> On Sun, 2016-03-13 at 10:35 +0100, Ruben J=F6nsson wrote:
> > >=A0
> > If the shield is connected to GND in your circuit, it will load the
> > data signal=A0
> > capacitively more than an unshielded cable which will affect the
> > signal and=A0
> > timing negatively. Use unshielded cable or connect the shield to
> > chassis GND on=A0
> > the master side.=20
>=20
> Given how slow OneWire is, I wouldn't expect that to be a problem; in
> fact, I might even consider some capacitance on the data line to help
> out with noise. =A0It isn't clear to me from the spec sheet, but it looks
> as if the sensor itself might provide considerable capacitance.
>=20

When your master should be able to read sensors both on a very small networ=
k=20
with short cable and few sensors and a large network with long cable and ma=
ny=20
sensors, the difference in total cable capacitance depending on if the cabl=
e is=20
unshielded, shielded or shielded with a floating screen compared to signal=
=20
ground, will make it harder to differentiate between a high level and a low=
=20
level in all cases because of the more critical timing. Eliminating as much=
=20
cable capacitance will make the network more reliable. The alternative is t=
o=20
limit cable length and number of sensors on the network.

But, you are correct, in the OP:s case it probably does not matter much.

The capacitance on the data line for the sensor is 25pF and the total=20
capacitance for a network should be less than 1nF. The tolerable load=20
capacitance can be increased using active pull up.

/Ruben



--=20
http://www.piclist.com/techref/piclist PIC/SX FAQ & list archive
View/change your membership options at
http://mailman.mit.edu/mailman/listinfo/piclist
.