Someone with RIMS expertise please help - last attempt before a change :(

[cxp073]

Ok, so I've posted about my problems a couple of times but I wanted to sum up my issues in a LAST DITCH ATTEMPT before I try HERMS.

I have the following system:

10 gallon cooler w/ rims tube mounted to the side of the cooler vertically.
SESTOS PID with pt100 sensor mounted on the output of the rims tube. Auto tuned.
3kw 240v LWD element running on 240volts
12v pump max flow 10l per minute with a ball valve on the output. The pump is approximately 1 foot below the output of the MT. The pump is definitely mounted correctly (output side higher than the input)
The wort return tube from the top of the rims tube is approximately 4 feet.

The problem:

At low temperatures, steps in temperature occur without issues. However, when I step from 62 to 70, the RIMS OUTPUT tube that returns the wort to the MT starts to shake ONLY WHEN THE ELEMENT TURNS ON. The shaking is even worse when I go from 70 to 78c for mash out and the wort returning to the MT starts to pulse in flow. The shaking and pulsing only occurs when the element turns on and increases in intensity the longer the element is turned on. This occurs both at a low flow rate (3 litres per min) and a high one (full throttle) albeit to a slightly lesser extent because the temp ramps quicker and hits its target faster.

The pump makes no noise difference or anyhing during these pulses

Do you guys have any idea of what I can do to fix this? As I said, I'm getting really tired of filling this thing up with water and trouble shooting for hours on end. If I can't find a solution quickly I will need to try another system help!

 

[chewse]

So the pulsing is in the tube that is connects the RIMS output port and the MT input port? If so, could it be that your silicone tube cannot handle the hotter temps or maybe the interior diameter is to narrow or the walls of the tube not thick enough?

 

[miahpage]

What is the diameter of your RIMS tube? Could the element be resting on the side of the RIMS tube?

 

[dyqik]

What cycle time is your PID set to? For a 3kW element it may need to be very short to prevent wort boiling on the element in the tube.

PIDs work by switching the output on and off for a portion of a cycle time. That time can be set to be very long or short. With an overpowered element like yours, if it's set to a long time, any demand from the PID for heating power will turn the element on for a good fraction of that cycle time and may boil some of the wort.

By my calculations, your element can boil a liter of water per minute starting from mash temps. At 3 l/min flow rate and 3kW output power, it could easily be boiling some of the water that flows in contact with the element, or causing cavitation. Cavitation will occur on heating elements at much below boiling, as anyone who's listened to an electric kettle can tell you. This is particularly true if any of the heated part of element is in a dead space in the tube where the flow rate is much lower than the total.

I'd try setting the PID control to as short a cycle time as it will allow. If that doesn't work, then either running your element on 120V (if you have that available, I suspect from the 3kW on 240V and using Celsius that you are on European/UK power) or using an SSVR to throttle back the power level would be a good idea before you give up on RIMS.

 

[cxp073]

What cycle time is your PID set to? For a 3kW element it may need to be very short to prevent wort boiling on the element in the tube.

PIDs work by switching the output on and off for a portion of a cycle time. That time can be set to be very long or short. With an overpowered element like yours, if it's set to a long time, any demand from the PID for heating power will turn the element on for a good fraction of that cycle time and may boil some of the wort.

By my calculations, your element can boil a liter of water per minute starting from mash temps. At 3 l/min flow rate and 3kW output power, it could easily be boiling some of the water that flows in contact with the element, or causing cavitation. Cavitation will occur on heating elements at much below boiling, as anyone who's listened to an electric kettle can tell you. This is particularly true if any of the heated part of element is in a dead space in the tube where the flow rate is much lower than the total.

I'd try setting the PID control to as short a cycle time as it will allow. If that doesn't work, then either running your element on 120V (if you have that available, I suspect from the 3kW on 240V and using Celsius that you are on European/UK power) or using an SSVR to throttle back the power level would be a good idea before you give up on RIMS.

I was thinking about cycle time. I can easily change this. How low do you think I should go at first? 10%?

Oh, and yes I am on UK power.

And, I don't think there is any dead space in the tube. The output of the rims is actually at the highest part of the tube ( I.e after the wort enters the tube, it goes straight up and out the very top.)

 

[BrunDog]

My guess is there is some local boiling going on. Are you sure the temp of the wort exiting the tube is being read correctly? It is possible that there is contCt between the element and tube as noted above and this is causing the shaking if the liquid in that particular section is boiling.

