Primary circ pump - playing with speed settings - your thoughts?

I don't have flow meter either. I have settled on low for the tank and medium for the boiler. Mostly to match the needs of the heat exhanger. I was pondering like you I figured that in the end more water at lower temps or less water at higher temps, the BTU's save may be the same so work things less, Plus the efficiency of my heat exchanger is better with the higher temp difference so that wins it.

Side note - I have the same pump as you have running my tank loop to the HX. If I set it to medium it is actually worse performance, If I set it to high the tank actually looses heat. I figure the turbulence of such a short loop creates so much head in the HX that I am actually moving less water than medium and much less than on low.

I'm very interested in this topic, and hope to have loads of hard data this season. Lacking hard data at the moment, I'll just make up some stuff and see if it sounds good.

My thought is that you want to adjust pump speed to manage temperature (or delta T) at various points in the system. Here's what I'm planning (I do some of this in a crude way with the poor man's variable speed pump):

1) Vary the boiler circ to get the desired outlet temp. The higher the boiler outlet and/or the hotter the inlet temp, the higher the boiler circ speed that you would need. Of course, your desired outlet temp might be different based on outdoor temp or storage temp or other factors.

2) If you use a circulator for inlet temperature protection like Econoburn does, then you want to vary the speed of that circulator to the lowest speed that provides your minimum safe inlet temperature. It might be nearly full speed if the return water is very cold.

3) In a primary / secondary loop or for individual zones, vary the loop or zone circ to get the desired temperature drop. No point in pushing water really fast and getting a five degree drop if the system is designed to operate well at a 20 degree drop. In some applications (storage would be one) you may want a very large temperature drop, and lower speeds will improve stratification at the expense of limiting heat transfer rates.

My thought is that almost all of these involve some sort of tradeoff. Optimizing the system is non-trivial, which is one of the reasons I've launched myself down the road of building a control platform that supports analysis and tuning. My plan is to gradually move towards more analytical tools and finally to an adaptive control system.

"Lacking hard data at the moment, I’ll just make up some stuff and see if it sounds good." I like that Nofo! I think I'll do the same...

It seems to me that you may need only lower speed at the beginning of the burn with the lower tank temp. Later in the burn as tank temp. nears boiler maximum you would want to maximize flow (higher speed). Of course this could reduce tank stratification in a pressurized tank, but not in one with a HX. Perhaps the circ. speed should be based upon tank temp.???
I think the goal should be to maximize temperature exchange, the electricity usage of one circ. between low and high speed is of little dollar value.

I really can't say as far as the primary loop speed, However I will say that my storage will gain more stratification when I run my 00R pump on low.
When on low the bottom gains only a few degrees until the top is nearing the output temp of the boiler, On high the whole tank seems to increase temp.
I will add that my stratification is not as great since NFCS took control and I believe this is because I am using differential temp to charge the tank instead
of a set-point temp. With a differential temp the pumps run continuously and with set-point control the pumps cycled on/off with much hotter water reaching
the top of the exchange coil. The jury is still out on the impact of this, I could always raise the differential temp.

Nofo brings up a good point... I have a baseboard zone that has about a 13-16 degree drop. When running from storage 165 into baseboard returns
150 degree or higher water to my tank. After a few hours my whole tank is within like 10 degrees. I would think slowing this loop would help but
how much and will it kill it's ability to use low temp water?

Good Thread and I hope to get some answers this year.

Hydronics said:

"Lacking hard data at the moment, I’ll just make up some stuff and see if it sounds good." I like that Nofo! I think I'll do the same...

It seems to me that you may need only lower speed at the beginning of the burn with the lower tank temp. Later in the burn as tank temp. nears boiler maximum you would want to maximize flow (higher speed). Of course this could reduce tank stratification in a pressurized tank, but not in one with a HX. Perhaps the circ. speed should be based upon tank temp.???
I think the goal should be to maximize temperature exchange, the electricity usage of one circ. between low and high speed is of little dollar value.

Click to expand...

I might like to politely disagree (I think).

