Pressure Drop / Head Loss for 1/2" and 3/8" tubing plotted for tubing length

[mattoak]

I'm just finishing up my very long project of building an all electric, arduino and processing controlled brewery. I'll eventually make a big post about it, but for my next planning phase I am ordering the tubing for my HERMS coil.

I read a lot about the debate between 1/2" and 3/8" stainless tubing. No conclusive answers. So I decided to calculate the head loss for both across many lengths of tubing. I am also going to be doing some heat transfer analysis to determine the exact length of tubing you need to have your HX out water be the same temp as your HLT at 1 gal/min of flow. I'll post an update when I finish that.

But for now, here is the plot. I used conservative numbers for the tubing roughness, so in reality the head loss might be slightly less than this, although I didn't account for the fitting restriction when going from 1/2" to 3/8". Y axis is pressure head in ft, and x axis is tubing length in ft. I use a wall thickness of 0.020" for each.

I am fairly confident in these calculations. If anyone is interested in seeing them, I used Mathcad and would be happy to share the file or post a screenshot of the calculations.

My conclusion: if you are going 50', you pretty much have to use 1/2". If you are going 25', as long as your pump can keep 1 gal/min at slightly over 5ft of pressure head, you should be good. But I don't know how much extra head the grain bed and connecting tubing has. I would say if you can pump 1 gal/min at 10ft from the pump stat sheet, 3/8" tubing at 25' would work.

 

[Dcpcooks]

Things I have learned over the past two years of building a HERMS system. I have found it better to monitor your wort temp at the outlet side of the herms rather than your hlt temp. You probably won't get the wort to the same temp as your hot liquor. Even with 50" of half inch tubing. If you don't use another pump to move water around the coil you'll have a large variance +/- 5-8 degrees F between your liquor and wort. Stainless isn't very efficient at heat transfer so plan on recirculating inside your HLT. Another thing to check is the opening at your compression fittings. It can be smaller than the interior diameter of your tubing. At 3/8 it can clog with grain causing a lot of frustration.
I use 50' of 3/8 with a chugger pump located 65" below my coil and I have to restrict flow to prevent a compacted bed. I use a 20 gallon Blichmann as my mash tun. I can raise temps 2 degrees per min at max power.

 

[mattoak]

I have an arduino controlled setup, so I monitor the temperature leaving the HERMS HX, as well as the temp leaving the MLT. I assume the grainbed is the weighted average of these two (keg is insulated), and I control the HLT temp so that (HX_out+HX_out+MLT_out)/3 = mash set point. So in reality, I don't really care how inefficient my HERMS coil is, since I heat the HLT until it is hot enough to get my desired mash temp. I could use 10' of tubing and it would probably just heat the HLT to near boiling. I do the weighted average because I don't want the water going into the top of the grainbed to be too hot which might happen if I just did a pure average of the 2.

Thanks for the advice though. I'm just interested in what the calculations will say. My very primitive setup had 1/4", 25' of copper tubing and the HLT had to be heated to around 180ish F to get the outlet of the HERMS coil to be 155 or so. I'm going to be using 25', just because I don't need to guess, but I'm still interested in the calculations (I'm a mechanical engineer, so its fun for me haha). Also, for me its better if the HLT stays a bit hotter than the mash temp anyway, since it will be quicker to heat it up to mashout and sparge temp after I'm done mashing.

And for your setup, if the pump is located 65" below the coil, but then that liquid uses gravity to send it back down to the pump (minus the losses through the the grain bed), it actually sees less than the vertical 5' of head loss. But assuming it didn't, if you look at this link:
http://www.chuggerpumps.com/media-center/chugger-pump-photo-gallery/397/

Which is the pump curve for chugger pumps (who knows if you have this model, but hopefully yours is close). At 10' of head from the tubing, plus 5' from the vertical height plus any losses in the grain bed, you're pump is still capable of 2.5 gallons / min, which is more than you would want and yes you would have to restrict the flow.

 

[mattoak]

Just wanted to add an update to this, regarding heat exchanger length.

For the heat transfer calculations I assumed a wall thickness of 0.035", and again a conservative SS roughness which probably makes the lengths a little longer than in real life. I calculate the length of tubing required, with an inlet temperature of 145F and a HLT temp of 155F, for the outlet wort temp to also be 155F, or the same temperature as in the HLT (for 1 gal/min, and a 10 deg F temp differential).

I would be happy to share my calculations..... At 1 gallon per minute flow rate, if you need to increase the temperature of water from 145F to 155F it would take between 1400 and 1500 W of power. In order to transfer that much energy from a non-mixed water bath at 155F (assuming uniform temp of 155F at all points of contact of the coil to water bath) to the wort flowing inside the coil of tubing which enters at 145F you would need 32 feet of 1/2" tubing, or 41 feet of 3/8" tubing.

I want to point out that one HUGE, and FALSE assumption is that the HLT water is all at 155F, around every surface of the HX coil. In reality, the top of the coil will be cooler, and the water between coils will also not be 155F but will be much cooler. It depends on how tightly coiled your HX is. I would say the one biggest improvement you could make would be to add a stirrer inside the HLT to keep the water mixed and at a uniform temperature.

But, there you have it. It would be interesting to do an actual experiment to see how close or far off these numbers are. Anyone have any first hand experience?