How to best maintain temperature

Constantin

Like motorized mixing valves, I doubt that the temperature can be as tightly controlled as called for by the test procedure while allowing a wide range of flow rates. That is the beauty of thermostatic mixing valves, they can react quite quickly. On the flipside, solutions like the Megatron are $$$ and require a minimum flow rate to ensure that the txv does it's thing. That minim flow is a significant portion of the expected heat gain to the chiller tank as even small recirculation flows of 1GPM add up over time.

Anyway, thanks for the tip. I think we've settled on the megatron foe the hot and the cold side tempering.

Constantin

How would you do it?

If someone tasked you with a challenge like the following:

maintain a domestic hot water supply at 135F +- 1F at 4 outlets with a draw of between 0.0 and 6GPM, what is the best method? My thought would be to use a precision txv and a continuous recirculation loop in conjunction with a big storage water heater at 140F but don't know of that is precise enough (thermal drift, etc)

also, if you have had a good experience with a product that fulfills this sort of need, please let me know...

Jamie Hall

Gotta love it

I can't think of any other way than what you have suggested to maintain a tight spec. like that. If you get really good mixing valves, which are themselves spec'ed to hold that tolerance, you should be OK with the system. The water heater, though, will have to able to deliver 36 gpm at 140 F indefinetly for the thing to work as advertised -- and that's calling for a lot of recovery.

Paul Pollets

Indirect or Direct?

My preference would be to use 2-VitoCell 120g tanks twinned together if a boiler is present. The boiler should have at least 150K btu's available for the HW coils.
If not, I'd use an AOSmith 100g "Cyclone" direct fired water heater. A quality thermostatic mixing valve works on either system.

To Learn More About This Professional, Click Here to Visit Their Ad in "Find A Professional"

Constantin

Thanks for the feedback!

Yup, the maximum flow rate is pretty close to those trophy showers in some homes. The loads are fairly similar in that appliances etc to be tested here do not usually have constant flow.

My thought is to use a high input water heater like the ef series from bw, etc as a 100 gallon storage tank in conjunction with a fairly high input rating to provide a reasonably-priced but effective hot water solution.

The mixing was a problem though as many txvs have mnimum flow rates to be effective and accurate. In the end, I think we'll settle on a Leonard product called the megatron1n because it will speed the lab completion, even if it's sold at a nice premium.

That solves the hot side for now. However, on the cold side, we discovered that the incoming water temperature is too high, necessitating a chiller system. We'll likely follow a simar approach, ie a big tank as a battery with a split chiller providning the cooling. Trouble is, I have yet to identify anything beyond the megatron which can hold a tight temperature specification at a low temperature like 60*F.

Powers seems to have a product called the powerstation that might fit the bill but if any of you know of a product (equivalent or better) please let me know. Also, if anyone has had a good experience with a chiller manufacturer, I'd love to get in touch with someone in that industry.

Cheers and thanks again. C.

Constantin

Thanks to you as well!

As usual, we are space constrained, so two tanks is all we get room for. To increase lab space inside, we have to keep the hot and the cold on the outside, ie one tank each.

To keep a tight tolerance on water temperatures across a wide spectrum of flow rates is an interesting problem. Also, I apologize for the many spelling mistakes, this site and my iPhone do not play nice, so I can't go back and edit out the worst mistakes. Oh well !

The Steam Whisperer (Formerly Boilerpro)

You may also want to look at....

the bigger 80 to 120 gal Triangle Tube indirect. These have 1 1/2 inch domestic water tappings to handle the high flow rates to help keep pressure stable to the mixing valve. The 120 gal model has 2 inch boiler side tappings that would allow the boiler side to be overpumped, helping maintain more even tank temperatures top to bottom to help keep the outlet temperature stable. Tie that to a Prestige 250 wall hung boiler with a sensor in the tank for feedback directly to the boiler so the boiler input can modulate according to load to further stabilize tank temperatures. The tank temperature setpoint and differentials can be set right at the boiler control.

