RIMS Exit Temp Higher than Setpoint

[jonathonhagerty]

Hi Homebrewers,

I am testing out my new RIMS tube with water tonight. I found I can hold 149 quite well.

To simulate a mash out, I set the pid to 168. After a few minutes, the temperature rose to 160. This is measured by a PT100 about 1.5 inches away from the end of the element inside the tube. However, my handheld thermometer read 175 at the sparge tube which resupplies the mashtun.

Why is the temperature so much higher during this ramp? Will it lead to scorching? Could it mean dead zones in my RIMS tube? I can’t imagine the PT100 is lagging so far behind the actual temperature when ramping, can it? I didn’t bother to wait to see if the entire mashtun stabilized. The rest of the liquid was still a few degrees below the pid probe reading (I’ll get the answer on this point tomorrow)

When I held at 149, the sparge tube temp was about 147. A few degrees lower due to the lack of insulation. (By the way I had to calibrate/correct my pid +5 to obtain this).

Thanks, Jon

 

[Vale71]

You must have a measurement error. There is nothing that can keep heating the wort up after the end of the element, it can only get cooler and once temperature has stabilized for a handful of seconds the reaction time of your probe plays no role at all.

 

[jonathonhagerty]

Hi Vale,
I see, thanks for the explanation. I’ve heard of ground loop issues, but I’m unclear on what exactly they are. I see my element is grounded to the tube. My probe is a three wire RTD, with Teflon coating (no metal shielding). How can I tell if I’m having a ground loop issue?

 

[Vale71]

I would test the resistance of the probe first. You should have two same-colored wires and one differently colored. The resistance through the different colored one and any of the two same-colored ones should measure the same. I've had failed probes where this was not the case and this can throw the measurement off by a lot.
If this checks out OK I would also check the actual value at different temperatures against a reference table such as this one.
If you suspect electrical interference through the element the simplest test for it is to just unplug the element while leaving everything else running and see if the problem goes away.

 

[processhead]

Test that your probe and thermometer are in agreement. Place them both in a container of hot water and see if they read the same temperature. Make sure they are both close to each other in the container. If you have a stir plate, that helps ensure uniform temps in the test container.

If the probe and your thermometer are not reading the same in the test container, you need a known to be accurate thermometer to find out which one is in error.

 

[jonathonhagerty]

Vale,
The resistance of red to white wires was 106.3 ohms at 75F. According to https://technitemp.com/rtd-pt100-temperature-resistance-table/, 75F should have a resistance of ~109.3 ohms.

Processhead,
I will check the difference in ice water.

I have two RTD cables I can try. One has metal shielding, one teflon.

UPDATE: Processhead, there is no difference in ice water, the probe matches my thermopen.

Vale, I just tried ramping up from 86 to 94F. I had very slow flow of 1 qt/min. I noticed the same temperature spike. The thermopen in the sparge hose read 110 while the probe read 91...92.... But when I increase flow quite fast (2 gal/min), there was not overshoot.

Why does my probe fail to see this overshoot?

The overshoot is happening because my probe is not detecting any overshoot. The probe still thinks 91...92, but the output is way higher. If the probe detected overshoot, it could throttle back the element.

 

[processhead]

Processhead,
I will check the difference in ice water.

I have two RTD cables I can try. One has metal shielding, one teflon.

Checking in ice water will tell if your probe and thermometer agree at that temperature. You are probably better checking them at a higher temperature closer to mashing temps, though.

Linearity issues can affect probes and thermometers where they can read accurately at the low end but be off at the high end. You really want your accuracy at higher temps closer to your actual process.

 

[jonathonhagerty]

At 150F, both the probe and my thermometer agreed.

At slow flow, could the hot water be trickling by (or sneaking by) the probe, like in a thin hot stream?

I want this slow flow.

What if I turn the RIMS exit port downwards? So the hot water may saturate the top of the tube instead of sneaking by the probe? Now it is facing up. Let me know if this idea is wrong. My RIMS diameter is 2''. Does 2'' diameter typically keep uniform temperature? Or are they prone to temperature stratification?

 

[Ahks]

Is this RTD in a thermowell?

What's the configuration of your tube? Any chance you have an air bubble preventing an accurate read in the tube?

 

[processhead]

At 150F, both the probe and my thermometer agreed.

At slow flow, could the hot water be trickling by (or sneaking by) the probe, like in a thin hot stream?

I want this slow flow.

