do I need to have anything connected to the E-stop group of terminals

Ne, NA, -- ?

Darn, so close and yet so faaaaaaaaaaaaarrrrrrrrrrrrrrrrr away!

Chris

[vroemm]

Hmm, I just can't seem to get the opto to trigger on the far side. I have dropped the resistor values down to 220 ohm (I don't have a 200 ohm).

What is the voltage on the parallel port pins measured with the scope when connected to the uhu servocontroller ?
And what is the voltage over the input pins of the opto coupler ?
Both for the step and dir pins.
This way you can figure out if any current if going to the led in the optocoupler.

You haven't put the optocoupler wrong in the socket did you ? :-)

[jrz126]

Hi Chris,
Looking at the datasheet, it appears that you are correct about the optoisolator. I'd say to try a 330 ohm resistor in there, (which should give you a current of 11 mA) and see if that works.
Also, my interpretation of the graph is that the controller is trying to hold the position at the center line. When you turn the motor by hand and release, it should try to move back to the center line. Then depending on how the controller is tuned, (underdamped, overdamped, ect.), it will display on that graph.
When the system is underdamped you should see something like this: (which it sounds like this is whats going on with yours)

[ulihuber]

Chris,

the estop terminal has an internal pullup on the board. You don't have to connect anything. If estop is active, the error LED is always on.

Are you sure you've put in 10 k pullups behind the opto ? What signal do you have there ? Gnd or Vcc ?

Uli

[jrz126]

Did you try triggering the opto without using the parallel port? Just apply 5V from the board supply to one of the v+ pins) and see if the opto triggers correctly.
I'm thinking that the parallel port isnt able to supply the needed current?
Which pins on the parallel port are you using?

Also, I'm still looking for the pcb layout in eagle (or any other free pcb creation software) so if someont has it, could you please help me out?

[Chris D]

HI Guys,

I believe the problem is with my breakout board. I don't think I am providing enough current based on some quick tests with a plain LED.

ULI, I am 99.8345% sure I have the resistors where they should be - but there is always that tiny chance I don't so I am going to check again :-)

The breakout board I am using is of my own design - please read that as I don't know what I am doing!

Basically, the outputs from the parallel port are swtiching a darlington pair inside the ULN2803 chip. I have +5 via a 10K resistor connected to the output of the ULN2803. When switched it is suppose to go to ground.

On a hunch, I tried replacing the 10K with a 220 ohm and that didn't help either. I am working through it slowly as I don't know enough yet to be sure of anything I am doing ;-----(

CHris

[vroemm]

Basically, the outputs from the parallel port are swtiching a darlington pair inside the ULN2803 chip. I have +5 via a 10K resistor connected to the output of the ULN2803. When switched it is suppose to go to ground.

Then you have a 10K resistor in series with the 220 ohm resistor.
Your using 10220 ohm, not 220 ohm.
The optocoupler input is not getting enough current.

Try the idea mentioned by someone ealier.
Disconnect the UHU from breakout board and computer.
Put 5 volt on the STEP input.
If you keep switching this 5 volt on and off the servo should step.

If this works why not connect the UHU direct to the parallel port ?
Without the breakout board.

[Chris D]

Latest update - problem is definitely the break out board. If I trigger the opto with 5 volts (at the Dir or Step connection), the far side of the opto operates properly.

I have to tinker some more with the resistor value of the breakout board and the ULN2803 chip. If that fails I am going to "Hot wire" the damn thing straight to the parallel port!

Wish me luck!

Chris

[vroemm]

I have to tinker some more with the resistor value of the breakout board and the ULN2803 chip.

Thats not the best idea.
The basic principle of using a dalington here is not good.

Wish me luck!

Allways :-)

[ulihuber]

Chris,

congratulations and Merry Christmas !

Uli