Thoughts on variable-speed pumping and air removal

DanHolohan

Have you considered the effect of slowing down a circulator on the removal of air from the system. The standard for keeping air from forming pockets has long been to keep the velocity above two feet per second. And to prevent air from coming out of solution, we've avoided velocities slower than six inches per second, although, I'm not sure how that plays into Henry's Law.



When we reduce pump head with variable-speed pumps, air will come out of solution (Henry's Law again). The reverse happens when we increase head, of course.



Have you had any experiences with any of this? Will variable-speed pumps change the way we remove air from hydronic systems?



Just thinking out loud here.



Thoughts?

Gordan

Perhaps the discussion should be broken up by control strategy for VS

For instance, it would intuitively seem that constant dP pumps would not affect the low limit on flow velocity.

DanHolohan

Good point.

I'm working on a book about air in hydronic systems (from lots of angles) and this is why I'm asking the question. I'm at the point where my mind is ranging over lots of things, variable speed being just one of them.  Is there a low-limit on velocity with these new pumps?

Paul Pollets

Test it

Why don't you have your buddy at Spirotherm hook up a VS pump to their Spirovent display and inject air at various speeds to observe?



I'd certainly like to know how the pump will affect the preliminary air purge of the system at slow pump speeds.

DanHolohan

Thanks, Paul.

That would be interesting to see, but I'm thinking about the slower flow rates out in the system, especially in a system that's not well balanced. Not that there are many of those out there. :-)

SpeyFitter

Are you pondering too much?

What happens when the system is stagnant and there is no flow (no call for heat)? Any Air, that was not expelled by a Microbubble air seperator will accumulate at high points, so why worry about low flow conditions?

DanHolohan

Because

I'm working on a book about air in hydronic systems, and I like to consider all the things that might affect the subject on which I am writing. Variable-speed pumps are relatively new to hydronics and I'm wondering what impact, if any, they'll might have on air in the system. It's how I work. Years ago, when I was working on Pumping Away, I asked similar questions and it led me focus on the new (at the time) high-head pumps and how their location might cause air-related problems.



The question you pose, I can answer, and will. It also goes to balance, by the way.



I like pondering, Scott.  

Steamhead

This might be related

Those of us who have been around a while will remember my putting forth the concept of slowing circulation in gravity-converted systems to increase their ∆T and make the water circulate better thru the boiler and rads. One of the questions asked at that time was how this slower circulation would affect air removal.



Gordon and I find that these systems do very well with smaller circs as long as the installation includes a good air separator. Once the system is filled and the water heated, any small stray air bubbles are absorbed back into the water as it cools. On the next cycle, the separator gets rid of it- as long as you're Pumping Away. This works with both diaphragm and plain steel expansion tanks.



This would probably work on newer systems as well. Pumping Away from a good air separator is key- but we know that.

bob_46

Ruminating

Good question, I would think that one of the worst case scenarios is cast iron radiatorswith TRV's or panel radiators with TRV's. Radiators look like they would be good air separators. When the modulating valve is just cracked velocity in the radiator must be extremely slow in existing systems and we don't seem to have a problem.

Robert O'Brien

Purely anecdotal

but I've noticed no difference between ECM and conventional circs in this regard. These have all been  loop and /or monoflow systems

SpeyFitter

Nothing wrong with Pondering

Hey Dan

Sorry, didn't mean to infer you ponder "too much," as I ponder about heating things ALL THE TIME. It's hard not to ponder, really.

Good luck with the book - You've sold one already.

Thanks,

Future customer.

Steve Ebels_3

Haven't found any difference Dan

I do think however that purging becomes more important along with a properly located microbubble type air eliminator.  We do not rely on the system circ(s) to perform purging duties. Not what they are designed for.  A good power purge with high pressure/high velocity water flow does wonders in a lot of ways.

DanHolohan

What sort of air separtor are you using?

Do you use the same type of air separator with both types of circulators? Thanks.

DanHolohan

I didn't take it that way, Scott.

No need to apologize. I was just trying to explain. It's a long process and I'm looking at a lot of things that happened in the past, as well as what's happening now. Lots of good history here.



Here's a good story for you.  Amtrol manufactured the first cylindrical Extrol tanks using an overstock of baby-carriage rims. When the competition arrived, Amtrol found small holes drilled into the competitor's steel hoops, mimicking what the competitor thought was a necessary feature.



