Heat source and load on same manifold.

[ewdudley]

I'm about ready to set the tank and start the plumbing and wanted to run a couple things past you guys.

I noticed in the Laddomat literature (not that I'm planning to use one) that they recommend not doing what I intend to do, which is to tie-in the load supply and return at points between the boiler and the storage. I don't understand why the prohibition.

Also I have eliminated a heat exchanger and pump that would have isolated the wood boiler to its own pressurized system. It just seems simpler and perhaps safer to just run the boiler output straight into the unpressurized tank. The expansion vessel will be a cistern with perhaps layer of paraffin/mineral oil, or maybe large inner tube modified to serve as an atmospheric bladder.

And the third thing is where I've incorporated all the main heating zones into the 'primary loop'. The only downside I can see is that if any of the main heating zones is calling, the pex zones only get series flow from whatever zones are active, which I can live with.

What can you all see that I might likely regret?

 

[sgschwend]

I can see that you have lots of good thought into your design. I have a question for you have you looked at this design? https://www.hearth.com/talk/threads/19086/


The reason I ask is because I assume you want to make your system as simple as possible and minimize the cost. For example do you really need the second mixing valve and pump near the Hx? They look redundant to me. The solution I am asking you to look at utilizes the physical location of the boiler output T to set the heat flow priority.

 

[ewdudley]

Thank you for taking time to have a look at this.

If we consider the atmospheric-pressure-to-the-right-of-the-heat-exchanger-circuits as a separate system from the pressurized-left-of-heat-exchanger-circuits, we can see the the right-side system is pretty much the same as the "Simplest Pressurized Storage System Design", except it's unpressurized and the heat source is physically between the load and the storage. That there must be two systems is due to the unpressurized storage being located at a lower elevation without my enjoying the option of installing an expansion-cistern-closet in the attic.

By 'pretty much the same' I mean that they share the essential feature of using a hot-side-manifold and a cold-side-manifold to allow the heat source(s) and heat load(s) to act independently. The Laddomat literature says don't do that, but again it's not clear to me why.

In answer to your question, the circ pump near the heat exchanger is necessary to induce flow through the heat exchanger, same as the 'Load Circ' is necessary in the "Simplest Pressurized Storage System Design". The mixing valve on the discharge side of the heat exchanger circ is a diverting valve to set the return-to-storage-temperature and maintain stratification, a hundred bucks well spent I should think.

Cheers --ewd

 

[Nofossil]

Interesting design. In this case, having the PEX zones downstream from the high temp zones is perfect - they don't need high temp water anyway, and that will give you a lower return temp to the right side.

I'm not sure that the diverter near the HX will do what you want in this case. I'll have to think on it, but it can't create lower return temps to storage than the temp of the water coming into it from the HX.

In this design, it would be nice to use VS circulators to match the flow to the heat load when heating from storage. That would give you best possible performance from the HX.

 

[jebatty]

You likely are aware that a plate HX works best in one direction (usually counter-flow: allows the cooler return water to stay in contact with the hottest water the longest to achieve maximum heat transfer) and less efficiently in the other direction. An approach temp of 10F in the "best" direction is a usual design standard. You can get a closer approach by upsizing the HX, and cost goes up quite a bit. Design for a plate HX is a bit complicated, as head (or pressure drop), circ selection, and gpm are pretty critical. Many plate HX's show approach tables with unrealistic gpm flow for most systems, so you need to select from the table based on your system gpm requirement. Perhaps you can get an approach table that also show approach in the opposite direction. Plan this carefully to get the performance you expect.

 

[ewdudley]

I'm thinking that the diverting valve should force all flow back through the HX until HX return temperature falls below the diverting valve SP, nominally 120F in this case. I'm looking at a Taco 5000 series 1" that should do the trick.

   .-----------.------
|           |
HX          |
|           C
|           |
.--(|>)---M-o-H----

--ewd

 

[Nofossil]

I'm thinking that the diverting valve should force all flow back through the HX until HX return temperature falls below the diverting valve SP, nominally 120F in this case. I'm looking at a Taco 5000 series 1" that should do the trick.

