Re: Heating design diagram (preliminary)

Thanks for your help to date. I've distilled the advice from various

> threads in this newsgroup to come up with a possible heating design > for my house. > > The drawing HD01 at
formatting link
shows the proposed > hydraulic design for a large domestic heating system. The boiler and > hot water cylinder is on the left hand side of the drawing, space > heating is on the right. (You might need to rotate the view so that > the drawing is in landscape orientation in your browser) >

Wonderful - but who's going to understand it when you're not around?

Incidentally, what's the 'top-up' tank in the primary circuit for - bearing in mind that it's a pressurised system with pressure vessel and filling loop?

Reply to
Set Square
Loading thread data ...

So you can get additives into the system easily?

Reply to
Rob Morley

I'll leave my inheritors the Google Groups link to this thread ;)

You're right - I don't really need the tank. I had half a mind to use it to monitor the volume of water that had escaped from the system so that I could assess whether there were any leaks.

Reply to
John Aston

Ooh, lovely. You'd said you were a plumbing amateur. Ha, ha! You've done this before, haven't you.

Suggestions;

1) Put a DCV between the drinking water and the softener. The DCV shown protects the mains, but not the drinking water from contamination by the softener. 2) 15mm should be plenty for the drinking water. 3) Softener. You'd need a duplex model, with two resin vessels, to ensure a softened supply at all times. A duplex softener would regenerate as required by a meter, so would minimize the salt consumption; costs more, though. The resin takes a 1 to 1.5 hours to regenerate. Softened water at 0ppm can pinhole copper hot water pipes in 2 or 3 years. A blending by-pass valve to give 40ppm is advisable, but you'd need a serious test-kit. 4) The PRV symbols are the wrong way round. The symbol is derived from steam PRVs which have small HP inlet pipes and large LP outlet pipes. Pedant mode off. 5) Don't fill the heating with softened water. The fill water will contain a minute amount of limescale, but softened water causes some problem with the inhibitors which I can't recall. 6) You don't want a PRV upstream of the accumulator. The idea is to accumulate a volume of water under pressure. A PRV downstream is OK. 7) I'd add the accumulator at a later date, if it proved necessary. 8) The PRV on the unvented HWS usually has a second outlet port for a balanced pressure cold supply. The strainer should be upstream of the PRV; often it's integral with it. The whole set of components comes with the cylinder 'package'. 9) Given the size of this system, I would consider supplying the DHWS cylinder with a pump (making 6) from the secondary side of the low loss header. 10) On the low loss header, the temperature sensor has to be immersed in the flow water and the primary & secondary pipes need to be off-set. This is a schematic drawing, so they probably will be. 11) There would be IVs on the flow connections to the UFH headers. These often have flow indicators to facilitate balancing. 12) On the heating flow connections, I'd put the CVs downstream of the IVs. It makes no difference to the function, but you could service a defective CV without draining the entire system. 13) You need IVs on the heating returns from the rad circuits, before they connect to the 22mm secondary return header pipe. I'd put all the IVs adjacent to the secondary header pipes. 14) Is 22mm pipe big enough for the secondary header pipes? There should be a negligible pressure loss at the design flow rate. 15) Why do you need a control valve upstream of the mixing valves? 16) Towel rails; another pump, I'd think, No.7. There's some EU regulation requiring low temperatures to towel rails, but I don't recall the details. 17) You don't need the top-up tank, you can add inhibitors through a radiator. A drain valve on a rad, after the rad valves, for this would be useful. 18) I don't think you need more than one pressure gauge.

Stick it on The Wall at Heatinghelp and invite comments. There are photos of similar systems posted there regularly. They appreciate this sort of thing. Tell them you're a first-time amateur!

It depends entirely on the heat losses. No other response would be sensible.

I have a schematic somewhere that I once drew of an existing heating system. It takes up an A0 sheet. Some details are difficult to read at that size.

Reply to
Aidan

It depends on the size of the resin vessels. Mine takes about 20 mins to regenerate a vessel but will comfortably supply softened water continuously. Salt consumption with a lot of water use is about 25kg every 3-4 weeks.