I'd also pump the recirc through the tank heavily to keep the tank well stirred.

You may also want to look at taking the chiller heat output and use it to help heat the hot water tank.With this combination, costs would be really low for waht sounds to be a potentially very enegy intensive application.

Boilerpro

To Learn More About This Professional, Click Here to Visit Their Ad in "Find A Professional"

hot_rod

That plus or minus 1F

seems like the challenge. Most mix valves operate in a 3F range. But with enough $$ I'm sure something is available.

hr

Constantin

Thank you...


I looked into indirects for both the heating and cooling portion... On the hot side, the units you mention make for a very nice system that will last many years. However, we have received pushback on anything that isn't fully assembled. So a separate boiler and storage tank is not a well-liked solution. Hence, I suggested they look into using a Bradford white ef100-250 which offers a all-in-one solution... with the price tag to match.

On the cold side, things are a bit trickier because the tank temp ideally is somewhere around 45*F. There are many chiller companies, but most of them offer product with unpressurized water storage (makes a lot of senece considering how many chillers are used). An indirect tank w/a glycol loop could work in this application except that chiller manufacturers typically aim for a 5*F delta-T between supply and return. In order to get good HX, the chiller loop would have to run glycol, etc.

What I proposed instead was using a 120 gallon electric water heater in conjunction with a split-system chiller. The indoor unit consists of a recirculator, a thermostat, and a refrigerant to water HX. The chiller loop would pull from the hot water connection and return to the upper element spud. The cold water fill tube would be removed so the warn city water enters at the top of the tank. I'd rely on cold water dropping as the chiller loop runs and warm water rising as the tank gains heat.

The reason I like this proposal is that it consists of simple components and the top 1/4 of the tank water mass is available for the chiller system to play buffer with. The cold water supply for the lab would be drawn from the drain hole. However, the folk in the lab want a fully-integrated solution. Where to find such a chiller system (80 psi, 100 gallon storage capacity, etc ) is a good question, however.

Constantin

A bit wider...

The tolerance calls for +- 5*F in the test procedure though we wnt to be as tight as possible. According to the folk at Leonard, we should be able to hold 3*F or better since their Megatron1n incorporates a recirculator to ensure constant flow through the mixing valve. A very premium solution!

Derheatmeister

Pri/sec Injection

How about using a Pri/Sec Injection type of control like a Tekmar on the recirc line ?
Good luck with this project... Richard

Larry C_13

A little late, but ....

Why are you not using mixing valves at the point of use? Wouldn't it make more sense to use individual lower flow rate temperature control valves instead of a sihgle valve that has to go from zero to a large flow rate?

Larry C

Baltimore/Maine Doug_9

For microwave oven testing of output,

the water was drawn from a large storage tank that was kept at the start of test water temp. The hot water resulting from the microwave test was returned to the tank where serious mixing blended it in. The stats in the tank were able to measure and control because stats do that well and the buffer was large enough to prevent swings from water in and water out.

There were no flow dependent regulating valves mixing water, it came from the tank at the right temp. Of course the requirement was for one temp water. I guess in your case you need water at two temps, one hot and one cold.

Are the ones who object to some of your solutions the ones that know know the least about how to solve the problem? ;-}

Constantin

Hey, I know that test!

The thing about the microwave test is that only 1kg is needed at a time, and the delta-T is but 10*C. Fond memories of melting a lot of ice in the basement... by the outlet underneath the main panel, the only place where the voltage deviated less than 1% despite 1900W+ being drawn by the microwaves under test.

The lab needs are a much higher, up to 100GPH at 35psi with up to 3GPM per station of either hot or cold water.

Baltimore/Maine Doug_9

Hey, I know that test!

Yup. On a large asembly line in Japan, U shaped. One leg for assembly and the other for testing. Impressive to watch.