Diodes, Rectifiers, SCR, Triacs, SSR Huh???

MartinMueller2003

Supporting Member
This thread is intended to help demystify some of the common questions we might have about Diodes, Rectifiers, SCRs and Triacs. I have seen a few threads around that indicate a bit of clarity (or maybe more mud) may be needed.

Voltage? Current?

In this explanation you will see a lot of talk about current as opposed to the normal chatter about voltage. The truth of the matter is that diodes, scrs and triacs are current flow control devices. They control current by impeding current flow through the device (much like a resister). The amount of current that flows through the device is dependent on the load and the voltage applied to the load.

What is a Diode.

A diode is a silicon based device that allows current to flow through the diode in one direction and prevents current from flowing through it in the other direction (up to a max rated voltage [This may seem to be an odd place to use voltage but at the time the diode is impeding the current flow, the entire supplied voltage can be measured across the diode]). When current is flowing through a diode, the diode tends to have a small voltage drop across it. While the voltage drop varies between diode families, the forward voltage drop tends to be between 0.7v and 1.5v. When current tries to flow backwards through a diode, the voltage drop across the Diode is 100% of the supply voltage up to the voltage at which the diode dies. This is not something you want to test.
NOTE: Voltage appears across any device in a circuit that impedes the flow of current through the device. A diode attempts to impede current flow as little as possible in the desired direction of current flow and attempts to impede 100% of the current flow in the undesired direction.

What is a rectifier?

In simple terms for uses in winky blinky, a rectifier is a diode (or set of diodes) inserted into an AC power stream with the intent of creating a current flow in a wire that is all in one direction. In an AC system, current flows through the wire in both directions, alternating at a fixed rate (60hz in some areas, 50hz in others). When a diode is inserted into a wire that is providing AC current to a load, the diode only lets half the current reach the load. The other half of the current flow is blocked by the diode because it is flowing the wrong direction. This operation is called rectifying the AC current. A single diode in an AC circuit performs an operation called ‘half wave’ rectification (half of the available current is delivered to the load and the other half is blocked). When four diodes are properly put together, they create something called a ‘full wave’ rectifier or bridge rectifier through which all of the available current is delivered to the load and all of the current is traveling through the load in the same (a single) direction. This ‘rectified’ current is often referred to as DC current (the voltage of a rectified AC supply is a pulsating DC voltage that varies between 0 volts and the max supply voltage in phase with the AC voltage swing). The interesting thing about the current flow in this scenario is that as the current through the load drops to near zero close to the AC zero crossing points) current flow will stop before the zero crossing point is reached. This is due to the minimum voltage drop across the diodes. Remember that the diode requires a minimum voltage for it to forward current. When the AC input voltage drops below the diode forwarding voltage, then current will stop flowing through the diode (this is actually an important fact to keep in mind).

OK So what is an SCR?

SCR = Silicon Controlled Rectifier. At some point in the experimentation to develop a better diode, someone discovered a formula for a worse diode (this is inevitable). A diode that did not conduct current as expected. However, they also discovered that if you applied a small voltage to the silicon junction (the silicon control part of the name) the broken diode would suddenly behave like a normal diode until you stop the current (as happens when the voltage drops to zero) It was also discovered that there is a minimum level of current that must flow through the diode in order for the diode to keep passing the current. If the current drops below the minimum level, the diode 'switches off'. This means that if you have a pulsating DC supply (like the one you get from a rectified AC supply), the diode will turn off at the zero cross point and remain off until the control signal is applied again, at which point the diode would turn to full on. Note that the diode has pretty much two states: off and on with not much in between, Yes you can run the diode in the grey area between too little current to stay on and enough current to be full on, but this is difficult and does not really give any significant benefit for winky blinky. Note that an SCR only allows current to flow in one direction. It is after all a diode that needs a kick in the butt to get it to work properly.

Finally we get to a Triac

At some point someone figured out that pulsating half wave DC is not what most applications needed. Controlling an AC load means the load needs AC power applied to it. An SCR only allowed current to flow in one direction. The solution: wire two SCRs together in an inverse direction. Now you can trigger one SCR when current wants to flow left to right and the other SCR when current wants to flow right to left. Getting the trigger voltage right had its own issues but that was quickly solved by the diac and opto coupler. The next step was to put the functionality of both SCRs into a single package and the TRIAC was born.

So how does dimming work?

Since a triac has only two states: ON and OFF, how do we dim things? We don't. In AC based systems we only turn triacs on and we wait for them to turn themselves off at the AC zero crossing point. However, by knowing when the zero crossing point is going to happen, we can choose to turn on the triac at any point in the AC cycle. The later in the cycle the triac is triggered, the less current gets through the triac before it turns off due to reaching the zero cross point. By varying when we turn on the triac, we control how much energy (current) is delivered to the load. Averaged over time, less energy means less light output on a string of lights. It can also cause flicker since the load ends up getting these tiny little spikes of energy instead of a constant low level of energy.

Why are LEDs such a problem for dimmers?

With some dimmers we see that the high and low (~20%) portions of the dimming curve do not work right. It is not the LED itself that is the problem. It is the high efficiency (spelled LOW current) of the LEDs (the very reason we want to use them) that causes the problem. Remember the need to meet a minimum level of current in order to keep the triac turned on? With LEDs, as you get near the zero crossing point (either at the start or end of the AC cycle) the triac will not have enough current flowing through it to keep it turned on after the trigger signal goes away. This is solved in a few ways. For example:

  • Add a greater load that keeps the minimum level of current flowing at lower voltages in the AC cycle (you will sometimes here people say add a C9 bulb to the output).
  • Keep the trigger voltage applied while the AC voltage is near the zero crossing point. Many lighting controllers have an option to run this way.

Leakage Current? This thing leaks?

When a triac is ‘off’ it suffers from two types of ‘leakage’

  • Gate leakage. This form of leakage results in the triac turning on at random times. This one is easily solved by adding a resistor between the triac gate and M1 leads. The sensitivity to gate leakage can also be reduced by the circuit used to drive the Triac. The opto coupler used in many of the SSR designs includes a built in resistor to prevent spurious triggering due to gate leakage. NOTE: Anything that could possibly conduct current to the gate can cause the triac to fire at unwanted times. Even flux or moisture can cause this. An external resistor in the design will help clean up these issues.
  • Forward Leakage. This is the type of leakage that will allow an LED load appear to be ‘on’ very dimly while the string is supposed to be off. This situation is a bit more difficult to ‘fix’. Prudent choice of triacs with a low forward leakage rating is the best way to go. Another possible solution is to add a resistive load that will drain the leakage current. The drawback to a resistive load is that the resistive load is still present when the triac is full on. This will cause additional power consumption and causes your display to cost more to run. Not as much as a string of incandescent lights but more than a string of LEDs.

I hope this information helps.

Martin
 
Last edited:
Thanks Martin,

I always enjoy re-reading posts like these. Great stuff, and I almost always learn something new (or re-learn something I forgot).
 
Another small addition may be to explain the "leakage current" that some experience with SSRs, when a small current flows and LEds glow without any trigger at all. What to do then?
 
Another small addition may be to explain the "leakage current" that some experience with SSRs, when a small current flows and LEds glow without any trigger at all. What to do then?

I have added information about leakage current. Thank you for the suggestion.

Martin
 
Perhaps this should go in the Wiki somewhere instead? Info tends to get lost over time in threads...
 
Back
Top