What if I turn the RIMS exit port downwards? So the hot water may saturate the top of the tube instead of sneaking by the probe? Now it is facing up. Let me know if this idea is wrong. My RIMS diameter is 2''. Does 2'' diameter typically keep uniform temperature? Or are they prone to temperature stratification?

Its good your thermometer and probe readings agree. Nothing is more frustrating than trying to troubleshoot a problem when your tools are lying to you

If you mean your heater tube, you want your exit port facing up. Generally speaking, in fluid systems you want the flow direction upwards as much as it is possible to minimize air entrapment in the system.

 

[jonathonhagerty]

Ahks,
My rims tube is 2 inches in diameter. It is horizontal. With the inlet facing down and the outlet facing up. The element is on the inlet side. 1350W foldback 30 watts per inch density. The probe is on the outlet side. The tube is tri clover. The probe is a tri clover pt100, 2.3 inches long. The element is 12.5 inches long. The entire tube length is 16 inches.

 

[processhead]

Is this RTD in a thermowell?

I was wondering about the thermowell too. They have enough thermal mass that they can cause some lag during temperature steps.

Any lag can cause temperature overshoot, but it should eventually stabilize and drop back to around the setpoint.

 

[LuckyBeagleBrewing]

Re-tune your PID. I ran in to that problem when I had to adjust flow rates and going through the tuning procedure again at the flow rate I was using seem to fix the problem.

 

[Vale71]

Ahks,
My rims tube is 2 inches in diameter. It is horizontal. With the inlet facing down and the outlet facing up. The element is on the inlet side. 1350W foldback 30 watts per inch density. The probe is on the outlet side. The tube is tri clover. The probe is a tri clover pt100, 2.3 inches long. The element is 12.5 inches long. The entire tube length is 16 inches.

Hard to say for sure without seeing an actual picture of your setup, but mounting the tube horizontally has the potential for creating stationary air pockets. If your probe were to be caught in an air pocket then it'd have no chance of measuring the actual wort temperature. If changing the flow rate changes the probe's behavior it sounds supiciously like an air pocket that gets dislodged only above a certain flow rate. Mounting the tube vertically is always the best solution to prevent such issues.

 

[Vale71]

Re-tune your PID. I ran in to that problem when I had to adjust flow rates and going through the tuning procedure again at the flow rate I was using seem to fix the problem.

Improperly set PID parameters can surely cause the system to overshoot the target temp but it would not prevent the PID from displaying the actual wort temperature, so that you would actually see the overshoot on the PID dispaly without the need for a separate thermometer.
What the OP has is definitely a measurement error, trying to tune the PID before fixing the problem would be just a waste of time and possibly make things even worse as auto-tuning relies on the system working properly to be succesful.

 

[doug293cz]

Vale,
I tried measuring the resistance from white to red on the same end. The multimeter said 'open' (infinite), but flashed a value for a split second before reporting 'open'. The split second values ranged from 120-180 (I tried many times).

...

Sounds like you have an intermittent connection in the RTD, at best. So, sometimes it works right, and sometimes not. Nothing should ever read open when measuring between any two leads of an RTD. The two same color leads should read close to 0, and both of the different color options should read about 100 ohms.

Brew on

 

[jonathonhagerty]

On the air bubbles, I pivoted my RIMS tube back and forth (facing vertical, then horizontal) a few times, it released some air bubbles. I hoped this would solve the issue.

Then I ramped from 73F to 80F on the PID. Monitoring the sparge tube still showed temperatures of 90-100F.

Doug, resistance of red to white wires, 106.3 at 75F. According to https://technitemp.com/rtd-pt100-temperature-resistance-table/, 75F should have a resistance of ~109.3 ohms.

Next I tried pivoting the RIMS tube while it was ramping from 80F to 100F. When horizontal, I saw the overshoot. When vertical, the probe detected the huge overshoot. The PID actually displayed the overshoot temperature. This seems to be a temperature stratification/zoning issue in my RIMS tube. Could spreading out the element slightly fix this? The delta of temperature stratification is about 15-20 F, quite a lot. I'm not sure if higher flow and bending the element can fix *that* much temperature difference.

Maybe I can get a longer probe that lies *just* inbetween the element wire.

 

[processhead]

I'm glad you posted a picture of your setup. It has me wondering if your temp probe is long enough to reach the flow stream exiting the top port of the tube.
Any orientation of the probe that positions it out of flow stream has the potential to record a temperature that is not representative of the actual temperature exiting the heater tube. Given a choice, I prefer plumbing the probe in a position further from the element and directly in the flow stream where it would be less prone to spot heating.

