Subject:
|
Re: bad motor output because of bad lego firmware?!
|
Newsgroups:
|
lugnet.robotics.rcx.nqc
|
Date:
|
Thu, 28 Oct 2004 17:36:15 GMT
|
Viewed:
|
5216 times
|
| |
|
|
In lugnet.robotics.rcx.nqc, Gunther Lemm wrote:
> Hi there,
>
> I've spent the last two days trying to write a function in NQC that transfers
> data to the Lepomux I/O-extender. The data transfer is normally done by
> connecting the Lepomux to one of the motor ports. Each transition from one motor
> state to another indicates the start of a new bit on the lepomux side. This
> includes transitions from forward to backward or vice versa.
> When I do something like this
>
> On(OUT_B);
> SetPower(OUT_B, OUT_FULL);
> while (1)
> {
> SetDirection(OUT_B, OUT_TOGGLE);
> }
>
> nothing happens at all.
>
> Something like this toggles between forward and backward:
>
> On(OUT_B);
> SetPower(OUT_B, OUT_FULL);
> while (1)
> {
> SetOutput(OUT_B, OUT_ON);
> SetDirection(OUT_B, OUT_FWD);
> SetOutput(OUT_B, OUT_ON);
> SetDirection(OUT_B, OUT_REV);
> }
>
> But the output waveform looks like this:
>
> FWD ____ ____
> | | |
> OFF |____ ____| |____
> | |
> REV |____|
>
> There is a gap of at least 3ms between the fwd->off and the off->rev edge. This
> makes it impossible for the Lepomux to recognize it as one edge.
> The reason for this behaviour seems to be that the bytecode interpreter (or even
> the ROM routines) don't directly access the motor output register in the RCX
> (two bit per motor for off,fwd,rev,brk).
The firmware (at least version 0328) intentionally prevents you going directly
from forward to reverse (or vice versa). I guess the designers felt this would
be too rough on the motors. Try inserting a Wait(50) in your first loop. It'll
work then, but there is a dead time between the forward and reverse states.
> Does anyone know a workaround or do I really have to patch the Lego firmware
> image?
I'm guessing a patch will be necessary.
Mark
|
|
Message is in Reply To:
 | | bad motor output because of bad lego firmware?!
|
| Hi there, I've spent the last two days trying to write a function in NQC that transfers data to the Lepomux I/O-extender. The data transfer is normally done by connecting the Lepomux to one of the motor ports. Each transition from one motor state to (...) (20 years ago, 28-Oct-04, to lugnet.robotics.rcx.nqc)
|
4 Messages in This Thread:
 
 
- Entire Thread on One Page:
- Nested:
All | Brief | Compact | Dots
Linear:
All | Brief | Compact
|
|
|
|
