How to size an expansion tank?

[RLWP]

For a horizontal one that will be true if the take-off is on one end a little down from the top surface. A vertical one will be full (assuming the take-off is from the top as per normal).

 

Chris

 

True. From memory our take off isn't at the top, and the PRV is probably the highest outlet although still mounted in the end. In the past I have jacked up one end of the calorifier and vented the prv manually to bleed air from it.

 

Richard

[casper ghost]

It's a matter of physics fact that water expands when heated. Over the range that the engine (or a diesel heater) will heat the water this will amount to about 3% of the size of your calorifier or around 2 litres for a 60 litre calorifier.

 

This has to go somewhere or your calorifier will split with catastrophic results. As you say, you have never checked the water in the calorifier area and you are just assuming it's due to the stern tube.

 

You won't see its gushing out of the PRV. As the pressure builds the PRV will weep and then close again as the pressure is released. Then the pressure will build up again and the PRV will weep some more.... and so on.

 

Chris

Ok, I'll check next time i cruise for a bit. These expansion tanks, are they just like fitting an accumulator?

[chris w]

Ok, I'll check next time i cruise for a bit. These expansion tanks, are they just like fitting an accumulator?

Yes exactly like it. You will need about an 8 litre expansion tank tee'd off in the hot water take-off from the top of the tank before the first tap. You should also fit a non-return valve (NRV) in the cold water inlet pipe to the bottom of the calorifier. Ensure that the expansion tank is suitable for hot water - some are cold only.

 

Chris

[casper ghost]

Yes exactly like it. You will need about an 8 litre expansion tank tee'd off in the hot water take-off from the top of the tank before the first tap. You should also fit a non-return valve (NRV) in the cold water inlet pipe to the bottom of the calorifier. Ensure that the expansion tank is suitable for hot water - some are cold only.

 

Chris

Cheers for that

[Charlie Breakfast]

It's a matter of physics fact that water expands when heated. Over the range that the engine (or a diesel heater) will heat the water this will amount to about 3% of the size of your calorifier or around 2 litres for a 60 litre calorifier.

 

This has to go somewhere or your calorifier will split with catastrophic results. As you say, you have never checked the water in the calorifier area and you are just assuming it's due to the stern tube.

 

You won't see its gushing out of the PRV. As the pressure builds the PRV will weep and then close again as the pressure is released. Then the pressure will build up again and the PRV will weep some more.... and so on.

 

Chris

 

Chris ....After a bit of thought I think the apparent contradiction between the Maffs and what is happening on my boat could be explained by what you say above.

 

My tank only gets heated to 60 degrees max ( old engine )...So water in my 55 litre tank is probably expanding by a fair bit less than 2 litres. As I use the low pressure sureflow and an accumulator set at about 20 lbs psi it may be that a combination of lower water expansion, low initial system pressure and a large dgree of 'reserve' in the accumulator results in the water expansion not pushing the water system pressure to the PRV release point ( 40+ lbs psi?).

 

One thing is for sure, the PRV does not leak a drop (the bilge dust is bone dry under the "leak off tube", and just like Casper when solo boating the taps dont get touched from starting engine to finishing ..sometimes 8 hours later.

 

There has to be a rational explanation and I cant think of a better one. Either that or the bacteria have got to me..... or the maffs are wrong!

 

Chas

[chris w]

Chris ....After a bit of thought I think the apparent contradiction between the Maffs and what is happening on my boat could be explained by what you say above.

 

My tank only gets heated to 60 degrees max ( old engine )...So water in my 55 litre tank is probably expanding by a fair bit less than 2 litres. As I use the low pressure sureflow and an accumulator set at about 20 lbs psi it may be that a combination of lower water expansion, low initial system pressure and a large dgree of 'reserve' in the accumulator results in the water expansion not pushing the water system pressure to the PRV release point ( 40+ lbs psi?).

 

One thing is for sure, the PRV does not leak a drop (the bilge dust is bone dry under the "leak off tube", and just like Casper when solo boating the taps dont get touched from starting engine to finishing ..sometimes 8 hours later.

 

There has to be a rational explanation and I cant think of a better one. Either that or the bacteria have got to me..... or the maffs are wrong!

 

Chas

Nope - the maffs are right!! Your expansion will be (20-60 degC) 1.5% ie:0.8 litres on a 55 litre calorifier or (10-60 degC) 1.64% ie: 0.9 litres on a 55 litre calorifier.

 

From the data you gave (low pressure shureflo and an accumulator set at 20psi), the accumulator is probably acting just as an expansion tank and not very well as an actual accumulator.

