millcityroasters.com: real roasters for real people

Calling all engineers - airflow calculations through a damper

Postby popeye on Tue Jan 22, 2013 11:41 pm

Hi everyone,

I'm trying to figure out a way to represent airflow through my roaster so I can understand and record what I'm doing. Basically, I want to figure out how the airflow is changing as I close the damper.

I'm thinking I don't need to figure it out in absolute manner, but just normalized from 0 to 100%.

My roaster has a fairly normal damper: it is a flat plate that rotates from parallel to the airflow to completely perpendicular, blocking the exhaust pipe. I want to relate the angle of the plate to the percent of airflow.

Here's what I'm thinking. I can do the math for the simplest model, but if anyone has any numbers or equations for the more complex models I would appreciate it. Models, from simplest to most complex.

1. The plate blocks airflow proportional to its cross section. This would be a simple sine function.
2. The above, with calculations for the angle of the plate and how that affects airflow around the plate
3. The above, with calculations for friction, laminar vs. turbulent flow, etc. at this point, it may be better to go to empirical data?

I know that fluid dynamics is a pretty extensively studied field, and I'm guessing the answers are out there somewhere. I have my old college textbooks, but that was a while ago. Can anyone here help me out or have some thoughts?
Spencer Weber
popeye
 
Posts: 225
Joined: Jul 03, 2006
Location: Corpus Christi

Postby chang00 on Wed Jan 23, 2013 12:00 am

Excellent question.

I thought about this a while ago, as other Mini500/Yang Chia 800 owners knew. I still don't have an easy answer.

Several solutions:

-Anemometer, which basically measures wind speed. Least expensive option.

-Magnehelic gauge, which measures pressure difference, since the pressure inside the roasting chamber is essentially negative. However, the reasonably priced gauge is only rated to 60c, and as we know, the roasting chamber temperature is well above that. I measured various temperatures along the exhaust, and between the roasting chamber and cyclone all temperatures measure above 60c.

-Mass flow meter. I still need a source of inexpensive mass flow meter. Perhaps an automotive junk yard.

The bean volumes changes throughout roasting, not to mention chaff which also changes air flow. I am inclined to use just the anemometer, although the magnehelic gauge is just as reasonable.

From what I heard, there is one roaster manufacturer in Asia which incorporated the mass flow meter to measure damper/fan speed, as an upgrade option.
chang00
 
Posts: 540
Joined: Jul 23, 2008
Location: WCR

Postby ira on Wed Jan 23, 2013 12:39 am

The easiest to work with is probably the Ford one that pops out of the intake by removing 2 torx screws. It needs 12V DC and gives you a 0-5V signal, I'd guess that in a roaster you would only get 0 to maybe 1 Volt if you're lucky. Some work unshrouding the elements so the airflow across them is increased might help a bit. With one of those in a 4" pipe you don't want to be anywhere near the intake when the voltage gets near 5V. It's been a while since I played on the flow bench but as I recall the curve is more sensitive near 0 flow.

Ira
I wrote RoasterThing
http://www.roasterthing.com
ira
 
Posts: 440
Joined: Dec 24, 2007
Location: los angeles

Postby TomC on Wed Jan 23, 2013 2:33 am

I could borrow a Wright Respirometer from work and probably figure out a way to adapt it to Henry's Mini 500. It would have to be tested with the heat source off though, since I'm pretty sure the super heated air would pretty much destroy the delicate inner workings of the devices mercurial seal. A timer would give you the volume of air moved per unit of time (flow) extremely precisely.
"Amazingness in coffee is latent, not intrinsic"
Christopher Schooley
User avatar
TomC
Team HB
 
Posts: 4510
Joined: Jun 06, 2011
Location: San Francisco

Postby hankua on Wed Jan 23, 2013 10:00 pm

I've used the hand held Anemometer on top of the chaff collector; just to get a rough idea how the air control flapper settings relates to air speed. It's not linear as one would expect; but then what? There are roasters that have Magnehelic gauges mounted in a visible place; likely on the cold air side. Now that sounds like something useful.
User avatar
hankua
 
Posts: 446
Joined: Apr 05, 2010
Location: Jacksonville, Florida

Postby popeye on Thu Jan 24, 2013 12:04 am

Thanks for all the ideas everyone. Anyone have ideas for the theoretical side (i.e. equations)

I dont think i can measure the airflow at the chaff collector because both airflow through the cooling tray and roasting drum combine prior to that point. I'd also rather avoid buying equipment to measure the airflow.

If I do buy stuff, I basically need a pitot tube, like an aircraft wing. The pressure would allow me to calculate airspeed.

But, I figure there's probably pretty accurate equations for fluid flow through a pipe with a roaster style damper. Fluid flow is important in a large number of industries, so I'm sure that there's an equation with inputs of density, pipe diameter and outputs of airflow.

I think the equation is probably pretty close to the sine function, but I'm concerned about how friction affects the drag when the damper is only slightly open.
Spencer Weber
popeye
 
Posts: 225
Joined: Jul 03, 2006
Location: Corpus Christi

Postby sversimo on Thu Jan 24, 2013 8:46 am

Im going to look into this at a later stage in my master thesis about coffee roasting, but hopefully i have found a way "around" it, simple use a regulating switch for the fan with a RPM-meter. I haven't done any research on it yet, but hopefully the airflow changes linear with the RPM. (going to ask a professor about it later)
User avatar
sversimo
 
Posts: 230
Joined: Dec 16, 2010
Location: Norway, Oslo
bootcampbarista.com: master the craft like never before
bootcampbarista.com: master the craft like never before

Postby Jisgren on Thu Jan 24, 2013 9:51 am

Jisgren
 
Posts: 29
Joined: Nov 21, 2012
Location: Houston

Postby popeye on Wed Jan 30, 2013 6:24 pm

Thanks, I've been messing around with that site and a few others and I've found the coefficients I need. Unfortunately, butterfly valves work most predictably in the middle of their range - stuff gets funky when you just crack them open slightly.
Spencer Weber
popeye
 
Posts: 225
Joined: Jul 03, 2006
Location: Corpus Christi

Postby JK on Sat Feb 02, 2013 11:12 pm

How about Smoke Pencil?
They use them for HVAC

You can run smoke through the machine and see how it flows if that will help..
-----------------------------
I'm on a Mission from God!
JK
 
Posts: 238
Joined: Feb 09, 2012
Location: Rex, Georgia USA