Need a Triac; haven't got one!

The impedance of 0.1u at 50Hz is 1/(6.28 x 50 x 0.1E-6)=32K at -90o phase. But this gets cancelled by the inductive reactance of the bell, but roughly that would mean L is 32E3/(6.28 x 50)=100H... seems high, possible but high.

Try moving the RC snubber to across the ringer coils instead of across the triac.

piglet

d n t y

Assuming now the bell has enough reactance to resonate almost all of the 0.

1u reactance, I made the bell 85H and advanced the line voltage phase to ac orresponding 80o. The transient resonance frequency is much less but the ov ershoot is now approaching 600V and this behavior lasts for about 5x line c ycles. So, assuming your bell, which must weigh 20 lbs, has that much induc tance, the main issue with the circuit is underdamping and not dV/dt. The o ne saving aspect is the current is flea power level of 3mA peak. So you'll need to use an 800V TRIAC or a 600V TRIAC shunted with a 400V TVS, then it should be controllable.

Version 4 SHEET 1 880 680 WIRE -32 176 -112 176 WIRE 80 176 48 176 WIRE 352 176 160 176 WIRE 352 272 352 240 WIRE -112 304 -112 256 WIRE 352 416 352 352 FLAG 352 416 0 FLAG -112 304 0 SYMBOL cap 336 176 R0 SYMATTR InstName C1

SYMBOL res 176 160 R90 WINDOW 0 0 56 VBottom 2 WINDOW 3 32 56 VTop 2 SYMATTR InstName R1 SYMATTR Value 4K SYMBOL ind 64 160 R90 WINDOW 0 5 56 VBottom 2 WINDOW 3 32 56 VTop 2 SYMATTR InstName L1 SYMATTR Value 85H SYMBOL voltage -112 160 R0 WINDOW 123 0 0 Left 2 WINDOW 39 0 0 Left 2 SYMATTR InstName V1 SYMATTR Value SINE(0 315 50 0 0 80 5) SYMBOL res 336 256 R0 SYMATTR InstName R2 SYMATTR Value 100 TEXT -144 336 Left 2 !.tran 0 100m 0 0.1U startup uic

You try it. All you've done is trade an underdamping problem to a dV/dt problem.

Try reducing the value of the cap to .033uF and up the resistor to 180 ohms. That ought to do it.

Argh! You're right, Jim. That was the problem; thanks!

Yes, I recognise the correct vertical format now you've reposted it, Jeroen, and it loads with no problem now. Many thanks indeed for that.

It does seem an awful lot, Fred, but would certainly account for why I've been unable to measure its inductance using any of my DVMMs (max range

20H). I've got a proper Leader inductance bridge somewhere if I can somehow find it among all the junk here... BTW, I got your netlist loading fine now. It's getting late here so I'll have to check it out properly tomorrow. Many thanks for that.

That'll merely bypass the ringer coil instead of the 'switch' so how can that help? He probably needs to firstly establish the correct values, then make 2 x snubbers and place one across the triac and the other across the coil.

Nothing. Except make sure the triac is rated for it. The max dv/dt is about 1V/us, so no problem there. The snubber values could be adjusted to reduce the peak voltage as suggested elsewhere in this thread. It is not a disaster, after all. Sorry.

,

Measure the ac current RMS through the line powered bell. If it's not somew here near 50mA or more, forget about using a TRIAC because you're not going to meet it's holding current requirements. Use a relay instead, and the re lay will need to be snubbed too.

If you drive the gate continuously you don't need to worry about holding current.

--sp

--
Best regards,  
Spehro Pefhany 
Amazon link for AoE 3rd Edition:            http://tinyurl.com/ntrpwu8

At HF the mains is very low impedance so the two configurations are effectively equivalent. You can think of the snubber across the coil as being a bit like the diode across a dc relay coil - it snubs the maker of the dv/dt rather than the victim of the dv/dt.

Snubbing ac loads instead of the switching device or triac is often very useful and avoids the troublesome leakage current through the snubber that the OP has experienced. I don't see the need for two sets of snubbers here, that strategy is not often done even in smps where dv/dt is more extreme.

piglet

o

l,

igh,

've

ll

mewhere near 50mA or more, forget about using a TRIAC because you're not go ing to meet it's holding current requirements. Use a relay instead, and the relay will need to be snubbed too.

Ummm...correction: You drive the gate continuously because you need to worr y about the holding current. How often does a bell get turned on/off anyway ? Driving one EM with another EM device is a more harmonious and simplified fit.

Much of the problem I'm having is that none of the standard textbooks I keep on electronics cover triacs in anything like sufficient detail. In fact even AoE has *nothing* on triacs at all! Nothing! Unbelievable. :(

Harmonious??? I've never seen that in a list of requirements. BTW, what then do you use to drive the new EM device?

--

Rick C

Why, a sub-miniature relay of course! :->

So it's relays all the way down? ;)

--

Rick C

Thanks, Cursitor! That's fixed it. Works like a charm now. :)

Strictly speaking I think you may be right. I tried feeding it with 35Hz from my HP LF gen and it was able to ring at a *much* lower voltage than

240V. But adding a 35Hz generator to the board would add an extra layer of difficulty, so it's going to work off 240V/50Hz from now on whether it likes it or not. :->