Dirt cheap RIMS heater

[BetterSense]

Just an update on my cartridge heater rims setup... I recently pulled the trigger on a 36" long 1800w cartridge heater and am building a long 1" stainless pipe rims setup. I believe the longer contact time from the long ULWD element combined with the low 2gpm flow of my 24v dc pump should allow me to gently raise the temps quite a bit in one pass while still not denaturing any enzymes. I'm still waiting on a couple pipe fittings to arrive but im just lengthening my current rims setup...I does work fine but I want to be able to step mash quicker and I'm really curious if this will work better in the way that a longer plate chiller chills better than shorter chillers with more plates can...


BTW my current 24" 1000w cartridge heater is completely buildup free and clean with over 12 brews on it....hmm

You want lower power, more surface area, and higher flow, not less flow, if you are trying to minimize enzyme denaturing.

Since heating elements dissipate power more or less evenly along their length, the situation is not really comparable to plate chillers, where the heat is transferred proportional to the temperature delta between the media. Mi

 

[augiedoggy]

You want lower power, more surface area, and higher flow, not less flow, if you are trying to minimize enzyme denaturing.

Since heating elements dissipate power more or less evenly along their length, the situation is not really comparable to plate chillers, where the heat is transferred proportional to the temperature delta between the media. Mi

I dont see why I would need more flow with the 1800w watts spread over such a long 5/8 diameter element surface area? the heat transfer will be more gradual than most of the rims setups I see here? and the pid and temp probe placement should ensure the wort is not heated over the setpointat any given time... as far as I'm understanding the lower flow will do more to ensure the element isnt staying on and trying to overwork itself?

I was under the impression that a longer plate chiller provides a better cooling effect because the cooling water temp is a bit more consistent throughout the length of the chiller and thought the same would be true of a longer even contact with an element which should prevent the element from having to fire longer and try to compensate with a hotter surface temp...

 

[PlexVector]

You want lower power, more surface area, and higher flow, not less flow, if you are trying to minimize enzyme denaturing.

Since heating elements dissipate power more or less evenly along their length, the situation is not really comparable to plate chillers, where the heat is transferred proportional to the temperature delta between the media. Mi

I dont see why I would need more flow with the 1800w watts spread over such a long 5/8 diameter element surface area? the heat transfer will be more gradual than most of the rims setups I see here? and the pid and temp probe placement should ensure the wort is not heated over the setpointat any given time... as far as I'm understanding the lower flow will do more to ensure the element isnt staying on and trying to overwork itself?

I was under the impression that a longer plate chiller provides a better cooling effect because the cooling water temp is a bit more consistent throughout the length of the chiller and thought the same would be true of a longer even contact with an element which should prevent the element from having to fire longer and try to compensate with a hotter surface temp...

There is a point where the flow could be too slow that would result in plugs of overheated wort. These plugs would be next to the heating element for laminar flow. Like a low frequency instability in an amplifier. I think it would be more likely if you used a bang bang controller, which I doubt your using. Although for that length it would help if turbulence could be generated through out the tube. Like a loosely fitted SS spring along the ID of the tube so it can easily be pulled and cleaned. Just my $0.02

 

[augiedoggy]

There is a point where the flow could be too slow that would result in plugs of overheated wort. These plugs would be next to the heating element for laminar flow. Like a low frequency instability in an amplifier. I think it would be more likely if you used a bang bang controller, which I doubt your using. Although for that length it would help if turbulence could be generated through out the tube. Like a loosely fitted SS spring along the ID of the tube so it can easily be pulled and cleaned. Just my $0.02

I thought about all that... honestly I'm using the same pump with a shorter version of the same rims tube now with a smaller diameter element which means less flow and I have had no issues.... a 5/8" diameter element inside a 1" opening should cause increased velocity over the 1/2" diameter element I'm using now and to maintain 152 degrees my rims pid is usually on about 20-30% duty cycle now.... again I get zero carmelization and buildup on the element which tells me its heating gently... and I believe the right angle entrance/exit ports, inline camlock fitting in the main tube along with the probe at the far end creates enough turbulence in the flow.

I also use a flow switch which kills the element power if flow were to ever drop below a certain point but with the triple stage filtering in the false bottom its never been an issue.

 

[PlexVector]

I thought about all that... honestly I'm using the same pump with a shorter version of the same rims tube now with a smaller diameter element which means less flow and I have had no issues.... a 5/8" diameter element inside a 1" opening should cause increased velocity over the 1/2" diameter element I'm using now and to maintain 152 degrees my rims pid is usually on about 20-30% duty cycle now.... again I get zero carmelization and buildup on the element which tells me its heating gently... and I believe the right angle entrance/exit ports, inline camlock fitting in the main tube along with the probe at the far end creates enough turbulence in the flow.

