Vacuum boost retrofit, For 1pipe and 2 pipe steam systems

AMservices

@P1385 The home I retrofitted had a 2 pipe air vented steam system. If you haven't heard of one, it's basically a 1 pipe steam system with a drain on the return side and air vents on all of the radiators. No steam traps.
Because of the unique steam system I was working with I had to run Vacuum lines to each of the radiators, I ran a vacuum main line next to the steam mains , I used ball valves as balancing valves on each of the vacuum lines at the radiators And at the branches where the main splits so I could balance the air leaving the system.
When the system starts the vacuum pump turns on. From a cold start the boiler is hot in less than 10 minutes. Within the 1st 15 minutes the radiators are nearly a 1/3 full. We control the vacuum pump with a vacuum switch set to cut out at 8HG, Cuts in at 5HG.
As the steam fills the system it compresses the air in the piping and the radiators towards the vacuum lines where the air vent would normally be. So The pump will cycle on and off a couple of times before all of the air is evacuated and the entire system is filled. You know if you have it right if all the radiators are hot The for steam makes it back to the pump.
When steam makes it back to the pump, a high limit control will shut it off and the rest of the cycle Is business as usual for a steam system. Pressuretrol controls the on off of the boiler, System runs until thermostat is satisfied.
When the thermostat is satisfied, The burner shuts off and the condensing steam create its own naturally induce vacuum, steadily lowering the boiling point of the water. whatever heat remains in the system will be moving towards the radiators that are condensing the steam the fastest. This is how the system becomes self balancing and there is a noticeable difference in comfort Throughout the house.

It can be even easier to retro fit a 2 pipe steam system that uses steam traps.
The return piping can be the vacuum lines And the vacuum pump Connects to where the main air vents are in the basement.
But who likes changing steam traps?
If I was designing a system from scratch, I would make it a 2 pipe steam system and I would throttle the supply valve to the radiator with orifices or balancing valves
And control the flow rate of the steam into the radiator so that It condenses the steam before it makes it to the return. Or I could do it with a one pipe steam system and save the money on the return piping and just run 3/8 vacuum lines to the radiators.
There are so many more options Designing the piping of a Steam system When using a mechanical vacuum pump. 1st, all of your mains and run outs can be 2 pipe sizes smaller If the steam, air and condensate are all flowing in the same direction. The drawback to that is You will always need the pump to have it running efficiently.
The good thing About the vacuum pumps we're using
Is they are a medical grade, oil free vacuum pump That can run for something crazy like 20000 hours without service. @izhadano Figured the amount that his vacuum pump runs, it would last 30 years. And that's only because his piping is old and it leaks.
New piping will be tight and possibly indefinitely hold whatever vacuum level you're making when the Steam condenses after the 1st cycle.
There are so many options with the controls we have these days, I could write all night about things I want to try With modulating burners, Controlling the steam by temperature instead of pressure and even zoning Larger systems.

To understand steam systems, you need to see how the air is being Pushed or pulled from the system, the way steam Is filling the system, How the condensate is returning to the boiler.

jumper

Well AMservices you've done it & I haven't. So maybe I'm missing something.If there's no traps then whole system sees same static (when no steam is generated) pressure. Why not evacuate air from anyplace you like? Seal off the exit point before firing the boiler.

Leon82

This is cool.

Are those California air compressor motors? I came across them looking for an air compressor. They were said to be the quietest ones and they look similar

mikeg2015

Water line will be balanced if you have an equalizer. The dry return connects to that pipe below the waterline, or you can have a hartford loop as well, but not much point if you have a single large dry return, where the only pipe failure you could have would the the piping in the hartford loop itself. sort of defeats the point.

I thought that generally, you had to draw vacuum from all the radiators by using vacuum vents that closed when steam hit them. They could have a simple needle valve on them to control vacuum flow rate. Then you have a vacuum pump on the main.

In theory, if the valve steams sealed almost perfectly *which is possible but takes repacking most all of them on an old house, you could draw a very, very high vacuum and just remove all the air. Then only fill the system with steam. If you fire the boiler at a higher rate, pressure increases, boiler temp goes up and so on. When it shuts down, vacuum naturally pulls down as it cools.

The only purpose of the vacuum pump is then to remove air that leaks in periodically. You also might be able to store vacuum somewhat with a storage tank, maybe install a steam trap on it and draw the vacuum off a vertical riser connected to the drain on that tank which returns condensate on a wet return.