There's at least a four-way tradeoff here with conflicting goals:

1) Maximize heat transfer rate (btu/hr)
2) Maximize of stratification (especially for less than 'fully charged' storage tanks)
3) Minimize electrical power consumption
4) Maximize boiler efficiency (higher outlet temps mean more flue loss)

My thought is that in the early phases of charging storage, you want as high a flow rate as you can get. The actual flow rate will be limited by boiler inlet protection since you're moving cold water from the bottom of storage. The goal at this point is to transfer every BTU that you can. This will give you better boiler efficiency and minimize the recharge time.

At some point, especially if you're not going to fully charge storage, you want to start bringing up the temperature going into the top of storage. All other things being equal, you do this by dropping the flow rate. Slower flow rate = more dwell time in the boiler = higher outlet temps. I think you choose a flow rate that gives you the temp you want in the top of storage.

I'm not sure the distinction is important, and perhaps the second route is the way to go. It's all influenced by other factors such as temps that might be required by other loads.

Great timing on this post Stee - I'll be watching for everyones thoughts too!

Another question (I think) related to this post is the use of a Tekmar differential controller...when not fully charging pressurized tanks, the bottom temp never comes close to the boiler outlet and the controller keeps the circ running and mixing the storage water until much of the potential stratification is lost.

"At some point, especially if you’re not going to fully charge storage, you want to start bringing up the temperature going into the top of storage. All other things being equal, you do this by dropping the flow rate. Slower flow rate = more dwell time in the boiler = higher outlet temps. I think you choose a flow rate that gives you the temp you want in the top of storage."

Good point, I'm thinking though that the actual temp supplying the storage only needs to be greater than the average tank temp. It doesn't have to be greater than the tank top. Maybe we should be thinking that the goal is to raise overall tank temp. Once the boiler to tank circ. stops, the tank will stratify on it's own.
True, less flow through boiler exchanger will produce higher temps but less flow could also produce less transfer to the tank. You may physically be moving less BTU's -higher temp yes, but less of it. Thoughts....?

I would think the way to go with a pressurized tank would be to charge the tank with whatever temp you want, Slow the flow down so you
only run the water through one time.(or less if not fully charged). Less electric and a tank of unmixed water. This may also be one of the most efficient
setups if the bottom temp is 140 or less.

]Good point, I’m thinking though that the actual temp supplying the storage only needs to be greater than the average tank temp.
It doesn’t have to be greater than the tank top. Maybe we should be thinking that the goal is to raise overall tank temp. Once the boiler to tank circ. stops, the tank will stratify on it’s own.

Click to expand...

The warmer water will only stratify up on its own if it doesn't mix to some average temp with the water it moves against. That's why turbulence is such a killer. It mixes the warmer and cooler water together and then it will sit right where it is. If the entire tank were all the same temp it won't stratify out into warmer and cooler again. But maybe I misunderstand what you mean.

There are some marvelously complicated solar heated storage systems plans I've seen on European websites using lots of hardware and electronics to put water of different temps into tall tanks at just the right point so it won't move vertically and mix up. But those looked like government funded (unaffordable) projects to me.

This is an intriguing subject for sure. I think there are a number of different best compromises depending on what kind of system you are using. At least a few different subtly connected teeter-totters.
One size will not likely fit all.

stee6043 said:

Nofo - check me here: Could one argue that a low flow rate in the early stages of a burn will allow you to more quickly get your boiler into the "sweet spot" of operational efficiency? My EKO really starts cranking at 165-170ish. Does a lower flow rate at the front end of a burn get you to the sweet spot more quickly? Does this impact overall efficiency?

Click to expand...

You're absolutely right, and in fact I do a bunch of things to help reach that goal. For one, as soon as my flue temp sensor detects that I've lit a fire (long before secondary combustion starts) I circulate water through storage and all zones so that the boiler won't see a giant slug of cold water from the zones when the circulator comes on. I usually leave out details like that to avoid confusion. For the purposes of this discussion, I was thinking about early in the storage heating cycle, when the boiler is likely running at full output.

Referencing another post - water will not self-stratify. It would be great if it did, but once a tank is mixed it will stay mixed. This isn't entirely true - as the tank loses heat, cooler water will flow down the sides and collect in the bottom. However, the temperature at the top of a mixed tank will not rise from self-stratification.

