Temperature-over-time charts for a well-functioning system?

tagb

Do you work with a well-functioning DHW system for a smaller multifamily building (like 10-50 units), and do you log the temperature at various points in that system? I'd be curious to see your temperature charts.

My building has a domestic hot water system with air-to-water heat pumps, a digital mixing valve, and a variable-speed recirculation pump. To try to learn about the system and help prioritize potential improvements, I'm using simple on-pipe temperature sensors with data logging, but it's hard to interpret these charts with nothing to compare them to.

I'd love to see what temperature patterns should look like in the following locations over time, such as over a typical 24-hour period:

1. Output from heat pumps (which goes into storage tanks)
2. Output from master mixing valve (which draws from storage tanks)
3. First fixture in each branch
4. Last fixture in each branch
5. Recirculation return line

Do you have anything like that? Or do know where I could find this kind of thing?

hot_rod

what type of mixing valve? Larger application are going with electronic type . So the outlet from the mixer should not vary more than a few degrees? Unless the storage drops below the setpoint?

Depending on on the balance valves at different loops, the recirc with a VS pump should not vary either. Thermo balance valves are the way to solve recirc temperature issues

tagb

Electronic mixing valve. No balancing valves on the individual branches, which is part of what I'm looking into (been reading https://idronics.caleffi.com/magazine/21-recirculating-domestic-hot-water-systems ).

hot_rod

Yes, with multiple branches you need to balance somehow. The problem with manual valves is they cannot sense temperature change.

The thermal valves notice a temperature drop and open, which allows the delta P circ to ramp up. The entire system should regulate within a few degrees anywhere in any loop.

Idronics issue 24 goes into more detail

tagb

Thanks! One thing I'm currently puzzling over is that our mixing valve seems to want a minimum of 5 GPM through the recirculation pump. We have two branches, so if we put Thermosetter 116s on both branches, it seems like combined return flow could go below 5 GPM. Seems maybe ok if we set the balancing valves fairly high (to only limit flow a little)?

(Not installing anything myself, would hire people for it, but found that it helps to try to learn as much as possible first.)

Homemade diagram (this is an overhead view of the horizontal runs; units are townhouse-style):

hot_rod

It would be good to know what gpm the two loops require. Is it copper pipe? Insulated?

The gpm has to do with the heat loss of the piping. The math for that is in Idronics 24

Do you know or have an estimate to the total gpm flow for the building, the hot side?

The mix valve wants to be sized close to that flow rate. Most often the valves are sized to fit the pipe and way over-sized to flow. So the valves cannot mix properly or need high minimum flow. With electronic valves this forces them to operate with flow choked way down and causes wear on the internals. Its called wire draw, erosion at a valve that is in the nearly closed position.

In large demand buildings it is often better to use two small valves in parallel. They respond better when loads are low, minimum flow requirements drop, and you have redundancy should one need a cleaning

With thermostatic mix valves you would use a Hi-Lo mixing station, a large, say 2" valve, with a 3/4 valve piped with it for low load accuracy.

With digital/ electronic just split the load.

More and more our reps are investing in portable ultrasonic flow meters. They connect to the hot line and monitor for a few weeks or month to get an accurate DHW load number. It is often much lower than assumed, especially buildings with elderly tenants.

The free IAMPO Water Demand Calculator is useful for todays DHW load calcs.

So… come up with a close DHW demand, check the mix valve sizing, and see if you can nail down required DHW recirc flow.

The nice thing with thermal balance valves and ∆P recirc pumps, they will find the sweet spot. You are using a temperature solution for a temperature problem. They are excellent problem solvers in buildings that have no data or have changed over the years.

Assuming the piping is correct, no cross connections upsetting the temperature conditions.

tagb

Copper pipe, mostly 3/4" foam insulation, runs at maximum 120F through spaces that average around 60F ambient. Estimating we need around 4-5 GPM total recirc currently, could be lower with better insulation.

I did a fixture count, checked out the IAMPO demand calculator, considered our community and our patterns/needs, estimated our max DHW demand around 25-27 GPM. That's slightly conservative but it could be even lower in practice. California has strong standards for flow restricted fixtures. I'd love to get some actual flow measurements.

Unfortunately, yes, the mixing valve was sized according to line size (2"), so it's very oversized. I do think a 3/4" hi-lo valve would be more appropriate, although it's tricky to justify replacing a very expensive piece of equipment that is only a year old. I've wondered if we could sell it on eBay or something to fund replacing it, but I don't know if anyone in their right mind would buy a used mixing valve…

But I think this is why my temperature readings are confusing and spiky: I haven't found any flow control valves at all on the return lines, no check valves or manual balancing valves. Our trunk return line has a swing check valve on a vertical line that flows downward, so that check valve is probably doing nothing. Then, right before the recirc pump, there's a 3/4" Thermosetter…not the best place for it.