Also, what is the approximate flow rate you are using during your tests?

 

[jonathonhagerty]

I'm aiming for 1 qt/min. I've never done this before. But I'd like to go slow.

Vertically oriented, I'm detecting the hotter liquid. This would be safer, it'd prevent scorching. I tried a test vertically. Ramp from 100 to 149. The PID reached 149 very quickly (maybe 5-10 minutes). I expected the actual tube temperature to catch up. The actual tube temperature only reached 140. I waited much longer, and it would not go any higher than 140. I was hoping the hot water on top would exchange heat to the flow path. But maybe it's being lost somewhere else. I wish I had gone with 1.5in tube.

 

[processhead]

I expected the actual tube temperature to catch up. The actual tube temperature only reached 140. I waited much longer, and it would not go any higher than 140.

So when you say the actual tube temperature, are you measuring the surface temperature?

 

[jonathonhagerty]

I mean the temperature of water coming out of the tube that sparges at the top of the mash tun.

 

[Vale71]

Your probe placement is less than ideal and is definitely causing these false readings. If the probe reads 149 the PID will start regulating even if the wort is actually leaving the tube at only 140 and you will never reach your target temp. With your setup probe and outlet should switch places and you should ensure that your probe is long enough so that the tip sits right in the middle of the RIMS element and is thus fully exposed to the flowing wort. A vertical placement of the tube is still the best option to prevent air pocket formation.

 

[Gnomebrewer]

Hotter wort will rise. With your tube in the horizontal position, the hotter wort leaves the tube, whilst cooler wort is near the probe. You will notice this more with slower flow rates and bigger temperature ramps, and it is significant. In the vertical position this issue is eliminated.

 

[Vale71]

There's no way you could get stratification when even with a slow (but still turbulent) flow you're still replacing the full volume of wort in the tube several times per minute. Air pockets on the other hand are a real possibility and should be avoided as in the worst case incomplete wetting of the element might cause localized overheating.

 

[Gnomebrewer]

There's no way you could get stratification

I disagree. Look at the photo - while I agree that the setup could cause some air pockets, there's no way the probe can be in one. You're most certainly not necessarily replacing the full volume of wort in the tube - with most RIMS setups there will be dead pockets, like backwaters in rivers, due to the flow of wort and stratification. The preferential flow in the horizontal system will be towards the top of the tube, above the element (the incoming wort is pumped upwards, heating causes movement upwards, and the outlet is at the top - all factors are leading to wort sitting high in the tube). The backend of the tube, where the probe goes in through to the outlet, will see very little flow. Hence the probe is partially in a 'backwater' of little flow, and also below the main heat and flow. In fact, looking again at the photo, if the tube is held vertical it would lead to an air-pocket at the top of the tube where the probe goes in.

 

[processhead]

Try two things.
1. Swap the temp probe and outlet port in their respective locations on the heater tube.
2 Orient the heater tube vertically with the outlet point pointing up.

Make sure that the tip of the temp. probe extends past the edge of the vertical run of the heater tube into the flow stream.
Do another test run and see how it goes.

 

[Vale71]

I disagree.

And I think that applying fluid dynamics that might work for a large body of water such as a river to a 2" ID tube is a bit disingenious. In the specific case being discussed you would actually have to have a 15°F gradient in the space of a few millimeters and that's also utterly unrealistic assuming a uniform medium, whereas an air pocket could very well account for that.
Setting that aside I think we can all agree that the best solution for the OP is to mount the tube vertically with the outlet on top and a properly sized probe inserted through the side port.

 

[jonathonhagerty]

Hi guys,
I appreciate your eagerness to debug more, you may be on to something. Unfortunately, I’m not game. It’s just hard to find the time. We just had a new baby . I’m going to return it and purchase their smaller tube to eliminate the issue altogether. Thanks Vale, processhead, and Doug.

Thanks, Jon

 

[processhead]

Hi guys,
I appreciate your eagerness to debug more, you may be on to something. Unfortunately, I’m not game. It’s just hard to find the time. We just had a new baby . I’m going to return it and purchase their smaller tube to eliminate the issue altogether. Thanks Vale, processhead, and Doug.

Thanks, Jon

First of all, Congrats on the new baby!

Second, Be advised that the smaller tube may not solve your issues unless it somehow addresses the fundamental issues we are discussing.
Sure hope you could see your way to try the recent recommendations.
After all, even new-borns gotta sleep sometime, although it may not seem like it.

