Gas Boiler Heat Exchanger Pressure Loss

[Johntheo5]

It shows the residual head

They just looks like typical pump curves to me where the pump head (as always) is proportional to the square of the speed.

The curve I showed above, and directly below if referring to a 7M pump shows that at 60% speed its able to pump over 900 LPH. 7X0.6^2 = 2.52M but this couldn't/cant pump anything as its the closed valve, dead headed, head??.

 

[JonathanM]

They just looks like typical pump curves to me where the pump head (as always) is proportional to the square of the speed.

I don't really understand pumps, still. But I'm pretty sure this is the pump that most boilers have in them these days. I wonder whether this would be what the curves would look like if they hadn't been through the boiler first i.e full head, rather than residual head. Please treat this with caution!

 

[vulcancontinental]

Viessmann have another graph I find confusing, but not quite as bad, and maybe it might help make sense of the Vaillant one. Maybe it's a German thing. It also has a horizontal line across it, like the Vaillant one. Viessmann call it Upper Operational Limit, and I have no idea what it means.

View attachment 283886

It means anything above that line you have to take other measures because it's quite a restrictive heat exchanger/boiler internals. e.g. Basically you have a metre and a half residual head hence low loss headers or closely coupled tees and a second pump. It may work well exceeding the line M but may not be shipping enough as the boilers minimum flow rate.

 

[Johntheo5]

Re post #17
Yes, but as I've often asked the reason for most A rated pumps behaving in a constant pressure (CP) fashion even though in constant curve (CC) mode, UPS 3 is a very good example.
It would require a ~ 12M or higher head pump to achieve the below if a true constant curve pump as the head starts falling as soon as the pump starts delivering flow.

 

[JonathanM]

It means anything above that line you have to take other measures because it's quite a restrictive heat exchanger/boiler internals. e.g. Basically you have a metre and a half residual head hence low loss headers or closely coupled tees and a second pump. It may work well exceeding the line M but may not be shipping enough as the boilers minimum flow rate.

Thank you. Great explanation!! Makes sense now.

Do many people have to fit those extra things these days?

I was worrying about my current pump being too powerful for a new boiler. Now worrying about it not being powerful enough!

 

[vulcancontinental]

They just looks like typical pump curves to me where the pump head (as always) is proportional to the square of the speed.

The curve I showed above, and directly below if referring to a 7M pump shows that at 60% speed its able to pump over 900 LPH. 7X0.6^2 = 2.52M but this couldn't/cant pump anything as its the closed valve, dead headed, head??.

View attachment 283894



View attachment 283892

Yes and no; I explain it as with no flow the pump will support a 2.5m column but no water is actually moved along as all the output is used to raise it. Alternatively imagine a pump on the outlet of a storage cistern with no pipework on it's outlet; at maximum speed it would move its maximum flow as it would have to overcome no resistance through a pipe to restrict flow.

 

[Johntheo5]

So, back to the start, how does one find out the pressure drop in any boiler Hx if not shown like below, does one have to get a couple of hoses, a isolating valve and a pressure gauge and run the water at say a 3M head through the Hx only??? seems ridiculous to me.

It would be a great aid to navigation in trouble shooting problems IMO since the modern gas boiler Hx makes such a wonderful strainer.

 

[vulcancontinental]

They just looks like typical pump curves to me where the pump head (as always) is proportional to the square of the speed.

The curve I showed above, and directly below if referring to a 7M pump shows that at 60% speed its able to pump over 900 LPH. 7X0.6^2 = 2.52M but this couldn't/cant pump anything as its the closed valve, dead headed, head??.

View attachment 283894



View attachment 283892

There are additional parameters to 41-100 whereby the pump willbe controlled by the boiler PCB to achieve a target delta T across the boiler thermistors. e.g.10, 15, 20°C

 

[fixitflav]

Viessmann have another graph I find confusing, but not quite as bad, and maybe it might help make sense of the Vaillant one. Maybe it's a German thing. It also has a horizontal line across it, like the Vaillant one. Viessmann call it Upper Operational Limit, and I have no idea what it means.

