ECM and Converted Gravity System

Lynnwill

My system is converted gravity hydronic. The gas hydronic boiler has 136,000 btuh output. Existing radiation is 1055 sqft with 23 cast iron radiators. Articles by such experts as "Steamhead" and others would indicate about a 25 gpm flow rate is necessary. Given the low energy usage I would love to use an ECM - Grundfos Alpha in fixed mode. No current ECM seems to produce the flow rates specified by experts. Question - Can an ECM work or be recommended on a high mass converted gravity system?

GW

Yes, look at Wilo Stratos, I have one at my home for a geothermal system. Grunfos makes a larger pump too. I'd recommend cleaning the goop out of tbe system and installing a magnetic device, and treating the water....these pumps aren't cheap.

icesailor

Lynnwill said:

My system is converted gravity hydronic. The gas hydronic boiler has 136,000 btuh output. Existing radiation is 1055 sqft with 23 cast iron radiators. Articles by such experts as "Steamhead" and others would indicate about a 25 gpm flow rate is necessary. Given the low energy usage I would love to use an ECM - Grundfos Alpha in fixed mode. No current ECM seems to produce the flow rates specified by experts. Question - Can an ECM work or be recommended on a high mass converted gravity system?

FWIW, that 25 GPM flow rate is water going in to the system, and water coming back out to the boiler. That's NOT 25 GPM throughout each and every part of the system. IMO, even with ODR, you need to pump hard to be sure that the properly heated water reaches every part of the system. How did you come up with the number that you need to pump 25GPM through the system? Are you confusing "Head" with Flow?

Maybe I am.

Lynnwill

See the link below:

https://heatinghelp.com/systems-help-center/sizing-circulators-for-old-gravity-hot-water-systems/

The 25gpm is water pushed into the system by the circulator at 3.5 head. I have the typical converted gravity system - 2" pipes, cast iron radiation, etc.

Paul48

Steamhead......Would re-piping this reverse return be of any benefit?

Gordy

There are very different characteristics between what was once gravity, and what now is pumped. Chances are the farthest radiators probably got the hottest the fastest. Changing to pumped the opposite happens. So orffices get flip flopped. If you read Stemheads work it is based on reality not theory.

And I dont think the dead men got all wrapped up in whether ther was a 10, or 30 degree delta t.

And you would be surprised at how clean these old systems can be.

icesailor

The difference is "Velocity". 3 GPM flowing through a 4" pipe on a standard un-pumped gravity system worked like it did because of a lack of restriction in the piping. A circulator overcomes what little restriction there is and the hottest water goes into the first radiators. Leaving the rest cold. The only thing that made it flow was the thermostat starting the burner, and the high limit shutting it off when the room with the thermostat was satisfied. When a system was properly sized and balanced, all the water in the supply was the same temperature. The water going into the return was the same, going back to the boiler.

Mother Natures Circulator. 7.5 degree rise in temperature.

Paul48

My system, with a 007, is right on the "hairy edge", pumping the farthest radiator. I don't doubt they need to be over-pumped. I was wondering if repiping off the far end of the return, back to the boiler would eliminate the need.

Gordy

where is 3 gpm comming from?

Paul48

Pumped or gravity, the water takes the path of least resistance. The convective flow in a gravity system is so slow, it doesn't allow large volumes of heated water up the supply risers. Leaving more hot water available for the next radiators in sequence. Knowing it takes the path of least resistance, pumping changes that dynamic. Much larger volumes are sent up the first risers. Those with the least resistance.

Gordy

We had variable pumping before there were even pumps on gravity systems, and they used zero electricity. The flow rate in a genuine gravity system was always changing during cycles, and in between cycles. How much depended on how big.

icesailor

Hatterasguy said:

Gordy said:

There are very different characteristics between what was once gravity, and what now is pumped. Chances are the farthest radiators probably got the hottest the fastest. Changing to pumped the opposite happens..

