Use Triac as SCR?

I'm thinking about reworking a 120VAC lamp dimmer so I don't have to turn the brightness to max to turn it off. I would rather turn the control to minimum and have the light go completely out and eliminate the switch.

The dimmer installed turns on at full brightness and doesn't quite go all the off at the minimum setting. There is an adjustment to set the low level, but it's not very stable.

This circuit from an old 1960s book uses a full wave bridge and SCR and can be adjusted to minimum so the light goes completely out. But I would like to use the existing triac and just add the bridge and other parts.

formatting link

Is there any problem substituting the triac for the SCR?

-Bill

Reply to
Bill Bowden
Loading thread data ...

Good morning, Bill. I'm not sure what you've got going with the dimmer (I'm assuming it's a store-bought item, and most of them work the other way, "off" being on the dim side of the dimmer pot).

I'm also not too sure you'll be happy with the circuit shown in the link. There are a couple of problems here, which might suggest a better circuit.

First, according to the text accompanying the schematic, the SCR will trigger when the cap reaches about 8V. This will mean the SCR gate is triggered by about 4 or 5mA of current (through the 1.5K resistor). This is enough for some sensitive gate SCRs, and there are sensitive gate triacs which will do this job (I would think something like the Teccor/Littelfuse L6004L3 would work, 4A(RMS)@600V, I(GT) = 3mA max in Quadrant 1), but if you want something beefier than that, I'm not sure. I'm sure you would also like a circuit that uses the triac you already have, which certainly isn't a sensitive gate triac.

Second, the triggering voltage itself is not going to be very consistent. You can visualize it working well on the first half-cycle, but the question is, what happens after that. There's no mechanism to discharge the cap every cycle or every half-cycle to ensure consistent triggering.

Third, the circuit itself is somewhat uneconomical, especially at higher currents. You have to pay for power diodes that will conduct the current of the load, just in order to get the benefit of control at low voltages without "snap-on hysteresis" (that's the effect you're seeing that's so frustrating when you turn the lamp dimmer down). Power diodes aren't a big problem with a 60 watt lamp (you can just use

1N4003+), but at 20 amps, it kind of whispers in your ear that there must be a better way.

There is a better way. I first saw this trick in the GE SCR Manual,

4th edition (another "old 60's book"), and it's just as good today. Not only that, but you can probably scrounge the triac, diac and the cap in your lamp dimmer, reducing your cost. Look at page 4 of Teccor AN-1003, "Phase Control Using Thyristors":

formatting link

and you'll see two little trick circuits on the lower right corner of the page that'll do the job quite nicely. The key is that they discharge the cap, leading to more precise control of load voltage at the low end. Circuit a) in figure 12 is just the ticket for lamp loads, because the two diodes only discharge the cap on the positive half-cycle. It only costs two diodes and a 1/2 watt resistor. This puts a major DC component into the lamp load, but there's no problem with that. If you're thinking about running something that will not like a DC component, try circuit b). It only costs two more diodes and another 1/2 watt resistor. And all these diodes are only responsible for discharging the cap -- they don't have to carry the load current.

Just for grins and giggles, I'd modify the app note circuit by adding a Quencharc snubber across the lamp load or triac, and also put a high current torroidal choke (you can scrounge these from PC power supplies) in series with the works, to minimize the possibility of RFI problems. You might also have the choke in the lamp dimmer -- look there, first.

You may want to consider having your switch in the circuit, just for safety purposes, if nothing else. When you turn the pot up to maximum resistance, the load is "off", but it's still "hot". This could make someone very unhappy if they're, say, replacing the bulb. Be sure to follow local code, and be careful about installing home brew line voltage stuff inside the walls without consulting someone who knows what they're doing. And of course, use appropriate safety precautions during construction and troubleshooting.

I hope this has been of use to you.

Good luck Chris

Reply to
Chris

formatting link

You probably could substitute a triac for the SCR, but that circuit is not very green - it constantly wastes 7W in the 2K resistor, and "needs some ventilation."

Can't you just reverse the outside connections on your on/off potentiometer so it starts from minimum brightness?

Reply to
Andrew Holme

You may be looking at the wrong diagram. I don't see any 2K resistor. It's the drawing titled "120VAC Lamp Dimmer"

Yes, thought about doing that, but I want to eliminate the switch and have the light go completely out at the extreme dim setting. And it won't do that, because the phase needs to be exactly 180 degrees, and it isn't stable enough to maintain 180.

-Bill

Reply to
Bill Bowden

Actually, the 1.5k resistor just establishes the trigger level. The two transistors will conduct at the trigger level and then discharge the cap through the gate at more than 20mA. The cap will discharge to well below the trigger level before the transistors turn off.

The triggering voltage should be consistent since the mechanism to discharge the cap on every cycle is the dual transistor arrangement that acts like a diac and becomes a very low resistance when triggered. I think it's called a SCS (silicon controlled switch).

It works well, I tried it with a 25 amp SCR with good results.

-Bill

Reply to
Bill Bowden

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.