Flat Plate HX works - My tank is charging!

[WoodNotOil]

Here is a link to pictures: (broken link removed to http://woodnotoil.googlepages.com/constructionlog) (scroll down to the bottom two pictures) This shows the piping by the pumps below tank level and where the pipes come out of the lid.

There is definite movement in the pipes when the ball valve is not closed because the are hot to the touch all the time when open and cold to the touch when hot. However, I will concede this is likely not the only thing going on. I am still not sure how well the probe well is indicating the tank temp and what is occuring in terms of mixing. These are not easy things to diagnose.

When I was first floating the idea of "dip tubus" with a flat plate someone in this forum warned me how easily these lines would start siphoning.

pybyr - Basically the siphoning in this case is caused by the hot water wanting to rise to the highest point which is in the pipes exiting the lid. That push starts a flow that keeps cycling.

 

[kabbott]

If it is siphoning you could use two weighted flowchecks parallel facing oposite directions, between the pumps.(as in not on the suction side)

On another note after looking at your website I notice the male adapters on the cpvc.
I have had nothing but trouble with male threaded adapters on cpvc on hot water lines, all seem to leak after time.
I use the unions that have a rubber washer like a hose washer, they come in cpvc/sweat or cpvc/npt or even compression for quicky repairs.
Keep an eye on them I would hate to see you trash a pump because it lost prime on count of a ten cent fitting.

 

[jebatty]

If siphoning, another easy fix are heat traps fashioned from pipe and placed on both the inlet and outlet side immediately above the tank. Hot water will rise to the top of the loops and stop at that point. Do not insulate the pipes until after the loops.

 

[WoodNotOil]

Thanks for the input guys. So I closed the valve and not only are the pipes cold to the touch but I now have 5* of stratification in the tank. As I said before, not sure how accurate the bottom tank reading is, but this is a good sign. I will have to trouble shoot mixing when I recharge this weekend. If siphoning is the culprit for that, it would be because when the fire in the tarm slows down periodically (like the end of a burn or during a burn before the wood falls to the bottom enough to continue gassifying) the charging pumps turn off and the siphoning starts. By the time it resumes charging heat has transfered from top to bottom. Not having to come up from cold tap water should minimize that effect on my next charge, so hopefully I will end up with more stratification.

I like your ideas to stop the siphoning. I have seen the weighted valve heat trap in other schematics. I have attached my translation of those to my piping. I think it would work. My questions are: Do weighted check valves add head pressure and where do you buy these? I see swing checks and spring checks, but none of those state whether they are weighted.

Jebatty - Your idea is simpler than check valves, but I am concerned that once the upward motion is started it will want to continue despite a drop. The water is already siphoning despite the 3' drop to the pumps and then the 3' rise back up.

 

[jebatty]

I think a swing check is a weighted check, and I also think if you go that route, you want the weighted check, not the spring check.

On the heat traps, if your rise in plumbing before the drop is 12" or more, that should be sufficient, although 18" might be better than 12". Hot water rises to the top of the trap and can go no higher, as the drop in the trap is cooler than the top.

Both weighted check and the heat trap should have minimal head effect; which is more or less I don't know.

Even if you're thermo-siphoning, stratification should be good. Hot water rise through the system, cools, and returns to the bottom of the tank. There should be no mixing.

 

[pybyr]

Even if you're thermo-siphoning, stratification should be good. Hot water rise through the system, cools, and returns to the bottom of the tank. There should be no mixing.

That- what Jim said (and that I was thinking, but hadn't articulated very well in my prior post in this thread about the siphoning) - is why I am still scratching my head about how some form of thermosiphoning outside the tank could be messing with stratification inside the tank

 

[WoodNotOil]

As I said before, I am sure there are other factors involved in this than the siphoning. My lines are insulated and 3/4" CPVC, so I don't think the siphoned water is cooling all that much before hitting the bottom of the tank.

I agree that mixing could still be happening and once I clear up the siphoning problem, I should be able to troubleshoot it a little better.

