Rapid on / off when approaching set point

[gbrewer]

I completed my first trial run using simple water. I am running temperature probe #1 associated with my propane gas valve. The temperature was set at 150 degrees during my test. As the temperature approached 150 degrees, the valve shut off and then clicked on and off several times during the course of a few seconds. Each time it clicked on, it wasn't for more that a half second or so. What is setting do I need to adjust so it doesn't enter this "rapid" on/off issue?

[np0x]

Under the "System Settings" menu you can adjust the "Output Update Freq (PID Sample Freq)" this is defaulted to 1 second, adjusting this value will change the frequency with which the pid will attempt to switch the circuit. I would be interested to hear what your experience is, I have an electric system, so this never was an issue for me, BUT I will also adjust mine next brew so I can see what the actual behavior is like. What the right value is for you is hard to say, I would be inclined to start at 10 seconds and see what an order of magnitude does for the situation.

[ECC]

I agree, experiment until you find a value that works for your system. I have a feeling that it'll be in the 2-5 second range.

[gbrewer]

Thank you. I will try this and let you know what works best. I probably would never notice if I had an electric element. It is really obvious when your gas valve runs a jet burner. I will start with 10 seconds on this.

[gbrewer]

Setting the output freq to 10 seconds worked well. During my batch, I was able to hold a steeping temp of 150 degrees +_ .8 degrees for 30 minutes on an extract batch. I was impressed at how well it worked. I was told by an electrical contractor for many years that the fluctuation may actually have more to do with the electrical circuit itself than the temperature of the water. He stated I should put an inexpensive UPS inline and I would see the temperature probe have a more stable reading.

[gbrewer]

I am trying to refine how often my burner fires on and off. Since V2, it seems to be doing the rapid on and off issue again. I really don't understand all the new options.

Example -

PID Sample Rate, Output period, Min Pulse %, Max Pulse Percent. I read the WIKI but don't fully understand this.

The gas is doing a great job keeping the temperature close but I don't like hearing it click on and off so quickly. There are numerous times when it open and closes the valve so quick that gas doesn't even flow. (A fraction of a second).

I want to make this so the gas will not turn on unless it will be on for at least two seconds. That is my goal . It seems to take 1-2 seconds to get the burner completely lit. If it won't stay on for at least 2 seconds, I don't want it to turn on.

Any ideas?

[ECC]

Yeah, I should probably document that a little better. As soon as v3.0 is released (hopefully soon), I'm going to spend some time updating the wiki.

First up, PID Sample Period. This is the time between updates to the PID algorithm. Its based on the time constant of the system. Its best to use an example to explain. If the PID Period is set to 10sec, the PID will read the temperature once every 10sec, and use that to calculate an output percent. Adjusting the PID period faster will allow the BCS to react faster to temperature changes. Setting it slower will keep the changes to a minimum for systems that don't react quickly to heat. For the sake of argument, lets say that the PID spits out 25%. It will hold the 25% until the next update 10sec later, at which time it spits out a different %.

Which brings us to Output Period. This describes how the PID percentage is translated to the actual on/off times. If the output period is set at 4sec, the BCS will use the output percentage of 25% to turn the output 1second ON and 3 seconds OFF. It will continue to run at this rate until it gets a new % from the PID.. In previous firmware revisions, Output Period and PID were tied together, such that the PID Sample Period directly set the pulsewidth of the outputs. However, many people requested that they be separated because there were systems that wanted the PID Period to be slow, but the actual pulse rate to be fast. For instance, think of an 20gal+ electric brew system, which would be quite slugglish, which perhaps had a PID period set at 20sec... But when the PID commands 10%, it might be better to have 0.5sec ON, 4.5sec OFF (out per=5) instead of 2sec ON 18sec OFF (out period=20sec=PID period).

So to answer your question, the Min/Max Pulse percent directly limit the PID output percent.. Min Pulse percent is exactly the setting that you need to adjust to get the behavior that you describe. It will set a minimum threshold to turn the output on. If its below that threshold, it won't turn on. If you want a minimum of 2sec, set the min pulse to your 2sec/Output Period. If you have an output period of 10sec, set the min pulse % to 20%. Because 20% of the 10sec period would be a minimum of 2sec.

It gets a little confusing, and I apologize for all of the vaguely documented settings. I want to make sure that every user can set the BCS to behave exactly how they want it, since every system is completely different.

[gbrewer]

Thank you for the explanation. I figured it had something to do with the percentages but couldn't seem to get that setting right. The more I played with it, the quicker it shut on and off. It appears that I didnt have the output period set high enough.

I will see if I can come up with some settings that work and write something up for other gas users.

It did work as is and worked well in maintaining the temp. I just want to fine tune.

[harpo]

I just got my BCS-462 set up in my control box and was attempting to get used to programming. My setup has the HLT and Boil Kettle under BCS control, both using low pressure propane to banjo burners through the honeywell vr8200 (IIRC) gas valves. Manual control has everything working. I tried to use the PID control to do a process I called "strike water". I left everything in the default state for the PID periods, min/max pulse etc. When it got to the temperature set point, it would cycle the gas on and off. Just enough to light the burner and then shut off.

So I re-read this thread.

I tried adjusting my sample period to 20, and my output period to 30 and my min pulse to 50. I really used no thought process for my decisions, but since I knew I wouldn't really understand how those three settings interact with each other until they are in motion, I just went for it. I just think the result was that the burner cycled on and off rapidly for about 3 or 4 cycles, then stayed on for what seemed to be 20 seconds, then shut off. Then it repeated.

I would like to slow this thing down a bit. let it cycle a little as it initially reaches the set point, then cycle about 20-30 seconds at a time to keep setpoint. Does this seem to be reasonable?

I guess I need even more simple examples to see whats going on. Maybe more related to my specific case? I just don't want rapid fire cycling for short bursts. gbrewer, did you figure anything out for gas?

Thanks.

Nathaniel.

[gbrewer]

The last setting I used before playing with the RIMS Tube was output period 10 with a min pulse % of 30. It seemed to work better than before.