How to size etc. CH expansion vessel

Joined
19 Aug 2004
Messages
372
Reaction score
4
Country
United Kingdom
Hi,

There seems a lack of god information on this so I thought I'd put together this post.

As hot water expands then in a sealed CH system you need an expansion vessel to allow for this.

C Factor
85 0.0324
90 0.0359
95 0.0396
100 0.0434

85C is the hottest the water should be but to play safe lets take the fault condition of 95C, 0.0396. You next need to know the total volume, either by draining it into a bucket if you are converting a vented CH, or from the sizes of the radiators, boiler, and pipe runs if it is a new system, or guessing. Let's say CH volume is 100l, so expansion is 3.96l. BUT, that does NOT mean you need a 4l expansion vessel.

Next we need to look at static pressure. If we assume 2 storey house, 3m per storey then total head is say 6m. 10m of water = 1bar so this would be 0.6bar. We want the highest radiator to be under positive pressure since negative pressure would draw in air. Typically this is rounded up to 1bar. The expansion vessel which is best sited at return near the pump, as that way the pump increases the pressure in the whole CH loop. As you move away from the pump return then you see a drop in pressure from vessel to pump return due to the water flow. Let's assume boiler is on ground floor so here we want 1bar. This is the precharge pressure we want in the expansion tank. If the boiler, pump and expansion tank was in the loft then it would be at the minimum pressure point so 0.5bar precharge would be all we'd need, just enough to ensure system has a positive pressure so doesn't draw in air.

But let's stick with ground floor boiler, pump, expansion, and 1bar. We cold fill the CH system to 1bar. Let's say we have an 8l EV. When water gets hot we have an extra 4l. That goes into the expansion tank and 8l air gets squashed to 4l. Pressure is the inverse of volume so 1/2 volume double pressure, 1bar -> 2bar. The boiler PRV is usually 3bar (when converting open vented to sealed you need to fit PRV BTW). Boiler spec usual is to say max pressure if 2.5bar so 8l EV works ok for 100l CH. If we fit 12l then volume is 2/3 so 3/2 on pressure, i.e. 1bar -> 1.5bar.

Some people say set cold pressure 10% over precharge in which case you need to allow for that, i.e. initial pressure is 1.1bar, 110/100 initial volume is 100/110, ~90.9%, or roughly 10% of EV is already used up, so a 12l becomes 10.8l at initial 1.1bar. 4l from hot CH gives 6.8/10.8 volume change, 10.8/6.8 pressure change, 1.1bar -> 1.75bar.

Setting a lower initial pressure works better of course so for a loft mount the 4l change means 0.5bar -> 1bar with the 8l EV.

The other debate is how to mount the EV, air or water at the top? The advice so potable water is air up, water at the bottom, otherwise crap can fall into the EV and it can't get out. Same applies to CH IMO so water pipe work should come up to EV.
 
Sponsored Links
The above was interesting.
How do boiler manufacturers know what size EV to put in there boilers, given that that don't know the volume of water in the system & where the boiler is sited.
Just interested.
 
What fits and assumptions I would guess. Instructions probably says installer should check EV is big enough and fit second one if not.
 
But let's stick with ground floor boiler, pump, expansion, and 1bar. We cold fill the CH system to 1bar. Let's say we have an 8l EV. When water gets hot we have an extra 4l. That goes into the expansion tank and 8l air gets squashed to 4l. Pressure is the inverse of volume so 1/2 volume double pressure, 1bar -> 2bar. The boiler PRV is usually 3bar (when converting open vented to sealed you need to fit PRV BTW). Boiler spec usual is to say max pressure if 2.5bar so 8l EV works ok for 100l CH. If we fit 12l then volume is 2/3 so 3/2 on pressure, i.e. 1bar -> 1.5bar.
Your gas physics is a little out there. At constant temperature gas volume is inversely proportional to absolute pressure. You need to add 1 bar to convert barg to bara.
 
Sponsored Links
Well spotted! That makes a big difference. Precharge would be say 1bar, then + 1bar for atmosphere, 2bar absolute. Reduce 8l to 4l doubles that from 2bar to 4bar absolute or 3bar relative once you take away atmosphere. So 8l doesn't really work for 100l CH (which is what manufacturer told me), unless you set precharge lower than 1bar.

In my case I'm thinking of putting EV higher up. That puts it a bit down the return pipe so I need to allow for the drop in that. I need to check the numbers but I think about 0.1bar for the flow with EV in airing cupboard, but that gives me a gain of 2m, so 0.2bar which offsets any flow loss. In that position highest rad is ~1.2m above, say 0.15bar, so I could precharge at 0.5bar. That works better, especially with 12l, 0.5+1bar precharge absolute, x3/2, 2.25bar absolute, 1.25bar relative.
 
its down to the installer to install the correct vessel & set the correct pre charge (either nitrogen or compressed air) and subsequent annual service tech visits to service the expansion vessel so its within the vessels manufactures requirement.