 

[srice]

Agree with ^^^^^. It sounds like you are boiling your wort with that much power being dumped into the small volume in your RIMS tube.

 

[kyle187]

I also agree, you are most likely causing some film boiling on the element, especially as the water temps get closer to boiling. Could you run your element on 120V (1/4 power) and see if you are still having the same issues?

At a much lower watt density i would guess you will no longer have an issue.

 

[GilaMinumBeer]

My first thought was boiling happening in the rims tube also.

 

[dyqik]

I was thinking about cycle time. I can easily change this. How low do you think I should go at first? 10%?

Oh, and yes I am on UK power.

And, I don't think there is any dead space in the tube. The output of the rims is actually at the highest part of the tube ( I.e after the wort enters the tube, it goes straight up and out the very top.)

The cycle time should be a setting in seconds on the PID. It's the time between the element coming on each time while it's maintaining a temperature difference. Looking at the manual for the SESTOS D1S (I don't know if that's your model or not, but all of the ones by this company will probably be somewhat similar), and assuming you are running an SSR to control the element, there's no reason not to set the "Ctl" parameter which controls the period to 0, which gives a cycle time of 0.5s. Factory default for the D1S is 4 seconds, so you might be able to find a sweet spot in between.

But be aware that when you add grain to the system, the achievable flow rate will be much lower than with just water. So you may want to find a way to lower the power of the element (SSVR and potentiometer is the cheapest option for you since you don't have 120V available) anyway.

 

[cxp073]

The cycle time should be a setting in seconds on the PID. It's the time between the element coming on each time while it's maintaining a temperature difference. Looking at the manual for the SESTOS D1S (I don't know if that's your model or not, but all of the ones by this company will probably be somewhat similar), and assuming you are running an SSR to control the element, there's no reason not to set the "Ctl" parameter which controls the period to 0, which gives a cycle time of 0.5s. Factory default for the D1S is 4 seconds, so you might be able to find a sweet spot in between.

But be aware that when you add grain to the system, the achievable flow rate will be much lower than with just water. So you may want to find a way to lower the power of the element (SSVR and potentiometer is the cheapest option for you since you don't have 120V available) anyway.

I'm just looking at the manual, and the default is 4 but if 0 = .5 seconds than 4 = 8s. So default would actually be 8 seconds, which seems more like how long it runs for before the light turns off.

I tried turning down the oPH setting before thinking that that was the setting to change for on/off time of the element.....

I think I'll give it a go and change the Ctl to 0. Should I run auto tuning again with this changed prior to tuning?

 

[BrotherGrim]

I agree with what is posted above. You may also wish to limit the max output on your PID. This will slow down the response somewhat, but it looks to me like you are over powered anyway.

A. shorten the cycle time to as low as you can make it so you have a more even application of energy
B. limit maximum output on your PID to 75% or so

Both of these options are assuming that you are getting localized boiling during the heat application cycles. It sounds like cavitation caused by boiling to me.

Good luck

 

[cxp073]

I agree with what is posted above. You may also wish to limit the max output on your PID. This will slow down the response somewhat, but it looks to me like you are over powered anyway.



A. shorten the cycle time to as low as you can make it so you have a more even application of energy

B. limit maximum output on your PID to 75% or so



Both of these options are assuming that you are getting localized boiling during the heat application cycles. It sounds like cavitation caused by boiling to me.



Good luck

Cheers

What does lowering the maximum output actually do?

I can see that setting - it's oPH On the SESTOS

 

[dyqik]

I'm just looking at the manual, and the default is 4 but if 0 = .5 seconds than 4 = 8s. So default would actually be 8 seconds, which seems more like how long it runs for before the light turns off.

I tried turning down the oPH setting before thinking that that was the setting to change for on/off time of the element.....

I think I'll give it a go and change the Ctl to 0. Should I run auto tuning again with this changed prior to tuning?

Yeah, definitely run autotune again. In fact, you probably need to run autotune every time you change the flow rate, either with the valve or by adding grain to the MLT.

 

[GotPushrods]

Cheers

What does lowering the maximum output actually do?

I can see that setting - it's oPH On the SESTOS

I'm not familiar with the Sestos, but is should mean your element will never come on for more than that percentage of your cycle time.

