I have never seen a triac as a slave to a relay before, but in this appli=
cation it seem the best thing.  So... what can go wrong...?

As a remedy for relay contacts burning i have put a TRIAC as slave to it.

Schematic:

    Existing relay contact that
    can=B4t handle the motor itself
       ,---o/o-----.
       |       |<|-' gate
   O---+-------|>|------------.
230VAC        Triac    Load=3DVery inductive motor
   O--------------------------=B4


Background: a local company makes an apoparatus in where they use a readi=
ly made controller, said by the manufacturer to handle motors to 1A, but =
breaks after a year or so with only 0,5A motor.  Very bad, but no alterna=
tive.  The design uses a ridicolously small relay marked 5A (resistive), =
and no spark suppression.

The motor seem to be an unusually inductive asyncronous design low rpm wi=
thout capacitor.  And do get hot, so it IS a bad lowcost motor, but anywa=
y this has to work together.  Sparks are pretty impressive when laboratin=
g with an opened relay...

Adding VDR across motor or relay did not help much.

RC for this high impedandce and voltage i think is bad way.  Maybe use RC=
 to limit dv/dt (thus avoid arcing), *plus* a VDR to protect the C from o=
vervoltage.

The ready-made controller is too crowded to put in a larger relay.  To co=
mplicated (expensive) to put in a optisolator and triac design.

Addnina large ecternal relay is too expensive in both material and work.

This is to be produce din hundreds of units per year an margials are lare=
ady low.

So simplest solution is just to add an external triac in shrink tubing.  =
And it works. =20

But, how reliable?

Worries:

1) Gate current.  The prototype use a R in series to the gate. (because I=
 was unceratin and did not want the prototype to blow up in my customers =
face ;) )  But this sould ot be needed.  Have now seen a design tip from =
ST where they use dual triacs in similar connection, a sensiteive driving=
 a large.

2) Spurious turn-on at transients; Transient on the supply might turn on =
the Triac spuriously.  I can not afford filter components (manual free-ai=
r build work cost)  While this prototype using TIC206D seem not to have p=
roblems, I think using a triac that can stand high dv/dt A2-A1 is preferr=
able.  (ST calls theese "snubberless" and as no other manufactirer use th=
at word it seem just that they have registered that as trade-mark..?)  An=
y experience out there about using triac unfitered?

(filter against its own (radiated) switching noise is not an issue as thi=
s design switched more cleanly than just the relay alone.)

3) Overvoltage behaviour. ( Relays are very good at transients. At *very*=
 high voltages they may arc, for that AC half-cycle)  What about Triacs? =
 I hope they will just turn on, temporarily, when exposed to overvoltage,=
 ant hus protecting itself.  (there will be no large current as the load =
is inductive)  I hope not to need a VDR.  Any experience?


(and before anybody ask; no I can not design a new better PIC based based=
 controller for this relative low series runs and comptete with the mass =
produced already used one)
--
Morgan Olsson, Kivik, Sweden


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