Isn't that delicious?



Stay tuned. :-)

Zippygdgddg

ALPHA on Spirotherm display

Hello-

I have replaced the pump on my Spirovent display and have run it with many different speeds of the ALPHA pump (high velocity, auto adapt, speed 1/2/3 and the constant head pressures. I have also played around with the ball valve (restricting flow, increasing, etc) in an effort to not remove air. Each time the variable speed pump has removed the air along with the spirovent. I have not been able to trick it yet. Dan, if you have an actual appplication, i would love to send you the variable pump to see how it performs with the air eliminator (spirovent or not)

DanHolohan

That's helpful, Scott.

Thanks for the input. I'm mostly looking for anecdotal information right now, and you've  done here what Paul suggested I ask my friends at Spirovent to do. It's good to see you guys were checking the same things I've been thinking about, but then, you work for a classy outfit. Thanks.

Glen

Well Dan -

may I play devils advocate? Most things in this world we call hydronics have been grossly over sized, over estimated, over pumped and and and ... What if our standards for fps and the notions of air removal are equally subject to hyperbole and exaggeration? Now  Mr Henry was one smart dude - and we are subject to his axioms every day; but I just found a little tidbit on this very subject the other day - which suggested strongly that the "very best" air removal was at a leisurely half foot per second. At that rate pipe sizing goes back to "gravity" sizing; so the discussion must focus not just on fps, but resulting gpm, pipe sizing etc, but still embrace our task - which is to deliver comfort  to the HO. Love your questions - always make me think.

DanHolohan

Thanks, Glen.

Can you share were that tidbit about velocity came from? 

Glen

It's at work -

I will scan Monday and send it to you.

DanHolohan

Thanks.

Every bit helps. 

hot_rod

looking inside

I cut a clear window into one of the brass micro bubble air removal devices. I can watch the way bubbles cling to the media at various flow rates.I suspect as long as the fluid moves the air along with it the eliminator will grab it. Temperature and pressure play into the equasion also.In drainback solar piping I've noticed 2 fps still moves the large air bubble along the pipe, even in a vertical drop. I built a clear tube version of a DB piping to watch that work.Caleffi has done a lot of developing and testing in the field of air elimination. If you have a specific set of conditions you would like tested let me know. hr

DanHolohan

Tanks, HR.

I'm thinking about a poorly balanced, two-pipe, direct-return system, where the water often slows in the circuits furthest from the boiler (top floors?). I think this is the place where low-velocity flow can cause air to come out of solution and gather to a point where it stops flow, especially in a pipe heading downward. I think the modern air separators thrive on the low flow.



Of course, with a single-speed pump, this can also be a problem (and is) in a poorly balanced system. Just wondering whether variable speed will change any of that.



Thanks for thinking with me!

NRT_Rob

when I read your post

I thought immediately... people really do the 2 ft/second minimum?



those of us in radiant don't, ever. that would be over 1 GPM in a 1/2" PEX loop, and very few people are using that as a minimum per loop. Heck, I'd say I almost never hit 1 GPM/loop.



with a microbubble resorber, we have had no air problems on any systems other than non-purging, no-flow situations or system leaks. I don't expect to see any on the variable speed pumping systems either.

Devan

LLH

I know your focus is on variable speed pumps, but is not a properly designed low loss header achieving air removal (and hydraulic separation) via the same principle. Potentially low velocity into the header, much like a very , very small gravity pipe system.

Glen Aspen_2

low flow -

Pgs 93/94 "Hydronic Heating and Cooling",(2007). Pipe Trades Joint Training Committee, United Association of Plumbers and Pipefitters. The half foot per second comes from an excerpt on a discussion of air seperators; it is within the seperator that the .5/ft/sec is deemed to be appropriate not in the piping system. Sigh - the danger of speed reading.

bob_46

Velocity low limit

Dan, you asked if there is a low limit for velocity on these pumps? There is a minimum head for the alpha when in auto-adapt mode , it's 4'. I noticed that the engineers that wrote the algorithm for the alpha use a factor called the K factor for system resistance.

They show K= flow divided by the square root of pressure. I am not familiar with this term.

I am wondering if it could be the reciprocal of head. I also wonder if the square root of pressure should be the 1.75 root. Maybe they are over estimating system resistance

like the syzer.