Air to Water Heat Pumps

John Ruhnke · January 26

What is the turn down ratio for that unit. The Vitocal HP has a 1.5 ton unit. I can see it specifying a small buffer tank. Still bigger is better than smaller when sizing a buffer tank. Turn down ratio can effect buffer tank sizing too.

hot_rod · January 26

https://forum.heatinghelp.com/discussion/comment/1838019#Comment_1838019

yes, this is the smallest unit, 1.5 ton that they offer.

Here is the COP chart they show. My design temperature is 5°, I'm running around 113- 115. So a far cry from the Heat Geek COP numbers😯

Typically winter temperatures run 20- 35° most of the time. So the numbers get better.

Screenshot 2025-01-26 at 2.08.20 PM.png

John Ruhnke · January 26

Viessmann makes great products. I installed a few Vitodens Boilers. Many Heat Geeks have used Viessmann in there projects.

Kaos · January 26

Not buying the bigger buffer tank theory. If you have filed data or simulation of some kind to back it, I would love to see it.

The 5ton unit (really 3 tons in our climate) has about a 1:2 turndown. It never cycles if temps are bellow 5C (41F). No buffer tank except for the volume of water in a pretty long primary loop. Even when it cycles, the runtime is pretty long on the order of 5 to 10min. This is with a reasonably aggressive ODR curve so some zones are running a good portion of the time.

Defrost is a non-issue, never seen the water temp dip bellow 20C during defrost. This was including on days where I though my unit sprung a leak as there was a stream of defrost water coming off it.

It is definitely much harder to mess up a system install with a big buffer tank, but I don't see it improving efficiency.

Over pumping secondary is definitely out there. Look at the number of installs with a bank of pumps for each zone.

John Ruhnke · January 26

Kaos,

If you size everything properly and don't use a buffer tank than the load for the heat pump becomes the system and not the buffer tank. The smallest zone has to be sized with enough volume or you need to add a volumizer to compensate. That is how the Heat Geeks achieve better efficiency. When looking at the buffer tank as the load then yes bigger is better but there is a limit to where the cycle length is plenty big enough that you achieve a max size. Then after that the stand by loss and the tank size starts working against you.

Sounds like you have a well designed system.

Kaos · January 27

The problem isn't a design one, the problem is commissioning. It took a bit of time to dial in the ODR and zone flow rates, not something I can see a tech spending time doing.

Didn't spend much time designing the system except for basic sizing of pipes and pumps, if anything my system is mis-designed as I violate some rules of primary/secondary plumbing. My calculations showed I would need a buffer tank, but since the cheapest option was a couple hundred, I figured to skip it and see what happens, so far nothing. If anything I would have needed one as the smallest zone is a bath floor heat zone which technically will cause the unit to cycle. In reality the bath heat zone overlaps the rest of the house zones enough to not matter. For the sake of science, I will try to run it in the summer to see what happens.

As for buffers, I'm still not convinced sizing helps.

If the unit is cycling, the mean space heat supply temperature will be the ratio of heat pump supply temp and duty cycle thus will always be less than what the heat pump is producing. Since the duty cycle is only a function of load, the size of the buffer only stretches out the cycle period, so mean and peak temps don't change*. It also means any time your heat pump is cycling, the COP will be less as it needs to make hotter than average temp water, again buffer tank does not fix this. Now this doesn't matter much in warmer weather as the COP hit to make this warmer water is not much, but if you are cycling when cold, COP will suffer.

*with a smaller buffer, the temps will rise to higher values at the end of an on cycle, so some minor COP effect.

John Ruhnke · January 27

With a smaller buffer you have shorter cycle lengths than a larger buffer tank. It takes longer to warm up a larger tank than a smaller one.

As far as water temps go. In my house I have a control that automatically sets the outdoor reset curve based off of indoor cycle length of the zones from the thermostats. It finds the coldest zone and works to lengthen its cycle to its limit. The control drops the water temps to lengthen the cycle length from the thermostat. Its called ODR with Indoor feedback. I don't even need to input a outdoor heating design temp. It automatically finds it. This control is twenty years old and works great to this day.

hot_rod · January 27

large buffers make a lot of sense for solid fuel boilers. I had a 500 gallon on my wood/ solar hydronic.