My improved stratification right now is likely the result of the water being left to stratify naturally and the effect of my DHW preheat coil introducing the coldest water at the bottom of the tank.

Jebatty - thanks for the minimum heights for the heat trap. I will keep that in mind as I design this. Would these swing checks work http://www.plumbinggoods.com/categories.asp?cID=3&brandid;= ?

 

[kabbott]

Sometimes just looking at a picture you don't get the "whole picture" but I wouldn't think It would even be siphoning.
Sure cant argue with the fact that shutting off the valve solves it though.

I think the flow checks should be between the two pumps, I wouldn't put them on the suction side in an open system.
Both pumps are pumping toward the heat exchanger correct?

 

[kabbott]

I am not sure if a swing check will stop it or not.
look at this link.
http://www.pexsupply.com/CategoryPre.asp?cID=275&brandid;=

 

[WoodNotOil]

Sometimes just looking at a picture you don't get the "whole picture" but I wouldn't think It would even be siphoning.
Sure cant argue with the fact that shutting off the valve solves it though.

I think the flow checks should be between the two pumps, I wouldn't put them on the suction side in an open system.
Both pumps are pumping toward the heat exchanger correct?

As I said before it was definitely siphoning. Hot pipes when valve is open, cold pipes when valve is closed. It is conjecture on my part however that it is also causing my higher reading at the bottom of the tank.

The pumps are actually pointed away from the exchanger. The heat trap I think is supposed to be at the top of the highest line and prevent flow at the highest point and allow flow in a line below it. I could be wrong though. Does anyone else have a heat trap in their system who can speak to this?

 

[jebatty]

I have the heat traps installed on my hot water heater in the basement, which of course feeds hot water to the upper level. The heat traps absolutely stop all thermo-siphoning in both the hot and cold water lines. These are in the basement, not the "high" point in the system. Hot water will exist from the top of the hot water tank to the top of the trap, those pipes will be warm/hot. The pipes beyond will be cool/cold, depending on the length of time that no hot water has flowed through the system.

Theory-wise, thermo-siphoning will occur between the heat source (tank) and the trap, which is limited to a short distance. Thermo-siphoning also will occur after the heat trap at various points in the system, with hot water rising to top points and cooler water falling. But this will not draw water through the traps, it occurs within the pipes. Hot water rising in the interior of the pipes and cool water falling down the exterior walls of the pipe.

The theory matches my experience.

 

[jebatty]

Just a bit of musing. In an open system, I'm wondering if thermo-siphoning can induce flow through the pipes, even with a heat trap set-up. Since hot water rises, it's pressure is higher than cold water, and in an open system, there may be induced flow. Anyone know?

In a closed system, I'm thinking that flow cannot be induced with a heat trap set-up, as pressure is equal through-out the system. Although thermo-siphoning may be able to create a pressure differential, like a circulator does, my guess is that this is very small, if any, and not sufficient to induce flow through a heat trap. This matches my experience with my hot water heater and use of heat traps.

If flow can be induced through a heat trap, open or closed, then a weighted flow check valve downstream of the circulator will be needed to prevent thermo-siphoning. A weighted flow check valve will increase the system head and therefore may impact system design. See the article referenced in
https://www.hearth.com/talk/threads/22206/

 

[WoodNotOil]

Just a bit of musing. In an open system, I'm wondering if thermo-siphoning can induce flow through the pipes, even with a heat trap set-up. Since hot water rises, it's pressure is higher than cold water, and in an open system, there may be induced flow. Anyone know?

In a closed system, I'm thinking that flow cannot be induced with a heat trap set-up, as pressure is equal through-out the system. Although thermo-siphoning may be able to create a pressure differential, like a circulator does, my guess is that this is very small, if any, and not sufficient to induce flow through a heat trap. This matches my experience with my hot water heater and use of heat traps.

