Using Mixing Valve on Tank to Maximize Buffer Use and Pellet Boiler Runtime

[__dan]

Here's a thought.

Without looking at your plumbing diagram, the zone circ / circs for the baseboards, if you have something like a standard 007, you could try replacing that with an ECM circ with fixed temp sensing. Then the circ sensor would go on the baseboard's return pipe / header and be set to something like 130 -140 F.

If it fits your application, it would be one of the easiest changes you could make. At cold start, the circ would max to achieve HW return of 130 F, but would quickly ramp down as setpoint is met. So you would have a variable load, high when you need it in the beginning so the boiler starts out firing hard, then quickly lightens up but stays on at a lower load, giving time for the boiler to turn down.

I cannot understand how the DHW tank does not become fully mixed with boiler water at the end of the cycle. But this way, the boiler circ should be flowing lots more than the load ECM circ, so the tank should see that flow. Sizewise, the tank should be invisible to the boiler, like putting an ice cube in a pot of boiler macaroni.

You do need to take out the loading unit, disconnect the zones / DHW tank, and fire the boiler, as making steam, condensate, popping the relief valve, seems to be the only thing that will keep you happy.

 

[velvetfoot]

I am using four B&G Ecocirc Varios for zones and boilers. Alas, they are ECM but not auto in any way, but are infinitely adjustable manually.
As I've noted, I feel that the cold water coming back to the boiler from the tank, ~105 at start of cycle, keeps the boiler from warming up.

I cannot understand how the DHW tank does not become fully mixed with boiler water at the end of the cycle.

The sensors on the outside of the tank generally show a 5* difference between top and bottom. So, ending up with something like 179 and 174.

 

[velvetfoot]

Good morning to fool around with this since it was -19F.
I tried lowering the boiler feed pump speed just a tad, and that may, or may not, have boosted temp.
Still, long recovery time from setback with low return temperatures. Would be nice if the zones were given priority and the tank was out of the loop.

 

[velvetfoot]

This is what I'm currently thinking. When a baseboard loop starts, the slug of cold water in the pipe takes little while to come back to the return header. I'm not sure where to put the mixing valve controller temperature sensor, before the zone return tees or after. The slugs of water are pretty transitory, and I don't know how fast the mixing valve would react. Before the return tees might result in a more smooth control, but I'm not sure.


I'm also a little unsure of the type of the two check valves I figure I'll need. I used a spring check valve on my oil boiler supply and that seems to work well. The system came with Taco FloChks on the zone supplies. I'm thinking spring check valves.

 

[__dan]

What I'm reading still makes no sense.

The response time of that type of valve, if it is a mixing type, is very slow. It will start moving almost immediately, but if the PID is tuned for hydronic, I would not expect it to settle into the final position for five or more minutes. If it's doing hydronic mixing, the controller will be slow to avoid hunting.

If it's something that is just driven open and driven closed, the cycle time could be just a few seconds.

The normal strategy in your application is to prioritize DHW, meaning the zones are off until some condition, DHW is satisfied or, maybe better with the pellet burner, boiler return temp is above setpoint.

I cannot see the 110 gal of tank being any kind of significant load that prevents you from getting heat to the zones. Even in a cold start first thing in the morning, the tank water is probably warmer than either the boiler or the baseboard water. So, what load can the tank present. It would actually give a shot of warmer water to the system, reducing the cold start time.

Also the DHW tank was part of a factory recommended and supported install. If the tank was part of the factory requirements in their installation instructions and not just a recommendation, removing it with the valve would constitute a code violation and may void your warranty, insurance. Code are adopted by State law, so it would be a violation of that requirement of the State law. What if you needed the insurance to pay and then explained to them how you disabled something the factory required then fired the boiler. Do you think you are covered for that?

 

[velvetfoot]

Five minutes is a long time. This would point to using this control would avoid the effect from where the zone returns tie in, dumping in periodic slugs of cold water, in my baseboard case.

As I've said, I have a reverse indirect tank and don't operate the zone pumps below a certain temperature to assure dhw.

I've observed the temps and the boiler return temp is lowered for a significant time before the boiler gets to, say, 168, even with a boiler protection valve in place.

When the temp of the tank gets cold enough the pellet boiler fires.

A rough calculation of 60 degree rise of the tank, 120-180 gives me 60,000 btus.

You know, I'm not sure if anything you've said is helpful at all. I'll leave it at that.

 

[velvetfoot]

On a more hopefully helpful note, I've started reading the Honeywell controller instructions. While mine is modulating outputs only, the models with relays can do 150 degree differential! That's more than the Rancos and Johnson Controls models usually cited here.

 

[velvetfoot]

Upon reading some more, the two check valves might be dispensed with. The controller has a terminal block on it called 'digital input' that's a dry contact that when closed, puts all modulating outputs to zero %. So, a relay can sense if the pellet boiler is 'off', and then close the tank bypass, which will allow flow from the tank to heat the zones.

My current plan is to have it serve as a mixing (rather than diverting) valve on the return, for hopefully better response.