With mod cons the sizing is based on the lowest turn down match up with the smallest load. So I have used anywhere from 6 gallon on up.

With heat pumps, the modulation has come close to eliminating buffer in some cases. The last Entertech HP webinar promoted tank-less HP installation.

So take your pick.

DCContrarian · January 27

Put me in the smaller buffer is better camp.

The key to heat pump efficiency is to have the heat pump controller see the system load and modulate accordingly, maximum efficiency occurs when the heat pump can match the compressor speed exactly to the load and run continuously. At low loads the buffer makes the load that the heat pump controller sees appear larger than it really is until the buffer temperature stabilizes, and then appear smaller than it really is — zero, actually — once the compressor has cycled off. A buffer tank causes the heat pump to run at higher output than the load, and then cycle off. With the variable speed compressor, higher output is lower COP.

Not only does running continuously at a lower output result in a higher COP than cycling at a higher output, it also results in better performance of the emitters because the water temperature is constant and at a higher temperature when heating and a lower temperature when cooling.

This is particularly important when cooling if you are relying on air handlers for dehumidification. Colder water means more dehumidification. Fluctuating water temperature wreaks havoc with dehumidification because when the coil temperature rises any water that is on the coil and hasn't dripped off yet will evaporate.

DCContrarian · January 27

https://forum.heatinghelp.com/discussion/comment/1838127#Comment_1838127

With a modulating heat pump the goal is not to cycle at all. Larger buffer tanks lead to cycling when it could be avoided.

John Ruhnke · January 27

@DCContrarian,

Fine if you are designing a perfect system in a perfect world maybe you won't have a cycle. Not one technician I have trained in the last three years understands what constant circulation is. It is a whole new philosophy. But I have to train lots of technicians with various backgrounds in hydronics. They all have there favorite piping method and they don't comprehend everything you teach them. You know the old saying. The average person retains 10% of what they learn. Not everyone wants a complete design from us. They just want to add a heat pump to an existing design. The more complicated you make things the faster they will get lost. Also the more likely they will screw something up.

The best way to make sure you get very good efficiencies (Not the absolute best) and don't have problems is by installing a buffer tank properly sized. The systems are problem free and last a long time and are easier to teach.

Not every installer is designing a system for his own house and wants to spend time adjusting delta tees, controls, pump speeds or water temps. Most techs want to set it up once and then forget about it.

If someone is doing new construction and wants to pay me a larger advanced deposit towards material I am willing to supply a design that is more complex. And yes I probably won't design with a buffer tank.

Also with the advanced training I have received I understand the advanced principles. If you understand advanced hydronics I will work with you. If I feel you are capable I will help you or provide material for a advanced design.

DCContrarian · January 27

@Kaos : "We have to look at what most people are fine with. That tends to be a single air handler for the whole house for cooling. Bigger house, maybe 2 zones (either as extra air handlers or duct zoning)"

The way to view air-to-water is that there are a few things it's good at, and you want to focus on those things. You want to stay away from the things that it's not good at, even if it's capable of doing them.

"Of course there are cases when you do want better control and I can't really argue with the flexibility of hydronic with smaller FCU but be ready to spend real dollars for it."

This gets right to a frustration I have with the FCU's available in the US market. Basically, this is a luxury market, they should look, feel and act like luxury products. They don't. It's a tough sale to get someone to put a painted metal box into their multi-million dollar house.

DCContrarian · January 27

https://forum.heatinghelp.com/discussion/comment/1838140#Comment_1838140

While I'm still learning and my thinking is evolving, at this point my feeling is a buffer tank is a necessary evil. The question is, what is properly sized buffer tank?

Chiltrix recommends 17 gallons for their two-ton nominal units. The unit is capable of modulating to 25% of output, or 6,000 BTU/hr. When cycling, they have a default of 7.2F swing. 17 gallons is 141 pounds, swinging 7.2F means a flow of about 1,000 BTU, which is one sixth of an hour or ten minutes worth of runtime with zero load.

That seems like a reasonable size, I don't see any reason to go bigger.

