Using a PID to control a pump in a HERMS setup? A no-no?

[The Pol]

Sure, I have emailed these spreadsheets to all of those who are building the system too, they are needed and make brew day mindless.

Total water needed: This comes from Promash, this is the total water needed to reach my target pre-boil/post boil volume. I have to buy water to brew, so this is handy.

First Strike Temp: This is the strike temp calculated by Promash for the protien rest. I add 15F to preheat my MLT.

HEX Initial Volume: This is the initial volume of water heated in the HLT.


1st Strike Volume: This comes from Promash, this is the protien rest infusion.


2nd Step Infusion Temp: This comes from Promash, this is the infusion temp to reach my desired sacc. temp.

2nd Step Volume: This comes from Promash, it is the qty. at the afore mentiotned temp, to reach sacc. rest.

2nd Step Recirc. Temp.: This is the HLT temperature for the sacc. rest recirculation. My differential in my system is 3F.

ADD Water temp: This is the temp of the water that is added to the HLT AFTER the step infusion to bring the mash to sacc. rest temp. This cools the HLT to the recirc. temp. This water is also NEEDED to reach the needed water for sparge.



It calculates the following:


1st Infusion HEX Post Strike Qty.: This is what is left in the HLT AFTER the inistial strike infusion, to verify the element stays wet.

Time to Reach Second Step TEmp: This is the calculated time to reach my sacc. infusion temp in the HLT.

2nd Infusion HEX Post Strike Qty: This is the HLT qty. after the SECOND infusion, the infusion to reach sacc. rest. This verifies the element is still wet.

HEX Add Water Qty: This is the water needed to add to the HLT AFTER the sacc. rest infusion to reach the TOTAL WATER NEEDED qty from Promash. This water also cools the HLT to sacc. recirc. temp.

Post Second Infusion HEX Temp: This is the temp of the HLT AFTER the afore mentioned cool water infusion to the HLT

HEX Time To Heat: This is the time that it will take the HLT to reach the desired sacc. recirc temp after the cool water infusion.

 

[Lil' Sparky]

I don't think I'll be doing all that. It's too complicated - even w/ a spreadsheet. If I do a HERMS, it'll be in a vessel separate from my HLT. I think it's best to separate the two to avoid the hassle. I wish I could automate my burners, etc. like the BRUTUS builds, but that's just too much $$ and fabrication for me.

 

[The Pol]

Even if your HEX is separate from your HLT, I think you will still find a 20 minute step up from protien to sacc. rest. The mash has A LOT of thermal mass. Much more than simple water.

I have yet to see a HERMS where the brewer did NOT overshoot the target temp. and WAS able to step mash with it.

 

[Lil' Sparky]

Here's what I'm planning. I think a smaller, separate HERMS (or RIMS) unit will be simpler (no complicated formulas) and provide a little more flexibility. For example, I'd like to be able to overshoot the target initially without fear of the entire mash overshooting the target. This means the HEX will need the ability to quickly shed the excess heat. A small HERMS or RIMS will allow you to do this. You may not agree with that approach, but it's valid if you want to step quickly. I've also got a direct-fired MLT that can effectively assist w/ steps.

 

[Seveneer]

My system uses a seperate heat exchanger and will raise the mash temperature for a 12 imperial gallon batch in about 10 minutes without overshooting.

/Phil.

 

[The Pol]

Here's what I'm planning. I think a smaller, separate HERMS (or RIMS) unit will be simpler (no complicated formulas) and provide a little more flexibility. For example, I'd like to be able to overshoot the target initially without fear of the entire mash overshooting the target. This means the HEX will need the ability to quickly shed the excess heat. A small HERMS or RIMS will allow you to do this. You may not agree with that approach, but it's valid if you want to step quickly. I've also got a direct-fired MLT that can effectively assist w/ steps.

Many people do use hybrid systems of HERMS and direct fire to step. That is just more equipment for me, and again, going back to gas when I have already gone all electric.

The infusion steps are why I installed the sight gauges on my vessels this past week.

 

[The Pol]

My system uses a seperate heat exchanger and will raise the mash temperature for a 12 imperial gallon batch in about 10 minutes without overshooting.