If flow can be induced through a heat trap, open or closed, then a weighted flow check valve downstream of the circulator will be needed to prevent thermo-siphoning. A weighted flow check valve will increase the system head and therefore may impact system design. See the article referenced in
https://www.hearth.com/talk/threads/22206/

Thermo-siphoning inducing flow is exactly what I am experiencing in my system! The fact that water needs to be able to move in both directions through the pipe is what makes stopping this with a heat trap a little confusing. Also, as you stated above, I do not want to increase system head at all.

I may be back to using some sort of automated valve that opens only when a pump is called for. Zone valves and EBV's increase head so they are out. The only option people have mentioned in the past that does not increase head is a ball valve with an actuator. That is because the opening in the valve is the same diameter as the pipe. Do people think and actuated ball valve will open and close quickly enough when a pump is called for? Is there another option I am overlooking here?

Jebatty - thanks for the musing. I think your thoughts are right on.

 

[WoodNotOil]

I emailed a rep at Taco and asked him about a low head zone valve option. He replied that the EBV's have the lowest head loss and calculated that it would add 4.9 ft/hd to my particular application. I checked a chart and I think that still falls within an ok range for my pump size. They are about $70 new. A heat trap with weighted flow checks would be cheaper, but I am not sure if one would solve the problem in this particular application. Thoughts?

 

[kabbott]

It may not be much cheaper, you need two flow checks, tee's, elbows etc.
I think the EBV will actually be less restriction than the flow checks you may want to confirm that.
Does the EBV have some sort of end switch that can control both pumps? I believe they take some time to open and wouldn't want the pump to start untill its open.

Either way it seems to me you should turn the pumps around, as they are now, no matter where you put the restriction(EBV or flow check) it will be on the
inlet side of one of the pumps.
If you face both pumps toward heat exchanger the head pressure of the plate exchanger,non running pump, and the EBV or flow check will be on the outlet side
no matter which pump is running.

 

[DaveBP]

In order for thermosiphoning to happen in your system the plumbing outside the storage tank needs to be losing heat. That makes the water denser as its temp decreases. When the water in your external plumbing is denser than the water in your tank ( lower temp) it will descend into the bottom of your tank, displacing the water in the tank up out the top. Simple enough to picture when you have an out-the-top-in-the-bottom loop but with your dip tube making the denser water rise up to about the same level as the hottest water I can't see the cartoon. A simple experiment would be to insulate the plumbing well to minimize the heat loss and see if that reduces the thermosiphoning. How to measure that might be a challenge. Maybe with a bunch of these.

(broken link removed to http://www.omega.com/Temperature/pdf/RLC-80.pdf)

 

[WoodNotOil]

It may not be much cheaper, you need two flow checks, tee's, elbows etc.
I think the EBV will actually be less restriction than the flow checks you may want to confirm that.
Does the EBV have some sort of end switch that can control both pumps? I believe they take some time to open and wouldn't want the pump to start untill its open.

After hearing back from Taco I was wodering the same thing about the restriction in the flow checks. I will check with Taco on the opening issue and see what they think. Are you thinking it would be bad for the pump over time or was it some other issue?

Either way it seems to me you should turn the pumps around, as they are now, no matter where you put the restriction(EBV or flow check) it will be on the
inlet side of one of the pumps.
If you face both pumps toward heat exchanger the head pressure of the plate exchanger,non running pump, and the EBV or flow check will be on the outlet side
no matter which pump is running.

I don't follow you on this one. The pumps must point in opposite direction no matter whether that is both toward the plate hx or both away. That would mean either way the EBV would be on the outlet of one pump and the inlet of the other. Am I missing something? Out of curiousity, why would it matter? Thanks for the input.

 

[kabbott]

author="WoodNotOil" date="1220068011"]

It may not be much cheaper, you need two flow checks, tee's, elbows etc.
I think the EBV will actually be less restriction than the flow checks you may want to confirm that.
Does the EBV have some sort of end switch that can control both pumps? I believe they take some time to open and wouldn't want the pump to start untill its open.

After hearing back from Taco I was wodering the same thing about the restriction in the flow checks. I will check with Taco on the opening issue and see what they think. Are you thinking it would be bad for the pump over time or was it some other issue?
My thought was that you wouldn't want the pump running before the valve was open.
Not sure how hard that would be on the pump,but cant be good.