I also use a flow switch which kills the element power if flow were to ever drop below a certain point but with the triple stage filtering in the false bottom its never been an issue.

What's the period of the duty cycle?
As long as the period of the duty cycle is much greater than it takes the wort to move from one end to the other then your golden. I trust your results, and I hope I'm not coming off as critical. This is more for the readers than for you.

I've not followed the thread completely, but I'd like to reproduce your RIMS if I may. Is the parts list complete?

 

[BetterSense]

I dont see why I would need more flow with the 1800w watts spread over such a long 5/8 diameter element surface area? the heat transfer will be more gradual than most of the rims setups I see here? and the pid and temp probe placement should ensure the wort is not heated over the setpointat any given time... as far as I'm understanding the lower flow will do more to ensure the element isnt staying on and trying to overwork itself?

I was under the impression that a longer plate chiller provides a better cooling effect because the cooling water temp is a bit more consistent throughout the length of the chiller and thought the same would be true of a longer even contact with an element which should prevent the element from having to fire longer and try to compensate with a hotter surface temp...

I was not saying that you actually need or will need more flow. Maybe you will have enough flow; I don't know. I was just saying that if you want to be gentle on your enzymes, you want higher flow and more turbulent flow. Your post made it sound like you thought slower flow would be better, but that's wrong. Think about the limiting case...zero or extremely low flow would denture and scorch the wort in the rims while sucking at heating your mash.

To summarize:

--Large heating surface area good because it promotes low watt density at the heating surfaces while enabling enough total power to be useful

--Low total heating power good because it promotes lower watt density (zero power=no enzyme denaturalization)

--High wort flow is good (to move the heated wort quickly through the rims tube before picking up too much heat, which both avoids denaturalization and helps with the system control)

--More turbulent flow is good (speed wort exchange at the heating element surface by avoiding boundary layers)

Full disclosure: IAAPEBIOPBOTI (I am a thermal process engineer but I only play brewer on the Internet).

 

[augiedoggy]

I was not saying that you actually need or will need more flow. Maybe you will have enough flow; I don't know. I was just saying that if you want to be gentle on your enzymes, you want higher flow and more turbulent flow. Your post made it sound like you thought slower flow would be better, but that's wrong. Think about the limiting case...zero or extremely low flow would denture and scorch the wort in the rims while sucking at heating your mash.

I get what you are saying but lets look a a coffee maker... extremely slow flow compared to my measured 2 gallons per minute, yet it stiill manages to produce nice pot of non scorched coffee... I guess im just saying if the element is low enough density you should be able to go with much less flow and longer gentle heating contact than I have now and still hold temps fine in the mashtun since its not like it loses temp that fast

 

[augiedoggy]

What's the period of the duty cycle?
As long as the period of the duty cycle is much greater than it takes the wort to move from one end to the other then your golden. I trust your results, and I hope I'm not coming off as critical. This is more for the readers than for you.

I've not followed the thread completely, but I'd like to reproduce your RIMS if I may. Is the parts list complete?

I use a mypin with actually cycles the element on and off for a % of each second so 30% on would be 30% of each second.... (from what Ive read auber units dont do this which makes me question how they could possible perform better in pid mode?)
There is a list of sorts with pics and links in my build thread linked below...(sorry for the disorganized thread) basically I used 1" stainless pipe fittings with a 1" stainless camlock for element removal for inspection and cleaning as well as the use of a 240v long single stainless steel rod cartridge heater.. these are cheap and can easily be mounted by just drilling the little bit of the base of a brass or stainless compression fitting so the element will pass completely through it and then tighten it down and thread it into the rims tube...

My first attempt was more a copy of alphaomegas design with copper pipe but I didnt use stoppers.... it works ok for the $50 I spent but the element was only 10" long and the space inside was tight creating a narrow dead space at the end with little flow and much opportunity for things to stick and scorching to occur and thats what happened when I brewed a heavy stout... heres some links of the version 2 rims I am now upgrading to the longer more powerful element for faster step mashing and goof up recovery if I mashin too cold.

 

[BetterSense]

I get what you are saying but lets look a a coffee maker... extremely slow flow compared to my measured 2 gallons per minute, yet it stiill manages to produce nice pot of non scorched coffee... I guess im just saying if the element is low enough density you should be able to go with much less flow and longer gentle heating contact than I have now and still hold temps fine in the mashtun since its not like it loses temp that fast

You are correct, if you have a large enough heating surface you can slow the speed to zero. Maybe your super long element will let you use low flow. As a first pass theory I would say doubling the element area will allow you to cut the flow in half. But low flow or high flow might cause you other control problems. Here is homework:

1) calculate area of heating element by measuring active area
2) calculate areal watt density of the element as a function of electric duty cycle
3) this step intentionally left blank
4) calculate the flow-normalized areal power density of your system. It will have dimensions something like J/m^5
5) compare that with flow-normalized areal power density of other systems that have been known to work successfully in the field

that will give you an idea if your larger element really is going to compensate for your flow rate

Or, try it and taste the beer.