/Phil.

Kudos, you are the only person I have met in three years that has a HERMS capable of doing this.

 

[Seveneer]

I make it very clear that my system uses a seperate heat exchanger.

People need to be aware that there is a way of conducting step mashes with a HERMS it just needs to be designed with that in mind. Using the HLT as the heat exchanger restricts this ability.

/Phil.

 

[Bobby_M]

I don't care where the heat is being applied. I think what you mean by "not overshooting" is not having the total mash go further than you wanted. However, the only way to get a 10 minute ramp from protein to sac rest is to have the output of the hex WAY higher than the target temp. The problem is that you're denaturing the hell out of your enzymes and running the whole mash through the hex several times makes it worse. Decoction mashes don't have this problem because you're only pulling a small portion out and it later acts just like a pure water infusion step.

Applying a well elevated temp to the mash inline like that is similar to heating the entire mash to that temp for a short time, then cooling it back down to your target temp.

 

[Seveneer]

I measure the heat of the recirculated wort as it leaves the exchanger. This is used as the input to the PID. It's not possible for the wort to go above the set temperature so long as the pump is running.

/Phil.

 

[The Pol]

I don't care where the heat is being applied. I think what you mean by "not overshooting" is not having the total mash go further than you wanted. However, the only way to get a 10 minute ramp from protein to sac rest is to have the output of the hex WAY higher than the target temp. The problem is that you're denaturing the hell out of your enzymes and running the whole mash through the hex several times makes it worse. Decoction mashes don't have this problem because you're only pulling a small portion out and it later acts just like a pure water infusion step.

Applying a well elevated temp to the mash inline like that is similar to heating the entire mash to that temp for a short time, then cooling it back down to your target temp.

Agreed Bobby... I dont see how you can possibly go from 122F to 152F in 10 minutes without returning wort to the mash that is ABOVE the target temp. That is 3F per MINUTE. That is like placing a 4500W element in the mash in fact, at full power.

If you are stepping from 122F to 152F... and you are using 152F wort returning to the MLT to do this warm up... I cannot fathom that it is even possible to do in 10 minutes.

If you place a 4500W element in 10 gallons of pure water, you will get 3F/minute increase in temp. I dont see how using 152F wort will do the exact same, in the same ammount of time.

 

[Seveneer]

All I can say is take a look at my web page and then try it for yourself.

/Phil.

 

[The Pol]

Bottom line, mathematically:

4500W applied DIRECTLY to 10 gallons of water (not as much thermal mass as a mash) will raise the temp at a rate of 3F/minute.

This is the same rate that you claim with your HERMS system, which is only using 152F wort to heat the mash.

Mathematically, I dont see it possible. 152F wort does not have the same heating capability of a 4500W element placed directly in the fluid at full power.

 

[Seveneer]

I'm not heating 10 gallons of water. I'm heating 10kg of grain and 25 litres of water.

/Phil.

 

[The Pol]

I'm not heating 10 gallons of water. I'm heating 10kg of grain and 25 litres of water.

/Phil.

Right...

SO you are heating 6 gallons of water

And 22 pounds of grain....

A total mash volume of....

7.8 gallons of VERY thick (1.1qt/lb) mash.

Which will have a much higher thermal mass than 8 gallons of water alone. The higher the thermal mass, the MORE energy it takes to change the temp of that volume. This was my point all along.

 

[eccsynd]

Here is an old chart datalog from my BCS-460 on a water test. I get about a 1.5-2deg rise / min, depending on the grain bill with 5gal batches and a 1500W element, herms w/ 1.5gal exchanger. I totally disagree that the HEX temperature has to be kept less than or equal to your desired temp. In the chart below, the PID temp probe is Trace1 (blue) which is placed in the wort flow on the output of the HEX, and the HEX temp is trace0 (gray). The output of the PID is the bottom blue, trace3, you can see it start switching as the temp approaches the setpoint, 154 in this case.

