Air to Water Heat Pumps

Some thoughts from another guru, on the John Manning method, bypassing top , straight through tank on bottom, would give you

Faster flow to the load, not passing across the tank

Less heat loss not keeping the tank hot on top

When flow is equal on both sides, the mass of the tank at the bottom is engaged

Eliminate any vertical movement in the tank

Could be the Heat Geeks could learn some from this side of the pond :) Explore all the options

An energy monitoring company is installing a number of systems of different brands, around the New England area and documenting them, similar to the Heat Geeks. It would be interesting to compare data. Especially that seasonal COP number?

If you have emitters that can modulate their output, so long as the heating load is above the minimum output of the heat pump you don't really need anything at all. You can just hook the emitters up to the heat pump, no buffer tank needed, and the heat pump should run continuously and modulate to adjust to variations in the load.

The buffer tank is for when the load is below the minimum output of the heat pump, and the heat pump has to cycle on and off. The buffer tank keeps the heat pump from short cycling by giving it a place to dump its output, as well as guaranteeing that a minimal level of flow is maintained if you're using zone valves and they all shut off. In addition, in heating mode many systems draw heat out of the buffer when defrosting, in order to defrost at a high COP without adversely affecting occupant comfort.

The goal in designing the plumbing is to achieve the necessary function of the buffer tank while avoiding negative impacts. So what kinds of unnecessary negative impacts should we worry about?

• Anything that conceals from the heat pump the actual load conditions. In particular, anything that causes it to run higher output than necessary and then cycle off rather than running continuously at a lower output.
• Anything that causes the water hitting the emitters to be not as hot as the water heaving the heat pump (or cold if we're cooling).
• Anything that adversely affects occupant comfort, ie, the ability to respond appropriately to changes in load, or the ability to dehumidify in cooling mode.

Surfing that UK forum, some of those engineers are going full in with controls. ODR, IDR, PID, solar irradiation sensors, sensors in every room, weather predictive input, wifi stats. My head hurts!

Will these third party control compliment or compete with the factory controls? Seems all the big names have their own control ideas.

It was mentioned that Dakin even has a buffer tank spec that you need to follow if you consider aftermarket buffers.

@hot_rod: "I suppose that is why the one control designer wants so many inputs. Start making corrections before a weather event arrives at your door step."

I find in my house it's not so much weather events, it's the sun. When it comes out from behind a cloud the heating load drops like a rock. When the sun sets in the afternoon it goes from zero to 100% in about 15 minutes.

Maybe it's my climate, but we don't get fast-moving weather events.

The only reason that heat pump controls would have anything to do with this is if your emitters are unable to scale their output sufficiently on their own. The only variable the heat pump can control is the water temperature. If, for example, you can't get the emitters to meet the current load with the current water temperature, you need to send a signal to the heat pump to boost the water temperature. But it's simpler to manage it all with the emitters if you can.

Note that all heating systems have the issue of trying to track changes in heating load. The reason it's particularly noticeable with air-to-water is that you want to minimize temperature and maximize runtime to maximize efficiency. With air-to-water you're much better off running continuously at exactly the right water temperature than cycling on and off at a higher temperature. And all other things being equal, for the same BTU output you're better off at a lower temperature, lower delta and higher flow than at a higher temperature, higher delta and lower flow, assuming your emitters could support either.

So in order for this to work it has to be used instead of a thermostat, because if a thermostat is doing its job indoor temperature is steady and the sensor is defeated. So you need to have either one zone for the whole building, or if you have multiple zones, one has to be designated as the "reference" zone, which doesn't have a thermostat and has the sensor instead. And you have to be sure that the reference zone truly is representative, otherwise the other zones may have trouble maintaining comfort.

Sounds good. Ideally you'd have one zone running at 100% cycle length, that seems like the optimal water temperature.

As I've mentioned a few times in this thread, I believe that a buffer tank is only necessary when the heat pump is cycling. I'll add the observation that any method of adjusting the water temperature to meet the load will work better the less water capacity the system has. Another thing I've been thinking about doing is creating logic to figure out whether the heat pump is cycling, and use zone valves to cut out the buffer tank otherwise and run direct to load.

This fits in nicely with the idea of trying to tune the temperature setting of the heat pump, because both ideas are about assessing the load.

So let me offer this method for how the controller would work:

1. Assess each zone. If there isn't at least one zone at 100% capacity, drop the water temperature until there is. If any zone can't meet the setpoint, raise the water temperature until it can.
2. If the total output of all the zones isn't more than the minimum modulation of the heat pump, open the buffer tank. If it is, close it.
3. If the flow through all the zones is less than the minimum flow through the heat pump, open the buffer tank.

The first one is pretty easy. The third one is straightforward. The second is a little trickier because the it's a moving target. But you'll be able to see it because the water temperature won't be steady, it will start creeping up.

For prosperity in case anybody makes it this far down into discussion. 80% of this thread was talking about tweaks and changes that make maybe a couple percentage points difference on a properly designed and set up system. Like with anything, there are also many ways of messing up the install, but it is also not that hard to get it right.

The two main configurations that will always work are the multi port buffer tank and direct to load with volumizer and differential bypass. The good news is that both of these configuration also work great with a modcon.

Air to water heat pumps are not hard or magical. They also don't need a Phd in controls to set up. A couple of rules of thumb, a bit of hand calc and a good dose of RTFM and it just works. About the only setup is getting a decent outdoor reset curve into the unit and making sure your flow rates are correct.

Hey, if it's worth doing, it's worth overdoing.

I'm with @Kaos, if you want a simple system that works, go with outdoor reset and a 3-pipe buffer. Focus on the controls of the emitters to ensure comfort. Skip the boiler backup and go with a heat pump water heater for DHW.

this seems to be a good description of SCOP and the standard that regulates it

https://www.h2xengineering.com/blogs/heat-pump-cop-and-scop-what-they-mean-and-why-they-matter/

I'd be leery to mention that 8 SCOP number to a potential customer. They may hold you to it.

Everything I find on line from engineering groups indicate 3-4.0 as a good SCOP number

@John Ruhnke : "Remember a graduate of the Heat Geek training has an average of 4.19 SCOP. Nothing coming out of America is even coming close to this."

I've never seen SCOP calculations from an American system so I don't know how you can say that.

SCOP numbers are a useful comparator but they are really dependent on the temperature profile of the location. Comparing SCOP numbers from the UK with the USA or even other parts of Europe is impossible.

That's exactly backwards. SCOP is adjusted for climate. They're designed to make it possible to compare performance across different climates.

It doesn't. But that's not what SCOP does. SCOP does that calculation and then adjusts for climate. We discussed this back on page 2 of this thread.

COP of 8 is stretching it. You are the point where installed cost of extra area makes the ROI pretty much never. As nice as to think of ways of improving systems you can quickly hit the law of diminishing returns.

The beauty of modern AWHP units is you don't need fancy controls. Nothing aftermarket will work as well as what is built in plus you are now adding cost, complexity and a heck of a lot more commissioning time.

When I'm looking at these systems, my first priority is how I can eliminate a part or controls. Simplifying the setup means cheaper install and much easier setup.

For example with a recent modification, I eliminate a relay module. Removed a couple of actuators for bathroom floor heat and adjusted the rest to allow some flow even when off. Set the manifold deltaP pump to run 24/7 as there was no need to control it from the thermostat. Bonus is the floor heat is much toastier (and happier management) and there is overall more circulation water through the emitters, thus lower RWT to the heat pump.