Switching inductive loads

[Perion]

Greetings fellow Earthlings,
I know pretty much how to design circuits for zero crossover switching
controllers for pure resistive loads but how is this done with inductive loads?
How accurate an estimate for the inductance of the load is required for proper
component selection?

Perion

 

[CFoley1064]

Subject: Switching inductive loads
From: "Perion" RazroRog@hotmail.com
Date: 10/27/2004 9:14 AM Central Daylight Time
Message-id: <K4WdnawYepTsMOLcRVn-iQ@comcast.com

Greetings fellow Earthlings,
I know pretty much how to design circuits for zero crossover switching
controllers for pure resistive loads but how is this done with inductive
loads?
How accurate an estimate for the inductance of the load is required for
proper
component selection?

Perion

And greetings to you, fellow Earthling. Since Klaatu is busy (man, we could
use him now), I guess I'll have to do.

I'm assuming you're talking about switching inductive AC loads with TRIACs.
This can be a bit of a problem because, at the moment of zero voltage crossing,
current is still flowing which prevents the TRIAC from turning off.

There are two approaches to your problem. One thing which is commonly done is
to place an R-C snubber across the TRIAC to get the power factor down a little
and help reduce the current going through the TRIAC at zero criossing. The RC
snubber is also good for limiting dV/dT, which can be responsible for false
turn-ons. This solution has limited usefulness, mostly because you have to be
very aware of the load inductance, and you're wasting power while the load is
off.

The other approach, which is much more commonly used, and which will truly
minimize the problem, is to use inverse-parallel-wired, or back-to-back SCRs,
and trigger each of them on the appropriate half-cycle. Since an SCR blocks
current when reverse-biased, it will turn off at zero crossing, and will stay
off (assuming the voltage across the SCR doesn't exceed Vr(max). The
difficulty with this is that it requires a more complex triggering system,
because you have to keep track of which SCR is next to be triggered, and the
gates are at different line potentials.



___ ___
o----|___|--UUU-------o-----.
| |
o--------------------------\-
| A
| -
- |
V |
- |
o--------------------/| |
| |
o---------------------o-----'
created by Andy´s ASCII-Circuit v1.24.140803 Beta www.tech-chat.de


One easy solution I've used successfully a number of times over the years is
Teccor Electronics alternistors. These are TRIACs which have been specifically
designed to switch inductive loads. They act the same as back-to-back SCRs,
but only have one gate. They require a little more gate current to turn on, and
only operate in quadrants I, II and III. But, you can use them with any
standard TRIAC circuit where you've got a positive gate pulse with a positive
voltage on the load (Q1), and a negative pulse when there's a negative voltage
on the load (Q3). They're made to control voltages up to 800V pk., and
currents up to 40 amps. Look up Alternistors in the Digi-Key catalog, and get
a one-chip solution. Here's the data sheet:

http://rocky.digikey.com/WebLib/Teccor/Web%20Data/Alternistor%20Triacs%20(
6-%2040%20Amps).pdf

For the standard newbie circuit of an RC charging circuit with a DIAC driving
the gate of the TRIAC, all you have to do is remove the TRIAC, insert the
alternistor, and you're good to go. (view in fixed font or M$ Notepad):


o----------o---------------------.
| |
| |
.-. |
| | |
| | |
'-' C|
| Inductive C|
R | C|
.-. Load |
| |<--. .-.
120 VAC | | | | |
'-' | | |
o----o----. '-'
| _|_ |
| DIAC V_A _|_
| | V_A
C --- '---------/ |
--- |
| Alternistor|
| |
| |
o---------o---------------------'
created by Andy´s ASCII-Circuit v1.24.140803 Beta www.tech-chat.de

It's still good to put a small snubber acrss the alternistor to help with
dV/dT, say 0.047uF in series with 100 ohms as a starter.

Good luck, and as always, be careful with line voltage
Chris