Good installation is the key,properties and installations vary :sneaky: Its not diy.
 
I came across this from Altecnic
AlternicEVcalc.png
The +1 to each of the pressures must be the atmosphere to change relative pressure to absolute. I guess the +0.3 is to give something in hand on the size of the vessel, or the setting of the pre-charge pressure, or both.
 
Whatever happened to 1L of expansion per radiator for domestic installations, think you are over thinking this lad, any one worth their salt will tell you on commissioning if additional expansion is required
 
hmm, well, rules of thumb are ok but I prefer to do the maths. The key element is how much pre-charge you leave in. EVs come set to 1.5bar it seems but you only need enough to ensure the highest point is under positive pressure. 1bar is 10m and typically EV might be half way up the wall too.

Perhaps it is easier to see if you drop the 0.3 from Altecnic formula which is an "allow a bit in hand" and then rearrange. That gives you Boyles law, or working form Boyle's law

V1 * Pa1 = V2 * Pa2

with absolute pressures (Pa), so calling atmosphere 1 bar gives on the gauge (Pg)

V1 * (Pg1+1) = V2 * (Pg2+1)

e*C in Altecnic formula is the extra water, so V2 = V1 - e*C, i.e. volume left after water takes some.

V1 * (Pg1+1) = (V1-e*C) * (Pg2+1)

rearranging

Pg2+1 = (Pg1+1)*V1 / (V1-e*C)

So taking the system as 100l, extra water is 4l. With a 12l EV you get

Pg2+1 = (Pg1+1) *12/8.

Which is what you expect from Boyle's law, reduce volume by 1.5 (12/8) and absolute pressure goes up by 1.5 (with a constant temperature for of course so we need a bit in hand to allow for the increase in pressure when the temperature goes up). If precharge is 1.5bar as the EV comes then you get

Pg2+1 = (1.5+1)*12/8
Pg2+1 = 3.75
Pg2 = 2.75

Which is too close to the 3bar PRV (and boiler maker probably says 2.5bar is max running too). But you probably don't need the 1.5bar precharge if you work out the actually head needed. If pressure is reduced to 1bar you get

Pg2+1 = (1.0+1)*12/8
Pg2+1 = 3
Pg2 = 2

Which is a lot more comfortable.
 
I'm seriously considering pressurizing my 14 rad heat only system hence doing a little digging to see what size EV the installer might recommend and to be sure we get one sufficiently generously sized.

And I discovered this thread. And spotted Malc's posts and saw the, quite impressive, calculations performed.

I then saw Ian's post (#8 ) suggesting a simple one litre per rad estimate (which, if I relate it to my own place, is spot on). As it happens, over the years I've come to learn that Ian's posts can be relied upon (along with a few other long standing members).

But I also saw in post #4 that fixitflav corrected Malc's mathematical error (apparently a quite significant error too, not that I would know coz it was all Greek to me).

I have to say I was tickled to then see Malc say in reply to Ian, and despite his arithmetical gaff just a few lines above ...... ' hmm, well, rules of thumb are ok but I prefer to do the maths ' I reckon that's a cracker ...... and he then goes on to produce a string of formulae that is presumably supposed to be of some practical help to readers but strikes me as a bit of showboating. Is there anyone on the site that understands that particular post coz it makes me feel a tiny bit inadequate.

The upshot is .... Ian's half a dozen choice words of advice were more helpful to me than that, quite remarkable display of apparent mathmatical prowess.
 
Not much maths involved here in my decades old calculations but might be useful for someone(s). Can't post the actual spreadsheet.

1708636063541.png




1708636027283.png
 
Have attached it .zip form.
 

Attachments

  • Very Little Maths Involved Rev1.zip
    8.5 KB · Views: 42
The installer's quote came today and he did, indeed, specify an 18kw EV.
 
Do you mean a 18 litre EV?, that should be fine IMO, you can't oversize a EV.
 
Yes, I mean an 18 litre Expansion Vessel not 18kw. Even I realise EVs are measured thus ... my mistake.
 

DIYnot Local

Staff member

If you need to find a tradesperson to get your job done, please try our local search below, or if you are doing it yourself you can find suppliers local to you.

Select the supplier or trade you require, enter your location to begin your search.


Are you a trade or supplier? You can create your listing free at DIYnot Local

 
Sponsored Links
Back
Top