 

[BrunDog]

Guys,

I have done a lot of research and testing on RIMs tube design. This isn’t a PID, probe, or calibration issue. The OP and @Gnomebrewer are correct. At slow flow in a horizontal layout, the heated liquid rises, escaping the exit without registering against the probe. The larger tube OD and slow flow exacerbated the problem. You likely got local boiling too.

RIMs tube’s should be mounted vertically (performance and safety purposes), the element as large surface area for wattage as possible (low density), the tube as small as possible. Baffles or some type of flow disturbance helps a lot (solid stainless wire interwoven with the element helps quite a bit.

Congrats on the baby!

 

[jonathonhagerty]

Hi Brundog. Thanks for taking a look. You have a great website on automation, and a great reputation on this site. I automate for a living, and damn! Good job.

When I tested vertically, the probe detected the peak temperature, and the actual exit temp (sparge tube in the MT) was lower. Lower than the probe temp. Am I just fighting a losing battle here with stratification? Will some stainless wire really unify a 15F delta? It'd have to be some thick wire. Does your quadzilla's baffles have some thick fins?

Thanks, she's rockin my world. Doesn't matter how my day went, it's amazing hanging with them afterwards.

 

[BrunDog]

Hi Brundog. Thanks for taking a look. You have a great website on automation, and a great reputation on this site. I automate for a living, and damn! Good job.

When I tested vertically, the probe detected the peak temperature, and the actual exit temp (sparge tube in the MT) was lower. Lower than the probe temp. Am I just fighting a losing battle here with stratification? Will some stainless wire really unify a 15F delta? It'd have to be some thick wire. Does your quadzilla's baffles have some thick fins? I'd love to go your quadzilla route, but man this RIMS concept is giving me issues. I'm considering using my CFC as a herms coil. And using a 240V RIMS to heat the strike volume inside the MLT to the higher temp which, after doughing in, will hit thermal equilibrium at my mash temp. Just worried I'm overcomplicating this as my first e-brewing-venture. Is it overkill to use the RIMS only for raising the strike temperature to hit dough in mash-temp spot on? I know you're anti herms. I want your honest opinion. Because I want to use my CFC as a HERMS coil (just so I don't scorch anything). I plan on mashing out, but not really step mashes. With only 4gal left in the HLT, reaching 170 from 149 should be quick with 5500W and HLT recirc.

Hi. Thank you for the compliments! What you are seeing is the same phenomenon - the hot liquid is rising along the element and heating the sensor. The cooler water slides along the sides, mixes with the heated liquid, and exits ad a lower temp. The biggest issue is your oversized tube - there is just too much deadspace. Maybe at high flow it would work, but at 1 QPM, you will get lots of stratification. The key is create as much mixing / turbulence as possible. I actually tested a re circulation circuit with a pump that took most of the exit water and pumped it back into the tube again - this worked great but isn't practical.

I used to use a 5500W element in a 1.5" tube to heat sparge water on the fly. I used a relatively thick stainless wire (maybe 0.063" or so), woven in a zig-zag around the element. This helped quite a bit compared to the element alone. There is a picture on my rig's build thread - I'll try to find it. But again, downsize to a smaller tube first. I'd love or you to have a QuadZilla but I think you can honestly do what you want with your current element.

That said, the QZ has thin baffles, but it has holes in it to permit flow but mix up the flow. It has three baffles - they are movable but we suggest one in the middle, one at the 75% mark, and another at the top. They are probably overkill but there is no stratification!

I wouldn't go HERMs - but then you know I'm already biased. I think you can make your setup work. How many watts is the element?

Edit: Here is the pic: https://www.homebrewtalk.com/forum/threads/brundog-50a-erig-no-hlt-for-me.542899/page-3#post-7285005

 

[jonathonhagerty]

5500W at 240V, or 1360W at 120V

 

[Vale71]

When I tested vertically, the probe detected the peak temperature, and the actual exit temp (sparge tube in the MT) was lower. Lower than the probe temp.

Which is what you should expect since you have the wort running through quite a length of silicone tubing with no insulation. Of course this might be coumpounded by measurement errors if your hand-held thermometer and your PID probe are not in agreement, so the difference might not be actually as large. You could try putting some insulation on the output hose (while you're at it you could also insulate the other hoses as you'll have heat loss in all of them) and see by how much the temp difference decreases. Using smaller ID hoses (i.e. smaller surface area for the same lenght) and possibly shortening them would also reduce thermal loss.