I would suggest stopping with inappropriate analogies and look at the physics. Coffee makers work through boiling...it's inherent to the design. ..boiling involves latent heat and enthalpy of vaporization...nothing at all like mashing.

 

[augiedoggy]

You are correct, if you have a large enough heating surface you can slow the speed to zero. Maybe your super long element will let you use low flow. Here is homework:

1) calculate area of heating element by measuring active area
2) calculate areal watt density of the element as a function of electric duty cycle
3) this step intentionally left blank
4) calculate the flow-normalized areal power density of your system. It will have dimensions something like J/m^5*s^2
5) compare that with flow-normalized areal power density of other systems that have been known to work successfully in the field

that will give you an idea if your larger element really is going to compensate for your flow rate

Or, try it and taste the beer.

I would suggest stopping with inappropriate analogies and look at the physics. Coffee makers work through boiling...it's inherent to the design. ..boiling involves latent heat and enthalpy of vaporization.

pardon me.... I guess im just not intelligent enough to know the reasoning and analogies I had in my head is all wrong and unrelated..
I share this things on a forum such as this to discuss and learn these things ...

BTW I am stepping up the wattage and size /length of my elements in small increments and brewing with each one to gather real world experience on what does and doesnt work well since I also lack the knowledge and confidence to do and trust in any math formulas pertaining to this... sometimes (not always) whats on paper doesnt have the same real world effects and one would gather.

 

[alphaomega]

Like a loosely fitted SS spring along the ID of the tube so it can easily be pulled and cleaned. Just my $0.02

This is a really good idea. I'm thinking that finding a spring like that may be difficult, but winding it yourself from SS wire around a pipe of slightly smaller diameter than the cartridge wouldn't be that hard.

 

[PlexVector]

This is a really good idea. I'm thinking that finding a spring like that may be difficult, but winding it yourself from SS wire around a pipe of slightly smaller diameter than the cartridge wouldn't be that hard.

Copper wire like some folks use in cross flow chillers would work and may be cheaper and more easily attained.

 

[mattd2]

...

BTW I am stepping up the wattage and size /length of my elements in small increments and brewing with each one to gather real world experience on what does and doesnt work well since I also lack the knowledge and confidence to do and trust in any math formulas pertaining to this... sometimes (not always) whats on paper doesnt have the same real world effects and one would gather.

I agree, the math will only get you to a closer starting point (or will allow you to rule it out as unfeasable before spending money). Even with the best mathematical models you still need real world testing to confirm that model.

 

[augiedoggy]

well the longer and more powerful element made a large difference... I get a consistent rise of 2 degrees per minute when testing by going from 72 degrees to 152 with about 5 gallons of water but now the pid needs to be tuned..

I brewed the other day and I noticed the temps kept fluctuating about 4 degrees up and down once the setpoint was met as the pid kept under and overshooting... I'm running an autotune now to correct the situation. I had the "D" setting turned off on the pid and am sure that had a lot to do with it.

 

[augiedoggy]

autotuning the pid made all the difference.... holds temps steady withing one degree variation now... honestly the temp seemed more consistent with the smaller element but step mashing wasnt practical and it would take for ever to recover if I lost too much heat during dough in...
I dont think one degree will make a difference anyway.

 

[Freskhu]

autotuning the pid made all the difference.... holds temps steady withing one degree variation now... honestly the temp seemed more consistent with the smaller element but step mashing wasnt practical and it would take for ever to recover if I lost too much heat during dough in...
I dont think one degree will make a difference anyway.

how do you mean by autotuning?
Thank you

 

[augiedoggy]

how do you mean by autotuning?
Thank you

in the instructions there directions to hit a button for a couple seconds and then the led starts flashing... it will then use the pid formulas to develop a "routine" to best reach and maintain the setpoint for that particular load... so if you always brew 5 gallons you would run the autotune with the same amount of liquid as you would normally brew with then it remembers these values and is able to reach and hold temps more accurately next time..

 

[Freskhu]

ok thank you

 

[Freskhu]

I got an heater of 3000 watt, the normal ones. I was thinking about using with a 200mm SS tube. Its to much? Its better to use a 1kheater with 20mm tube?
I want to do mash step
Thanks

 

[Freskhu]

The heating element i got
http://www.ebay.com/itm/Water-Heate...55?pt=LH_DefaultDomain_15&hash=item2c9b8616c3

 

[stever1000]

Lots of good information here... adding this to my list too