 

Chris

[Smelly]

OP, I'd be ringing BES before ringing swindlers... 12l EV for 20 quid or so... stick "expansion vessel" into the search engine. BES

 

I'll probably be going myself soon as now I've modded the central heating again there's a possibility once the fire's going full bore that the pump will start cycling overnight so baby EV for fresh water side and a BIG EV for the hot side (twin 60l cal's) would be appropriate.

 

I've not read the rest of the thread in detail as, TBH I've heard it all before, but suffice to say, when we heat the water up from lukewarm we end up with a few mm of lukewarm water in the bath (the PRV drains into the bathroom waste) and I'm getting bored of it...

 

As to bacteria, I sorely doubt, at the kind of temps our cal's regularly get to that any bacteria survive... Our engine runs at 85 deg C according to the book and the water certainly gets close to this after a couple of hours' running. I think for a system on a boat where a relatively small amount of water is kept warm and cycled through recently, then there's very little risk of any bacterial growth unless you supply a medium for it to grow on and don't heat the water above 60 deg C. Methinks Allan's allegation of Urban Miff stands well next to the Maffs being touted around...

[Colin Smith]

Chris

 

The figures will be psig.

 

The immediate problem I see is that your pump cut-off pressure is far too high. At 45psi it's going to blow the PRV everytime it switches on because the PRV will be set at about 42psi.

 

A friend of mine fitted the same Flojet pump and the PRV started opening and closing at rat-a-tat speed everytime the pump activated. I diagnosed the problem for him and he had to change it for a Jabsco 30psi cut-out version (20psi cut-in) as the Flojet does not have an adjustable pressure switch setting. If someone sold you a 45psi pump (rather than your specifying it) then take it back and swap it. They are wrong to have specified it knowing that PRV's are invariably 42psi (to protect the calorifier).

 

If you follow the maths, you will see that actually you want to turn the pump down to about 28psi to be absolutely sure (with a reasonable safety margin) that the PRV will not open owing to expansion. A 10 litre expansion vessel and a 2 litre accumulator will be appropriate in terms of size.

 

Chris

 

Hi Chris,

 

Thanks for your response. I’ve been off and re-checked everything: I take your point regarding the pump pressure etc. However the calorifier is stainless steel with a design max pressure of 8bar and it’s fitted with an adjustable PRV. The calorifier manufacturer recommends a maximum of 6bar so I believe I should be OK using the Flojet pump and the PRV set at 4bar.

 

I know most boats are fitted with a copper calorifier and they can are susceptible to ‘balooning’ or failure at pressures over 3bar – hence the PRV’s set at 3bar.

Thanks again for the calcs and your recommendation for sizing the expansion vessel and accumulator tank.

 

Colin

Edited May 18, 2009 by Colin Smith

[RobinJ]

Nope - the maffs are right!!

I have no doubt that Chris's figures are right, the problem is it is often difficult to pin down what they mean in an inidivudal case!

 

When looking at accumulators, the maths is fairly straight forward, in theory dealing with changes in pressure and volume at constant temperature.

 

Turning to calorifiers and expansion vessels, the maths gets complicated, since the expansion of water actually varies with temperature. As a result it is necessary to know exactly what temperature you are starting from and going to. A calorifier could be full of cold water just out of the tank, or warmish water thats been sat in an engine room for hours. It could be heated to 80 by an engine thats been run all day or heated to 90 by an immersion heated plugged into the bank.

To demonstrate the problem, water heated from 20 to 40 will expand by nearly 2%, heated from 40 to 60 it will expand further by over 2.5% and going from 60 to 80 by over 3.5% more, in total around 8%.

Clearly what is important when doing these calculations is to use maximum values, so any expansion tank is big enough to cope with the worst case.

 

I suspect that the comments made by people who are saying my PRV nevers leaks and those who are saying my PRV always leaks, cover the variation in the variables of each individual case, including whether or not a NRV is fitted.

 

As a guestimate (taking 90 as a typical engine hot end limit) I would allow for 10% expansion of heated water.

[RLWP]

A bit more from me.

 

I needed to take our hot pump out of our system on Saturday (it leaks) and so had to let the pressure out of the hot side. we have the simplest of simple systems (Tank, filter, pump, horizontal calorifier, tap). With the pump off and the tap open I got about half a small washing up bowl full of water from the hot tap before it stopped running.

 

I guess that there is quite a lot of air trapped in the top of our calorifier.

 

Richard

[chris w]

To demonstrate the problem, water heated from 20 to 40 will expand by nearly 2%, heated from 40 to 60 it will expand further by over 2.5% and going from 60 to 80 by over 3.5% more, in total around 8%.

Clearly what is important when doing these calculations is to use maximum values, so any expansion tank is big enough to cope with the worst case.