Referencing yet another post - if the only goal is to add heat to storage then all we care about is that the water being put in the top is warmer than the water that we're drawing out the bottom. That will maximize total heat transfer at the expense of stratification.

There are many schemes to increase stratification and to introduce intermediate temperature water at the right level. Not all of them come from Europe, even. I do that with the solar coils in my storage tank. A simple mechanism is a vertical pipe (large diameter, maybe 6") with holes along the sides. Introduce water at the midpoint and it can rise or sink as it needs to, and flow out through the sides at the right level. I've never found anyone who can supply engineering details about flow rates, diameter, hole sizes, or anything. I'm sure that it works, though. Hard to add to a pressurized tank.

ESBE makes a valve with a brain set up for boiler temp control and charging storage much like what is being discussed here.

http://www.esbe.se/nav5174/?prodid=4031

DenaliChuck said:

Great timing on this post Stee - I'll be watching for everyones thoughts too!

Another question (I think) related to this post is the use of a Tekmar differential controller...when not fully charging pressurized tanks, the bottom temp never comes close to the boiler outlet and the controller keeps the circ running and mixing the storage water until much of the potential stratification is lost.

Click to expand...

How about if you raise the tank sensor off the bottom of the tank to the middle. I did this and stopped the mixing with the Tekmar.

I have a dumb question on slowing down flow.
Can I slow down flow by throttling ball valves? Will this increase velocity?
I'm thinking of kinking a garden hose, velocity is increased.

kabbott said:

I would think the way to go with a pressurized tank would be to charge the tank with whatever temp you want, Slow the flow down so you
only run the water through one time.(or less if not fully charged). Less electric and a tank of unmixed water. This may also be one of the most efficient
setups if the bottom temp is 140 or less.

Click to expand...

I'm basically doing this utilizing a motorized two way valve. I have a bypass loop that goes from the supply (hot) to the return (inlet) of the boiler that circulates the water all the time the boiler is on with a 007. After the bypass loop I have the two way valve that only opens after the water hits 188 degrees to allow water to go to storage tanks. My inlet water to the boiler from the bottom of my storage is usually 110 degrees and it mixes with the bypass loop water and enters the boiler at 140ish. This is working well as I can watch the 190 degree water slowly work its way down the storage tanks as I'm charging. I only run every gallon through the boiler once this way and it keeps the tanks stratisified.

chuck172 said:

I have a dumb question on slowing down flow.
Can I slow down flow by throttling ball valves? Will this increase velocity?
I'm thinking of kinking a garden hose, velocity is increased.

Click to expand...

No dumb questions here. Any question is a good question because someone else probably has the same one.

Technically, the valve used to throttle flow is a circuit balancing valve which has ports that allow actual measurment of flow. It is much like a ball valve in construction. Probably the best off the shelf type of valve for that application is a good globe valve which has a round configuration that moves perpendicular to the seat. They will last for years in a partially open setting. A gate valve and even a ball type valve will wear with time and get to the point where it won't shut off completely.

I’m basically doing this utilizing a motorized two way valve. I have a bypass loop that goes from the supply (hot) to the return (inlet) of the boiler that circulates the water all the time the boiler is on with a 007. After the bypass loop I have the two way valve that only opens after the water hits 188 degrees to allow water to go to storage tanks. My inlet water to the boiler from the bottom of my storage is usually 110 degrees and it mixes with the bypass loop water and enters the boiler at 140ish. This is working well as I can watch the 190 degree water slowly work its way down the storage tanks as I’m charging. I only run every gallon through the boiler once this way and it keeps the tanks stratisified.

Click to expand...

This may NOT be the most efficient as far as boiler/wood but would seem to me to be the tops as far as pumping/ charging tank. No sense running
the water through 5 times when one will do.

I do not have a gasser yet and really can't say how this will effect the heat transfer efficiency of the boiler. As long as you can keep the return temp
as low as possible without going below 140(or whatever one feels is safe) I Wouldn't THINK it would be much different than a very high flow that
only raises them water a few degrees.

However as usual "Much to learn I have".

Guess I have to make a "donation" to Nofossil and try out that fancy variable speed thingy. ;-)

Two thoughts, both more theory than experimentally demonstrated...

