Need a Triac; haven't got one!

Since you need isolation between mains and telephone line, the other alternative would be to use two opto-SCRs in antiparallel and putting the isolator inputs in series. Alternatively, use ordinary SCRs with two opto-couplers.

Reply to
upsidedown
Loading thread data ...

In message , Jim Thompson writes

It's possible to make SCRs and indeed triacs ( and unijunctions) out of discrete transistors .

formatting link

I've done SCRs and unijunctions that way, but not triacs.

TBH with next day delivery from a lot of component suppliers, I'd just order the proper thing and have done with it.

Brian

--
Brian Howie
Reply to
Brian Howie

Two SCRs will readily outperform a triac but are not a drop-in replacement as the gate triggering will need total redesign.

How were you planning on triggering the triac, if you had one?

The old GE SCR manual is a worthwhile addition to an electronics bookshelf.

piglet

Reply to
piglet

MOC3061 is multi-sourced, cheap and has 600V PIV.

piglet

Reply to
piglet

I should perhaps mention that isolation for the phone line is NOT an issue here. None of this will be connected to the phone line in any way.

Reply to
Julian Barnes

No I don't!!! I'm using this phone bell for completely different purposes nothing to do with the phone system!

Reply to
Julian Barnes

For breadboarding purposes, I'm using a Solid State Relay instead of a triac. The control voltage on the SSR is either 12V or 0V, it never goes negative, yet the SSR has no issues turning off. Mind you, a SSR might be more 'intelligent' than a triac for all I know. "Why don't you just use that SSR", then I hear you say. Well, it's too good for this app. It's an expensive 40 amp device and this bell only requires 60mA.

Reply to
Julian Barnes

Argh! I do wish you hadn't introduced this *non-existent issue* into the thread. :(

Reply to
Julian Barnes

Cher Piglet,

The easiest solution would be to order a suitable triac of course, but deliveries around here are not undertaken lightly hence I have to adapt what I already have to fit.

The triac would be triggered direct from a 555 ideally: 0V or 12V. Shouldn't be a problem.

Reply to
Julian Barnes

Put the SCR inside a full wave bridge, is one way.

Reply to
bloggs.fredbloggs.fred

Thank you for that kind suggestion, Mr. Bloggs. The matter has now been resolved satisfactorily. Thanks to all who assisted.

Reply to
Julian Barnes

That is OK as long as th SCR gate voltage is derived from a circuit between the bridge positive and negative terminal. If any external control is required, some isolation would be needed, However, the OP stated that isolation would not otherwise be required.

For a triac, the gate control voltage can derived between mains Live and Neutral, thus simplifying control voltages relative to the Neutral.

Reply to
upsidedown

AC input }}------LOAD-----+ | | ^ ~ / \ _ / \ \ /| D D \| / / \ - / \ +--< - + >----+ | \ / | | \ / | | \ D D _ _ | _| \ / /| V SCR | \ / / - | v |\_____ Trigger | | ~ | | | | }}------(--------+ | AC input | | +-------------------+

Can't really draw the diodes so well. Indicated polarity with arrows.

--

Rick C
Reply to
rickman

take a triac out of a wall lamp dimmer

m
Reply to
makolber

A local radio ham was eventually able to help me out with one from his junk box. Do they require some form of snubber when driving an inductive load? If this coil has a DC resistance of 4k ohms there must be hell of a lot of turns on it!

Reply to
Julian Barnes
[about a triac]

If your trigger is a pulse at zero-crossing, then yes a snubber is recommended. Something like an R-C series combination parallel with the load. This ensures the triac turns ON before your pulse ends.

A continuous gate current, on the other hand, doesn't require any snubbing on the load side, it should just work with a resistive or inductive load.

Reply to
whit3rd

Didn't you say the load was a bell? That works by a switch opening and closing so there will clearly be breaks in the current which will allow the SCR to stop conducting when the trigger signal is removed.

I think a triac will have the same problem with an inductive load, but then the bell will provide the same fix.

--

Rick C
Reply to
rickman

The bell works off 250VAC. There's just the one coil and no switches. The magnetic poles on the coil's core just swap polarity on alternate cycles which makes the hammer wiggle.

Reply to
Julian Barnes

.

The TRIAC commutates ideally when its terminal current becomes zero. The cu rrent through the TRIAC is the same current through the bell coil. Because the bell coil is inductive, its current lags the circuit source voltage, li ne, by an angle ATN(wLcoil/Rcoil). This means that when the current passes through zero, the source voltage is Vm*sin(ATN(wLcoil/Rcoil)), which, in th e absence of snubbing, all gets instantaneoulsy applied across the TRIAC te rminals. Most line TRIACs are only guaranteed to withstand 30V/us rate of r ise in voltage and still remain off under this kind of situation. So you ne ed to parallel the TRIAC with a series R+C to limit the turn off voltage ra te of rise. The simplest way to determine adequate C is to just run a sim o n it. In the end you select a C 10x larger than the sim would indicate. You r circuit is complicated by the fact the bell magnetic circuit has mechanic al motion in it, meaning probably the Lcoil is not constant and varies as a function of some kind of striker position. It's not worth fretting over, j ust over design.

Reply to
bloggs.fredbloggs.fred

Thanks for that very clear explanation. I was recently wondering what the claim, "dv/dt rated" means on some of the power MOSFET datasheets I've seen lately. I'm guessing something similar...

Reply to
Julian Barnes

ElectronDepot website is not affiliated with any of the manufacturers or service providers discussed here. All logos and trade names are the property of their respective owners.