As you can see, the HEX temp is 20deg higher than the wort temp. And it rises linearly. If I had a bigger element, it would rise faster. I don't overshoot because my PID works as it should. As soon as the mash temp nears the setpoint, the hex temp starts trailing off. It keeps the HEX high enough to counteract any thermal losses, but not too high such that it creates an overshoot.

This specific log is a water test, but normal mashes behave similarly based on past experience, although the mash/hex differential is less because the pump doesn't run wide open.

 

[The Pol]

You know what, HERMS is well suited for step mashes. I just did some other calculations, and it checks. HERMS is excellent for step mashing.

 

[DonD13]

Well, not having brewed an AG yet, but considering building a HERMS and having a decent understanding of thermodynamics, this thread is really interesting to me.

Without any specific measurements (purely hypothetical HERMS setup), you could concievably have a big enough coil that the wort temp at the output of the HEX is very close to the HLT temp. In this case, you would get the entire mash up to temp in whatever amount of time it takes to recirc the entire mash. With a typical march pump (lets say 2.5GPM to account for the flow restriction of your HEX and valves, plumbing etc. and to make the calcs easy) you would recirc an entire 10gal mash in 4 minutes, which I would think is plenty fast step time.

The question is: how big of an HEX would one need to get the output temp almost to HLT temps? Has anyone done any experimentation to determine this? Like I said, I'm looking at building a HERMS, so this is fascinating discussion to me.

Later,
Don

 

[The Pol]

Well, not having brewed an AG yet, but considering building a HERMS and having a decent understanding of thermodynamics, this thread is really interesting to me.

Without any specific measurements (purely hypothetical HERMS setup), you could concievably have a big enough coil that the wort temp at the output of the HEX is very close to the HLT temp. In this case, you would get the entire mash up to temp in whatever amount of time it takes to recirc the entire mash. With a typical march pump (lets say 2.5GPM to account for the flow restriction of your HEX and valves, plumbing etc. and to make the calcs easy) you would recirc an entire 10gal mash in 4 minutes, which I would think is plenty fast step time.

The question is: how big of an HEX would one need to get the output temp almost to HLT temps? Has anyone done any experimentation to determine this? Like I said, I'm looking at building a HERMS, so this is fascinating discussion to me.

Later,
Don

A March 809 will pump about 2.5 gallons/min at FULL throttle, whether or not your filtering device, false bottom or braid will allow this without grain compaction, would be a question.

You left out one thing... the grain. You arent heating the grain in the coil... you are passively heating that with wort.

The temp at the output will depend on pumping rate, heat applied, coil length, coil ID etc.

 

[The Pol]

If you want to pump 2.5gal/minute through a coil and heat it in that coil from 122F to 152F, a 30F rise... you will need... wait for it.

The equivilent of 8000 WATTS of power applied directly to 2.5 gallons of water will bring it from 122F to 152F in ONE minute.

This is 27,300 BTU applied to that water. That is an incredible ammount of heat applied directly to the water.

So if you take the coil itself out of the picture, you still need A LOT of heat, even if directly heating, to make the 30F rise at 2.5gal/min that you are speaking of.

The assumption that circulating the 10 gallons through the HEX in 4 minutes will allow you to reach your step temp. is incorrect. In practice, this does not work, as the grains absorb A LOT of heat, and getting the output of the coil to your target temp in ONE pass will take incredible heat and transfer of heat through that coil.

But again, whatta I know.

 

[eccsynd]

Without any specific measurements (purely hypothetical HERMS setup), you could concievably have a big enough coil that the wort temp at the output of the HEX is very close to the HLT temp.

I wouldn't worry too much about the heat transfer differential, i.e., the difference in HEX temp and the wort temp. The graph that I posted above shows that the transfer characteristics of my HEX are TERRIBLE, 20deg difference. I could make it better by adding an agitator, and more copper coils.. But why? The only down side of a big differential is response time, but I think the time/money is better spent getting the HEX volume lower, which REALLY affects response time. Let your PID do the work.

 

[kladue]

What is so hard about raising wort temps 30+ degrees, here is a shot from the old (2003) system with 35 degree rise across the mixer chamber Picasa Web Albums - Kevin - Old Brewing S....