You can get these from Hach. Cost is about £30 but will do several hundred tests.

Some softener valves have a small amount of bypass built into the valve - I believe the newest Autotrol ones do. Mine (Kinetico) doesn't, so a small bypass was introduced consisting of a screw turn service valve opened a fraction. It does the job and the ppm value is consistent over a range of flows through the whole softening set up.

Reply to
Andy Hall

No, honestly! It's really the result of pouring over the manufacturers' instructions and trying to piece it all together.

OK. I was worried about a drop in the pressure/flow rate. My incoming supply is only 2.8 bar static and 30lpm.

OK, thanks

The softener I had in mind was a Kinetico 2020c HF which uses two resin-filled cylinders alternately to give a flow rate of 51 lpm at 2 bar. I never knew its effect on copper pipes! I'm surprised they don't mention it ;)

Thanks, I'll change the above in the second draft.

The upstream PRV is part of a combination valve that is supplied with the Dualstream cylinder and accumulator. I guess it's set at 3.5 bar to comply with regulations. It might be easier to leave out the downstream PRV. The mains static pressure is only 2.8 bar anyway.

Yes. If the vendor allows me to buy an accumulator without the cylinder, I'll plumb for it as a future addition.

You're right about the package. My drawing has shown the individual components really just for my own understanding. I'll reverse the position of the PRV and strainer.

I did think about this but (a) The cylinder is less than 8m pipe length from the boiler so I hoped that the boiler pump could get the water there on its own. (b) The cylinder water will be pumped from the boiler at 80°C and the water through the heating circuits will typically be at 50-60°C. I didn't want the header to cycle between the low and high temperatures when the boiler heated up the cylinder. I thought that that might be less efficient.

Yes, you're right. The header is a pre-fabricated assembly.

I'm hoping that these come as part of the manifold from the UFH suppliers

Thanks, I'll change their position in the second draft.

OK. I'll add these

Pipe diameter is an area of uncertainty for me. Also, you can see on the diagram that I've written MAX. DISTANCE? I can't get any definitive answers (or even rule of thumb) from the boiler manufacturers.

The UFH heating circuits have their own individual themostats. If one room gets up to temperature, it can close down its circuit's valve, but still let the other circuits on the manifold heat up.

Yes, another pump is probably the way although, with the way I've drawn it, in Summer the boiler would have to heat up the header just for the towel rails. I can live with this, I guess.

OK. I think that they come as part of the UFH manifold.

I'll make the changes and do this.

The heat losses would be 31kW. I'm hoping that the additional 7kW will give me some headroom plus a bit for the domestic hot water.

Thanks, Aidan. You're a gentleman.

Reply to
John Aston

Lots of deliveries this time of year.

Reply to
Dave Plowman (News)

Deliverables are in order.

Reply to
IMM

No. To protect the unvented cylinder. A heat bank with a plate heat exchanger can go to about 10 bar.

Reply to
IMM

Aiden has highlighted some points, so I will not go over them. Some observations and Qs:

Firstly, what drawing package did you use, Visio?

The boiler appears to be a Veissmann with an in-built outside weather compensator with the temp sensor in the low loss header. If the compensator slope is set to the UFH, it will not be suitable for the rads. You would require the low loss header to be on the minimum temp that the rads take, which means a higher temperature for the boiler to operate on making it less efficient. If you set the compensator slope for the higher temp rads, each the UFH zone will control itself on its own mixer controls, set to maximum of 55C.

You could use a dual temp boiler as the Eco-Hometec, or a simple boiler maintaining a hot low loss header at a high temp. This is what is done with non-compensating boilers where the return temp "has" to be high. So, as you have it the rads will not go about 55C.

The mains water from the accumulator. As you have done in splitting the DHW supply to the cylinder after the accumulator, and cold water. But! have all cold taps off one leg. On the other supply only the cylinder. With the exception of just before the cylinder have the cold supplies to the showers only.

