[Nut-upsdev] driver for IVT SCD solar controller ?

[Rainer Fuegenstein]

Arjen de Korte wrote:

> Although we're not getting the reply we expected, at least we're getting 
> something (otherwise you would see an error or timeout message), so this 
> isn't too bad. Most likely the controller is (unexpectedly) sending out 
> empty lines before the reply. Please try the latest version from the SVN 
> trunk, that is supposed to skip over replies that are shorter than 10 
> bytes. When running in debug level 4 (-DDDD), it will also show whatever 
> we got in the reply.

yesterday I tried to track it down, mostly in serial.c (with some custom 
printf()s) with the following results:

- the F gets sent.
- the low level receive function returns 41, which is (estimated) the 
length of the response record from the SCD with CR+LF.
- but the receive buffer seems to be empty. maybe some stray 0 byte in 
the start of the buffer ?

> Since we're getting *something*, the connection seems to be fine.
it works without problems in minicom and with a simple perl script (see 
attachment), so it shouldn't be a wiring problem (while testing I  made 
sure that no other app was accessing the port and the driver has r/w 
access to ttyS0).

I'm now writing this from memory, therefore sorry for the lack of 
detail; I'll check the latest version from svn in the late evening when 
I'm back home again and let you know my findings.

thanks & cu
-------------- next part --------------
An embedded and charset-unspecified text was scrubbed...
Name: scdlog.pl
URL: <http://lists.alioth.debian.org/pipermail/nut-upsdev/attachments/20090730/a4bd5dac/attachment.asc>