 

[jonathonhagerty]

Brundog,

If I run the RIMS at higher flow, and vertically, I avoid this stratification.

I'm considering using my CFC as a herms coil for single infusion mashes. HLT would heat all water to mash temp. Then RIMS to heat the strike volume the rest of the way so dough in will hit mash temp. Would this work? I'd then switch back to the CFC to recirculate everything at one temp. Do others do this? Would it work? I have the MoreBeer copper CFC.

 

[BrunDog]

Brundog,

If I run the RIMS at higher flow, and vertically, I avoid this stratification.

I'm considering using my CFC as a herms coil for single infusion mashes. HLT would heat all water to mash temp. Then RIMS to heat the strike volume the rest of the way so dough in will hit mash temp. Would this work? I'd then switch back to the CFC to recirculate everything at one temp. Do others do this? Would it work? I have the MoreBeer copper CFC.

Yes, because there is more turbulence and mixing. 1 qpm is very slow, so lots of heat gets dumped into a small volume of liquid.

It could work but sounds awefully complex. I think I have heard of hybrid systems but I honestly dont see the benefit. I'd go all HERMs or all RIMs. I personally really enjoy the best of all worlds with a high power RIMs element - heat water at 5500W and wort at 1350W. Clean, simple, fast, and no downside IMO. Again, that 2" tube is too big though.

 

[processhead]

At the lower flow rates he is running, I can see the potential benefit of a smaller heater tube with less volume.

 

[postalbunny]

I can think of only two reasons your PID would be showing lower temp than a thermopen at the mash tun where wort is returning... there is no way the wort is actually gaining heat after it exits the tube. Your PID must not be able to grab the temp

1) You may have air pockets... if there is an air pocket over by the probe it may not be reading correctly. You may be able to fix this by tilting your setup so that the right/entrance is lower than the left/exit side. That way any air is coaxed out if it's in the middle. I'm not a fan of the probe being on the far left, you may end up with an air pocket there.

2) the wort by your probe is possibly not in the flow... air pocket or a current may be taking wort straight out and not replacing/swirling where the probe is. This is probably what you have if changing the flow rate makes everything work, and slower flows stop working.

3) Just a thought, if you're using a thermowell then the RTD may not be getting good contact. You can put some thermal paste/grease in the well that will help conduct the heat between the walls and the probe. You can verify this by yanking your RTD probe out and using your thermopen there. If the pen reads the same as the PID at the thermowell then the issue has something to-do with that thermowell and/or it's placement.

Depending on your probe length... maybe do what I've done? I put the probe where you normally have the "egress" port and put the exit for the wort on the end of the tube. I titled the whole thing and that way air flows out the exit and ensures no air pockets. My probe is just long enough to stick out into the middle of the tube and is about 1/2" away from the end of the element. Therefore i'm measuring the temp of the liquid directly at the element. This helps if there is a pump failure, the probe will detect the heat from the element and reach set-point... turning itself off and preventing scorching. (i also have a flow switch as an extra precaution)

*Mash enters left side horizontally through flow switch, entire tube is tilted slightly up, just after element is the RTD probe on the right, then the "top" of the tube on the far right has a 90deg exit back to the mash tun.

 

[augiedoggy]

View attachment 623809

Hotter wort will rise. With your tube in the horizontal position, the hotter wort leaves the tube, whilst cooler wort is near the probe. You will notice this more with slower flow rates and bigger temperature ramps, and it is significant. In the vertical position this issue is eliminated.

I dont have this issue... I have a horizontal rims with 1.5-1.8 measured flowrate. my probe is only about 1/2" from element though. its mounted in the end of the tube not out the top. I believe this is because along with the length /additional heating time, the inside diameter of my rims does not have the dead space to allow as much if any uneven heating.

I think the OP is on the right track to think a 2" tube with that short element is leaving a lot of unheated liquid past and temps just arent even at the probe area.

the element is 36" long to temps are totally even and reached by the time the liquid gets from one end to the other.

 

[BrunDog]

Augie are you saying your flow is 1.5 - 1.8 gal per min or qt per min? I’ll assume gals. At this flow rate, you are right - there is a lot of mixing and turbulence that assures heat is more evenly distributed into the liquid. Especially given your cartridge design.

Now contrast that with a very slow flow rate at 1 qpm. The liquid is moving so slowly it is not mixing and totally stratifying. This is the issue the OP had.