John Ruhnke · January 27

A thirty minute run time is better than a ten minute run time. So again why would anybody say that a smaller buffer tank is better? You will get better COP's with a larger buffer tank and the HP will last longer.

Kaos · January 27

https://forum.heatinghelp.com/discussion/comment/1838145#Comment_1838145

Not sure about that, I have not seen any info out there about this. The closes I have found was a study looking at residential AC unit and there minimal effect from cycling.

If you have better data would like to see it.

And like I said earlier, 10m or 30m, you are cycling, so you are taking a COP hit.

John Ruhnke · January 27

From what I learned in the Heat Geek courses is that they like to size the HP to the system and have the HP run as long as possible. There is a loss in energy or COP right at start up. As the HP has been running for a little while the COP gets better. It takes a little time to get warmed up. This has been observed through monitoring. Less cycles are better thus longer cycles are better because it reduces the amount of cycles. I will have to go through my course videos again. I can't remember exactly how long the HP needs to run for to get to that optimal level. It not very long though. It is better to avoid this start up period. Thus reduce cycles.

Kaos · January 27

Closest I have found is this video:

https://www.youtube.com/watch?v=0oszaqP0TI4

If you look at the graph here:

There is no major changes in power when the unit cycles. I guess there is a bit lost at startup and shutdown doesn't look like enough to matter.

John Ruhnke · January 27

That is a great video. John Cantor is considered by the Brits to be great with hydronics. He is someone to be learning from. I do know that in general turning things on and off a lot reduces the life of things. Adam Chapman talks about this loss of power at start up in his training. So Adam has seen this. The Heat Geeks are definitely designing to avoid unneeded cycling. At 13.28 in this video John Canter talks about the loss of power at start up with a ASHP. In this video the HPs are running at 40F which is pretty warm out, a mild winter day. What happens at 10F? I still need to do my network phase of my training. I need to talk with some of these Heat Geeks in the U.K. I have my first live video meeting February 4th. I would like to see more videos like this. We don't have anything like this here in North America. From this video it seems that some HPs aren't effected much by cycling.

He talks about a thermostat adding unneeded cycles. I have seen this a lot with boilers in my hydronic projects from the past. It is a good reason to avoid using hot air thermostats in hydronics. In that case the cycles from the thermostat was ruining performance and is something a buffer tank would avoid.

I am new to studying graphs like that. I have a lot more to learn. The courses in the U.K. from the Heat Geeks were intense and I learned a lot. But I also need to ask a lot of questions and I need to get into a network with Heat Geeks. I have not gotten that far yet.

clammy · January 27

Not to melt any of the subject but am i right in believing that the british government is also tossing a decent amout of money for those who are doing these upgrades some of which i have just viewed on line and most talk about the amount of money the government is kicking back . I was astounded by the price of these retro fitting jobs and realize that most aside from the ultra wealthy would never afford and the wealthy can more then afford outta pocket but like all freebies they go for it being there getting some back .

As for oversized panel rads and sizing for lower swt well i ve size downto a 135 for a 10 oat and when doing the math it can to about 40 to 50 % higher and thats for a 135 water temp .I can only image the cost of a much lower supply water temp it starts to get crazy plus do all house have that much area for a panel rad that big .

As i enter a phase of stepping back it will be fun to see who pays for all the mistakes being our society is over run w the smartest of the smart and the best of the best in following instruction and proper procedures and quality workmanship it should excellent to see hear and witness all the unsatisfied customers who over seas have been so happy w all there wall hungs over the years wonder how this will settle across the pond on our side . Cant except to float to the top when your stuck on the bottom w a weight on your ankles .

Makes me wonder why they add a electric heating accessary and well heater s in the piping . Guess not enough emitters.

peace and good luck clammy

DCContrarian · January 27

@John Ruhnke : "What happens at 10F?"

One of the things we've been talking about is letting the heat pump be able to sense the load so it can match its output to the load. We should probably talk about the other part of that, which is that in order for that to work the emitters have to be able to vary their load. With air handlers, the way you do that is to have a thermostatically controlled multi-speed fan in each unit, the output is determined by the fan speed. With passive emitters like radiators or radiant panels the gross control is outdoor reset, the fine control is either zone valves or TRV's.

