PID Tuning - HELP!

[Walker]

The only anomaly I've seen is that my current reading of the BK(MT) is a degree higher than the output of the hex, which is odd!

Did you calibrate all of the sensors and enter any temp offsets into the PIDs?

Your might just have one sensor that is reading a little off.

My first probe had a static error of 3.5*F. The new one has a static error of 1.5*F. If I were to use those with out accounting for the errors, I would see temp differences 2.0*F between the two probes.

 

[onelegout]

I'm using one sensor attached to the hex output on the PID and sensing the other temps using a calibrated digital thermometer.
I'm running a new test with the set point at 45*c so that I can test the PID's ability to hold a temperature and complete an autotune session. We'll see how that goes!

I've set the correction to make the sensor correct using an ice bath and boiling water to get it right and checking it at different temperatures with my digital thermometer, but I think my homemade thermowell in the HEX output takes a while to adjust to the correct temperature, so part of the reason I'm testing the brewery at the mo is to see what the correction should be. Right now it looks like the sensor inside the thermowell is around 3*c colder than the liquid in the hex.

 

[Walker]

Ah.... I didn't realize you were using a thermowell.

I put a DIY thermowell in my MLT (for a small probe thermometer) and that thing always reads about 1*F lower than the actual temp inside the MLT (as measured with the same thermometer), so that might have something to do with it.

 

[onelegout]

yeah I think mine takes a minute or two to catch up; I'll calibrate it as soon as I can get hold of a new thermometer - mine decided to die today half way through the test!
The PID seemed to work ok but I think I may have to tune it myself manually. I've been looking into PID tuning and have a basic idea on how to do it. The main focus for now is making sure that the PID reads the correct temperature, as I still have not got confidence in the accuracy of the sensor.

Tomorrow my goal is to calibrate the sensor with the thermowell.

 

[onelegout]

Ok a quick update on this

I've been very suspicious about the PID's I had. They were bought from ebay and were from Italy. The manuals I had for them were all in Italian, and when they sent me an English one I realised it was for a slightly different model. I finally got one of the PIDs working properly with a sensor but when I ran a brew test on it with water, it didn't track the temperature properly, and kept getting stuck at certain temperatures for ages.

I eventually got really annoyed and posted a rant on facebook about how I was fed up with temperature control systems, to which a friend said 'you know I work for a PID company right?'

So I'm now the proud owner of two new PID controllers They're West Control Systems '6100+' and '6700+' models, and will hopefully do a much better job than the italian ones did

I'm going to set them up and try them out in a bit, will have some inital results by this evening.
H

 

[BrewBeemer]

So I'm now the proud owner of two new PID controllers They're West Control Systems '6100+' and '6700+' models, H

This is what I come up with using a different address for that PID if this helps;
http://iseinc.com/process_and_temperature_controls.htm

 

[onelegout]

I finally sorted the new PID controller - had to use a battery pack circuit to use the relay output of the PID to switch an SSR (the PID had no SSR driver).

My first test of the new PID was odd - the temp kept rising and the output wouldn't turn off - then I realised something! There are two modes, 'direct' and 'reverse' acting. I had assumed that the 'direct' mode was for heating and the 'reverse' mode was for cooling - its the other way around!

 

[Walker]

I finally sorted the new PID controller - had to use a battery pack circuit to use the relay output of the PID to switch an SSR (the PID had no SSR driver).

Ouch. Mechanical relays will wear out over time. If you ever have to replace it, look around for one that can drive an SSR directly.

My first test of the new PID was odd - the temp kept rising and the output wouldn't turn off - then I realised something! There are two modes, 'direct' and 'reverse' acting. I had assumed that the 'direct' mode was for heating and the 'reverse' mode was for cooling - its the other way around!

Hmmmm... If the PID was set to act as a "cooling" device, wouldn't that mean that the heat would not have even been turned on at the start of te test (when the measured temp was already lower than your set temp)?

 

[onelegout]

I used the manual mode to ramp the temperature - and it appears that I now have it working!!!

The temp had risen to 75*c, I set the mode to 'reverse acting', and the output turned off. It dropped to 65.9, the output turned on. It rose to 66.1, the output turned off!

It's held 66*c within 0.5 degrees now for 20 minutes, and occilates by 0.2*c every minute or so. Fingers crossed it will stay like it

Oh, and I'm trying to get my hands on an ssr driver board for my PID, as the company does make them and they're a simple switch.

 

[onelegout]

SUCCESS! We finally have fully functional PID control!

it's been within 0.2 degrees of 66*c for half an hour, apart from once when I had to turn off the pump for 30 seconds to fix a leak, when it dropped to 59.5*c

The problem I have with measuring it now is that it oscillates quicker than I have been measuring it. I'm now going to sit with it for 15 mins and record the temp every 30 seconds or so.

 

[cdbrown]

Reverse mode is the one you want to let the PID do the work. When doing autotune it's best to be at the temp you normally want before running it so let the system run for a while and then hit the autotune setting. Run the autotune for a couple of batches and then let it do it's job. It's great to be able to walk away and come back an hour later to see it ramping up to mash out temps.

 

[onelegout]

I'm not sure auotuning further is even necessary as I measured the oscillations and it doesn't deviate by more than 0.3*c, which I thing is more than good enough for brewing I'm going to clean the whole thing out and get brewing on it asap will be great to see how it does with a proper mash nit just water