Heat compensation circuit design

Thread Starter

kvsingh21

Joined Apr 15, 2008
63
Hi all, i am trying to simulate the circuit below using multisim but its is proving to be a bit tedious(This picture is from a reliable journal). As can be seen, the omp amp is connected directly to 2 voltages and there is no feedback?? surely there has to be feedback for the circuit to function.
Additionally even if i insert a resistor(between op amp -ve terminal and output) to provide feedback and make the circuit functional, the current through poly_sensor Resistor is not constant anymore.
So basically I am trying to get this circuit to function whilst keeping current through poly_sensor Resistor constant. Any suggestions??
 

t_n_k

Joined Mar 6, 2009
5,455
There is negative feedback - the op amp compares the control voltage with the voltage developed across Rpoly_T_sensor. The op amp drives the control FET such that the voltage across Rpoly_T_sensor tends to equal Vcontrol - by virtue of the associated degree of heating produced (as the heating current is adjusted) in Rpoly_heater.

The current source is there to ensure the voltage drop across Rpoly_T_sensor is primarily (if not solely) related to its temp coefficient of resistance.
 

Thread Starter

kvsingh21

Joined Apr 15, 2008
63
There is negative feedback - the op amp compares the control voltage with the voltage developed across Rpoly_T_sensor. The op amp drives the control FET such that the voltage across Rpoly_T_sensor tends to equal Vcontrol - by virtue of the associated degree of heating produced (as the heating current is adjusted) in Rpoly_heater.

The current source is there to ensure the voltage drop across Rpoly_T_sensor is primarily (if not solely) related to its temp coefficient of resistance.
Thanks for your reply mate.
When the Vcontrol and voltage across Rpoly_T_sensor are similar, the output of op amp is just the supply voltage of op amp, which means the FET is always allowing maximum current through it due to a fixed high gate voltage. Am i missing something obvious??
 
Last edited:

Thread Starter

kvsingh21

Joined Apr 15, 2008
63
I have deduced that the op amp basically switches the FET on or off, depending on voltage across Rpoly_T_sensor. So as the temperature increases, the voltage across the resistor increases to switch of the FET off. When the resistance falls back down, the FET switches on again. So basically pulsing the FET.
Is that correct??
 

t_n_k

Joined Mar 6, 2009
5,455
Sounds about right.

This is a simple on-off or "bang-bang" type of control - nothing particularly fancy but nevertheless effective. The op amp is functioning as a voltage comparator.

One of the features of this type of control is a tendency to cycle continually between on and off states - rather than having some degree of "dead-band" where there is a small temperature band in which nothing will happen - either with a falling or rising temperature. Sometimes one might add hysteresis to the comparator to accomplish this. This would be done only if the continual cycling was undesirable for some reason. One such reason might be the tendency for the FET to begin operating in a quasi-linear mode in which it's losses (self heating) became problematic. Perhaps that's the reason for the inclusion of Rdeg in the FET source. I'm not sure.
 
Top