Where to take temp reading in RIMS setup

[Bobby_M]

Definitely nothing wrong going with the flow sensor.

 

[Shred]

In case it isn't totally obvious, I'm no electronic engineer. I understand the concepts being discussed and the practical application, but when it comes to the actually design and wiring, I'll probably be doing a lot of reading.

I have a couple questions if you more knowledgeable/experienced folks don't mind helping out...

I like the idea of a second PID as a sort of "fail safe" at the hottest point (for me based on my whole 1 run, that's the return into the mash tun AFTER the outlet on the RIMS, for whatever reason).

With that in mind, what type of relay should I put between the PIDs and would it be the same relay used to tell the element to turn off if the pump is off? I guess I'm basically asking for a simple parts list.

 

[Shred]

I should probably mention, I'm running at 120V (1650W element), 15A with an external GFCI mounted to my brew stand.

 

[brewman !]

In case it isn't totally obvious, I'm no electronic engineer. I understand the concepts being discussed and the practical application, but when it comes to the actually design and wiring, I'll probably be doing a lot of reading.

I have a couple questions if you more knowledgeable/experienced folks don't mind helping out...

I like the idea of a second PID as a sort of "fail safe" at the hottest point (for me based on my whole 1 run, that's the return into the mash tun AFTER the outlet on the RIMS, for whatever reason).

With that in mind, what type of relay should I put between the PIDs and would it be the same relay used to tell the element to turn off if the pump is off? I guess I'm basically asking for a simple parts list.

I'm not building one of these, but I'll help you if nobody else does.

You don't need a relay. You need 2 PID controllers, one with at least a relay output and one with the fancy SSR PWM output.

Parts list

- 2 PID Controllers, with temp sensors
- 1 SSR
- 1 heating element.
- misc hardware, wire, crimp on ends, etc.

My PID controllers are JLD612s, which are an Auber knock off. Not sure if the terminal numbering is the same as the Aubers or not.

 

[Shred]

I'm not building one of these, but I'll help you if nobody else does.

You don't need a relay. You need 2 PID controllers, one with at least a relay output and one with the fancy SSR PWM output.

Parts list

- 2 PID Controllers, with temp sensors
- 1 SSR
- 1 heating element.
- misc hardware, wire, crimp on ends, etc.

My PID controllers are JLD612s, which are an Auburn knock off. Not sure if the terminal numbering is the same as the Auburns or not.

I really appreciate it. I've got the RIMs tube with element already, stand built, 2 pumps and plenty of tubing... My current control unit is STC-1000 based, which is obviously not appropriate.

Can you clarify for me what SSR is? I've had a hard time finding a literal definition and a lot of the PID controllers I've looked at seem to incorporate SSR somehow.

 

[brewman !]

I dont see what the problem is here?
I have my temp probe for my rims pid at the exit end of my rims tube and it keeps consistent mash temps all through my mash time within 2 degrees variance at the most. Once it has had about 10 minutes to stabilize the temps are consistent throughout the whole mash...(unless I ramp them up)
From this experience I would say something else is amiss if this setup isnt working for someone and that person should correct the real problem and not move the temp probe around in hopes that it will compensate for whatever is really wrong....

Whats the flow rate through the rims tube? I find that 1.5 gallons -2 gallons a minute works great for me.

If it aint broke dont fix it...

All things equal, the rate of temp change is way higher if you heat the returning mash liquid higher than the set point for the MLT. Lets say you are doing a step from 144F to 154F, ie the new setpoint of the MLT sensor = 154F. If you set the RIM chamber output setpoint to be 156F it will take a lot longer to reach than if you set the RIM chamber output setpoint to be 165 or 170F.

All things being equal, the heat transfer from the RIMS element to the MLT is proportional to temp difference between the mash liquid and the MLT. As the MLT approaches the mash liquid temp, the rate goes to zero, in fact.

By keeping the RIM chamber setpoint quite a bit higher than the MLT setpoint, mash steps will be much, much faster. Its also way easier to control the MLT temp when the heat input is faster.

The reason this matters is when you are doing mash steps you don't want the temp to dwell too long at the lower temps or you'll get a thin, highly fermentable wort. You want the steps to be fast and crisp, especially if you are trying to generate a maltier, complex wort.

 

[brewman !]

I really appreciate it. I've got the RIMs tube with element already, stand built, 2 pumps and plenty of tubing... My current control unit is STC-1000 based, which is obviously not appropriate.