The only time you should be cycling is when the load is less than the minimum modulation of the heat pump, which is typically 25% of full output. That's only going to happen in warm weather. In cold weather the expectation is that there's always some load, it varies and the heat pump modulates to meet it.

So a classic hydronic system with one thermostat in the first floor hallway controlling all the radiators in the house is not going to be good match for an air-to-water heat pump, you're going to have to do something to give more gradations of output.

John Ruhnke · January 27

@DCContrarian,

What you said makes a lot of sense and I am gradually grasping these principles. There was a lot of advanced math in the Heat Geeks classes. The level of math was advanced beyond my current skills. I had learned it back in high school. I got straight A's back then. I am 60 years old now and that was over forty years ago. So most of it I couldn't remember. Lucky for me my 18 year old son is currently in college taking advanced math and getting A's as we speak. So he tutored me and I was able to grasp it again and pass the courses. I learned a lot of what you are talking about in the Heat Geeks classes.

So I can do the math and make all this work. They talked about TRVs and I have installed them before and balanced out systems. Some old houses from the 30's before circulators were invented had old gravity fed systems. I installed TRV's on these radiators with ODR and got everything balanced. I did a few older houses like this and the homeowners were very thank full. Our AHU's modulate and have a real good turn down ratio. I can make them work with ODR. The FCU have varying outputs with different water temps and I can make them work with ODR too.

DCContrarian · January 27

@John Ruhnke : I feel like this is the barrier that air-to-water has to overcome before it becomes mainstream, it needs products and methods so it can basically be installed using rules of thumb rather than having to engineer every installation.

Kaos · January 28

@clammy I'm heating a 2/3 uninsulated double brick with a heat pump and original rads. Definitely possible even during the recent cold snap. The key is reducing the load which does take work, a fair bit of it was a lot of elbow grease.

The good thing is that reducing the load saves heating costs for any house, so really this should be the first step no matter your fuel source. I know this is hard to believe but you can get an old house without insulation to under 20btu/sqft.

@DCContrarian The more I look at how best to run this, the more I'm leaning towards direct to load and TRVs. The common setup here with zone valves that turn On/Off doesn't work well. This is not just for heat pumps but also for modcons. The nice part is that it is actually a simpler setup to build when starting from scratch as most of the controls bits are eliminated. My unit even comes with an indoor temp sensor that it can control off of instead of a thermostat.

Not easy to implement, but the band aid to zones turning fully on/off is to have a bypass that allows some flow even when off. You still get control from the thermostat but can deliver some heat all the time.

DCContrarian · January 28

@Kaos : I want to make sure we're all on the same page.

My preferred plumbing is a three-pipe system like this:

The buffer tank is there in case the load presented by the emitters is smaller than the minimum modulation of the heat pump, so the heat pump can cycle on and off. In particular, if all the zone valves are closed the buffer tank provides a bypass to keep the heat pump from dead-heading while it finishes the cycle.

With TRV's, it seems you would need to have a similar bypass, or you would have to keep them from shutting off and have a minimum flow level through all the emitters all the time. With passive emitters that would mean a minimum output level. I think you'd see a situation similar to what is often seen with multi-splits, where there is a similar minimum output requirement, which leads to complaints when unwanted heat is being put out.

A layout like I've drawn above works fine with a heat pump. You want to have at least three zones so that you get variation in output and not just on/off behavior. I could see that TRV's would help because they would provide more variation in output.

Kaos · January 28

Like this:

Two port buffer tank if needed in series so there is no mixing. Diff bypass in case all TRVs are closed.

AWHP monitors air temp at reference zone and is used as house thermostat.

You have to be careful with pressure loss or add an extra booster pump controlled by the AWHP.

DCContrarian · January 28

I'd rather put the buffer on the return side so there's no delay in conditioned water hitting the emitters.

DCContrarian · January 28

Does the reference zone have a zone valve?

hot_rod · January 28

the buffer is just a wide open tank, nothing inside? If so I don’t see it adding any value in series like that?

DCContrarian · January 28

https://forum.heatinghelp.com/discussion/comment/1838434#Comment_1838434

It adds enough system volume to prevent short-cycling.