Done well, I'd argue this is better than hot water because there's no circulator, smaller pipes, smaller radiators, and the system still has a 5:1 turn down ratio on radiator output. You could increase economy further and possibly use gravity hot water in some rooms.

PMJ

@AMservices ,

My issue with what you describe is your hurry up aspect to fill the entire system on every call for heat. I agree that evacuating the system first will fill it faster. But filling the entire system on every call for heat cannot produce particularly even heat. My entire effort is to avoid ever filling at all or at least definitely never quickly. The most even heat is when rads are partly full to an amount that exactly matches the loss and never change that amount of fullness + or - quickly. In the original coal system nothing could ever change quickly. Running any boiler to a pressure stop or continuously to a satified Tstat will always overfill for the conditions on every cycle. The room temperature swings may be tolerable but certainly not desirable.

AMservices

@jumper You can't pull the air from anywhere in the system for the same reason you want the air vents on a steam system at the ends of the mains and 1/2 way down the far side of a radiator.
Let's say we have a vacuum pump make 15HG in the system before the boiler starts, there will still be 50% air remaining in the system. As the steam starts Moving into the supply piping It will be compressing the remaining air In the mains, the run outs and into the radiators. You would probably have to Pull 90% Of the air out of the system and give the remaining air somewhat of an expansion tank to collect in. If yo u could maintain at least 26 to 28HG In you could practically have your house work like a Heat Pipe.
You could fill it up once with purified water, do everything in brazed copper, But then again steel wouldn't rust with almost no air in the system, Pull a vacuum of 28 hg, Put some birthday candles under the boiler and there you go.
I'm sure the dead man must of tried But experienced problems with air leaking back into the system, Maybe between the boiler sections or the radiator valves, making the system unbalanced.

AMservices

Another thought about a heat pipe style steam system.
You wouldn't even need a vacuum pump, you could fire the boiler and purge the air through coin vents on the sides of the radiators And if all your piping was tight, it would Always maintain a vacuum.

AMservices

@PMJ Yes hurry up and fill the system. The faster I get heat to the space, the sooner the boiler can shut off in the natural vacuum can do its thing. Let's say a radiator does heat up quicker than I want it to, all I have to do is Throttle down the ball Valve on the Vacuum line to that radiator and It will heat slower on the next cycle. I explain that to the homeowner and he had his system working Just the way he wanted it after the 2nd day.
The vacuum pump gives you a lot more control over The lowest point of pressure. By controlling the lowest point of pressure I can direct the steam to fill whatever radiator I want as fast as I want.
Just because I can fill a system in 30 minutes doesn't mean that it needs to fill every time the thermostat calls for heat.

PMJ

I'm still not clear then. You said before that the pump maintains 5-8 "HG when the boiler is not running. So on a call for heat with the burner on and much less than 5 " in the system the pump runs until it sees steam all the way thru at least one rad somewhere. I am reading this to be on every new cycle. If you have adjusted things so that all rads are approximately equally full then at the front end of every cycle you have overfilled the entire system for the conditions and will overshoot. There will be a longer off period to get back to tsat cut in and then you will do it all over again.

I grant you that your vacuum pump speeds even faster this hurry up and fill process. The vacuum will also even things out room to room. But then leaving the boiler itself to simply pressure controls and/or tstat call will overshoot the target each time. The overshoot may not be much in degrees, and most people with steam are just used to it. But by definition the shorter the time from the start of the call to the tstat hitting cutout, the more the overshoot. I find that most people are trying to do exactly this - satisfy the call as quickly as possible. That is a choice everyone can make for themselves. It is clearly not the most even heat. I work at doing the opposite. My goal is to have calls that are as long as possible - hours. And I will point out that coal may have had undesirable smoke, but the heat was steady and even. Hurry up of any kind with the original design of these systems was quite impossible. I am trying to approximate that. Capacity control and reduction is absolutely required for this - either through modulation or controlled cycling.

AMservices

@PMJ I never said the pump maintained 5-8 th when the system is not calling for heat. The pump turns on when the thermostat calls for heat.
The vacuum switch cuts out at 8 hg and cuts in at 5hg.
The pump runs every cycle because the old piping has leaks and can not hold a vacuum, so air needs to be evacuated every cycle.