If you are able to set a lower cycle, say 1 second, then it will never pulse for more than 750ms every second, (off for at least 250 ms)

If you are stuck at 8 second cycle times, it would max out at 6 seconds on and 2 seconds off during each cycle.

 

[Gameface]

So I just want to clarify, the problem is the shaking? That's the issue you're troubleshooting? Or do you believe the shaking is the cause of some other problem?

I mean it does sound like boiling, but I can hear my element in my HLT (HERMS) making boiling noises when I never go above 160F in the HLT.

 

[dyqik]

So I just want to clarify, the problem is the shaking? That's the issue you're troubleshooting? Or do you believe the shaking is the cause of some other problem?

I mean it does sound like boiling, but I can hear my element in my HLT (HERMS) making boiling noises when I never go above 160F in the HLT.

Boiling noises will start at pretty low temperatures in the HLT (again, compare with electric kettles), due to boiling of local pockets of water on the element. But even if the water in the HLT was locally boiling, in a HERMS system the wort only sees the average temperature in the kettle.

In a RIMS tube, the wort is in direct contact with the element, so local boiling will de-nature enzymes and possibly scorch the wort. The shaking is a sign that local boiling is happening in this tube, to a pretty severe degree. Certainly enough to scorch wort at the lower flow rates that will occur when there's grain in the system.

 

[Gameface]

Boiling noises will start at pretty low temperatures in the HLT (again, compare with electric kettles), due to boiling of local pockets of water on the element. But even if the water in the HLT was locally boiling, in a HERMS system the wort only sees the average temperature in the kettle.

In a RIMS tube, the wort is in direct contact with the element, so local boiling will de-nature enzymes and possibly scorch the wort. The shaking is a sign that local boiling is happening in this tube, to a pretty severe degree. Certainly enough to scorch wort at the lower flow rates that will occur when there's grain in the system.

Right, so what he's troubleshooting is the possible denaturing of enzymes, not a shaking RIMS tube. I mean, I'll let him confirm, but yeah, if the problem is shaking then there isn't really a problem.

 

[cxp073]

Boiling noises will start at pretty low temperatures in the HLT (again, compare with electric kettles), due to boiling of local pockets of water on the element. But even if the water in the HLT was locally boiling, in a HERMS system the wort only sees the average temperature in the kettle.

In a RIMS tube, the wort is in direct contact with the element, so local boiling will de-nature enzymes and possibly scorch the wort. The shaking is a sign that local boiling is happening in this tube, to a pretty severe degree. Certainly enough to scorch wort at the lower flow rates that will occur when there's grain in the system.

Thank you to everyone. I set the cycle time to .5 seconds and noticed an improvement. I also set the max output to 35 and this also improved things.

It is now working flawlessly.

Dyqik - thank you especially. Weeks of trouble shooting was largely solved by your comments. Await a mention on my blog http://grizzlybearloveskolsch.com when I do my how to write up.

 

[rudylyon57]

This is an interesting phenomenon! Here are a couple of possible causes and fixes.

1) If your pump is downstream from the RIMS tube, it may be surging due to pulling in a two-phase mixture of liquid and vapor when your heater powers up. Your heater power seems high for the pump flow rate and could easily boil. Solution might be to move the pump upstream of the tube where it will not encounter the vapor.
2) Silicone tubing becomes very flexible when hot. Maybe when hotter liquid is flowing the tube wall can collapse until pump pressure forces it open. This could happen in a cyclic mode. One solution would be to restrict the flow slightly at its entrance back to MT to maintain a slightly elevated pressure to keep the tube open.

Good luck

 

[Indian_villager]

What cycle time is your PID set to? For a 3kW element it may need to be very short to prevent wort boiling on the element in the tube.

PIDs work by switching the output on and off for a portion of a cycle time. That time can be set to be very long or short. With an overpowered element like yours, if it's set to a long time, any demand from the PID for heating power will turn the element on for a good fraction of that cycle time and may boil some of the wort.

By my calculations, your element can boil a liter of water per minute starting from mash temps. At 3 l/min flow rate and 3kW output power, it could easily be boiling some of the water that flows in contact with the element, or causing cavitation. Cavitation will occur on heating elements at much below boiling, as anyone who's listened to an electric kettle can tell you. This is particularly true if any of the heated part of element is in a dead space in the tube where the flow rate is much lower than the total.