1. The pump curves vs power draw listings I've seen for various three speed pumps all show very significant jumps in power consumption for relatively minor increases in flow. Thus I would tend to say that to minimize power consumption, you want as low a speed as you can use and get the least flow required... As an example, I looked at the specs on a Taco 0010-FC... Zero head flow rates vs power were Low - 21.5gpm / 85W (3.95W/gpm) Med - 26gpm / 100W (3.84W/gpm) High - 28.5gpm 125w (4.39W/gpm) - on this pump it looks like medium is the most efficient speed in terms of cost per gpm at the same head, but it's still a significant jump in power. Given that increasing flow increases the head resistance, a real world install would probably not see as much flow increase so the cost per gpm would probably be closer to the same, but just more watts. High is definitely a more expensive speed, even before the increased flow resistance penalty.

2. I see very little benefit to ever increasing flow resistance if it can possibly be avoided... (maybe others can explain why it might be a good thing?) The more flow resistance one has, the more work a system has to do in order to shove the BTU's around the loop... I don't see how this can ever be desirable if there are less restrictive alternatives. The only place where it is kind of tolerable is in balancing the flow through parallel loops, and that is because in most cases the restrictions involved are fairly minor, and there really isn't a better way to deal with the fact that even supposedly identical loops can have slight differences in flow resistance.

Gooserider

I would think the way to go with a pressurized tank would be to charge the tank with whatever temp you want, Slow the flow down so you only run the water through one time.(or less if not fully charged). Less electric and a tank of unmixed water. This may also be one of the most efficient setups if the bottom temp is 140 or less.

Click to expand...

Theory may be good but doesn't work in the real gasser world. The gasser wants and needs to operate at high burn until the load is exhausted. If you slow the flow down too much, the gasser will produce more btu's than you are using, boiler water temp will rise, and the gasser will idle at somewhere around 185-190F. IMO the goal is to get all the btu's the gasser is producing out into the zones or storage to prevent any idling. This flow rate will depend upon the delta-T. At below 140 system return, there will or should be some boiler return water protection mixing occurring to keep boiler return at 140 minimum (some say should be as high as 160). So the flow has to be sufficient to provide the required mixing plus take the btu's to system/storage to keep the boiler from moving into idle. This is pretty easy with low system return. Example: 140F system return, 180F boiler output, delta-T = 40, gpm to system only has to be 5 to handle up to 100,000 btu's. Keep in mind that in this example some gpm's and btu's are being used for boiler return water protection. A 15-58 on L likely will do this, depending on system head.

But as delta-T closes, it can get much more difficult. If system return is 160F or higher, no boiler return water protection is needed. If boiler output is 180F, delta-T = 20, now 10 gpm's to system are needed to move 100,000 btu's, and since no gpm's or btu's are being used for return water protection, you likely need more than 10 gpm's on a 140,000 btu rated boiler to prevent boiler temp from rising and the boiler going into idle. A system which handles 5 gpm's will see about a 3x increase in pump head as gpm's move to 10 gpm's. Depending on system design, it may take lots of circ power to achieve this.

So, once you calculate needed circ flow rate, you also need to calculate pump head at that flow rate, and then select a circ that can deliver the required flow at the calculated head. Simply increasing circ speed on a variable speed circ may not be enough. One person may have 1.25" lines and storage located right next to the boiler, maybe 50 feet equivalent pipe length. Another may have storage and boiler located quite a distance apart, and maybe 1.0" lines. There will be a radical difference in circ and speed selection in these two circumstances.

Not to be forgotten in this mix is flue temp. Although the gasser needs to operate at high burn, beginning to end, you do have some, or maybe a lot, of control over flue temp, and boiler btu output, by adjusting the draft fan damper/control. Full draft on high burn may produce flue temps of 600F and higher for some, which also means higher btu output; but slowing the draft down for a flue temp of 400-500F (some may go less) also results in lower btu output. Therefore, I think you also have to consider your target flue temp on high burn. My Tarm on very dry aspen/pine mix typically operates 400-600F flue temp over a burn with my current fixed draft damper adjustment, which is closed a moderate amount. If the draft damper was wide open I would see flue temps of 700F and for a short time even higher during highest burn periods.