Have only one pressure reducing valve as mixers require equal pressure on each inlet line. Using an accumulator means you only need cheap shower mixers.

If you assess the heating requirement to 32Kw then stick to this, or the nearest to, depending on price. No need to go over for DHW as you have a priority system. This diverts all the boilers heat to the cylinder.

Make sure the cylinder is quick recovery. Unvented cylinders are never as quick as vented or thermal stores. The coils are restricted so as not to generate too much pressure inside.

Do you still intend to have two stage heating? UFH with rads boosting? Yiou have 5 UFH ziones. Where will the rads fit in relating to these zones?

Back to having differing temps for rads and UFH. As it is, efficiency is compromised by the high temp rad circuits. You have an efficient expensive boiler not performing to maximum potential. Look at the low loss header on the diagram. Replace this with a heat bank/thermal store. Off the bottom UFH section of the thermal store have the UFH circuits. Off the high temp top section have the rads and DHW. You may want to have three sections: top DHW, middle rads, bottom UFH. Then you have all circuits coming into a neutral point, the heat bank.

Now you have greater control of temperatures, dividing and ruling, which means the boiler will not be running at too high a temperature to suit only the rads compromising efficiency. An outside weather compensator can be on the UFH section to keep this part of the store at the ideal high efficient low temperature and prevent boiler cycling. You may want a compensator on a heat bank mid section serving the rads (UFH & rads have different slopes).

Using a heat bank, a far cheaper and simpler boiler may be used.

Having three sections means that in summer, only the DHW top section is heated, not the whole store, saving on standing losses.

Using a heat bank immersions may be fitted in the different temperature sections. So, if there is a boiler outage you can run the whole system, heating and DHW off electricity. You can't do that with a boiler connected via a low loss heater.

Towel rails: These can be teed in before the diverter valve at the boiler, between valve and pump and direct to the return. They will then work in summer, but only when the cylinder or heat bank is being re-heated, which is fine for summer use. If you take a shower and the boiler kicks in the re-heat you will find the towels are hot on the rails.

Accumulator: If they will not sell unless they supply the unvented cylinder (they would sell one separately to me), keep a tank in the loft and have a booster pump serving a heat bank. This is a cheaper an simpler option too. The tank does not need to be in the loft. It can be anywhere as it is pumped.

It would be interesting to see what the Yanks say. They don't do thermal stores in a big way there and thinking tends to be 1950ish, so only regard what they say as interest only.

Simpler Alternative:

  1. Low Temp Circuit: One dedicated condensing boiler serving the UFH only on a weather compensator. Simple, separate and sorted.

  1. High temp circuit. Another boiler using a 3-way diverter valve serving DHW and rads. You may want a weather compensator switching the boikler to give the ideal temp for the rads. When DHW is called the boiler runs up to max temp. This is similar to normal domestic setup.

  2. A controller staging in the UFH and rads to give precise control of room temps. UFH 1st stage.

  1. Backup: Now you have heating backup if one boiler drops out. Electrical backup for DHW.

  2. Cold water storage tank instead of an accumulator with a booster pump. Tank can be fitted anywhere.

  1. The cylinder can be:

a) An unvented version,

b) A DHW only heat bank, such the DPS Pandora, which does requires an overflow so can be fitted anywhere in the house.

You will find that two condensing boilers can be had for less than the price of the Viessmann, and lots of change too.

Reply to
IMM

compensator

I forgot to mention. Insert two low loss headers would not cure this problem. A priority system would need to be in place favouring the rads. The problem is that the headers do not contain enough mass. In a heat bank the top rad section could re-heated rapidly and left for while for the rads to extract the heat. One up to temp it reverts to UFH temps and heat the lower section. So it would switch from UFH to rads with long intervals between. This cannot be done with a small mass header. In effect a heat bank is a very large header

Reply to
IMM

CorelDraw 6. Not a great program but I'm used to it.

compensator

The compensator slope is set for the rads. The UFH is on all the time with a flow temp of 55°C maximum, modulated down by the thermostatic mixing valve on each manifold. The radiator temperature is determined by the outside temperature.

