"** no stamp or unable to identify"
Archiver
Posts: 46,084
Bugfix for stampbc - see the files section for stampbc_123.tgz
Some people reported that they were not able to identify stamps with
stampbc -j - nor
being able to make stampbc to communicate with the stamp at all.
Recently someone
said his stamp would reset when he merely plugs in the serial cable.
I think that this
now led me to the actual problem.
The reset at stamp pin ATN is formed by pulsing the serial DTR high
for some ms.
But in pre 1.2.3-versions stampbc just forced DTR physically low and
DTR physically high
(logically it's the other way around with RS232). The DTR (and thus
ATN) was held high
afterwards.
If your stamp connected DTR directly to ATN (DC coupled) the stamp
would be reset and held
in reset state until you pull the plug. ATN is merely an inverted and
level converting
connection to RST.
I usually connect the DTR to ATN via series and parallel capacitors.
This makes the ATN
line more resistant to noise and thus to unwanted resets. With this
circuit the level of
DTR is not important, the pulse is formed by the caps when changing
from low to hight.
DTR
+
||
ATN
| 100n
_|_
--- 100n
|
_|_
This has now been fixed in stampbc 1.2.3 (the only change to 1.2.2)
the DTR(ATN) pulse
is now formed as it should be. DTR is foced low, high for some ms
(RESET_DTRPULSE_US in
defs.h) and then brought low again.
Some people reported that they were not able to identify stamps with
stampbc -j - nor
being able to make stampbc to communicate with the stamp at all.
Recently someone
said his stamp would reset when he merely plugs in the serial cable.
I think that this
now led me to the actual problem.
The reset at stamp pin ATN is formed by pulsing the serial DTR high
for some ms.
But in pre 1.2.3-versions stampbc just forced DTR physically low and
DTR physically high
(logically it's the other way around with RS232). The DTR (and thus
ATN) was held high
afterwards.
If your stamp connected DTR directly to ATN (DC coupled) the stamp
would be reset and held
in reset state until you pull the plug. ATN is merely an inverted and
level converting
connection to RST.
I usually connect the DTR to ATN via series and parallel capacitors.
This makes the ATN
line more resistant to noise and thus to unwanted resets. With this
circuit the level of
DTR is not important, the pulse is formed by the caps when changing
from low to hight.
DTR
+
||
ATN
| 100n
_|_
--- 100n
|
_|_
This has now been fixed in stampbc 1.2.3 (the only change to 1.2.2)
the DTR(ATN) pulse
is now formed as it should be. DTR is foced low, high for some ms
(RESET_DTRPULSE_US in
defs.h) and then brought low again.