The theory of this discussion is good for everyone interested, but the application will vary widely depending on the circumstances of each installation.

Some other thoughts. If system return is 140F or lower, until boiler temp gets up to at least 160F (maybe higher) at the boiler output port based on gravity heating inside the boiler, I don't see any need to operate the primary circ at all. Any operation below 160F will mostly recirculate boiler water through boiler return water protection. Nofo says he provides some system circulation so the boiler won't see a shock of cold water. This probably makes sense as he uses bang bang return water boiler protection. I don't see that need with the Termovar (or similar), as any cold water shock seems to be extremely brief, the Termovar rapidly limiting output to system and mixing to achieve 140+F return water protection.

As to artificially increasing flow resistance, I agree with Goose, the exception being to provide required boiler return water protection.

As to flow rate and impact on stratification in storage. This is subject to lots of variables, depending on each installation. For me, I have flow rates of about 2 gpm minimum (system return below 120F and boiler output 160-165F) and up to about 10 gpm on the high side. The 2 gpm is not surprising, as 2 gpm = 100,000 btu at bottom of tank 70F and boiler output 170F. The higher flow rates occur as bottom of tank temps (return to system) gets above about 160F. Flow rates increase "automatically" over this range as the Termovar increasingly opens to provide less boiler return water protection. I see very little impact on stratification in my storage tank over this flow range, and at the high end it also doesn't matter much. If bottom of tank is 160+F, top of tank input is going to be 175-195F, and at 10 gpm (600 gph), mixing the tank up to 180-190F will occur nearly as fast as trying to continue to stratify as the tank heats up to 180-190F.

In addition to the "automatic" increase in flow rates, I have a 007 in series with the 15-58. The 007 is triggered by a separate aquastat, and I currently turn on the 007 when boiler output hits 185F. This gives the gpm boost I need to handle the btu's and keep the boler from going into idle. The 007 cycles on and off as needed over a burn and does a very good job at limiting any idling. Currently, I experience idling only as bottom of tank return to boiler is about 170F +/-, meaning a delta-T of 20 or less. Since my maximum flow is about 10 gpm, delta-T = 20 = 100,000 btu, which is less than the Tarm puts out, depending on the stage of the burn. I usually do not charge storage to above 160F at bottom of tank. Top of tank will be 180-190F in this circumstance.

jebatty said:

I would think the way to go with a pressurized tank would be to charge the tank with whatever temp you want, Slow the flow down so you only run the water through one time.(or less if not fully charged). Less electric and a tank of unmixed water. This may also be one of the most efficient setups if the bottom temp is 140 or less.

Click to expand...

Theory may be good but doesn't work in the real gasser world. The gasser wants and needs to operate at high burn until the load is exhausted. If you slow the flow down too much, the gasser will produce more btu's than you are using, boiler water temp will rise, and the gasser will idle at somewhere around 185-190F. IMO the goal is to get all the btu's the gasser is producing out into the zones or storage to prevent any idling. This flow rate will depend upon the delta-T. At below 140 system return, there will or should be some boiler return water protection mixing occurring to keep boiler return at 140 minimum (some say should be as high as 160). So the flow has to be sufficient to provide the required mixing plus take the btu's to system/storage to keep the boiler from moving into idle. This is pretty easy with low system return. Example: 140F system return, 180F boiler output, delta-T = 40, gpm to system only has to be 5 to handle up to 100,000 btu's. Keep in mind that in this example some gpm's and btu's are being used for boiler return water protection. A 15-58 on L likely will do this, depending on system head.

Click to expand...

I should have been a little more specific. What I was thinking was to use two pumps, one for the tank/storage loop and one as a bypass for return temp
protection. Now with the right combination of flow between the two you SHOULD in theory be able to always guarantee above 140 to the inlet of the boiler
and control the outlet temp. Obviously it wood need to flow enough on the high side to keep from idling and the outlet temp must always be lower than
the shutdown temp of the boiler.The system I have now NEVER idles ( It can't it's manual) so being able to move all the BTU's is a necessity, But that's easy
for me to say I only have to move ~80,000 BTU's.