View attachment 283886

I assume the curves are at a range of pump speeds. It's just a guess, but I'd say Residual head is the actual head minus the boiler losses i.e. the head available to push the water round the rest of the system (rads plus p/w). So if you want a flow say 550 l/h, and the system loss at that flow is 10 kPa, you would run the pump on curve E. If you plot the system curve on it, the intersection point is the system pressure on that particular pump curve.
For the Upper operational limit, all I can think of is that if the head were higher, the flow would be higher (along the curve) and perhaps the pump would draw excessive current.

 

[vulcancontinental]

Do many people have to fit those extra things these days?

Increasingly as systems have been extended, sometimes tacking radiators onto the ends of circuits raising the head loss if the index circuit and the mass flow rate. I usually read about them used with Intergas, Vaillant and Viessmann boilers but obviously any boiler may need it if the pump can't cope with flow rate and resistance.

 

[Johntheo5]

Viessmann have another graph I find confusing, but not quite as bad, and maybe it might help make sense of the Vaillant one. Maybe it's a German thing. It also has a horizontal line across it, like the Vaillant one. Viessmann call it Upper Operational Limit, and I have no idea what it means.

View attachment 283886

Still trying to figure this out.
Assuming pump running on curve L, 6M, and assuming a lowrate of 900LPH, 15LPM, then a 6M pump should have a total head of ~ 4.5M pumping at 900LPH, residual head at this flow is shown as 2.5M, this means that the boiler Hx requies a head of, (4.5-2.5), 2M at a flowrate of 900LPH, which seems reasonable enough.
Assume pump now changed to curve H, 4.2M, assuming same flowrate at 900LPH should result in the pump having a total head of ~ 3.2M, the Hx head loss will be the same at 2.0M so leaves a residual head of (3.2-2.0), 1.0M but above is showing a residual head of 1.7M??

 

[fixitflav]

Still trying to figure this out.
Assuming pump running on curve L, 6M, and assuming a lowrate of 900LPH, 15LPM, then a 6M pump should have a total head of ~ 4.5M pumping at 900LPH, residual head at this flow is shown as 2.5M, this means that the boiler Hx requies a head of, (4.5-2.5), 2M at a flowrate of 900LPH, which seems reasonable enough.
Assume pump now changed to curve H, 4.2M, assuming same flowrate at 900LPH should result in the pump having a total head of ~ 3.2M, the Hx head loss will be the same at 2.0M so leaves a residual head of (3.2-2.0), 1.0M but above is showing a residual head of 1.7M??

Where does the ~ 4.5m come from? I assume one of the graphs you're looking at is the one in #15, but which other(s)?

 

[Johntheo5]

Yes, curves L and H, curve L assuming a 6M Pump, the head will fall to - 4.5M when pumping 900LPH, curve H at 4.2Mwill fall to -3.2M when pumping 900LPH.

 

[fixitflav]

Yes, curves L and H, curve L assuming a 6M Pump, the head will fall to - 4.5M when pumping 900LPH, curve H at 4.2Mwill fall to -3.2M when pumping 900LPH.

Curve L shows 2.5m at 900LPH. I appreciate that is residual head, not total pump head, but I can't see any graph showing 2m boiler loss at 900LPH, though as you say that's reasonable.
But unless the system resistance (at a given flow) changes, if you reduce the pump from L to H the flow doesn't stay constant at 900LPH. If you draw a curve through 2.5m/900LPH, with head varying as flow-squared, the flow and head at speed H are where it cuts the curve. I make it approx 2m at 800LPH.

 

[Johntheo5]

You can't see the the Hx pressure loss, you have to derive it, as i said originally a 6M pump pumping 900 LPH will have a total head of ~ 4.5M (assuming a true CC pump), at 900LPH the residual pump head is 2.5M therefore the Hx head loss is (4.5-2.5) or 2.0M.
We are informed, below that when on curve H that the residual head is 1.7M @ 900LPH so of course its for a system that needs a lower residual head to give the same flowrate, one system might need 2.5M to circulate 900LPH, another may only require 1.7M to circulate 900LPH, the Hx head loss will remain the same at 900LPH, but because a 4.2M pump head will fall to ~ 3.2M when pumping 900LPM then the residual head is only (3.2-2.0), 1.2M (made a typo of 1.0M, above) or else a 4.2M pump will only fall to a total head of (1.7+2), 3.7M in pumping 900LPH. Its still amazing why they don't just specify the Hx head loss, all you need is the head loss at any one given flowrate, the rest can then be calculated very accurately because head is proportional to flow squared.