SLY 2362 or EZBoil?

[Dert]

Thoughts?

I just built 2 control boxes with one each and ended up getting 2 more ezboils to replace the Watlows I initially installed as the the 2nd PID in each box...

The ezboils are super ez to program. But haven’t tested either with water yet.

View attachment 601107

 

[doug293cz]

I'd always pick an EZBoil over a PID. All reports I have seen say that EZBoils hold temps extremely well, with little to no over/under shoot. I have seen lots of reports of folks having trouble getting PID's tuned to work as well as an EZBoil. I think the control algorithm used by the EZBoils works better for brewing applications than the PID algorithm. Plus, I don't think there is a PID with "auto power turn down" on approach to boil, which the EZBoils have (greatly reduces chance of boil over at hot break.)

Brew on

 

[golfindia]

I can (and have) put almost any combination of PID parameters into my software based controllers and they keep/hold temps just fine. I can't wrap my head around how one device could have a better "algorithm" than any other. PID is PID. There is no magical set of parameters needed to control the temp of a large mass of circulating liquid. People autotune 100x, get a 100 different sets of values and get wrapped around the axle. It really doesnt matter that much. Getting a PID to run a smoker is a PITA though.

 

[augiedoggy]

I'd always pick an EZBoil over a PID. All reports I have seen say that EZBoils hold temps extremely well, with little to no over/under shoot. I have seen lots of reports of folks having trouble getting PID's tuned to work as well as an EZBoil. I think the control algorithm used by the EZBoils works better for brewing applications than the PID algorithm. Plus, I don't think there is a PID with "auto power turn down" on approach to boil, which the EZBoils have (greatly reduces chance of boil over at hot break.)

Brew on

I agree, the solution to boil overs with a pid for me was to set the pid for 208 and then when it reaches that it stops before possible boilover. At this point I would switch to manual boil mode. Honestly the slow down feature was not well loved in my ezboil setup since it slowed down the heating process by what seemed like quite a bit at the time.

Hands down though the ezboil worked better for temp control.

 

[augiedoggy]

I can (and have) put almost any combination of PID parameters into my software based controllers and they keep/hold temps just fine. I can't wrap my head around how one device could have a better "algorithm" than any other. PID is PID. There is no magical set of parameters needed to control the temp of a large mass of circulating liquid. People autotune 100x, get a 100 different sets of values and get wrapped around the axle. It really doesnt matter that much. Getting a PID to run a smoker is a PITA though.

once you try both as I have you would see the difference. I'm not saying its enough to have any real effect on the beer but the ezboil reaches and holds precise temps way better fo different varying amounts of liquid in different circumstances with no tuning needed.

I have graphs with brucontrol which show the performance of my rims in real time and theres a large difference in how well a pid performs as far as constant over or undershoot depending on the tuning parameters..

 

[golfindia]

I've never graphed mine, because the PID parameters don't make a difference (within reason). Never had overshoot or undershoot, large volume, small volume....

I'd be interested to know what the magical PID settings are in the EzBoil. Because those, the little knob and some memories are the only difference between it and a $7 inkbird....

 

[augiedoggy]

Anyway... again as someone who currently owns and has used both I stand by my comments above.

 

[doug293cz]

I've never graphed mine, because the PID parameters don't make a difference (within reason). Never had overshoot or undershoot, large volume, small volume....

I'd be interested to know what the magical PID settings are in the EzBoil. Because those, the little knob and some memories are the only difference between it and a $7 inkbird....

According to the documentation, EZBoils are not PID's (i.e. they don't control based on a PID algorithm.) The algorithm is proprietary (they don't disclose it), but in any case there are no P, I, and D parameters to compare. As @augiedoggy notes, the differences in control performance are observable.

