PID Burner Control - Lost Duck!

I have read more post than I can count! I need Help! Please!
I need all the help I can get with PID/Logic and Latter to control it.
I have a burner with constant material flow. Material (rock and sand) vails through the flame to dry. My problem is when the PID reaches setpoint my flame shuts down. I need to maintain a flame of atleast 15% to dry the material and keep the output temp up. I am using PC control and a relay to control the valve.

Thank you! in advance

Randy

If I only understood you correctly...

Your PID control activates a relay that, in turn, operates a gas valve? This is not going to work, period.

A PID controller is an analog one: it reads the current value of whatever you are trying to control, compares it to the setpoint value, does some math, based on the difference of the two as well as how long did that difference exist and how rapidly was it changing... And it outputs another analog value - the output. It could be 0 to 32767, -32768 to 32768, 0% to 100%, 4mA to 20 mA, -10 V to 10V - whatever. If it is at 50% of its range, that means that there should be 50 per cent of the gas flow at this particular moment of time.

Now, most PID controls also provide a discrete output, which is width-modulated according to the analog output value. In other words, it pulses on and off more or less rapidly. This is what is usually used when you want to operate an electric heater: a solid-state relay can pulse the heater at different rates thus affecting its temperature.

With the gas valve... well, the system has just decided that it is the time to pulse it a little bit, but the very first time it has to go off for a while - that's exactly what happens: your gas supply goes off. End of story.

You must have a proportional gas valve and an analog output from your PC control. Then, instead of shutting the gas off completely, PID will just turn the valve, thus changing the gas flow gradually . Of course, you will want to keep the existing on/off valve so you could shut the system off in case of emergency.

If I misunderstood you, please disregard this post...

I would have to disagree that PID control must be analog. My processes call for discrete outputs from the PID temperature controller to actuate a burner control motor. The flame is always lit, so temperature is varied from low fire to high fire by movement of the motor.

So, perhaps a new way of controlling your temperature would be in order. MaxPak burners are the one brand that comes to mind.

Good luck.

If this motor drives what, in essence, amounts to a proportional gas valve, then there is no disagreement. Except that, in this case, you need two discrete outputs instead of an analog one. This might be a better, less expensive solution.

What's this?

firealarm said:

I would have to disagree that PID control must be analog. My processes call for discrete outputs from the PID temperature controller to actuate a burner control motor. The flame is always lit, so temperature is varied from low fire to high fire by movement of the motor.

So, perhaps a new way of controlling your temperature would be in order. MaxPak burners are the one brand that comes to mind.

Good luck.

Click to expand...

Firealarm, think what you want, but a PID output is definitely analog. In your application the analog value gets converted to PWM. This must look a lot like a digital on/off to you.

Randy, in your application this relay can't be closed more than 15% of the time and the cycle time must be short to keep that flame from going out. You should consider addding 15% to the PID output and scale the output so it's range is only 85%. It is also possible that the valve response is not fast enough to switch as fast as you wish.

Randy,

Does the flame safeguard have an interrupted pilot? In other words
once you have a pilot and the main gas is on- does the pilot interrupt and the main gas valve keep the flame established?

Yes JRW

Then you answered your problem. I am with Ladderlogic, no gas- no flame.

No I did not. The burner will run low flame.

What is your definition of "low flame"
What is the difference between "low flame" and zero output on your relay?

1 to 2 %

Randy,

I used to design textile ovens (gas fired) and we used UV scanners and flame rods. Of course I dont know your system but that doesnt sound like much to me, if you had a flame strength meter (or do you?)
What does it do? It just goes in series with your flame rod (scanner)

Anyway- Good Luck

Maybe I'm missing something, but when I read...

randyintejas said:

when the PID reaches setpoint my flame shuts down. I need to maintain a flame of atleast 15%

Click to expand...

...I hear a system that has all Proportional and no Integral control active.

When Error = 0 (i.e., "reaches setpoint") the Proportional part of the equation is zero ("flame shuts down"). Only the Integral part, with it's memory of error over time, would be able to keep the system at 15% (assuming proper tuning) when Error = 0.

I know he says he uses a relay, but with the talk of 15% minimums, and 1-2% "low flame", there's got to be an analog component to his system somewhere.

Fix da problem!

In addition to fixing your system and giving it the proper parts to accomplish the PID task, you should read Peter's PID presentation. It is very good and many of us can learn a lot from it. Good show Peter!

Did I read correctly? Are you cycling the gas valve? If you are using a firing rate control motor, what type input does it require? (4-20, 0-135 0hm, voltage with or without spring return.)