I'd try setting the PID control to as short a cycle time as it will allow. If that doesn't work, then either running your element on 120V (if you have that available, I suspect from the 3kW on 240V and using Celsius that you are on European/UK power) or using an SSVR to throttle back the power level would be a good idea before you give up on RIMS.

I believe I told this same gentleman the same thing on reddit.

 

[cxp073]

Alright - posted pics and build process on my blog if any of you are interested!

 

[MrVix]

In my opinion this is a cavitation issue. I think that the pump is cavitating causing air pockets or voids in the system. When these air pockets reach the RIMS tube heat they expand causing the pulse. Before you discard the RIMS system, replace the pump with something more substantial. I am not prompting a debate on the performance of those little 12V pumps. I have one...and I know that the tolerances are inconsistent. Maybe the heat from the wort combined with the 4' of lift is exploiting this. I would suggest that you buy or borrow a Chugger pump to do a temporary test.

 

[augiedoggy]

This is an interesting phenomenon! Here are a couple of possible causes and fixes.

1) If your pump is downstream from the RIMS tube, it may be surging due to pulling in a two-phase mixture of liquid and vapor when your heater powers up. Your heater power seems high for the pump flow rate and could easily boil. Solution might be to move the pump upstream of the tube where it will not encounter the vapor.
2) Silicone tubing becomes very flexible when hot. Maybe when hotter liquid is flowing the tube wall can collapse until pump pressure forces it open. This could happen in a cyclic mode. One solution would be to restrict the flow slightly at its entrance back to MT to maintain a slightly elevated pressure to keep the tube open.

Good luck

I know this is old but I dont know why someone would put the pump downstream from the rims?.... pumps are meant to push liquid not pull. and mounting these small dc pumps down low like a chugger just hurts thier performance and doesnt help anything... Yes I agree you would have a possible negative pressure on the silicone hoses if the pump was strong enough in this case.
I used the 12v and now 24v pumps with my rims along with 3ft of head height to the top of my MT from the rims location but since the pump it mounted directly to the base of my MT with a camlock and the sparge arm goes down int the MT a ways the pump does not really see that head pressure.. with my flow meter I measure 2.1 gallons per minute thorough my rims with the small tan dc 24v pump and this is with enough grain for a 10 gallon batch to filter through... yes my heating element is only 1000w so boiling of scorching is not a concern (no carmelization whatsoever occurs on my element)

 

[rwc_617]

Since you are in the UK, your voltage is set, but are there lower wattage elements available? Power is what is needed to boil water, so that 3000 W element is overkill and can get your wort boiling FAST. Can you find something like a 1500 W stainless element for your voltage and thread size? That should be cheaper than converting to HERMS and allow a broader range of controller settings.

 

[cxp073]

Since you are in the UK, your voltage is set, but are there lower wattage elements available? Power is what is needed to boil water, so that 3000 W element is overkill and can get your wort boiling FAST. Can you find something like a 1500 W stainless element for your voltage and thread size? That should be cheaper than converting to HERMS and allow a broader range of controller settings.

I didn't want to have to get another element, really. I'm sure I could find something but really wanted to use the parts I had. I managed to use the PID And tune it so that the 3kw element works flawlessly. I just posted a full tutorial / experience in my blog if your interested!

 

[crane]

I've had this same problem on my setup when I switched from a chugger pump to a 12V DC pump. My setup is similar to yours. I also use a 5500w element running on 240V. For me it happens when the grain bed compacts too much and flow rate goes down causing my 12V DC pump to struggle. I found that the shortest amount of silicon hose works best for the DC pumps, especially on the inlet side. I ended up directly mounting it to the mash tun. I've also tried a few different 12V pumps and it's more prone to happen on lower flow rate pumps.

View attachment 1433008368095.jpg

 

[cxp073]

I've had this same problem on my setup when I switched from a chugger pump to a 12V DC pump. My setup is similar to yours. I also use a 5500w element running on 240V. For me it happens when the grain bed compacts too much and flow rate goes down causing my 12V DC pump to struggle. I found that the shortest amount of silicon hose works best for the DC pumps, especially on the inlet side. I ended up directly mounting it to the mash tun. I've also tried a few different 12V pumps and it's more prone to happen on lower flow rate pumps.