Brew on

 

[golfindia]

LOL. They are PIDs. They use the same feedback sample loop logic that every other PID uses. All Auber did is take away the ability of the end user to infinitely adjust p, i, and d. Because as I have been saying, these aren't that critical for keeping a kettle of liquid at 150 deg. They let you monkey with overshoot and undershoot, which is just simplification of the "D" part of a regular PID. Not rocket science. There's no way it can be better or worse at feedback loop sampling than $7 PID, whether you "believe" it is or not..... but hey, not my $50....

Edit, D, not I.

I also revied a 2362 manual ( i have one) it has a damping setting that looks like the same thing as the ezboil over/undershoot adjustment with a different name

 

[doug293cz]

LOL. They are PIDs. They use the same feedback sample loop logic that every other PID uses. All Auber did is take away the ability of the end user to infinitely adjust p, i, and d. Because as I have been saying, these aren't that critical for keeping a kettle of liquid at 150 deg. They let you monkey with overshoot and undershoot, which is just simplification of the "D" part of a regular PID. Not rocket science. There's no way it can be better or worse at feedback loop sampling than $7 PID, whether you "believe" it is or not..... but hey, not my $50....

Edit, D, not I.

I also revied a 2362 manual ( i have one) it has a damping setting that looks like the same thing as the ezboil over/undershoot adjustment with a different name

Just because you have closed loop feedback control, that doesn't mean the control algorithm is PID. PID is an algorithm that determines how the control output is calculated from the measured parameter input. There are lots of potential control algorithms, and PID just happens to be the most common. Without Auber disclosing the EZBoil control algorithm, or reverse engineering it, you cannot know what the algorithm is.

From Auber: "The EZboil® can also be used as a temperature controller for mashing. In mash mode, its temperature control function utilizes an artificial intelligence algorithm optimized for mashing application. It is easier to understand and more convenient to operate than a PID controller."

The PID algorithm is definitely NOT an artificial intelligence algorithm, as it was originally implemented with analog circuitry. There is a possibility that Auber's statement is just marketing BS, but I personally doubt that.

Brew on

 

[golfindia]

Alright. Artificial Intelligence it is then :^)

 

[brumateur1]

Most likely they are talking about fuzzy logic algorithm.
Some basics about it - https://www.controleng.com/articles/temperature-control-pid-vs-fuzzy-logic/

 

[doug293cz]

Most likely they are talking about fuzzy logic algorithm.
Some basics about it - https://www.controleng.com/articles/temperature-control-pid-vs-fuzzy-logic/

Could be, but not enough information is available to tell. Auber states that their PID controllers employ fuzzy logic, so what they say about the EZBoils implies something else, or in addition to fuzzy logic.

Brew on

 

[Gameface]

Maybe the EZ Boil uses PEDs, thus the unfair advantage.

 

[golfindia]

The SYL2362 says it's "arificial intelligence" technology too. Fuzzy logic isn't technically "PID", if we're going down the A.I. vortex.

Let's just call it magic. Hopefully the EZboil won't refuse to open your pod bay doors.

 

[Gameface]

 

[doug293cz]

The SYL2362 says it's "arificial intelligence" technology too. Fuzzy logic isn't technically "PID", if we're going down the A.I. vortex.

...

Hadn't noticed that on the 2362. It may use a similar algorithm as the EZBoil. Won't really know unless Auber wants to tell us.

Brew on

 

[4of7]

I like the ezboil ..it works for me.

 

[etk29321]

AI is marketing speak for a feedback loop in most cases. ‘Training datasets’ used in big data basically just set the algorithms equivalent of the p,I,d params

 

[augiedoggy]

Yea well the point is theres a large difference in how well an ezboil mainatins temps out of the box and it does not need tuning period. all the pids I have used have never worked as well to hold a temp.. again I dont think its enough of a difference to actually matter for brewing but it might help for rims or herms were a few degrees does matter.

 

[golfindia]

And lest the horse not be dead.... these are the "AI" settings I use. Pulled directly out of my tookus, based on what I guessed they should be for a giant kettle of liquid. They work great.