 

[The Pol]

What is so hard about raising wort temps 30+ degrees, here is a shot from the old (2003) system with 35 degree rise across the mixer chamber Picasa Web Albums - Kevin - Old Brewing S....

At .83GPM?

We are talking about 30F at 2.5 gal min.

There is a small difference, I would think, when you tripple the flow rate.

Sure sure... you can raise a 10 gallon mash 30F in 4 minutes with a HERMS, I give in... sure you can. Build it.

 

[DonD13]

If you want to pump 2.5gal/minute through a coil and heat it in that coil from 122F to 152F, a 30F rise... you will need... wait for it.

The equivilent of 8000 WATTS of power applied directly to 2.5 gallons of water will bring it from 122F to 152F in ONE minute.

This is 27,300 BTU applied to that water. That is an incredible ammount of heat applied directly to the water.

So if you take the coil itself out of the picture, you still need A LOT of heat, even if directly heating, to make the 30F rise at 2.5gal/min that you are speaking of.

The assumption that circulating the 10 gallons through the HEX in 4 minutes will allow you to reach your step temp. is incorrect. In practice, this does not work, as the grains absorb A LOT of heat, and getting the output of the coil to your target temp in ONE pass will take incredible heat and transfer of heat through that coil.

But again, whatta I know.

I wasn't suggesting that this hypothetical system would actually be buildable, and you're right, I totally ignored the mass of the grain, which would make a HUGE difference. 8000 watts in general doesn't scare me, since I'm looking at propane heat for the HLT, but actually transferring all that heat through the HEX without heating it to denaturing temperature would require an enormous HEX.

This is exactly why I come here. No matter what I think up, somebody here has already tried it and found the practical limits.

 

[Lil' Sparky]

I wonder if some BRUTUS users could chime in on their experience stepping (without overshooting) on their direct-fired system. They're running their pumps at full throttle and the burners are pid controlled so that the output from the MLT is at the target.

 

[The Pol]

Direct fire will be completely different.

You are dealing with A LOT more BTU potential, you are also firing right at the MLT, so you are heating EVERYTHING at one time, you arent heating a small stream.

Direct fire will be MUCH different than a HERMS. Direct fire heats everything with 2x-5x the BTU potential to do so.

 

[Bobby_M]

...but you still cannot apply too much heat in a direct fire RIMS because you can't really pump more than 1GPM without grain compaction (Yes, even with a FULL diameter false bottom). At best, the direct fire is helping to offset heat loss by warming the vessel itself, but the flame has to be low enough not to heat the 1 gallon under the FB to higher than your target temp. I have to agree with Pol that the hypotheticals being thrown around here might as well be talking about heating with pixie dust.

 

[kladue]

Well i will have to put the new system to the test and see what it can do, it has twice the boiler area of the old system. The only test data for the new boiler so far is is at 1 gpm Picasa Web Albums - Kevin - System Testin... and at .5 gpm water Picasa Web Albums - Kevin - Boiler Test.

 

[Lil' Sparky]

Yeah, I know direct-fire is different. I was just wondering, for sake of discussion, if we had any hard numbers by the brutus guys who use similar electronic control logic to ensure the wort doesn't exceed the target temp.

Bobby, how did you come up with the 1 gpm maximum on recirculation?

 

[conpewter]

...but you still cannot apply too much heat in a direct fire RIMS because you can't really pump more than 1GPM without grain compaction (Yes, even with a FULL diameter false bottom). At best, the direct fire is helping to offset heat loss by warming the vessel itself, but the flame has to be low enough not to heat the 1 gallon under the FB to higher than your target temp. I have to agree with Pol that the hypotheticals being thrown around here might as well be talking about heating with pixie dust.

Hey guys, why have all of these talks been about not heating the returning wort over the target temp? I know enzymes will denature at higher temps but this is not an instant thing (Heck even Alpha amylase will denature at 153 after 2 hours) So even if your wort hit 170 before being put back in the tun you'd not have damage the alpha amylase too much in that short amount of time, I'm not positive about beta.

Edit: This BYO article has some of the temperature dependence talk of the amylase enzymes at the bottom