Which type capacitor is appropriate for a triac driver snubber?
I'm using a sensitive-gate triac (Teccor L601E3) with an optically coupled triac driver (MOC3011 equivalent) to control a solenoid valve coil that draws about 75 mA at 120V with a PF of about 0.15.
According to the application literature and calculations, I plan to use a
0.068 uF 400-600V capacitor and 3300 ohm resistor as a snubber for the triac driver. The app lit indicates that this should also be sufficient to protect the triac.
I'm leaning toward the metallized polypropylene dielectric caps, such as the DigiKey BC1879 and/or BC2218. Would these be appropriate, or would a different type be better?
Any type that is rated for across the line filtering should work. It looks like the first one you list (the only one I looked up) has a 630 volt DC rating and and a 220 volts AC rating, so it should handle the occasional line voltage spike pretty well.
I am wondering why you do not drive the solenoid directly with the MOC3011. It can carry about 150 ma of load current. The snubber you connect across the load will lower the total current a bit also (power factor correction).
Hi, Neil. When you're choosing line voltage snubber caps, you really should pick caps that have been rated as "across-the-line" capacitors. These caps are self-healing -- that is, they're made so that in the event of an overvoltage punch-through, the short will fuse itself, preventing a short across the line. You can key in UL1414 as the standard to look for in the US, and CSA C22.2 No.
1-94 for Canada.
In the Digi-Key catalog, one good choice would be the Panasonic Interference Suppressor Caps on p. 903 of the current catalog (T043). One good choice might be DK P/N P10729-ND for an 0.068uF cap rated for 250 VAC. They're cheap ($0.20 USD ea.).
I'm kind of curious about the 3300 ohm series resistor, though. Don't you think that's a little conservative? You've got a TO-92 triac that can handle up to an amp, and you're only switching 75mA. Wouldn't something like a 330 ohm 1/2 watt carbon comp resistor be a better idea? Your snubber will work a lot better and do a better job of protecting the logic triac if the cap can soak up energy faster.
If you choose a snubber cap rated for across-the-line, then you can pick a series resistor small enough so the cap can actually accomplish something. Oh, yes -- don't forget the fuse.
This particular design from the Motorola app note is actually intended to protect the triac driver, which is triggering the gate of the triac at about
3 mA. The triac driver sees the capacitor through a 180 ohm resistor. (See Figure 22 in the app note AN3008 at
formatting link
(The Motorola app note is no longer available on line.)
I'm also looking at the off-state current flow around the triac through the snubber, and trying to balance the two. Don't want the solenoid buzzing when it's supposed to be off.
Thanks for the pointer to the capacitor. I hadn't run across that one yet.
The problem I am seeing here is that the snubber might pass enough current to keep the solenoid pulled in once it is activated. The hold current and the pull in current can be vastly different. I have seen it happen. I would even hazard that the original value might still be a bit low, but testing can bare that out. The 330 with .068uF might even be enough to pull it in from power up.
Hi, Jim. Check the impedance of an 0.068uF cap at 60 Hz. The cap limits the
60 Hz current, not the resistor. The purpose of the series resistor in a snubber is to limit instantaneous current when the switch closes. The balance between rise in voltage from switching an inductive load off, and current inrush through the switch when switching on, is the whole art of choosing a snubber for inductive loads. And 3 mA shouldn't be enough to hold a 75 mA solenoid on. Generally you figure that less than 1/10 rated voltage or current is pretty much failsafe to turn off a solenoid off (although you might get something of a buzz -- if so, you might want to lower the capacitance rather than increase the resistor)
From a later post, though, the OP says that the cap is actually protecting an opto-triac rated for 150mA. In that case, a 3300 ohm resistor makes more sense.
Also excellent advice. The resistor is necessary to let the TRIAC turn on without exceeding its peak current capability, but only degrades the snubbing operation when the TRIAC turns off.
If you put the snubber across the TRIAC, it will snub just fine, but does not lower the average current any, via power factor correction. It also allows a leakage current to pass through the load when the TRIAC is off.
Putting the snubber network across the load, instead, lowers the total load current but might allow an occasional half cycle of fault conduction if a line spike occurs.
No need to achieve resonance to reduce the current a bit. Just reduce the resistor till the peak current possible (turn on at peak voltage) starts to get appreciable. If the cap passes 1/8th of the solenoid current, it can reduce the solenoid current by about that amount. This could get you well below the thermal limit that keeps you from driving the load directly with the isolator. Large pads on the isolator pins (without violating voltage rules) helps there, also.
If the source impedance is essentially zero, it works either place, since it keeps the inductive load from producing a sudden voltage change when the TRIAC turns off at zero current but the source is not at zero voltage because of the inductive lag. The risk of putting it across the TRIAC is that if the capacitor shorts it keeps the load energized, and otherwise leaks current through the load any time the TRIAC is off. The risk ot putting it across the load is that if the cap shorts, it overloads the TRIAC, and if there is a line voltage spike, it goes through the cap and risks false triggering the TRIAC, though this is often inconsequential.
I had played with the idea of putting a cap across the coil for the reasons you mention. I figure that a 1.6 uF cap would make the solenoid coil a parallel resonant circuit at 60 Hz, bring PF to 0 etc. But, I would also expect some ringing at shutoff..... Plus, caps of that value are large and expensive.
In digging through a fairly extensive library of books (I taught electronics for 17 years) I found very little on snubbers except in some RCA and Motorola databooks. In all of them, the snubbers are placed across the triac. No mention is made of putting it across the load.
Neil
should
caps are
overvoltage
line.
No.
Very good advice.
Interference
might
($0.20
handle
330
work a
can
Also excellent advice. The resistor is necessary to let the TRIAC turn on without exceeding its peak current capability, but only degrades the snubbing operation when the TRIAC turns off.
If you put the snubber across the TRIAC, it will snub just fine, but does not lower the average current any, via power factor correction. It also allows a leakage current to pass through the load when the TRIAC is off.
Putting the snubber network across the load, instead, lowers the total load current but might allow an occasional half cycle of fault conduction if a line spike occurs.
The reason for the snubber is to limit the dv/dt across the triac so it does not turn on ( either when the mains are applied or when the inductive load creates dv/dt from the residual current after zero current crossing ( which is not zero voltage crossing )). It sounds like your relay has such a high impedance that it cannot cause a very high dv/dt across the capacitance of the triac to need a snubber at all. Having said that you may need a very small snubber if the triac is a low dv/dt or have a low voltage rating. If you select a triac with a voltage rating much higher than required the need for a snubber diminishes.
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.