That's a nice little setup you have there! I guess it really depends on whether the pump becomes strained when the wort inside the tube heats up. I can't see that being the case, though, because the pump is only running 1/2 open. I can open it, to 3/4 for example, and the flow would increase without problem so the grain bed couldn't have been the issue. It would be interesting to try a more powerful pump, and change nothing except for the percentage of time during each cycle that the heater can be on. If I ever decide to get a bigger pump I will give it a go.

 

[augiedoggy]

I've had this same problem on my setup when I switched from a chugger pump to a 12V DC pump. My setup is similar to yours. I also use a 5500w element running on 240V. For me it happens when the grain bed compacts too much and flow rate goes down causing my 12V DC pump to struggle. I found that the shortest amount of silicon hose works best for the DC pumps, especially on the inlet side. I ended up directly mounting it to the mash tun. I've also tried a few different 12V pumps and it's more prone to happen on lower flow rate pumps.

this would make sense because the flow through the 12v pump is usually only about 1.5gpm this way (I measured with my 12v pumps) I now use the 24v version and flow is 2.1gpm when pulling through the grain bed and pushing through the rims.... I dont use a 5500w element though.... way too much power for this setup which is why it will boiland cause cavitation with such little flow... mismatched components... I mentioned above that im using 1000w element myself but am in the process of switching to a longer 36" 1800w cartridge heater... If I have problems I plan on adding a simple ssvr to the element to dial down the power...

a 5500w element at 240v is best suited with a much larger AC pump.... This is what 99% of everyone else with them does.... Simply put lower flow pumps should work just as well with lower wattage elements.

I am very surprised your not denaturing your enzymes and getting low efficiency/conversion.

 

[crane]

this would make sense because the flow through the 12v pump is usually only about 1.5gpm this way (I measured with my 12v pumps) I now use the 24v version and flow is 2.1gpm when pulling through the grain bed and pushing through the rims.... I dont use a 5500w element though.... way too much power for this setup which is why it will boiland cause cavitation with such little flow... mismatched components... I mentioned above that im using 1000w element myself but am in the process of switching to a longer 36" 1800w cartridge heater... If I have problems I plan on adding a simple ssvr to the element to dial down the power...

a 5500w element at 240v is best suited with a much larger AC pump.... This is what 99% of everyone else with them does.... Simply put lower flow pumps should work just as well with lower wattage elements.

I am very surprised your not denaturing your enzymes and getting low efficiency/conversion.

My efficiency comes in at 85 to 90%. I use a period of 1 second on my PID control loop. I disagree about 5500w being a mismatch. All that means is that the proportional gain is set to a lower value than if I were running a lower Watt element. This lower proportional gain gives the same response as running a lower Watt element at a higher gain. the lower gain results in the element not being on as long per cycle. My rims tube is 2" in diameter and 26" long. There is enough thermal mass to absorb the shorter burst of heat without causing boiling.

The benefit I gain by this is that I time share my 30A supply between 2 heating elements. During mashing the rims is on at a very low duty cycle leaving most of the rest of each 1 second period for the second element to heat my spare water.

 

[augiedoggy]

My efficiency comes in at 85 to 90%. I use a period of 1 second on my PID control loop. I disagree about 5500w being a mismatch. All that means is that the proportional gain is set to a lower value than if I were running a lower Watt element. This lower proportional gain gives the same response as running a lower Watt element at a higher gain. the lower gain results in the element not being on as long per cycle. My rims tube is 2" in diameter and 26" long. There is enough thermal mass to absorb the shorter burst of heat without causing boiling.

The benefit I gain by this is that I time share my 30A supply between 2 heating elements. During mashing the rims is on at a very low duty cycle leaving most of the rest of each 1 second period for the second element to heat my spare water.

It doesnt matter how long your rims is if the element is only 12" long... the watt density and flow are the two key elements to make it work... more energy release needs more flow or it boils too quickly and scorching on the element surface can easily happen if not denature enzymes.. ... less flow works better with less energy or energy spread out in a more indirect way such as a ULWD element...
As I mentioned I also only have 30a service... my 4500w element only draws 18amps... leaving plenty for my 1000w and soon to be 1800w element (7 amp draw) to run at the same time.... no special timing needed...

some pids such as the auber units cannot be adjusted down to fire the multiple times a second you would need for what you are saying to work and if you are trying to step mash you would likely denature or scorch for sure.