Chris Loiacono wrote:

> I already have current sensing on two of three load legs. I am reading the
> load current, comparing it to a dipswitch-set limit and closing the loop by
> limiting the firing angle so as to not to exceed the set limit. What can
> this tell me that can prevent out-of phase gating, or is this kind of
> mis-firing even possible in this scenario?

All sorts of things are possible if you are driving an
inductive load. :o) Have you considered putting your
thyristors on the secondary of the 3-phase transformer,
and using SCRs instead of triacs will act as your
rectifier too, saving significant $$ and removing your
phase problem.



> This is already an option that can be selected if the current limit feature
> I described above is not. It displays in text whether an open or
> over-current situation has been detected and allows the choice of whether to
> notify only, or disable the controller.

I would really think about monitoring all 3 legs,
for total overcurrent and imbalance.

> > Commercial controllers of that power
> > level are very expensive, you are finding out why!
>
> Yes, for sure. This is a commercial product that had an original goal set
> that only included use with large resistive loads. As such, it took over a
> year and a budget literally as big as a house to develop single and
> three-phase versions with analog and digital command input, current limit,
> dynamic current tracking and the like. It has been serving very well in
> commercial resistive load applications.
> I am just hoping it can be adapted for use with inductive loads without the
> same kind of expense.
> I haven't been able to find a competing product that is as precise and has a
> similar set of features that will work equally well with resistive and
> inductive loads. They are now being assembled in lots of 50 & 100, but the
> distributors are getting interested, and I would like to include this
> capability in the next go round.

The inductive loads are going to be an issue if
you are putting triacs before the load, ie as you
originally mentioned using it to control 3-phase
transformer to rect to resistive load. In that case
it should be predictable at least, you may be able
to use a lookup table of phase firing angles to give
you a ball park secondary load current, providing the
transformer will be the same on each unit produced,
then use your feedback circuit to fine tune it. That
should give some safety and stability.

Since the end load is predictable (resistive) and
the transformer properties are constant you might be
able to get away with just monitoring secondary current
and limiting to a slow rate of change. Hopefully
that would reduce problems caused by current lag at
the thyristors. In the worst case i'm sure you could
do it by monitoring the primary current (and current
lag) but you still really need to monitor secondary
so it gets complicated. :o)
-Roman

--
http://www.piclist.com hint: PICList Posts must start with ONE topic:
[PIC]:,[SX]:,[AVR]: ->uP ONLY! [EE]:,[OT]: ->Other [BUY]:,[AD]: ->Ads