As a guestimate (taking 90 as a typical engine hot end limit) I would allow for 10% expansion of heated water.

I don't know from where you are culling your data, but even if you heat water from 0degC to 100degC it will expand by 4.05% and not the 8% to which you allude.

 

Over the likely maximum range of say 10degC to 90degC it will expand by 3.4%.

 

Chris

[RobinJ]

I don't know from where you are culling your data, but even if you heat water from 0degC to 100degC it will expand by 4.05% and not the 8% to which you allude.

Values for β vary from -0.000203 @ 0degC to 0.00212 at 100degC, so over that range it will be as high as 12%. I suspect you are looking at the average linear expansion coefficient (α)!

[chris w]

Values for β vary from -0.000203 @ 0degC to 0.00212 at 100degC, so over that range it will be as high as 12%. I suspect you are looking at the average linear expansion coefficient (α)!

No, I am looking CORRECTLY at the table of density changes. You're the one using the wrong parameters.

 

clicky

 

Chris

[ChrisPy]

Values for β vary from -0.000203 @ 0degC to 0.00212 at 100degC, so over that range it will be as high as 12%. I suspect you are looking at the average linear expansion coefficient (α)!

 

Chronic abuse of maths somewhere.

The plumbing industry works on 3%.

The total expansion, from 4degrees C to boiling, is about 4.5%. Your typical range is much less than this.

[chris w]

Chronic abuse of maths somewhere.

The plumbing industry works on 3%.

The total expansion, from 4degrees C to boiling, is about 4.5%. Your typical range is much less than this.

Sorry to be pedantic but it's 4.07% from 4degC to 100degC.

 

I know you already know this next fact but, for the avoidance of doubt in others' minds, water increases in density from 100degC down to 4degC and then decreases in density again from 4degC to 0degC by a small amount (.013%).

 

It's this latter weird property of water, different from all other liquids, that allows fish to survive the winter in a pond because when the surface water is frozen, the "warmer" water sinks to the bottom of the pond. Otherwise the pond would freeze from the bottom and the fish would be forced to the surface to die. It's the same reason that icebergs float.

 

Chris

Edited May 19, 2009 by chris w

[RobinJ]

No, I am looking CORRECTLY at the table of density changes. You're the one using the wrong parameters.

Thermal expansion coefficient

Not interested in how heavy it is, just how big it is!

Chronic abuse of maths somewhere.

The plumbing industry works on 3%.

The total expansion, from 4degrees C to boiling, is about 4.5%. Your typical range is much less than this.

Is that because they only look at expansion down a small bore pipe?

[chris w]

Thermal expansion coefficient

Not interested in how heavy it is, just how big it is!

Now don't be a silly-billy........ do you not understand what causes the density to change? Do you not understand what density is?

 

What you are quoting is the thermal expansion coefficient which is 1/ρ.dρ/dt where ρ is the density and T is the temperature. So at 4degC the thermal expansion coefficient is zero. You're confusing what the various maths bits actually mean.

 

Chris

Edited May 19, 2009 by chris w

[RobinJ]

Now don't be a silly-billy........ do you not understand what causes the density to change? Do you not understand what density is?

Err... Heat maybe?

What you are quoting is the thermal expansion coefficient which is 1/ρ.dρ/dt where ρ is the density and T is the temperature.

And as the density goes up the water expands in all three dimensions?

[chris w]

Err... Heat maybe?

 

And as the density goes up the water expands in all three dimensions?

You've got it the wrong way round....as density increases its VOLUME (expansion) DECREASES ie: density is inversely proportional to volume. I suspected you didn't understand the term density.

 

ie: mathematically, density = K x 1/Volume and what you have quoted is K

 

Chris

Edited May 19, 2009 by chris w

[RobinJ]

You've got it the wrong way round....as density increases its VOLUME (expansion) DECREASES ie: density is inversely proportional to volume. I suspected you didn't understand the term density.

I just knew I was going to fall into that one

[nb Innisfree]

Innisfree stated that he didn't have any expansion vessel. Is he now saying that he has an accumulator? He said that his additional 4 metres of pipe is absorbing the expansion. That's total crap. How can a pipe that is already full of water absorb more water? It's nonsense. The heat will get transferred along it but not the expanded water.

 

If you are able to follow my maths above then you will understand that just using an accumulator to do both jobs (ie: accumulator and expansion vessel) is a real compromise unless your accumulator is even bigger (~10 litres) than a normal expansion vessel. It'll partially work but the PRV will still open. What is probably happening is that someone uses a hot tap before the water is fully expanded and that relieves the expansion pressure before the PRV opens.

 

Chris

By the way I never stated that I didn't have an expansion tank I said I had 2 accumulators, read my posts