It talks about throttling range and integral settings and recommends keeping derivative at zero. This is what it says about integral and throttling range settings:

NOTES:
1. The Integral time is factory set for 400
seconds. This is a good middle range and
should satisfy many applications. The integral
time can be increased for applications where
sensed response is slow, and can be
decreased for applications where sensed
response is fast (e.g. discharge air control).
2. As a starting point, an optimal integral time for
discharge air typically ranges from 12 to 200
seconds. An optimal integral time for room
control typically ranges from 60 to 2,500
seconds. The purpose of integral action is to
reduce or eliminate the offset from setpoint
during steady state control that is often seen
in proportional only control.
3. Keep in mind that control is most sensitive to
throttling range. Adjust the throttling range
first before any adjustment to integral time.
Adjust throttling range to be as wide as
possible to start since this will provide the
most stable control. Remember that the
integral will eliminate the steady state error so
you do not need to have a small throttling
range to have accurate control. (Integral
action allows for controlling to a setpoint even
with a wide throttling range)

 

[velvetfoot]

Several other interesting things about the controller:

-The output can be set to a minimum percentage. I'm thinking in case I want the top of the tank to gain some heat earlier for dhw purposes, or if the boiler run times are getting too long .

 

[velvetfoot]

Well, I managed to hook it up electrically. Wasn't entirely intuitive for me but got it working. What they call 'digital input' also worked. While 100% modulated, (100% bypassed), like if the return temperature is below X*, when the dry contact 'digital input' leads are joined, it goes back to 0% modulated (free flowing through the tank). This would allow reverse flow without check valves and extra pipe, so that the tank can heat the zones when the boiler is off.

I have to say the display is bright. It never dims or shuts off though. They could've put a timer on it like my boiler, I think.

 

[velvetfoot]

I just piped up the valve and I'm filling things back up now. Hopefully, no leaks. Knock on wood.

 

[velvetfoot]

Came home from movies-small leak, only right near the tank. Crap. Draining tank. Hopefully can drain water from pipe w/o taking apart. Crap.

 

[TCaldwell]

How's it going?

 

[velvetfoot]

I hate to jinx it, but good. There, I said it. I drained things again, tried to isolate the tank, sucked some more water out of the bottom of the tank with a shop vac, applied a little flux to the outside of the leaking joint, heated and added solder. Filled things up, repressurized and, knock on wood, no leaks.

I also relocated the Spirovent closer to the heat source; I had to move it anyway since it was near the tank and was interfering with new piping.

I've mounted the actuator on the valve and the controller on the wall. I've powered the controller up and installed the temperature sensor on the pipe a relatively short distance downstream of the valve.

I've checked the control voltages coming out of the controller and it seems right: controller shows 100% bypass, 10 volt output, valve SHOULD stem rotate all the way clockwise.

This morning I will read up some more on wiring up the actuator (really don't want to burn anything out), and also what my preliminary controller settings will be; hopefully the settings won't be a long drawn out thing. I'll complete the wiring this morning and fire up the oil boiler and see what happens.

Pictures might be difficult because of close quarters, but I'll try.

What's a shame is the beautiful and expensive glued-together 1" thick foam insulation doesn't remove that easily. I do have some spare pieces though.

Anyway, you asked.

 

[TCaldwell]

Just installed a 3 way mix valve ( shunt valve) on my system to accommodate outdoor reset, will get into that with another thread. The valve/ actuator is a belimo 0-10vdc with a 100 sec stroke speed, the controller is a Tekmar 423 reset module, there aren't adjustable pid parameters, rather tekmars own learning algorithm, I'm surprised how precise the control actually is. Once dialed in you'll be happy.

 

[velvetfoot]

Not happy now. Leak started again as it was heating up. Very depressed. Right at tank. I might try one more time. I really don't want to break it open, and not sure of how to proceed if I do. Very bummed.

When it was heating up the controller it was doing its job but maybe working the valve a little too hard, I don't know.

 

[TCaldwell]

Shark bite fitting?

 

[velvetfoot]

Do you think one more round of attempting to add solder would not be successful?

 

[TCaldwell]

Doubtful, l would take it apart clean up fitting/pipe and re sweat. Your almost home to testing something you have put a lot of effort into.

 

[velvetfoot]

Yeah. It's going to be a pain to take things apart though.

 

[TCaldwell]

Cut the pipe in the middle, between the valve and tank fitting, unscrew witch side is leaking weather from the valve or tank, put it in a vice, heat and remove the bad joint from the fitting, clean it up and resweat. Then screw fitting back in and connect with a sweat coupler

 

[maple1]

Yep. Been there. Looks like a fix that starts with a recip saw, and ends with a sweat coupling.

Edit: Although it does look like a tight spot. Which might mean more than one cut & coupling.

 

[velvetfoot]

Thing is, it's so tight there might not be enough length for a slip coupling. I'm also reluctant to unscrew the adapter from the tank-who knows if that might not leak. I might try sweating again tomorrow, but I'm not optimistic.

 

[Bob Rohr]

Thing is, it's so tight there might not be enough length for a slip coupling. I'm also reluctant to unscrew the adapter from the tank-who knows if that might not leak. I might try sweating again tomorrow, but I'm not optimistic.

You may end up over-heating the thread dope and create a leak there. best to start over, once you have a bad spot inside a sweat fitting, desoldering is almost impossible.

Do you have a union connection in there somehow?

Perfect spot for a press fitting and tool

 

[velvetfoot]

I was thinking of a union, but they've leaked on me too.