Nubotics WW01 Servo Encoder

[Guest]

Has anyone used the Nubotics WW-01 encoder and the example program posted on their site? I can't get it to work on the 16f877 and I am trying to port to a 18F4520.

Also when using a crystal oscillator as a clock does it matter if it puts out a square wave or sine wave?

Thanks

[Guest]

So I take it nobody has worked with this encoder. So how about my crystal oscillator question? Does it matter if its a sine wave or square wave?

Thanks

[PCM programmer]

Yes, the external clock should be a square wave.

Look at this figure in the back of the 16F877 data sheet:

[Guest]

Ok, I cant get the sample code below to work. I am able to go through the encoder test and when I manually
turn the wheels I see positions change. Then I hit enter to see the individual
wheels turn but nothing turns. If I comment out the #define
LEFT_CLK_ON_TIMER_1 the wheels just turn constantly. After zeroing the wheels the program still work correctly. On the PIC itself B1 and
B2 go directly to the servos B4 and B5 are connected to the dir lines of
encoder, C0 left servo clk A4 right servo clk. I did try the servo.c file and was able to turn the servos correctly using another sample program but I want to be able to use the encoders. I was going to try to port to a pic18F4520 since it has built in clock but its not working even with a 16f877.

Thanks!

Sample code:
http://www.nubotics.com/support/ww01/code/ccs_c/ww01_picc_square.c.php

[Guest]

Can anyone help with this problem?

[PCM programmer]

Their sample code has a large number of #ifdef's that are used to configure the options. The software also expects an external H-Bridge circuit that is identical to their MarkIII sensor board. There is a #define
statement in the code that is currently enabling that mode. There are
tons of other #define statements that must be analyzed to decide if
they are appropriate for your hardware.

Also, you've been in contact with the author of the sample code.
I can tell that by looking at the April 2009 Gzip'ed archives here:
http://list.nubotics.com/pipermail/wheelwatcher/
Do you expect us to know more about the program than he does ?


c:\windows\desktop\ww01_picc_square.c

[Guest]

No, I dont expect you to know more than the programmer but I was hoping with your expertise you would be able to spot any problems I have missed. You do not need an hbridge at all if you using it in servo mode. I think there just may be a problem with the routines used to set up the pulses for the servo.

[PCM programmer]

I can try it. Post your criteria for a successful test. What do you want
to see the code do ? (That can be observed with an oscilloscope).
I don't have their specific H-Bridge hardware.

[Guest]

OK, I would just like to be able to run through the test and have them be successful. Using RC servo mode and all debug test enabled. The HBridge will be commented out since I am not using it. B1 and B2 go to servos and B4,B5 connected to direction lines and C0,A4 are clk lines.

With the LEFT_CLK_ON_TIMER_1 uncommented I can manually turn the servos to test encoders and that works. But the next test to test motors fails. The servos do not turn at all. Talking to the author the code was originally compile with version 3.234.

With LEFT_CLK_ON_TIMER_1 commented out the servos turn in opposite directions constantly. After calibrating the servos they just work sporadically.

Also I am not using a markIII board at all just a 16F877.

Thats why I think there is something wrong in how it generates the pulses for the servo.

Thanks

[PCM programmer]

I'm not sure how I'm going to do this since I don't have your servo.
I can look at output pins on the PIC with a scope, but without the
encoder input signals, I don't see how I can test anything.

You need to list your connections between the WW01 board and the PIC.
You need to list your Revision of the WW01 board.