A SCR heater controller

J

Thread Starter

José Claudeni

I´m a electrical engineering student and I urgently need some help. I need a cheap circuit (schematic with components values)that warms a little furnace using a power wirewound resistor in 60Hz line. The heat level (90 -250 graus celcius) is controlling by a output of a PLC (S7300), using voltage output 0-5V or/and 4m-20mA and a feedback inptut of the same PLC from a temperature sensor. The SCR´s firing must be made using PWM modulation (frequency carrier=1KHz).

 
José,

You may be a little confused about the way SCR's work and so I'll try to offer some assitance.

SCRs (Silicon Controlled Rectifiers)are essentially latching devices, that is to say that once they are triggered into conduction (by the gate input rising a certain level above the cathode) they will remain conducting so long as the current is maintained through them (the holding current). Put simply - once you turn them on, they stay on until you cut the power or the voltage goes through a zero crossing point(ac).

If you were to put an SCR in series with a light bulb, put it across a DC source and then trigger the SCR on, it would remain on, even after the gate signal was removed.

In AC applications, SCRs are normally controlled by one of two methods, Phase angle control, or burst firing.

Phase angle control turns the SCR on part way through the cycle and it turns off when the voltage reaches zero, this therefore varies the time the that the load receives power. This is how most common light dimmer circuits work, (although they typically use a Triac).

Burst firing turns the SCR on for a complete cycle (or number of cycles) and is normally used with resistive heating loads where they usually have a relatively long thermal time constant (they stay hot for a period). These are usually controlled by zero-crossing devices - ie. they turn on when the voltage is near it's zero point, so as to reduce RF interference.

You say that you want to control the SCR firing using PWM, but with SCR's as soon as the first PWM pulse was sent it would trigger the SCR into conduction and it would remain that way until it's next zero crossing. Therefore I can't see how PWM could really be used with SCRs on a 60Hz line.

If you have to use PWM for some reason (eg. if this is a requirement of the project) then you would need to look at using something like an IGBT, Mosfet, or possibly even Bipolar transisitors. This would also require rectification to a DC bus.

With most heating applications, the thermal time constant is very long in relation to the control timing. Put simply the heating elements (and whatever they are heating) will not change temperature much if the power is removed for a short period. This is why methods like burst firing work.

There may be some applications where the thermal mass is fairly low and the temp needs to be kept very stable (Crystal ovens for example) but these are less common.

In most cases PWM would probably be overkill for a heating application like this and may possibly be more complex for you than a simple phase angle setup or burst firing.

You really need to work out what the requirements are - does it *have* to be PWM? What line voltage will you be dealing with, what are the power requirements, what is the application like as far as thermal mass, etc. ? Commonly these sorts of applications can best be handled by a simple / cheap / safe Solid State Relay (SSR). These are available in Zero crossing or instantaneous turn on, are very easy to fit / wire and can be controlled typically from a low-power 3-30V source. They are opto isolated, so are safer too. Prices start from about $30AUD (Aussie $) so I suspect about $15 US. You will need to bolt the SSR (or IGBT or whatever) to a suitable heatsink and ensure it's correctly fused for safety!

If it's not essential that you build the project from discrete components, then I'd suggest this is probably your best bet. Opto22 would probably have a good line up, and I have used a number of devices from RS components (http://www.rscomponents.com) for small low-power apps like this and they have been great. They have various types of heating controllers and / or SSRs(check in 'semiconductors - Discretes' section, or look in the 'relays' section.


If you must go with PWM, then you'll need to get a suitably rated MOSFET or IGBT (and possibly an Opto for isolation / safety). You can get Mosfets in +5V (logic Level) triggers, but you would generally want to isolate it for the PWM controller via an opto or Mosfet driver/isolator chip. You'll also need to rectify the AC to DC.

If using a power resistor you will need to ensure it can handle the temperature / power or it will soon burn out - best to check the manufacturers specs and derating curves.

If you can give some more specifications and requirements for your project, we can probably offer some more advise and / or circuits

Regards,

Fred :eek:)
 
J

I've got a few questions for you, if you don't mind.

I am also trying to control a heating element using some device. I originally thought the SSR would be the best bet. The one I have picked out for the project is by Crydom, P/N CX240D5(R). I now understand that I will not need it to be controlled by PWM, based on your article, although it will still be somewhat PWM, in that I'll be switching it on and off, however, the time base will be much longer.

The heating element runs off of 110VAC, and I would be switching only the "hot" side of the circuit. Here are my questions.

1. What is meant by the zero crossing circuit? Does this have something to do with the zero voltage crossing of the AC source? Why would I need it?

2. This part is also available in a "random turn-on switching version." Is this just the opposite of the zero crossing design, in that it just comes on right when you switch it, instead of waiting for the zero crossing?

Which would you recommend for this application? Or would you recommend a different part for the job. I'm planning on using a PIC microcontroller for the controller. Any advice would be great.
 
Top