strange start on the latest test… letting it run to see how it evolves
I noticed that I did not update the recommended settings for 0.4.1. I have updated them now.
Brian: uninstall your cooler when using for mashing, so it will not be activated.
HDTV devil: I think you’ll have better results with heater1 Kp set high, for example 50. Because with the proportional gain at 50, the actuator will clip to 100 faster and the integrator will not wind up while it is clipped to 100.
Also please take a look at the new recommended settings. To prevent overshoot you can also increase Td.
If you want to manually set the heater PWM value for boiling, this is supported by the controller.
The web interface just does not have a button for this yet. You can use this hack until it is implemented.
- Set the mode to OFF. This will prevent the PIDs from setting a PWM value
- Go to /sockettest.php
- Send a message with messageType = writeDevice, message = {“i”:, “w”:<0-100>}. You can refresh the device manager with ‘read values’ checked to verify that the PWM value was set.
1 Like
I have a cooler? I thought it was some sort of default behavior. I will have to look into that.
(-B
I have released an update for the script that fixes the control algorithm tab not working properly. That will give you a bit more info on what the controller is doing.
updated to the latest settings and updated to 0.4.2. Starting a testrun.
On larger heat increases, the overshoot is a lot more than on small differences. Here it overshoots 1.2C. Strangely enough, it decides that it still needs to heat although it is above the set temperature.
This is with Heater 1 Kp set to 20. Changed to 50 for the next increase.
Btw, this is just with water.
Both fridge and beer temperature are above the set temperature but still heating:
and full run graph:
Observations: Heater 1 KP = 20 gives a smooth graph but overshoot a lot. Heater 1 Kp = is a more staggered graph but not much overshoot. Will see if I can do a wort test tomorrow.
If your kettle is at 65 degrees, you will have losses to the environment. It will still be heating to compensate that, even if it has reached the setpoint. The integrator does this.
What happens is that during the ramp, your actuator will be maxed by Kp. The integrator will be zero. When it reaches setpoint, the output is not maxed and the integrator is allowed to rise. The overshoot is because the integrator built up too quickly (during the last bit of the ramp). You will have better results if you increase Ti (so it takes longer for the integrator to build up, this reduces the buildup during the last bit of the ramp). And you will have less overshoot if you increase Td (further ‘lookahead’ which reduces output).
Can you show the command you sent?
The message type shoud be writeDevice and the message for example {"i":0, "w":50}
@elco I was copying and pasting from your instructions but I’ve worked out the issue. I reset all to factory defaults and then added just the PWM actuator and it worked! I then added a temp sensor and it continued to work but as soon as I added a second sensor it stopped working.
Penny dropped, The “i” value refers to the device slot and I hadn’t realised this. It works fine now I’m not trying to set the setting against a sensor! 
What settings should I use if I am using immersion heater inside mash tun and therefore do not have two temperature sensors?
1 Like
Try fridge constant mode, with the same settings for heater 1.
Just a note of positivity for anyone thinking about utilising brewpi to control their brewery.
It’s really great!
Even with my shakey wifi, crappy skills with all things computers, and a strange heat exchanged biab system I’ve brewed 3 60L batches the past few weeks with great results. Mash temps hover within 0.5 of set point, boiling works fine, software appears really stable. Again… It’s great.
The future holds cool things for this product.
I just have to work out how to get 2x 60L conicals into a fridge now so I can use it for ferm control!!!
1 Like
Hello, can you share your settings and what your setup is like? Mainly interested in your heat exchanged biab system.
Thank you
Here is my settings for a recirculating BIAB mash. https://community.brewpi.com/t/simple-biab-mash-with-brewpi-spark
The setup is a 55L pot with a homemade grainbag, a 2kw electric heater and a chugger pump.
I have the temp sensor placed in a tee at the pump.
I have still only used it once, it worked pretty ok: Using BrewPi 0.4 for mashing
Have you noticed if temperature of the actual mash was higher than what is measured at the tee? I noticed a couple of degree celsius difference, mash was higher than tee. Just curious if I am the only experiencing this or if it is normal in this type of setup.