PID Coffee Experiments

Kelsey

I thought we might start a thread for discussing PID related matters. I understand the variables at play (different coffee bean roasts, different coffee machines, PID reporting temps that are perhaps not absolute) make it very difficult to get a true baseline on matching espresso flavours between members, but I'd still be interested in other's reports.

Brett in particular has found some interesting correlations between temperature and his Java Kalistat roast (which I miss terribly) and I'd be interested to see what others say.

I'd also like to find out a little bit more about the functioning of my PID. When I turned it on this morning I had a bit of an odd experience - all zeros showing on the display for a while. I hit the 'up' button and then it started counting up - but given the 'boiling' sounds from within, the numbers on the display didn't match the temp. I hit the 'up' button again, which set the display to 'off' and boiling stopped.

I wound up pulling the old IT trick - turned her off and on again - and all was well with the world. I'd like to find out more about how my Bezzera BZ07 PID is supposed to work.

UNM

that description sounds like gremlins to me. without knowing the internals it is just a guess, but probably a minor timing related coding bug. once the controller got stuck where it shouldn't, a reboot would be the best fix.  Chances are it will not happen again. if it does note down exactly what is the exact  sequence of events and try to duplicate. Displayed PID temperatures are of limited value  for comparing different types of machine and even subject to minor variation between machines of the same model. a lot of manufacturer installed PIDs don't accurately display measured temperatures - they often have a frig factor built in that is intended to display the extraction temperature. How well they do this varies a bit between machines and how they are used. My home installed PID is accurate as I can make it to the actual thermocouple temp, but this is a tad higher than brew temp I think. What I find a PID gives me is precise repeatable control of temperature, no dicking around with timers and rising/falling temp gradients required. Limiting overshoot and fast recovery are areas that might also be worthy of discussion, but I just used auto tune on mine. If someone tells me a bean benefits from a slightly higher or lower extraction temperature, I set the PID accordingly relative to my 'default' setting.

Brett H

The Pid function is really cool and can sometimes make some really interesting changes to flavor.  My close friend swears by his nespresso and when I tasted it I couldn't hide my dislike.  His shot of choice was something roasted really dark with lots of funk and tar flavors.  Enter my machine where I ramped up the temperature and made him a coffee with a blend of Robusti and Indonesians I had roasted very dark.  I called it my "nexwretcho is sh!thouse" blend and made us double flat whites.  I was surprised by the sweetness that was produced by a 1 degree difference with my machine and his blend.  It was actually drinkable and he was loathed to admit, though he did that it left the pod burning money pit on his bench for dead.  What fun.

UNM

one thing I am waiting for is an espresso machine with very low thermal mass and a well tuned PID. Think of the fun you could have temp profiling. big heavy groups are really a very primitive low cost solution to a problem that PIDs overcame a long time ago.

Kelsey

Now that's a truly interesting concept. It'd probably need to be a total power hog tho...

UNM

less power than a traditional machine due to less thermal mass to bring up to temperature. PID tuning might be tricky - I am not an expert on tuning PIDs, but maintaining at idle AND a user specified profile while pulling shots could require different algorithms. just thinking 'out loud' really.