Can you clarify for me what SSR is? I've had a hard time finding a literal definition and a lot of the PID controllers I've looked at seem to incorporate SSR somehow.

SSR = solid state relay. You'll need one that handles at least 15 amps. And probably a heat sink for it.

I'm happy to help, but it isn't optimal doing this over the Internet if you don't have a bit of experience. Do you have any electrical nerd friends to look over your shoulder ?

Here is the manual for a JLD612.
http://mythopoeic.org/misc-files/JLD612_Manual.pdf

 

[Shred]

I'm pretty handy, but this is a touch past my previous experience. I have a friend who will be helping as well. He's a "handyman-type".... I know this sounds like an accident waiting to happen, but we're pretty thorough and don't leave anything to question.

The PID should have the SSR built in or is this a component? As far as I can tell from the PIDs I've been shopping, they seem to be part of the unit... heat sink, fan and 15 amp minimum understood.

 

[preinke]

The idea of cascading PID's for for the offset of temperatures between RIMS output and Mash temperature is mentioned in several threads on this site. A more "conventional" approach is to have second PID's output be the setpoint to the first PID. This method is used widely in process industry to control temperature in a vessel with a heating jacket and many other scenarios.

I am in the middle of creating a PLC based cascaded PID setup based on the Allen Bradley Micro 800 series controller. It has a IPID function and an auto tune function. It's not as advanced as the PIDE function in RSLogix5000 but I think with a little glue logic I will be able to solve bumpless transfers and "ramped clamps" for the step mashing process.

Is it necessary ? Probably not but the cost at this time to add another PID is just programming time. I will post my project as it progresses. I find this site to be a wealth of information and an inspiration to up my game. I hope to return some help to others

 

[brewman !]

A more "conventional" approach is to have second PID's output be the setpoint to the first PID. This method is used widely in process industry to control temperature in a vessel with a heating jacket and many other scenarios.

Please explain, especially how you do this with 2 sub $50 PID controllers.

PLCs are great, but we are trying to use the KISS method here.

 

[brewman !]

I'm pretty handy, but this is a touch past my previous experience. I have a friend who will be helping as well. He's a "handyman-type".... I know this sounds like an accident waiting to happen, but we're pretty thorough and don't leave anything to question.

Good. I don't want to be semi responsible for an accident.

The PID should have the SSR built in or is this a component? As far as I can tell from the PIDs I've been shopping, they seem to be part of the unit...

Its a (slang) terminology thing. Solid State Relays are their own entity, separate from the PID controller. Where you are probably getting confused is that some of these devices have SSR "outputs", meaning you can program them to output a signal that works well with an SSR. The SSR output is what connects to the SSR itself.

Don't worry about all the tech in the wiki page below. All you need to know is what it does. An SSR is a switch that is controlled by an electric signal. In this case the PID generates the electric signal.

http://en.wikipedia.org/wiki/Solid-state_relay

 

[Shred]

Good. I don't want to be semi responsible for an accident.



Its a (slang) terminology thing. Solid State Relays are their own entity, separate from the PID controller. Where you are probably getting confused is that some of these devices have SSR "outputs", meaning you can program them to output a signal that works well with an SSR. The SSR output is what connects to the SSR itself.

Don't worry about all the tech in the wiki page below. All you need to know is what it does. An SSR is a switch that is controlled by an electric signal. In this case the PID generates the electric signal.

http://en.wikipedia.org/wiki/Solid-state_relay

That is exactly the clarification I was looking for. Thanks so much!

 

[preinke]

Do they have the capability to accept a remote (analog) setpoint ? I come from the industrial world where most PIDs have this functionality. Cascading works better if the two PIDs communicate when they reach clamps or are taken out of cascade mode. The STC1000 has been repurposed to schedule fermentation setpoints. There maybe a variant that uses the two temperatures in a cascaded PID mode.
I like the building part of this hobby and may venture back into the embedded world into a BrewPI setup. I like the idea of open systems like the Allen Bradley where the systems are expandable, the user community is large, parts are of the shelf and reliable.
The cost per loop on the AB will be around $25
2 variable speed motors, 4 PIDS, 2 flow meters, 1 level controller

 

[brewman !]