You keep saying this system will be unbalanced or it will over heat the way I'm running it. Let me tell you again, that's not how it works.
Every room is as hot or cold as the home owner wanted to make it. For example, he likes his Room cold, so he keeps his vacuum line off. His in laws who are in the room next to his Like it to be 95ﾟ, They keep their vacuum line all the way open.
There are 3 generations of people that live in this house and every one of them has told me that there is a noticeable difference in comfort.

When the room with the thermostat is satisfied, the boiler turns off and the steam moves to the radiators that are colder then then ones in the room with the thermostat.
Its that whole self balancing effect we've been talking about.

I did say there are a lot of control options.
I never said pressure controls are the only way to go.
Simple usually works and works Reliably.

You said
"leaving the boiler itself to simply pressure controls and/or tstat call will overshoot the target each time."

Do you really think every one controlling steam systems with thermostats and pressure controls over heat?

Is it always best to run a steam system in timed on off cycles.

You want to keep imagining this system in your own way and trying to tell me, the person standing in front of it what's it's really doing.
Your not testing these things for yourself.
I think you need a vacuum pump to play with. Then you would understand.

PMJ

@AMservices,

I am sorry you have received my comments as an attack on your work. I apologize as I did not mean to do that at all. I am quite convinced that your pump system does in fact add significant additional control options and has indeed improved the heat in that house.

The subject I was addressing is the theoretical one of even heat as it relates to boiler run time. It is without question that a boiler running continuously at a rate capable of heating the house on design day from the call for heat to cut out of the thermostat will fill rads more than is required for the conditions and overshoot the temp setting some amount. This small amount is what most people have simply become used to and in fact is quite tolerable. But the rads will go quite warm to near room temp every cycle. It is certain they will not be steady state.

There are those of us who wish to do better than that and not have our rads go so hot and so cold every cycle. Quite a few on this forum have contacted me directly wanting to do better with this very issue. So the point I am making is that truly even heat means rads never change temperature quickly. Truly even heat means the call for heat can't be satisfied quickly. Truly even heat means no hurry up about anything ever - like the coal fire. I was just reacting to your comment "yes, hurry up and fill the system". I never want to do that.

So yes, I am saying truly even heat does require some timed control of the fire to reduce the net capacity of the boiler. Two stage valves and modulation gets you closer too but at higher cost. Timed control is the most adjustable and cost effective way to address this issue.

mikeg2015

I agree, that timer control is a cheap and simple way to reduce burner output. A very low frequency pulse width modulation solution if you will.

For a quick and dirty solution, I was just going to add a timer that does maybe a 10 minutes on, 5 minutes off cycle. That gets my NET output pretty close to matching my EDR with header and boiler jacket losses included.

acwagner

I also agree on the timer, especially on 1-pipe systems. My boiler is not that oversized (maybe 10-15%). It only becomes an issue when we recover from a setback. Nonetheless, I think the system runs better with the timer.

For those that want to try adding a timer, I recommend this:

http://www.galco.com/buy/Macromatic/TR-6512U

I believe I saw a post from PMJ on using this type/brand timer relay, so he deserves the credit. Just don't forget to order the socket it mounts into as well (rookie mistake on my part). It takes any power input and the timer ranges from seconds to 100 hours.

I just put it in series after the thermostat. I'm currently at 18 minutes on and 12 minutes off.

PMJ

Thanks @acwagner and @mikeg2015.

We do get that this probably isn't for contractors without any proven off the shelf controls to buy. Ecosteam did address all this. I don't know much about it other than from the theory side it was dead on. Anyway, for those of us that are live in steam enthusiasts timing control is an easy way to enhance things a lot.

For those heading down this road some things to think about:

Moving up to a PLC platform allows for more than one % run/wait time combination. One firing rate does not fit all conditions if you want even heat - the whole point I have been pushing. I currently have 3 levels and I'm considering a 4th. On every new call the PLC looks for steam at the end of the main and runs continuously until it sees that - worst case is maybe 25 minutes from dead cold everything. After that it runs the first level - 7min on 13 min off. After one hour of that (including preheat if there was one) it goes to 8/12. After 2 hours of that (3 hours call total) it goes to 10/10. Usually takesdown into the teens outside to get that far. But when it is even moderately cold(20F) calls are 2-3 hours long and not much off time as there is minimal if any overshoot at tstat cutout. Rads never cold or full hot the whole time. The best thing about the PLC is you are free to do it your own way. Very enjoyable stuff to be able to have a big impact from just your laptop.