Well, let's take a look at that.

If a gravity system heats the farthest radiators first at 3 GPM (not sure why that would ever occur)..............why would the nearest radiators heat first at 6 GPM pumped?

Yes, lets.

Gravity systems were all "Cold Start". The boiler and system was cold or at a equilibrium system temperature. When the call was made for heat, it wasn't ever hotter in the system than the last call for heat when it shut off, and started cooling. Because of the low velocity of the heated water, in a properly piped system, no radiator could ever get hotter than another. Just like a steam system, properly vented. All radiators should get the same amount steam pressures.
The problems started when they put pumps on something that worked perfectly before.

Gravity systems are like the lake with the ice on the top. The ice can't cover the entire lake until the water at the bottom is 39 degrees. The top is 32 degrees. The bottom of the ice is 32 degrees. The top of the ice plate is the OAT. If the bottom of the lake is 40 degrees, that 40 degree water will rise and melt the ice. The bottom of the lake can be colder than 39 degrees. The water just under the ice will be 32 degrees. Its that way the world around and has been for billions of years. Its what makes life work.

If you have a gravity system that is pumped, put a Taco "I: Series 4 way mixer and let the mixer control the system temperature with the thermostat and the ODR. There'll be no hotter radiators. It will all be working in Siggy's "equilibrium".

Even in a pumped system, it is physically impossible to have the last radiator get hotter than the first.

Paul48

The flow varied according to the load.

FranklinD

My experience: 100 year old house. Can't say it was gravity for sure, as the runouts to every radiator are only 3/4". All steel pipe, mains start at 3.5" and the last radiator would represent the last 'rung' of the ladder (direct return type system). At the last radiator, the mains are 1-1/4". The supply main 'steps down' a pipe size at each tee out to a radiator (while the return main 'steps up' a size at each tee on it's way back to the boiler).

After my new boiler was installed, I put in a bumblebee on delta T mode. I have 428 sq feet EDR, my heat loss calc came up with 67k, and my boiler is 78k net, 68k doe output.

I found that the last radiator in the ladder, so to speak, wasn't getting nearly as warm as others in the system. That rad is in the living room with the thermostat. As a result the upstairs was always 6-7* warmer than downstairs.

I read Steamhead's article and set a constant speed of about 10 gpm and noticed an instant improvement. So I stepped it up a notch to 12.2 gpm (displayed) and holy cripes is the heat more even now. Apparently I just have to pump hard enough to assure a good temp at that last radiator, too slow and most of the hot water will shoot up the runouts to the second floor rads.

The last boiler we had (installed in '81 or so) was a WM VHE with a BG Series 100
pump. You could hear the water moving through the system with that one. Pretty much guaranteed even distribution that way, in my mind.

This spring I'm repacking all my hand valve stems and am going to try to rebalance the system to utilize the lower flow rate...at a 20* delta I should be able to run about 7gpm I believe. Right now the temp difference is around 10-11 degrees. The valves need to be done regardless as I think a few are way out of whack.

Lynnwill

I have to admit I may be more confused. With one way of thinking the current ECM pump rates are way too low for my system and should be therefore excluded. However with low GPM theory, ECM pumps appear to be a perfect addition.

Gordy

Has anyone actually read Steamheads article, other than FranklinD??

Lynnwill

I posted a direct link to it earlier in this chain. I also emailed Steamhead asking him to chime in.

icesailor

Like Steamhead (must have said) and FranklinD said, when you pump it properly, to match what was designed in the beginning, it thinks and acts like one of those big old round coal boiler conversions with all the big pipes coming out of the top, and big pipes returning into the bottom. A coal boiler output was controlled by how fast you burned the coal and how much air you gave it. When you converted it to oil, if you put 150,000 BTU's into the boiler, it was 150,000 BTU's whenever the burner ran. It overheated. With the coal, you just cut back on the air and let it burn down.