Do they have the capability to accept a remote (analog) setpoint ? I come from the industrial world where most PIDs have this functionality. Cascading works better if the two PIDs communicate when they reach clamps or are taken out of cascade mode. The STC1000 has been repurposed to schedule fermentation setpoints. There maybe a variant that uses the two temperatures in a cascaded PID mode.
I like the building part of this hobby and may venture back into the embedded world into a BrewPI setup. I like the idea of open systems like the Allen Bradley where the systems are expandable, the user community is large, parts are of the shelf and reliable.
The cost per loop on the AB will be around $25
2 variable speed motors, 4 PIDS, 2 flow meters, 1 level controller

Here is the manual for one of the PID controllers that is commonly used.
http://mythopoeic.org/misc-files/JLD612_Manual.pdf

Do you see any inputs for a remote set point ?

So AB sells a $100 PLC, complete with 2 temperature inputs ?

Its great you have BrewPi aspirations. I don't think that fits with the KISS principle the OP is looking for. I'm building a steam powered brewery that overcomes all the RIMS scorching limitations. I didn't bring it up until now because it wasn't pertinent to this discussion.

 

[Clankenbrew]

I like to take my reading coming into the tube. I feel I need to know where it's at.... not where it's going. If wort coming into my tube is 152 then I'm guessing the rest is that temp.

 

[Gameface]

I like to take my reading coming into the tube. I feel I need to know where it's at.... not where it's going. If wort coming into my tube is 152 then I'm guessing the rest is that temp.

But if your inflow is at 148 and you're trying to reach 152 and you use that inflow temp to drive your element you may be heating wort much hotter than 152, in fact you might boil it before you start to see the temp increase on your inflow side.

It's fine to monitor temp in places other than immediately after the RIMS tube, but you should ALWAYS control your RIMS element based on the temp immediately after it. You can then decide if it needs to be set a little higher in order to get your actual mash temp where you want it.

 

[Shred]

The part that is still confusing me about my current setup is why the temp reads nearly 10 degrees warmer at the end of the return hose than directly over the element. Is it normal for the temp to rise that much in the 5 feet of hose after the RIMS tube or is my thermowell insulating the probe so it's reading lower than what is actually happening in the tube?

For what it's worth, I ordered new temp probes for my STC-1000. These are 1.5" stainless ends. Not sure if it will make a difference, but they look a lot nicer than the little, black piece of rubber.

 

[Clankenbrew]

So what do you do when your temp probes reading hot wort off the element and shuts off and the wort container is still low in temp? Your elements shutting off before your at temp.
I like my reading coming from the wort..

 

[brewman !]

So what do you do when your temp probes reading hot wort off the element and shuts off and the wort container is still low in temp? Your elements shutting off before your at temp.
I like my reading coming from the wort..

The controller running the RIMS loop should be running in SSR PWM mode and thus turning down the element power, not shutting it off.

There is also no reason for the element to be shutting down while the bed is cool unless the denaturing temp has been reached or the wort is being scorched. Since the denaturing temp is a lot higher than mash temps and scorching even higher than that, this should never be a real world problem.

 

[Gameface]

So what do you do when your temp probes reading hot wort off the element and shuts off and the wort container is still low in temp? Your elements shutting off before your at temp.
I like my reading coming from the wort..

You increase the set temp or duty cycle for the element until whatever temp it takes there gets you the desired temp in the mash. It's fine if your desired mash temp is 152 and you're reading higher than that coming off the element as long as you're not denaturing the enzymes. You run a MUCH higher risk of denaturing enzymes by controlling off the infeed temps, as well as having a system with tons of lag.

 

[Clankenbrew]

Im going to have to test and monitor with more probes inside the mash and adjust if necessary.
Im not convinced it makes a big difference as long as you have good flow rate.

 

[Shred]

I thought this too but I had scorching issues when I forgot to turn on the pump or turned it off without killing the rims heat.... Now that I have spent the $20 total investment on the flow sensor and wiring components I have peace of mind that this will not happen again.

Do you have a link to your $20 flow sensor?

 

[BBBF]

The part that is still confusing me about my current setup is why the temp reads nearly 10 degrees warmer at the end of the return hose than directly over the element. Is it normal for the temp to rise that much in the 5 feet of hose after the RIMS tube or is my thermowell insulating the probe so it's reading lower than what is actually happening in the tube?

For what it's worth, I ordered new temp probes for my STC-1000. These are 1.5" stainless ends. Not sure if it will make a difference, but they look a lot nicer than the little, black piece of rubber.

I have the same problem. It wouldn't be gaining 10 degrees in the hose. I'd expect it to lose a degree or two.

I manually check the temperature of the wort going back into the MLT and lower my PID temperature account for the difference. I could set the PID to account for the error, but after the mash I plug in a different probe, which it reads correctly, and use the panel to control my boil kettle.