On setbacks - on top of the above operation I recover from setbacks gradually using the digital thermostat's wake,leave, and return functions with sleep being the setback itself. So I don't even attempt to recover the whole thing at once. I don't use a large setback, only 66 - enough to have the boiler off 2-3 hours at bedtime. I step it back up thru the morning in 3 steps a degree at a time. No one seems to notice the lower temps on the slow ramp up as rads are always warm with "live heat" as grandmother called it. Clearly more efficient overall as the house spends so many hours each day a couple degrees below the normal set point that total losses to the outside are clearly lower. Again, the savings is not the reason I setback - only for comfort.

acwagner

Back to the vacuum discussion. I was insulating my near boiler piping this afternoon and was watching my boiler in operation. I put a 0-32" Water Column pressure gauge on my boiler a few weeks ago. When I first got it, 0" WC was atmosphere. Now, atmosphere is more like +0.5" WC. Anyways, after my boiler shutdown I noticed the gauge went down to 0" WC again, suggesting it was in vacuum. For the first 5 minutes after it shut down I could hear the water still boiling in the vessel. The gauge stayed at 0" WC for at least 20 minutes after the boiler was off. More than an hour later it was back at +0.5" WC.

So a question for you vacuum guys: what gauge to you recommend I get if I want to investigate what kind of vacuum I'm seeing? A compound gauge that shows both above and below atmosphere?

Fred

I notice my gauge, on my one pipe system does that too. What I found is that the vacuum is actually in the pigtail, between the water seal and the gauge.

jumper

Ideally a heating boiler does run on a schedule (timer). And the schedule is determined by outdoor temperature. (And maybe wind?)

Naturally induced vacuum cannot eliminate as much air as a pump will.Not even close.

Does anyone here disagree?

acwagner

@Fred

So you somehow independently verified on yours that the boiler vessel wasn't in vacuum? I can't think of a situation that the air in the pigtail would be at a different pressure than the vessel unless either the pigtail was clogged or the boiler water level was above the pigtail.

Fred

@acwagner I really can't verify with 100% certainty but what I can surmise is that when the steam in the boiler is collapsing, that vacuum pulls on the water in the pigtail loop and that pulls that gauge into what looks like a vacuum. My logic being the gauge looks to be in vacuum long after the main vents have been fully open and some of my radiator vents never actually closed because the radiator wasn't full so the system itself can't be in vacuum.

AMservices

@PMJ the radiators are cast iron. They will heat up faster with the vacuum pump, but they release the heat slowly for awhile after the thermostat satisfies.
Your under estimating how much control you gain when you control the lowest point of pressure.
I can control the vacuum level and how fast it will exhaust the air from each radiator.
If you like low temp and slow heating, that can be achieved.
I have no problem with PLC Or on off cycling To maximize heating output and efficiency.
The problem I have as a very busy contractor is the time it takes to get them timed out right.
Most homeowners don't wanna be thinking about adjusting the cycling. They just want what they know and a thermostat is easy to understand.
I'm sure if you put a vacuum pump on your system, you have the knowledge to have it do amazing thing.
Ask Santa for one. Maybe it will be under your boiler Christmas morning

Pumpguy

The usual compound gauges fitted to the vacuum condensate return pumps for 2 pipe vacuum return systems that I deal with are graduated 30-0-15. I frequently see compound gauges with this graduation range on boilers too.

I sometimes see 30-0-30 gauges used, but the pressure side of the scale takes up too much of the gauge's total pressure measuring range, and the range for measuring vacuum suffers as a result.

Ideally, I would like to see 30-0-10 compound gauges used, but I don't think such an animal is available.

I understand more sensitive pressure measuring devices are available, but don't know too much about them. Photohelic is one brand that comes to mind.

acwagner

I see what your saying, @Fred. It's not that the gauge is giving a false reading, it's just what's going on in the boiler vessel isn't representative of other parts of the system.

