Mash Temp Control - the devils in the details

[ABrother]

Ok, I'm new to this forum but not to brewing. I've been extract and crude all grain brewing for 15 or 20 years. I've finally made the jump to designing and building a brew stand. I've fallen in love with the HERMS process but want to use propane heat. I guess that makes it a p-HERMS. The plan is a 3 keg system all on one level using two pumps and burners under the HLT and BK.

Anyway, I thought I had a good understanding or idea of how to monitor control and adjust mash temperature. I was planning to use a PID to monitor HLT temp measured in the HLT and turn on and off the burner. The temp, guessing now, would be set to 180 F and maintained. A second PID would monitor mash temp in the MLT and with a re circulation pump running continuously open and close solenoid valves to divert wort through the HEX or a bypass loop depending on if heat is needed or not. I thought the denaturing of enzymes due to the high temp HLT would not be significant enough to worry about but I'd get much faster response in changing temps for step mashes and mash out.

Now I have seen several alternatives like turning the pump on and off, or using only a single PID monitoring mash temp but controlling the HLT burner. Many of the posts that I've read just don't give quite enough detail for me to feel like I fully understand. I'm curious about the details of how some of you experienced HERMS brewers are doing it. This is your chance to brag about your system and all it's pros and cons. Details on how and where you monitor and adjust temp as well as why you choose that method would be great.

Thanks

Thanks

 

[yogensha]

Mine's based on the system at theelectricbrewery.com. The HLT is controlled to hold at your target mash temperature and the wort is continuously circulated through the HEX. No worries about overheating any part of the mash, and no cycling of pumps. I don't think the heat source really matters, as long as it can be controlled to maintain the HLT at mash temperature.

 

[LordUlrich]

Maintaining the hlt at your mash temperature eliminates several parts, which is an advantage as there is less to go wrong. The RIMS brewers don't have huge issues with denatured enzymes, I would not be worried especially at 180. That is not a instant reaction anyway.

 

[ABrother]

Thanks for the info, I have a few followup questions.

I definitely like the simpler approach of maintaining an HLT temp at my target mash temp. Sounds like a single PID setup.

The two things that come to mind are how hard is it to step mash or mash out? That is, does the HEX stabilize the wort at the HLT temp fairly quickly and are you able to change HLT temps with the volume needed from protein rest to sacrification, to mash out in a reasonable amount of time? I'm expecting to prepare about 12 gallons in the HLT.

Also does the HLT have to be set a few degrees hotter then the target mash temp to allow for heat loss in the plumbing coming from the HEX back to the MLT? I assume your at least monitoring the mash temp as well as the HLT even if mash temp isn't part of the control system.

 

[MikeRussell]

In my setup I have the HLT and the tun each at their own setpoint. When the tun needs heating it just turns on its dedicated pump to recirculate through the HLT. If the tun is at the correct temp, nothing happens, it just sits. The closer you set these two the more recirculation you get.

If you have two setpoints you will be able to make your steps happen faster but that's about the only advantage all I can think of right now. I suppose with two setpoints you can get your HLT ready for sparging while the tun is mashing out. I think that would happen about the same with a single setpoint though.

In my system tests, I noticed that the tun lags behind the HLT a fair amount. I don't have exact figures but I noticed it took a few (around 10?) minutes after the HLT hit its setpoint. A couple variables to consider for that are tun insulation/heat efficiency, batch/grist size, length of HEX coil and pump throughput. I'm using very small pumps (12v solar) to pump wort from the tun so that could be a big part of the problem in my setup.

All in all, you don't really need to set the HLT higher than you want the mash unless you want the tun to get to its setpoint faster. In my system there's pretty minor heat loss to the air.

Sounds like you're on the right track!

 

[yogensha]

The two things that come to mind are how hard is it to step mash or mash out? That is, does the HEX stabilize the wort at the HLT temp fairly quickly and are you able to change HLT temps with the volume needed from protein rest to sacrification, to mash out in a reasonable amount of time? I'm expecting to prepare about 12 gallons in the HLT.

The time it takes varies with mash size, the volume of water in the HLT and your heating element. On my system, going from saccharification to mashout (~67C to 75C) takes no more that 15 minutes on a 10 gallon batch. I don't add any water to increase the temperature, just raise the set point on the HLT PID and let the mash rise with it.

Also does the HLT have to be set a few degrees hotter then the target mash temp to allow for heat loss in the plumbing coming from the HEX back to the MLT?

I don't have to set the HLT hotter, but some people have to. You just have to see how your system behaves.

I assume your at least monitoring the mash temp as well as the HLT even if mash temp isn't part of the control system.

Absolutely!