So I will avoid the non-compensating boiler and consider a boiler where the flow temperature is modulated between 50°C and 70°C, say (this can be adjusted). Therefore, the radiator flow temperature is between 50°C and

70°C. The thermostatic mixers keep the water through the UFH at 55°C or less.

Yes. UFH as the primary heat source, with the rads on when it's cold. The colder it gets, the warmer the flow through the rads.

There are two UFH manifolds (one for the ground floor, one for first floor + attic). The rads are in every room. When it's reasonably warm outside, only the UFH is on. When it gets cold, the rads come on. As it gets colder still, the temperature in the rads starts to increase.

Here are two situations to demonstrate the weather compensation for my house at 20°C inside:

When the temperature outside is -3°C, the boiler temperature is 70°C, the mean water-to-air temperature in the radiators is (almost) 40°C and the water into the UFH manifold is 55°C. Under these conditions, a heat loss of 31kW from the house of is met by the UFH output of 17kW plus the combined radiator output of 14kW.

When the temperature outside rises to +3°C, the boiler temperature falls to

55°C, the mean water-to-air temperature in the radiators is (almost) 25°C and the water into the UFH manifold is 55°C. Under these conditions, a heat loss of 24kW from the house is met by the UFH output of 17kW plus the combined radiator output of 7kW. (I derated the rads by 50% for a 15° fall in mean water-to-air temperature. I hope that's suitably conservative.)

This means that the temperature of the water flowing into the radiator is equal to to the temperature of the water flowing into the UFH when it's above 3°C outside. I'm hoping, therefore, that most of the time I won't have two different temperatures serving the two types of heat sources.

I understand. I guess this assumes that the header has been replaced by the heat store. (The header would be a very low resistance compared to the towel rails.)

Thank you for taking the time to compose this reply, I appreciate your thoughts. I believe that I understand the principle of the heat store and the principle of a modulating boiler. They both have their advantages. I'll get some prices together and put my best foot forward.

Reply to
John Aston

is only 2.8 bar static and 30lpm.

It would probably be OK to just relocate the DCV shown. I think the Water Regulations require a DCV on a softener inlet. The mains pressure varies inversely with demand in the neighbourhood during the day, so you'd work on the minimum. The flow rate achieved at the minimum pressure should indicate whether you might need the accumulator.

It's not something the manufacturers publicise and I don't have any definite information. My understanding is that it's something that can happen, but that the softener isn't always the guilty party. Erosion and bad pipe-fitting have some role. My understanding is that water softened & blended to about 40ppm hardness is fine for all practical purposes and doesn't create the corrosion problems. Softened water can be unpleasant stuff in the wrong place. I think the dissolved calcium salts are mostly changed into sodium carbonate (washing soda?). I've no experience of Kinetico.

The last test kit I used was made by Hach, supplied by a company called CamLab. It was a titration test kit (colour change). I've found dip strips can be misleading in some hands (RTFM again).

out the downstream PRV. The mains static pressure is only 2.8 bar anyway.

I think your original detail & IMM were right (Ow, that hurt!) & I was wrong. The upstream PRV would be set to limit the pressure to the design rating of the accumulator.

I once had a discussion with an HSE inspector who expected to see every PRV accompanied by a correctly-sized safety/pressure relief valve to prevent the equipment being over-pressurized in the (quite likely) event of a PRV failure. We were talking about compressed air, but his point is still valid here.

(a) The cylinder is less than 8m pipe length from the boiler

This seems to be a compromise between the conflicting requirements of supplying the cylinder with sufficiently hot water (to get the stored DHW above 60degC), keeping the boiler return temperature low and avoiding the need for lots of mixing valves. It's outside my experience, so I'll shut up.

You could fabricate one, it's just pipe & fittings.

suppliers

Only if you specify them. The flow-rate indicators are an optional extra. They usually become unreadable after a few years, so are only useful for initial balancing. You need to make a record of the settings and keep it in a safe place.

definitive answers

There is no maximum, so long as the pump(s) can handle the resistance at the flow rate.