 

[thekraken]

I used to be of the mindset that the best place for the temp probe was at the RIMS outlet, I'm starting to reevaluate that position. If you are using a generic PID control (like most here use) then I still think it's the safest place.

I'm currently working on a custom RIMS controller using an Arduino. It really is eye opening to graph hard data on temp differences between the RIMS tube and the MLT. I don't feel the differences are insignificant.

Through some trial and error, and help from another forum member BigBlock, we're narrowing in on a cascade controller that uses the MLT output/RIMS input to control the temperature.

I'm using one temp probe at the RIMS output (becoming superfluous), one probe at the MLT ouput, and a cheap flow meter with pretty good results. Given liquid input temperature, flow rate, we can estimate the power required to heat the input to a given output temperature. We then use a simple PID routine to adjust the RIMS output setpoint. This way we can prevent scorching, denaturing AND know/control the temp of the actual MLT.

If folks are interested in the theory, code and development I can start a thread and link back here.

 

[thekraken]

I thought this too but I had scorching issues when I forgot to turn on the pump or turned it off without killing the rims heat.... Now that I have spent the $20 total investment on the flow sensor and wiring components I have peace of mind that this will not happen again.

Agreed, as far as I'm concerned a flow switch/meter was required for my system. But in all fairness shouldn't your control panel be wired to cut power to the element if power to the pump is off?

 

[thekraken]

The part that is still confusing me about my current setup is why the temp reads nearly 10 degrees warmer at the end of the return hose than directly over the element. Is it normal for the temp to rise that much in the 5 feet of hose after the RIMS tube or is my thermowell insulating the probe so it's reading lower than what is actually happening in the tube?

For what it's worth, I ordered new temp probes for my STC-1000. These are 1.5" stainless ends. Not sure if it will make a difference, but they look a lot nicer than the little, black piece of rubber.

Are you using the STC's probe to measure both of these temperatures? If you'r measuring the output of the hose with a thermometer there may be calibration discrepancies. It could be that your thermowell is too large and your not getting a good transfer of heat to your probe. You could try some thermal paste or even some toothpaste (very temporary) between the probe and thermowell to try and see if that makes a difference.

 

[thekraken]

Do you have a link to your $20 flow sensor?

Not to speak for augiedoggy, but because I sub'd to his build thread, I think he's using one of these: http://www.ebay.com/sch/i.html?_fro...RS0&_nkw=stainless+steel+flow+switch&_sacat=0

 

[preinke]

I used to be of the mindset that the best place for the temp probe was at the RIMS outlet, I'm starting to reevaluate that position. If you are using a generic PID control (like most here use) then I still think it's the safest place.

I'm currently working on a custom RIMS controller using an Arduino. It really is eye opening to graph hard data on temp differences between the RIMS tube and the MLT. I don't feel the differences are insignificant.

Through some trial and error, and help from another forum member BigBlock, we're narrowing in on a cascade controller that uses the MLT output/RIMS input to control the temperature.

I'm using one temp probe at the RIMS output, one probe at the MLT ouput (becoming superfluous), and a cheap flow meter with pretty good results. Given liquid input temperature, flow rate, we can estimate the power required to heat the input to a given output temperature. We then use a simple PID routine to adjust the RIMS output setpoint. This way we can prevent scorching, denaturing AND know/control the temp of the actual MLT.

If folks are interested in the theory, code and development I can start a thread and link back here.

I am also going down the cascade route with variable MLT tuning based on flow rate with "clamped" RIMS output temperature setpoint based on MLT setpoint for stepped mash profiles. I am going the route with a Micr830 Allen Bradley but would be interested in your control structure and happy to share anything I am doing - mine is a work in progress. Please start another thread

 

[brewman !]

I'm currently working on a custom RIMS controller using an Arduino.

<snip>

If folks are interested in the theory, code and development I can start a thread and link back here.

Please do.

If you write code, could you open source it ?

And could you make the control loop a generic self threaded library written in C or C++ with generic function calls like getMLTTemp() and getRIMSTemp(), so that it can be reused ?

I'm probably going to build a controller too, though I'll probably do mine on the RPi. Though Arduino isn't a bad choice either.

 

[augiedoggy]

Agreed, as far as I'm concerned a flow switch/meter was required for my system. But in all fairness shouldn't your control panel be wired to cut power to the element if power to the pump is off?

Yeah thats one way of doing it but that wont help in a stuck sparge situation and that was one of my main concerns. plus my pumps are on pwm speed controllers so even if they are on the speed could inadvertently be set too low for proper flow arcoss the element and scorch.