I may investigate, regardless. I use Gorton vents on my radiators. I pulled out a spare one and I can blow air thru it but I can't suck air thru it. So, even though my vents don't close during the heating cycle, I don't think they let air in. I also read somewhere that Gorton main vents only open to break the vacuum below a certain temperature, like 140 degrees F or something. So it's possible it's holding a vacuum until the temperature of my main vents drop and they open again.

Just me spit balling here.

mikeg2015

What are some good, inexpensive PLC's? I've used Allen Bradleys for several years in a previous job. Pretty fluent in ladder logic. But don't want to break the bank. But I agree that just using relays is limiting.

PMJ

> @AMservices,

As I said before I understand why this is not a contractor thing. The owners that will pay for anyone to fiddle with vacuum or anything beyond conventional pressure controls will be few. That is a shame but there it is.

My wife will not tolerate the noise of a pump. Currently the loudest noise in the system is the burner. I am firing my burner into 4-5" HG of natural vacuum 90% of the time for free with zero noise. It is so much quieter since I stopped venting and went to natural vacuum the total lack of noise has become an important feature.

Let's just say I am operating with things a contractor really can't. And I do understand completely why that is.

PMJ

@mikeg2015,

Look up Teco at Factorymation.com. $100+/- and you are in business. I have AB at my factory too. No need for all that.

P1385

A question about natural vacuum; I'm working my way through The Lost Art of Steam Heating and have read where Dan Holohan mentions it, but I think I'm missing something. In a 2 pipe system are the vacuum breakers removed? Is it a matter of when the steam condenses a natural vacuum results and the absence of vacuum breakers ensures the natural vacuum is held? If so, how is the "straw effect" avoided on the return side of a radiator? I'm imagining a slug of condensate suspended in the pipe after the radiator trap.

PMJ

P1385 said:

A question about natural vacuum; I'm working my way through The Lost Art of Steam Heating and have read where Dan Holohan mentions it, but I think I'm missing something. In a 2 pipe system are the vacuum breakers removed? Is it a matter of when the steam condenses a natural vacuum results and the absence of vacuum breakers ensures the natural vacuum is held? If so, how is the "straw effect" avoided on the return side of a radiator? I'm imagining a slug of condensate suspended in the pipe after the radiator trap.

Yes, in 2 pipe if everything is closed up when the burner goes off the collapsing steam creates the vacuum. There is no straw effect anywhere. No return pipe(except the wet return) is ever full and no column of water can be suspended in them. You need to envision the system as normal - just with the entire thing at a different(lower) pressure. Gravity is still the only significant force in where the condensate goes same as above atmospheric.

P1385

Another question; for mechanically induced vacuum how is steam kept from the pump when all air has been evacuated? I'm envisioning a pump that cuts out in at 8 IN/HG and out at 12 IN/HG, or some similar band. Once all air has been expelled from the system the steam would have a certain "pressure" if even below atmosphere. It seems like this would trip the pump to run. Am I thinking about this wrong? Does the thermostat end up cycling the boiler within this range and the pump doesn't trip on?

AMservices

If steam gets anywhere near the pump a high limit control will cut the power.
I use a simple Honeywell aqua stat.
The vacuum switch and the aqua stat a wired in series so either one of them will turn the pump off.

P1385

PMJ said:

Yes, in 2 pipe if everything is closed up when the burner goes off the collapsing steam creates the vacuum. There is no straw effect anywhere. No return pipe(except the wet return) is ever full and no column of water can be suspended in them. You need to envision the system as normal - just with the entire thing at a different(lower) pressure. Gravity is still the only significant force in where the condensate goes same as above atmospheric.

@PMJ So, to the untrained eye, it would seem that a well designed two-pipe, gravity-return, natural vacuum system is a straightforward way to have the efficiency benefits of vacuum without using a pump. The only components, outside of the boiler and piping, are traps, air vents, and some type of control at each radiator, correct?

PMJ

P1385 said:

@PMJ So, to the untrained eye, it would seem that a well designed two-pipe, gravity-return, natural vacuum system is a straightforward way to have the efficiency benefits of vacuum without using a pump. The only components, outside of the boiler and piping, are traps, air vents, and some type of control at each radiator, correct?

Yes exactly, simple and straightforward. To start you need only one vent place on the dry return with a check valve on it. No steam there so no worries at all - just a regular check valve to get started. You will want to go to something with zero cracking pressure down the road but not necessary at the beginning of your work. No vents required on supply main. And though I have the original throttle valves on all rads the balancing effect of the vacuum has allowed me to forget about those and I just leave them all wide open.

