I agree. It is difficult to get you when you keep contradicting yourself. Saying PID is "total overkill", then saying they are OK to use. Which is it?
-BD
That is not contradictory at all.
Let me make up an analogy for you. Say you need to move a piano. A van would be best suited, but you don't have access to one and no rental near where you live.
Now, you have a buddy that has a trucking company and he'll shift it for you. It is a bit difficult to get the truck up close to your drive way to load up the piano, but once setup loading the piano is no problem since the truck is 10 times bigger than necessary.
The truck works fine. It is overkill, but it is what was at hand.
The drawbacks of PID control here, does not justify investing a lot of time and/or money to overcome, so you might as well just use PID.
I'm not against using a PID here. I'm just saying you don't *have* to use PID control.
Look, I'm getting tired of debating this. And I don't think we really disagree either. And even if you don't agree then fine, let's agree to disagree.
I think hes trying to say a simple thermostat controller like an stc 1000 will work fine as well?
Actually no, I don't think thermostat control is a good solution either. You probably could use it, if you either have a low power element or use it in conjuction with a power controller. That is not an improvement over just using a PID.
Like I said before, if I were to 'roll my own' solution I'd go with fuzzy logic.
Simple and intuitive, and can work really well. You want to take advantage of what you know about the system and process, and try to mitigate the influence over control by change in flow rate and such.
This device alone is pretty powerful in non-oscillating power control but your statement that using it in concert with a PID is not a good idea is unsubstantiated. It will literally control the power of the element. The only downside is that the PID's tune will be affected. No bother though, you figure out what output wattage is most approprate for the flow rate you choose, then auto tune the PID at that power level.
It's a lot tighter control than just letting a PWM run without any temperature feedback especially since it has no protection for flow problem overheating.
Yeah, I know about the Auber power controller, and I agree it is very slick
Yes, the tuning will be affected, but probably it will push it further in the stable region and just make the PID slower to react.
But the thing is, you go to all that trouble and still don't have good control.
You say you want a high powered element to be able to ramp quickly. But you can't allow the PID to just run it, you must choke it when not ramping and you don't trust because it might overheat if flow stops.
Well, I'm sure you're fine with it as is. Personally in this case, I'd definitely build my own controller instead, to really make it do what I wanted. Adding a tempprobe or two to the RIMS tube itself, to prevent overheating. Arduinos are cheap. But of course this is not for everyone.
Seems to me, that you now have two controllers instead, to some degree working 'against' each other rather than in unison. And it seems fiddly. You need to change setpoint and tweak the power output, while also keeping flow rate in check.
There is no way you can tune the PID more restrictively (maybe more P and less I) and skip the power controller?
This is the problem. RIMS is not really good for step mashing. Yeah, before @brundog and everybody else jumps at me again, that is not to say it can't be done, it's just harder get right (as shown here).
Personally, I avoid step mashing, if I have to, I do infusions or dection.