Either way it seems to me you should turn the pumps around, as they are now, no matter where you put the restriction(EBV or flow check) it will be on the
inlet side of one of the pumps.
If you face both pumps toward heat exchanger the head pressure of the plate exchanger,non running pump, and the EBV or flow check will be on the outlet side
no matter which pump is running.

I don't follow you on this one. The pumps must point in opposite direction no matter whether that is both toward the plate hx or both away. That would mean either way the EBV would be on the outlet of one pump and the inlet of the other. Am I missing something? Out of curiousity, why would it matter? Thanks for the input.

Sorry for the confusion,If you look at your drawing in post #28 the top pump runs to charge tank.All restriction's are on the inlet of that pump.
Same for the other pump when drawing heat from the tank.

If you turn both pumps around so they are facing the plate exchanger then the restrictions would be on the outlet of the pump.
You would then run the bottom pump to charge the tank and the top to discharge.
If the wires are long enough you can simply swap top and bottom pumps and rotate them end for end.

Just seems to me that you would have less problems with air or cavitation that way, although you may not have trouble now if you add flow checks or EBV
on the suction side it could cause a problem.Really simple to switch while your adding EBV or whatever anyway.

 

[WoodNotOil]

Cavitation was a new one for me (had to look it up). You make a good point and probably I will turn them around when I put in an EBV. The only problem is that facing the other way there is not enough space to the wall and there a soldered connections that would have to be redone in order to move everything away from the wall.

I charged the tank today and instead of measuring the bottom tank temp with my well, I put the probe on the copper portion of the piping that draws water from the bottom of the tank (as suggested in this thread). I wrapped it with foil and put a piece of pipe insulation around it to make sure it would read accurately. This paints a much better picture. The bottom of the tank is definitely lower than the top. In the early part of the charging the differential was as much as 25* between top and bottom. Now toward the end of the charge it has narrowed to a 15* differential. This shows the signs of stratification and hopefully closing the valve after the charge is done will allow it to stay stratified. I can take this to indicate that I need to design a new well that is not copper straight down. What is considered a good stratification differential?

The first 4 hours of charging ran with no idling at all. The next two after that had a couple of short idles and now that the tank is 150* I am getting a lot of idling. I am not sure if this is an indication that the plate is sized too small or if the flow is to high. I am running 3 speed pumps on both sides on high. I wonder if turning them down would help any. I think it will be a different situation in the winter as well when there is a heat demand on top of the transfer of heat to the tank. Any thoughts or experience here is appreciated. However, I am in no hurry to buy a larger plate. I would rather lose a little efficiency for the time being and see how it perfroms in all seasons before taking that leap.

 

[WoodNotOil]

I have an update on my research. I emailed Taco about the issue of the EBV valve opening at the same time the pump is energized. Here was the response:

"If EBV valve opens before, or at the same time as the pump, that's a good thing. If the valve opens slightly after the pump energizes, it will dead head the pump for that amount of time, but with no detriment to the pump."

I think, if I am reading this right, he is saying it is fine opening at any time. Before, sametime=good after=dead head pump, but no detriment.

Sounds like it would not be a problem. Now, I think these are 24vac which means I need to something to cut down the current from 110vac. I guess either a transformer in a metal box or 24vac fed from an existing box through a DPDT relay that is open and closed by 110vac aquastat, but only runs 24vac through the relay. Are there other options I am missing? Thanks for the input.

 

[Nofossil]

I dead-head small circulators all the time. As the Taco guy says, they don't mind. Larger circs with really hot water might add enough energy to cause boiling and cavitation.

I'd check my temp sensors by buying one of those cheap indoor/outdoor thermometers that has a small sensor on a 6' wire. Dangle it down your sensor well at different heights and see if it records different temps than what you're seeing. By the way - use cotton or modeling clay or something to seal the sensor well so that you don't get air circulation in and out of it.