If you pipe it out of, and back into a tea pot boiler, you need to overcome the restriction of the highly restricted boiler. The old one had little to no restriction.

Water is like ME's lazy Brother In Law. He doesn't care how he rows in the lake, as long as he can get quickly back to shore, where he started from. He'll never travel along the entire shore of the lake unless he is forced to do so.

Or so it seems to me.

FranklinD

I admit it took a little while for me to wrap my brain around the concept of pumping *more* than the calculations called for, but in my case it does make perfect sense.

Too low a flow rate and it reverts to a gravity feed (where the hottest water seems to favor the highest radiators). Especially in the relatively cavernous main pipes I have. The heated water will stick to the top of the main and take the first runouts it comes to, leaving the last one cold (pumping too slowly = laminar flow? I'm in over my head here and possibly embarrassing myself in front of the pros). Pump it harder, more turbulence (mixing) in the big mains, hotter water to the last radiator. I'm assuming that I always had flow thru that last radiator, just not hot enough water.

icesailor

You really don't understand the concept. The input of the heat into the building matched the outflow of heat,

Gordy

I think the flow was a little more vigorous than 3gpm. That is 17.85' of pipe for 3 gal of water. Thats saying water was only moving 17.85' for 2" pipe per min thats only .29FPS.

Harvey Ramer

I will try my attempt at explaining how it works.

First, back to the original gravity. The weight of the water in the returns is what caused flow. Because the return water is cooler than the supply water. The higher the radiator is from the bottom of the boiler, the heavier the water column on the return and the faster it will want to flow. That's why a two story gravity system has orifices on the second floor to restrict the flow and balance it with the first floor.

Got that?

Enter the pump. The pump has one job, move water. That's all it does. And the water, when it leaves the discharge of the pump, will find the quickest easiest way to get back to the inlet of the pump. On that I am sure we all agree.

Now let's look at what occurs in the system whenever the pump turns on. The water will find a couple rads that offer the least restriction to flow, zip through them and into the return, through the boiler and back to the inlet of the pump.

Now, what just happened?
We have killed the gravity system and here's why. We now have heated water in the supply and the return pipes at basically the same level. Any rad that did not participate in the path of least resistance, now has an equal amount of water weight hanging in both the supply and the return pipes. Kind of like an old balance scale with an equal amount of rice in each tray.

Will they ever heat up? Yes, eventually. The main is slightly sloped and like a turtle, convective currents will begin to establish in the mains and slowly move warmer water out towards the end. As this begins to happen, it's like taking one kernel of rice at a time, out of the tray on the one side of the balance scale.

Now if we put in a pump that moves enough water, it will pump through all the rads on the first pass. Reason is, we are now moving enough water so that even the path of least resistance offers enough flow restriction to ensure we are getting flow through every other circuit.

Hope that makes sense.

Lynnwill

This all started with my having uneven heat. My 1930 system was gravity but was converted (who knows when) to a closed pumped hydronic system. Ii is now cold start with 150-200 cool gallons dropped in to my 5 gallon boiler. I am working with a local professional to establish constant circulation among other things. He is recommending an ECM (Bumblee or Alpha). The current 9 yr old system was installed with a Taco 0011 on a single heating zone - 7 degree DeltaT and high noise. With constant circ an ECM makes all kinds of sense. My conflict however was with the high GPM flow recommended in some writings for my high mass, high water content, low head system against the relatively low flow provided by ECM's.

Harvey Ramer

Hatterasguy said:

We're talking gravity feed?

Precisely.

Harvey Ramer

Constant circulation will work at a lower flow rate because the returns will all eventually assume the same temp and the same weight.

If the boiler and circ turn on and off from a call for heat, you have to use the higher flow circ.

Gordy

Thanks Harvey. And would you say in the end a true gravity system was a variable flow constant circulation system?