   .-----------.------
|           |
HX          |
|           C
|           |
.--(|>)---M-o-H----

--ewd

Unless I'm really slow this morning (a distinct possibility, by the way) you may have an issue with temps in the circ -> diverter -> hx -> circ loop.

Let's assume that you start out with a cold system and hot storage. You get a nice slug of 180 degree water hitting the hx, and the water on the left side is nice and cold. Circulation starts and all is happy. Eventually, you start getting warmer water back to the HX on the left side. With everything going, it might be 150 degrees (could be more if the left side circ is running faster).

When the 150 degree water hits the HX, the return water coming off the HX on the right side will now be around 160.

The diverter valve gets into the act and starts recirculating near 100%. That drops the temp going into the right side of the hx down to 160. That in turn drops the primary loop supply temp to 150, and the primary return to 130 (assume not as much drop at lower temps). The right side HX return is now 120.

The diverter is now satisfied. I think the system will stabilize with a primary loop temp that results in a primary return temp of 130 to the HX. That may not make you as happy as you'd like to be.

But I could be confused. Time for another coffee.

 

[ewdudley]

(The signal-to-noise on this forum is just amazing.)

Excellent point nofossil.

If I don't reduce the heat supply flow rate into the HX, then as demand decreases on the load side the system is doing its best to prolong the time it takes to satisfy demand.

I'd have to think about it, but it might even reach an equilibrium in some cases and both circ pumps would be running 'forever'. Presumably not a good thing but maybe not the end-of-the-world.

I'll have to look into the whole multi-speed/variable-speed pump control thing. If I can figure out a way to easily switch over to a limp-mode when the VS controller goes tango-uniform, the VS route does seem better thermodynamically. Even as control systems engineer, I was hoping to keep this system all mixers, diverters, timers, relays, aquastats, and thermostats; but maybe the time has come to introduce a micro-controller or two.

Thanks for your thoughtful review.

--ewd

 

[sgschwend]

I believe your earlier question about the placement of the boiler is due to the need to have a second pump to circulate.

Why do you want the boiler in the middle?

Physical location does influence the plumbing but the plumbing can also be different. So why not:
Keep the plate heat exchanger and install it in the boiler loop with the low side going to storage (about where your number 4 is on your print).
Keep the mixing valve for the boiler input regulation, consider connecting the H port to the Hx low side.
Drop the second pump mixing valve.

More radical:
Drop the plate heat exchanger, and install a copper tube one in the storage tank.
Use one pump in one loop on the boiler side.

It only takes me 3 hours to recharge my 500 gallon storage tank,

What you hope to do is give the home heating priority remainder goes to storage, but your design does not switch loads except for the use of mixing valves (which in my limited experience seem to be a little iffy). Why not let your house load switch the boiler output?

 

[in hot water]

Why a supply and return up to the open expansion vessel? Looks like all the boiler flow will take the path of least resistance to that tank. What and how much flows right or left at that 4 way connection, if any.

hr

 

[Nofossil]

Further pondering:

You'll have some difficulty managing flow in and out of storage when the wood boiler is running. If the wood boiler circ and the right-side circ are not matched to heat production and heat demand, then you may get unintended flow in and out of storage. Since the effective flow rate of the boiler circ depends on the inlet protection mixing valve and the effective flow rate of the right-side circ depends on the diverter valve, it may be hard to predict and/or understand what's happening.

If the effective flow rate of the right-side circ (by the HX) is more than the effective flow rate of the boiler circ, then you will be drawing water from the top of storage in addition to the water that's flowing through the boiler. Not a big deal, but in the simplest storage design the number of open zones determines the flow rate in the load loop so that in most cases the flow rate in the load loop is lower than the flow rate from the boiler. You may want a VS circ there as well to provide higher temps to the loads, especially during boiler startup with cold storage.