I am bit skeptical on using PID since this system is a slow response one and considering the complexity of tuning a P,I and D ...


View attachment 77721

PID was developed by watching ship captains steer their ships into dock. PID systems can be tuned for systems of almost any speed. Remember the phrase, it takes a long time to turn a battleship? PID can help with that.

Also, if you buy an arduino max board and a MOSFET, you can use the FREE software from the 3d printer group REPRAP. Their system has a self-tuning PID controller for a bed heater - the bed is the flat plate that they use to print the part into.

It would be best to have a 100k pull-down resistor fir the mosfet gate and a small value resistor between pin and gate.

 

I want to maintain the surface temperature of the table at 80 deg centigrade. so i hope with this design and the generated temp it will take more time to heat the table to 80 deg centigrade (considering the table's initial temperature is 25 deg) with the present wiring scheme.

You want to add an extra circuit in your plan: an over temperature cutout. Using a second temperature probe, if the temperature goes above the limit, it will cut out the power to the heater. This way, you have a controller for the heater, and if it fails, this circuit will prevent damage to the table. You will probably want an indicator to show when this circuit is blocking the power to the heater.

 

Setting up a PID controller is not all that hard, there are only three numbers to work with, plus most come with a secondary alarm output that can be used to trigger a relay to override the main power if the temperature gets too far out of range.

Once set right a PID controller would keep your table within a few degrees of its set point which you will never achieve with a normal capillary tube or snap action bimetal disk type temperature control system.

 

It is more clear now. I think i ll go with the PID option. I ll try to tune the system adjusting the P,I and D if it doesn't work then hope i can use the auto tune function. If nothing works then i ll go with the ON/OFF control.

 

I have two 500w (230V, 2.17A) strip heaters and each having max operating temp of 300 deg centigrade. If i wire them serially will i be able to reduce the heat generated to 50%?.. Some one plz advice.

You can reduce it even further on a purely resistive load by half-wave rectification of the mains.

The rectifier has to handle the full peak current even with the heater cold, and withstand the full peak reverse voltage - including any noise spikes that might appear on the mains.

 

What I do when I need to very finely control an AC run resistance is use a zero-crossing detector, and afterwards calculate time elapsed for a 120 hz frequency (2 x 60 hz) so I can later turn on a PWM Nmosfet back-to-back pair, turning it off again at zero-crossing detection... this way I can use a fraction of half a wave to power the resistance, don't know if I'm explaining myself well here...

 

You can reduce it even further on a purely resistive load by half-wave rectification of the mains.
The rectifier has to handle the full peak current even with the heater cold, and withstand the full peak reverse voltage - including any noise spikes that might appear on the mains.

hi ian,
The only problem with that method is that it will create a DC offset current in the mains distribution transformer, the local power authority will not be happy.

I guess he could use two rectifiers to power a split an equal load, one diode for positive half cycle and the other diode for negative half.

E

 

hi ian,
The only problem with that method is that it will create a DC offset current in the mains distribution transformer, the local power authority will not be happy.

I guess he could use two rectifiers to power a split an equal load, one diode for positive half cycle and the other diode for negative half.

E

A few decades ago, practically every TV had a half wave rectifier - then they discovered they could use cheaper dropper resistors by half-wave rectifying the heater chain as well.

The electricity companies whinged even more when SMPSUs came along - drawing narrow blips of high current at the AC peaks.