FranklinD

Hatteras -

I'd like to find that out too. I'm all for experimentation. Unfortunately I can't find a circ that will pump that low of a flow. I've looked. The 'Bee, at the time, was the best I could do, and maybe sometime I'll be able to try out that new Viridian circ that's replacing it. I think I read that it has an even lower minimum flow rate.

I firmly believe that once I rebuild the hand valves I will be able to successfully rebalance the system at a lower flow rate. I believe all the 2nd floor valves are wide open at this point.

It's difficult because as piped, I have minimal head, and any circ seems to go wide open. All the new near boiler piping is 1-1/4". The installer used the 007 that came with the boiler and WOW you could hear the water running through the whole house. I never even measured the delta T because I couldn't tell a difference at the boiler by touch. But at 3.5 of head, I think a 007 flows like 18-20 gpm.

I have to do some repiping this spring anyway (pumping away, bypass line, re-do the connection between the East and West mains, add an indirect, etc), as my installer wasn't as knowledgable as he claimed to be. The system works, but can certainly work better. I'm toying with the idea of P/S piping and constant circulation, and despite being cast iron my boiler has an outdoor reset feature and sensor...I haven't decided yet. At 0*f I'm running an AWT of 125-135*...way less than I (or my installer) ever figured on. But at a 10* delta T as I have now, I'm still in the 'safe' return temp area for this boiler (110* minimum).

If I repiped all the radiators as home runs with pex into manifolds, I'm sure my response time would be phenomenal... But I like my big pipes just like I like my plaster walls: thermal mass. Plus they are all that heats my basement

Lynnwill

I think the last comment by Mr. Ramer provides convergence.

FranklinD

Harvey just stated everything that was floating around in my head, but that I couldn't adequately describe. Thanks!

Harvey Ramer

Gordy said:

Thanks Harvey. And would you say in the end a true gravity system was a variable flow constant circulation system?

Absolutely. The flow and water temp go hand in hand.

Gordy

Depends how the boiler is piped. No? Bypass boiler will not see full flow. P/S same, Direct pipe not good no matter the flow. for a gravity conversion.

Harvey Ramer

@Hatterasguy
"Constant circ is perfect for such a system. Use the ECM at the lower flow rates (to get a DT of 30 across the rads) and you'll be more than pleased with it."

I am not sure how you propose to do constant circulation while maintaining a 30° Delta-T. That would require a marvelous feat of engineering while giving the laws of physics and the rules of mother nature the bird.

None of us propose to pump cold water into the boiler return. As Gordy said, we use a bypass or thermal loading valve.

Lynnwill

It is amazing to have a discussion like this among seasoned professionals. It is even more apparent as I further examine my existing setup that everybody who has a truck/card does not necessarily understand what they are doing! Thanks to all. Constant circulation and an ECM are in my immediate future.

BTW- The boiler system is currently P/S piped. No air elimination however.

Gordy

As for the air elimination it depends on how your x tank is set up. Compression, or diaphram/bladder tank.

Lynnwill

FYI - I have ODR thru a Tekmar 260

Lynnwill

One last question regarding the ECM. Delta-T (Bumblbee) or Delta-P (Alpha) . Since it will be used as a single zone circulator with fixed flow and head requirements does it matter? The Alpha tends to be a little cheaper.

Lynnwill

Sorry - fixed head, flow may be adjustable

Harvey Ramer

Lynnwill said:

One last question regarding the ECM. Delta-T (Bumblbee) or Delta-P (Alpha) . Since it will be used as a single zone circulator with fixed flow and head requirements does it matter? The Alpha tends to be a little cheaper.

I would choose the Vario. It has a knob to control infinitely variable speed. Might help get you flow rates tuned in better.

Paul48

I sit in the balanced flow camp, as opposed to the overpumped camp. It is simply a manifold system, on the grandest of scales. They were designed with balancing devices, at gravity flow.

Paul48

They knew it would not be easy, but then again, several weeks of tuning an ODR is not "easy".

Paul48

Does your name refer to the boat, or just NC?