Yes, but I was querying why you need a motorized control valve adjacent to the UFH manifolds' mixing valves. If the pipe stat temperature was exceeded, you could stop the pump and/or set the mixing valve to 0%. Are the mixing valves thermostatic or electric?

me some headroom plus a bit for the domestic hot water.

An additional 7kW should be lots, check the cylinder manufacturer's spec. You won't need additional capacity if the HWS has priority, as IMM said. I don't like the diverting idea, but I see the need for it.

A couple of other points;

The DOCs on the heating zones return pipes should be upstream of the IVs. They should drain the zone when the IVs are shut, but will drain the whole system as shown.

You have to ensure that the primary flow through the low loss header is greater than the secondary flow at all times. If the secondary flow is greater, then some of the secondary return flows back up the header and you then get a reduction in the secondary flow temperature.

The drain pipe from the tundish on the cylinder has some rules to determine the pipe size, depending on the pipe length and number of elbows. It's usually not a problem, but needs to be considered if the proposed cylinder position is a long way from the final "safe & visible" outlet outside. Otherwise, you can end up with a huge drain pipe or it can be impractical. The cylinder can't be in a basement.

A pleasure.

Reply to
Aidan

easier to leave out the downstream PRV. The mains static pressure is only

2.8 bar anyway.

can't get any definitive answers

It should be the reverse, even Viessmann say that. The idea is that the secondary circuits will take all of the flow from the header with little going back to the boiler fom the boiler flow (short circuit). The idea is that all the hot water coming into the header from the boiler will be sucked into the secondary circuit. What returns from the header to the boiler should be cooled water from the secondary circuits.

Reply to
IMM

It is as I suspected not taking full advantage of the boilers efficiency. The UFH is running all the time and the rads occasionally, yet the boiler is set to for maximum efficiency on the high temp rads.

Typo on my part. Should have been non-condensing. A weather compensator can be on the rads circuit too. A weather compensator can be an external controller rather than an internal one (integrated with the boiler). Danfoss Randall make one for around £160.

Weather compensation.

output of 7kW. (I derated the rads

Having two different temperatures for rads and UFH is the way.

Yes. The store also serves the DHW too. So it replaces the header and serves DHW, UFH and rads and prevents boiler cycling.

That is so.

In your case with having three differing functions of different temperatures a heat bank is the ideal way to supply those temperatures promoting maximum efficiency from the boiler.

The most efficient, and easiest, is having two boilers as described above.

Reply to
IMM

This is a poor way to do it because these controllers work by turning the boiler on and off for variable periods.

The cycling issue is a corner case and is irrelevant for the types of boiler under consideration.

Reply to
Andy Hall

It is not a poor way of doing it. Anti-cycle control is incorporated in most boilers and the compensator. I have had one switch a boiler for eons and there is no excessive cycling at all. Coupled to a large mass of water, like a heat bank, and cycling will be minimal to the point it is not an issue.

If the UFH heating is being run directly from a boiler dedicated to that function then there are boilers around a lot cheaper with integral weather compensation as an extra, that will modulate down on the compensator control.

Once using a thermal store this expensive boiler is then not an issue. A cheaper simpler boiler can be used, that is one of the selling points of thermal stores/heat banks.

In this case with three functions: DHW, UFH and rads, all operating on different temperatures, a heat bank/thermal store is by far the best solution, maximising boioer efficiency.

Reply to
IMM

Cycling a boiler at full output is a very different issue than doing so at 3kW in terms of efficiency.

Since the discussion is based around a modulating boiler, this type of controller is irrelevant.

It is if you're DPS and trying to flog heatbanks.

However, if you are the purchaser, the equation may be quite different,.

You haven;t explained how all of this can be done with the operating parameters of the boiler type under discussion.

I think that you are simply bunching together all of the claimed advantages and assuming that they all can happen concurrently. I don't believe that that is possible.

Reply to
Andy Hall

temperatures

IMM, leaving my particular case to one side, may I ask in what application would you recommend the use of a weather-compensated modulating boiler?