So I suspect we need to remove the thermal balancing valve from the return line, put check valves and thermal balancing valves at the ends of the two branches (ok if we need them set relatively high), and replace that swing check valve with a spring check valve. Then figure out if we need to do anything else in the near term. Medium term, upgrade the insulation on the supply lines, especially for the longer branch. Longer term, replace the mixing valve.

Kaos

This is for house, but similar issues.

For the two branches, you need flow checks on each otherwise the flow can reverse under high draws. As long as the two branches are somewhat balanced, you don't need much balancing. The ideal setup is a two thermosetters, one on each branch.

Recirc flow should be set to maintain a reasonable temp at the last fixture at the loop. This is a very squishy number, I find 100F in the ballpark for me. Less than that and you are waiting for hot water, more than that and the recirc runs too much and you have much larger standby losses and the heat pump runs more often.

With heat pump as the source, you don't want to be mixing water down much. The only reason to mix is if you need extra capacity and the buffer tanks are not big enough. Best way to reduce power use is to bring the heat pump output close to what your load actually needs. It also can't be too cold as you need to run at pasteurization temps (or at least periodically bring it up to it).

If you are using a CO2 refrigerant heat pump, those need a highly stratified tank. If you look closely at the diagram HotRod posted, the recirc goes the middle of the tank not to the bottom. This is what you want to avoid mixing the buffer tank and reducing stratification.

hot_rod

https://forum.heatinghelp.com/discussion/comment/1854737#Comment_1854737

a 1” Caleffi Legiomix would easily cover that, maybe even a 3/4 so that drops the recirc gpm issue way down from the 5 gpm you have now.

If there are legionella concerns you would need to run the system up higher. It’s a time and temperature game with legionella, the hotter the temperature, the shorter the time. There was some talk about 160f for 20 minutes. But point of use protection POU, becomes so critical at that extreme high temperature disinfection time period.

Flow velocity in a 2” DHW supply will be very low with 25-27 gpm, so probably some scale build up and bio film in those lines. That is where the bacteria lives and thrives. Another reason to consider an ocassionally high temp blast. That is the big feature of the Legiomix, the ability to open up for a timed period and allow the higher temperature. Not ideal for the HP however😗

tagb

Thanks! Storage tank setpoint is 140F. My understanding is that should help keep Legionella under control. I don't think we have any point-of-use scald protection, so we'd need to do a lot of work before we could run routine disinfection blasts through the distribution pipes. Would be nice though.

hot_rod

https://forum.heatinghelp.com/discussion/comment/1854749#Comment_1854749

it just depends on the legionella comfort level concerns. It is the slime in the piping where the bugs live

Chemical treatment is another option if it is even a concern?

140 tank temperature will hit your HP efficiency a bit as @Kaos mentioned

Sounds like a recirc mix valve upgrade will fix many of the issues

tagb

My understanding is that higher storage temperatures are good for efficiency with our multi-pass heat pumps, since it makes them run longer cycles. Here's an example of temperature logging on the pipe from the heat pumps to the storage tanks, showing the gradual increase during heating cycles:

@hot_rod Interesting, it seems to me that improving the flow control is the most important first step, to see whether that gets us more consistent delivery, then figure out what to do about the mixing valve. Curious if you think differently?

hot_rod

Think of arecirc system as a hydronic loop. You want to dial in just the correct amount of flow to cover the "load" theload in your cse is adequate temperature at theof the loop. Generally if you leave the mix valve at 120, expect around a 10° drop to the last fixture. Your pump speed (flow rate) determines that temperature drop.

That is what a thermal balance valve looks for, the loops end. Then it monitors and adjusts accordingly.

The issue you may have is the mix valve wants 5 gpm, but you are circulating 2 gpm. The mix valve will "hunt" around trying to find the exact mix. It may be just right, then a bit too cold, or a bit over-shoot.

So all things need to be on the same page, accurate mix, correct recirc flow, long and short loops balanced. When everything is correct you get an efficient system both from gas usage and power to run the circ. Happy tenants, and a trouble free system.

In large buildings cross connections can throw a wrench in the works. Someone connects a Y hose, a single handle mixer crossing over, even kitchen faucets with electronics can cause cross over.

pecmsg

https://forum.heatinghelp.com/discussion/comment/1854749#Comment_1854749

140* for 1 hour kills the legionella bacteria.