This should mean a very steady temp supplied to the tank, low flow at first as boiler starts up, increasing as the boiler gets cranking. Lets assume for
minute that the whole tank is 140 degrees, This will eliminate the bypass from the mix to simplify. If the gasser shuts down at 190 then just vary the flow to maintain
185 at the outlet throughout the burn, inlet temps in this case will not change because the whole tank is 140. Oversimplified I know but with 2 variable speed
pumps and 3 sensors, 2 on the inlet and 1 on the outlet this should be possible. This should also create as nice strong thermocline and keep the water from mixing no matter what size/shape the tank is.

The biggest problem In my case is my baseboard, with 180 out to baseboard it will return at like 165, so unless I used some system to return the water to
different levels in the tank it's gonna mix anyway. Radiant would be soooo much better.

The ESBE website that Heaterman referenced has a Solid Fuel Products manual available that is the most thorough I've seen yet. It covers Euro-style systems using boiler directly coupled to pressurized storage with thermostatic return temperatue protection. This goes directly to it:

http://infoweb.esbe.se/files/52901/Solid fuel products_GB_99501264_A_LR.pdf

On page 5 there is a simplified graph that demonstrates very well what Jebatty is talking about. As the burn continues along with the high temp difference between the boiler and tank temp it's easy to get the BTUs out of the boiler at a moderate flow rate of water through the boiler. But further along as the tank approaches the boiler output temp that deltaT starts to close and if the boiler is still cranking along the only way to get the BTUs out of the boiler is to increase the flow rate through it. It you don't you will start to see the flue temp rise and that's just losing heat where only Nofossil's brother can get it back.
I would guess that the computer controlled boilers (Froling, KOB, etc.) are able to efficiently lower the burn rate to help keep the heat transfer to storage on an even keel without dumping extra heat up the flue or going into idle. But the manual models that most have over on this side of the Big Water don't do that on their own.
Keep in mind if you're looking over this manual that it applies to boilers hooked up to pressurized tanks with thermostatic return temp protection. Most systems here will be adding other variables that would complicate the analysis.
Specifically they're modeling on the loading units they sell. They look to be the same design as the Termovar Loading Unit and the Laddomat units.

My argument for that is if you charged the tank with the highest temp water you can without idling the boiler, when the bottom tank temp rises sharply
as on page 5 of the referenced pdf, the tank is charged. You either need a bigger tank or shorter burn. As a general rule I never fully charge my tank, I
just charge it enough for the next day when it's cold out. This way I could draw it down to a reasonably "low" enough temp over the next day to start
the whole process over again.IOW have enough cool water in the bottom of the tank to keep boiler from idling.

Again my boiler does not idle so the only way to control temp is to vary the flow. I now do that via bang/bang but variable speed would be much better.
It just seems to me it would work for a gasser to. This all depends on being able to move the BTU's which I could see may be a problem on larger systems.

Anyway I don't want to derail the thread with tank charging theories any more than I already have. Sorry Stee6043. :cheese:

Kris

Awsome thread here! I think I am keeping up with it quite well.

I have the Laddomat installed in my sytem. I am hoping it operates much like SRoberson's 2 way valve set up. I just wish I could see the flow rates it provides during tank charging. All I have to read is a few temp gauges through the primary loop and top/middle of my tanks. I am sometimes troubled by reading that the hot water going into the top of my tanks might be at a lower temp than what the tank is currently at. So, would this then mean that I am actually cooling the tank water, or is the flow really slow at that point (Eko is still heating up) and it's barley mixing with the tanks water.

I would have to agree that you should be able to control your stack temp with water flow, but I don't have a way to try it for myself, particularly when my Laddomat is at its top speed (or fully open) when the Eko is peaking. Also, for some reason I am convinced that my laddomat cools my tanks as the eko is out of heat, because once again I see lower temps going into the top of the tank(but maybe the flow is just slowing way down?) My tanks can lose heat even when I am not using any, but yet again, this could be an insulation issue.

I really like the idea of having that second primary pump to kick in just as the eko is about to idle. For me as well, I usually only get 3/4 of my tank charged up before idleing occurs, and this might help with that- but someone said that might be a good thing.

I never saw the answer to the question about manually turning a hand valve to slow down flow rates. I was doing just that on a radiant loop to try and provide the loop with less heat, but it did not seem to effect its performance.