Reply to
John Aston

Firstly, having a dedicated boiler for distinct functions is the ideal method: DHW, UFH, rads. Separate dedicated functions easy to control and design. That was the way in ye olden times. But in ye olden days boilers were expensive and very large, so quickly they found ways of joining up all the various circuits: DHW, rads, fan coil unit circuits etc. They devised methods using headers, buffers (thermal store) etc. All this was a compromise to tap into one source of heat, the boiler. It worked well because the boilers had to operate at high temperature because they were non-condensing. High temperatures water was on tap and the varios circuits tapped off it blending it down when necessary for the various circuit functions.

In commercial applications they could then have two sequenced boilers, bringing in both or just one, or none, depending on heat demand, and a spare boiler if one is down. All was fine and dandy and things went along like that for decades. These sort of systems are still in the mindset of many designers today, designing the total system to the highest temp required in a sub-system.

Then condensing boilers came along offering high efficinencies the lower the return temp. This required a re-think. With a simple two function system of say: DHW and UFH you could have the boiler on a weather compensator serving the UFH to the lowest temps for high efficient operation. If DHW is called, a 3-way diverter sends all the boilers heat to the cylinder (must be quick recovery) while ramping up the boiler temp to maximum for the high temp DHW requies and rapid DHW re-heat. When more than one function is thrown in, with all three, or more, requiring different operating temps, as in the rads of your system, matters become a little complicated when the aim is to keep the boiler running at the lowest return temperature for maximum efficiency.

Also, boilers became smaller and smaller and cheaper and cheaper. This then made the dedicated function boiler a cost effective reality. To implement a condensing boiler system to maxiumum efficiency, running it at the lowest temps for most of the time, it becomes more expensive in ancilliry equipment and controls, and complicated, as you have seen in attempting to get a solution. The best solution was thermal store/heat bank with dedicated sections, or zones, for the various functions (DHW, UFH, rads) down the store with these sections providing exactly the the right temperature for the functions. The boiler only heating one of these sections at a time to the lowest temperature for that function - ideal, and sorted.

Back to your question. "what application would you recommend the use of a weather-compensated modulating boiler?"

If you have one function, say just rads or UFH then it is ideal. It is also ideal for two functions, but a priorty system is required, with the highest temp having priority with boiler cycling eliminated. A thermal store can supply a buffer of water to eliminate this, as you don't want the system cycling from one function (UFH) to the other (rads or UFH) every couple of minutes.

The answer:

  1. One dedicated function (like rads or UFH)

  1. Two functions like UFH and DHW (only if the system is designed properly to a priority system).

To go further.........

As I have said, in your case it is worth costing up:

  1. Low Temp Circuit: One dedicated condensing boiler serving the UFH only on a weather compensator. Simple, separate and sorted.

A number of makers supply integrated weather compensated boilers, as an extra. Viessmann, MAN, etc, are very expesnive and will probably have far more control functionallity than what you need. They are very good and the RRs of boilers, with price to match. Very good boilers are available from reputable makers much cheaper.

  1. High temp circuit. Another boiler using a 3-way diverter valve serving DHW and rads with DHW priority. You may want a weather compensator maintaining boiler to the lowest temp the rads require. When DHW is called the boiler runs up to max temp. This is similar to normal domestic setup.

  1. A controller staging in the UFH and rads to give precise seamless control of room temps. UFH 1st stage, rads 2nd.

  2. Backup: Now you have heating backup if one boiler drops out. Electrical backup for DHW in the cylidner or heat bank cylinder.

  1. The cylinder can be: a) An unvented version,

b) A DHW only heat bank, such the DPS Pandora, which doe not require an overflow so can be fitted anywhere in the house and will take high pressures way above 3.5 bar.

You will find that two condensing boilers can be had for less than the price of the Viessmann, and lots of change too.

Reply to
IMM

HomeOwnersHub website is not affiliated with any of the manufacturers or service providers discussed here. All logos and trade names are the property of their respective owners.