As mentioned some time ago I am using a max6675 connected to various K-Type
thermocouples

During initial testing there was always an error of approx 1.5 degrees
compared to a range of other test equipment, in some cases using the same
probes for comparison.

I am now investigating the discrepancy.

Initially I thought it was because the connector (the other junction) was
not co-located with the max6675 and therefore the max6675 could not
accurately  do  could junction temperature compensation.

But as best as i can physically set up, the two devices (the thermocouple
connector and the max6675) are at the same (similar) temperatures and now
mounted about 20mm from each other.  So I have to dismiss this is the
cause, at least for now. The is nothing that I can see that is generating
any localised heat.

Disconnecting the Gnd wire from the Neg side of the thermocouple connector
and the readings seems to track nicely with all the other test equipment
that I have.  However, the max6675 is no longer able to detect when the
probe is disconnected.

The maxim manual states "The Alumel Lead of Type-K Thermocouple. Should be
connected to ground externally."  Not 100% sure what they are trying to say=
..

I understood this to read  as "connect as close as possible to the
thermocouple connector."  i.e Dont just simply short the Gnd and T- pins
together.

If this is not the case then I would need a three terminal connector to
connect the Gnd externally.

In summary, not grounding the T- path gives good results but unable to
detect a disconnect.

Any suggestions experience on how this should be wired.

Justin
--=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
.