P1385

PMJ said:

you need only one vent place on the dry return with a check valve on it

The check valve you mention would be in the outlet of the dry-return air vent, or only a check valve (no air vent)? Would this one vent (or check valve) handle all air venting responsibility for the entire system?

PMJ

P1385 said:

PMJ said:

you need only one vent place on the dry return with a check valve on it

The check valve you mention would be in the outlet of the dry-return air vent, or only a check valve (no air vent)? Would this one vent (or check valve) handle all air venting responsibility for the entire system?

When I started this I replaced the dry return vent with a check valve only and capped the steam main vent nipple. Anyone with steam in the dry return of two pipe has big problems that need to be addressed before any of this. Just this one vent handles the whole system. It seems to be a surprise to many but if you simply don't let the air back in to these systems then venting ceases to be any kind of issue at all. I have a 1000EDR system and vent the entire thing though one 3/4 pipe nipple. On cold start it takes longer to heat pipe than vent air through a small hole. When you actually need heat there aren't any cold starts! After the system is warm if you just don't let the air back in there is next to nothing to vent. There is never enough volume coming out of my nipple to make a sound even on the cold start. The effort and expense of vents to horse air in and out of a two pipe system on every cycle has never made any sense to me. That's what I moved into 26 years ago and finally did away with all that.

Koan

@PMJ : Just (finally) ordered the SG2 ULINK cable to program the TECO. I'll PM you about the data file. Just a note looks like TECO is discontinuing the SG-2 series. I got the Dwyer pressure sensor and the solenoid. It would be great to talk through were you put the temperature sensor ( I think you said at the end of the longest steam run). Thanks!

P1385

@PMJ Upon startup, assuming a tight system, is the system under vacuum from the last time it was operated?

If I do go ahead with a natural vacuum steam system on my new house, I might use Niron PP piping (similar to Aquatherm) which is all fused and good to 258F @ 90 psi. There would be very few leak paths. It would seem this could hold vacuum well.

PMJ

P1385 said:

@PMJ Upon startup, assuming a tight system, is the system under vacuum from the last time it was operated?

If I do go ahead with a natural vacuum steam system on my new house, I might use Niron PP piping (similar to Aquatherm) which is all fused and good to 258F @ 90 psi. There would be very few leak paths. It would seem this could hold vacuum well.

That's just the thing most people seem to miss in this @p1385. The system does NOT have to be that tight if you run spaced cycles. 3cycles per hour the firings are only 20 minutes apart. My system is not that tight - 60" water vacuum would be totally gone in say 4 hours. But after 20 minutes the vacuum is 50+ and still deepening. So all my firings are into the deepest point of the vacuum I ever see and boiling is almost instantaneous. There is a chart on one of my other posts that shows all this - worth some study. Everyone seems hung up on that first cold cycle being under vacuum. But when heat is actually needed there aren't any of those! All the rest of my time in my leaky system I take max advantage of great vacuum all for free and incredibly quiet. My other point which doesn't seem to get much attention is that with conventional pressure controls even with a mild overshoot the off periods are longer so more natural vacuum is lost before the next restart. Even cycling wins again.

Thanks for bringing this up.

PMJ

@p1385,

Obviously if you starting new you would build as tight a system as possible.

I was just taking the opportunity to speak to those considering this with old leaky systems like mine.

MilanD

@P1385

Curious about the Niron pipe you mentioned. I can't find data sheet showing 258*F for temps. Mind sharing which pipe it is?

P1385

So, it seems that Nupi has reigned in their recommended operating range. The literature I have (I work for a distributor) claims continuous operating at 230F and max ~258F. After your question, I checked on their site for a PDF version and now they suggest max operating temp at 203... I'll have to call tech services and see what their thoughts are. The system would never see more than ~218F at 2# so it would seem possible... we shall see.

jumper

PMJ said:

@p1385,

Obviously if you starting new you would build as tight a system as possible.

I was just taking the opportunity to speak to those considering this with old leaky systems like mine.

Exactly. And with a manifold,preferably overhead,pipes need to be only large enough to service one terminal. Dead men paid dearly for fittings. Different century,different world.

The same advantages hold for closed gravity HHW. Fittings are eliminated. TRVs customize each room and automatically balance the system