Kaos · January 28

The reference zone is always connected. The AWHP modulates outlet temp to keep the reference zone at the thermostat setpoint. The rest of the zones on TRVs should be sized to provide more heat so the TRVs are always dialing flow down a bit. You can replace TRV with zone valves, but I think a modulating valve works better.

The buffer tank is there in case the reference zone is not big enough to limit cycling. In most cases it should not be needed as long as the reference zone is sized right.

JakeCK · January 28

This is a schematic I drew up a few years back, something I thought about doing to my old boiler. Looks awfully similar to some of the schematics of a2whp's posted above. I wonder how one would work in this configuration with a boiler? Or by its self.

Note: this is shown using a reverse indirect buffer, this was before I installed a hpwh. So that can be removed from diagram.

This also shows a boiler protection loop which can be omitted on a modcon or a2whp

Another strategy I thought of was omitting the zone valve for the largest zone(main floor) and relying entirely on ODR to maintain temperature for that floor.

DCContrarian · January 28

@Kaos : "The AWHP modulates outlet temp to keep the reference zone at the thermostat setpoint."

Is that a feature of your heat pump? The closest I could do with mine would be to have an outdoor reset curve but you'd need something on the reference zone to modulate the output.

With differential bypass I don't see why you couldn't put TRV's on every zone. If all the zones shut the bypass opens, the buffer tank heats up and the heat pump shuts off.

John Ruhnke · January 28

https://forum.heatinghelp.com/discussion/comment/1838417#Comment_1838417

Hi Dc Contrarian,

you are missing a pump on the return back to the heat pump. It is dangerous to pipe this way unless you know what you are doing because according to what I learned if the flow rates and delta tee aren't right you get some bad things happening.

DCContrarian · January 28

@John Ruhnke : my diagram doesn't show the circulator that is internal to the heat pump.

Kaos · January 28

https://forum.heatinghelp.com/discussion/comment/1838441#Comment_1838441

I haven't tried it yet but it the ORD config there room temperature limits, not sure how they would effect the reset curve. I can also set room temp target both in heat and cool with a delta T.

TRVs in every zone does work but then you don't have a thermostat. I think there are TRV actuators with programmable thermostat which would be a better option than on/off zone valve control.

If the heat pump doesn't have room temp sensing, I think it would still work.

My setup is all zone valves, for giggles, I cranked up the target temp on the outdoor unit when it was warmer. This meant the zones were quickly satisfied and the unit started cycling way more than I'm used. So something to watch.

John Ruhnke · January 28

This works great efficiency wise if you have slower flow rates on the secondary side. You will get a lower return water temps that keeps the HP cooler and the storage tank will take longer to heat up and cool down. Almost like having a bigger storage tank. It is critical to have lower flow rates on the secondary side vrs the primary side. This is the best pipng method out of the three pipe buffer tank connections.

John Ruhnke · January 28

https://forum.heatinghelp.com/discussion/comment/1838426#Comment_1838426

The buffer tank in this application is called a volumizer. It is better on the return piping than the feed piping. Its job is to add volume to the piping and radiation when you don't have enough volume. If you have multiple zones and a small baseboard zone comes on you will have the system volume for this to work. Also the volumizer can be sized big enough to handle the defrost cycle. It will keep the HP at lower temps and allow you to control the cycles to the system as the load. This is the best way to pipe up a buffer tank. To use the buffer tank as a volumizer. Again you need to understand some advanced U.K. hydronics to make sure things are right.

hot_rod · January 28

I've heard Dakin is suggesting a minimum of 18L between the monobloc and distribution. They are working with a 5 gallon under counter tank. If you can fine small tanks with large ports it would be nice. Well insulated if used with chilled water.

3 pipe brings the best of two and four pipe options. 3 pipe with the header across the top of the tank for direct to load. I modified one of the Discals to go into the top of a 2" port. If you find a tank with a center top tap.

2 pipe,manning3pipe.png

John Ruhnke · January 28

@hot_rod

According to Heat Geeks. If you use a three pipe tank design with the bypass on the return it allows for lower HP temps and increases the cycle length of the tank. Higher COP and less cycling. The Heat Geeks like to pipe it such as below.

tim smith · January 28

what are natural gas prices in Boston?

Air to Water Heat Pumps

Comments

Categories