 

[ABrother]

Would there be any advantage to a single PID controlling HLT burner heat based a sensor monitoring mash temp? In this scenario the wort would continuously circulate. I know the initial runs would probably overshoot badly but with an auto tune PID it would learn the system and correct right? Of course every batch is slightly different; volume, grain bill, ambient temps, even propane tank pressures; so maybe the PID wouldn't be able to learn the system fast enough. What's everyone's thoughts on this scenario? I'm sure it's been tried but I've not seen anyone advocating it so it makes me suspicious.

 

[LordUlrich]

Would there be any advantage to a single PID controlling HLT burner heat based a sensor monitoring mash temp? In this scenario the wort would continuously circulate. I know the initial runs would probably overshoot badly but with an auto tune PID it would learn the system and correct right?

Autotune is not self learning, autotune allows you to run an autotune cycle to get PID turnings, they do not change over time unless you change them. My experience with autotune has been bad, it gave me way too much integral.

 

[ABrother]

Autotune is not self learning, autotune allows you to run an autotune cycle to get PID turnings, they do not change over time unless you change them. My experience with autotune has been bad, it gave me way too much integral.

Ok, I obviously don't understand the details of how the PID does its job. I had heard that some PID's had better auto tuning then others but have no idea how to know which is which or if I even need or want auto tune. Do you have any suggestions where to get a better understanding of how the PID does its job? I'm sure I heard a PID learns the system somehow by looking at the past, present and future to improve its response. Is that somewhat accurate? Would the PID monitoring mash temp but controlling HLT heat eventually be able to maintain the set point or is there just two much lag?

Thanks everyone for all the info and help.

 

[yogensha]

Would there be any advantage to a single PID controlling HLT burner heat based a sensor monitoring mash temp? In this scenario the wort would continuously circulate. I know the initial runs would probably overshoot badly but with an auto tune PID it would learn the system and correct right?

I suspect you are correct and that it would introduce lag in the system response and would tend to overshoot too much.

Ok, I obviously don't understand the details of how the PID does its job. I had heard that some PID's had better auto tuning then others but have no idea how to know which is which or if I even need or want auto tune. Do you have any suggestions where to get a better understanding of how the PID does its job?

http://en.wikipedia.org/wiki/PID_controller

The auto-tune function built into some controllers applies process input, measures the system response and attempts to compute reasonable coefficients for each term. It doesn't adjust anything after the tuning is complete.

In your example, during the auto-tune the controller would likely excessively overshoot the target temperature due to the increased system response lag and compensate by selecing coefficients that would result in a slower approach to the target temperature. On brew day, your heating times would be longer.

FWIW, the design at theelectricbrewery.com is extremely popular, I suspect because it works quite well.

 

[ABrother]

FWIW, the design at theelectricbrewery.com is extremely popular, I suspect because it works quite well.

I have been looking at theelectricbrewery design but there's a lot there so I'm still wading through it. So far it seems that all the mash temp PID is used for is to monitor and alarm. Not sure a PID is needed there in mine if all I do is monitor and control the HLT. Also the BK PID only seems to be run on manual to adjust the heating element duty cycle. For mine I planned to just manually adjust gas flow on the BK to maintain the boil. I'd like to optimize the BK to apply just enough heat to maintain the boil I'm just not sure how yet.

The reason I'd like to monitor mash temp is I'd like to use a ramp soak PID so I could input all the steps and times and it could run the entire mash profile from grain in to mash out without any input from me (i.e. more sitting and drinking home brew time) I'll do some more research and see if a PID on the mash could communicate with the HLT PID to tell it when to make changes. I'm afraid I'm getting into PLC territory and I'm pretty sure I don't want to do that. At least not yet.

 

[yogensha]

I have been looking at theelectricbrewery design but there's a lot there so I'm still wading through it. So far it seems that all the mash temp PID is used for is to monitor and alarm. Not sure a PID is needed there in mine if all I do is monitor and control the HLT.

Correct. In fact, I just stuck a cheap thermometer on the MLT output to verify that it's where I want it instead of buying a PID, thermocouple and requisite cabling.

Also the BK PID only seems to be run on manual to adjust the heating element duty cycle. For mine I planned to just manually adjust gas flow on the BK to maintain the boil. I'd like to optimize the BK to apply just enough heat to maintain the boil I'm just not sure how yet.

Are you talking about automatically achieving and maintaining boil? Not sure anybody has tried that, but it would be pretty cool.

The reason I'd like to monitor mash temp is I'd like to use a ramp soak PID so I could input all the steps and times and it could run the entire mash profile from grain in to mash out without any input from me (i.e. more sitting and drinking home brew time) I'll do some more research and see if a PID on the mash could communicate with the HLT PID to tell it when to make changes. I'm afraid I'm getting into PLC territory and I'm pretty sure I don't want to do that. At least not yet.

I think you're right about that, but there are lots of options these days that make this level of automation a lot easier. I haven't caught the automation bug myself, but I enjoy reading this forum and seeing what folks are up to.