Auto bypass always open - help!

[doitmyself]

I can't help wondering if the system has been balanced for the correct temperature differential.

The British Gas 330 boiler has a maximum output of 30kW but is set, at the factory, to give 18kW. As there are only 13.5kW of rads, I would assume the output has not been changed. The boiler is designed to have a flow-return differential of about 20°C. With a 20°C drop the head loss through the boiler, at 18kW, is about 0.9m. This would leave about 3m available from the 15-50 on speed 3 for the head loss in the radiator circuit.

If the system has been balanced for a differential of about 10°C, the flow rate will double and the boiler head loss alone will be about 3.6m (four times greater), which is more than a 15-50 can deliver. This would not someone trying to balance for an 10°C, even if they could only achieve something higher.

 

[doitmyself]

Could the output of the boiler affect things? IE if it is adjusted to put out the full 31kw? I believe its maximum output can be limited.

If it has, it should be indicated on the data label attached to the inner door.

 

[Chris Clarke]

I can't help wondering if the system has been balanced for the correct temperature differential.

The British Gas 330 boiler has a maximum output of 30kW but is set, at the factory, to give 18kW. As there are only 13.5kW of rads, I would assume the output has not been changed. The boiler is designed to have a flow-return differential of about 20°C. With a 20°C drop the head loss through the boiler, at 18kW, is about 0.9m. This would leave about 3m available from the 15-50 on speed 3 for the head loss in the radiator circuit.

If the system has been balanced for a differential of about 10°C, the flow rate will double and the boiler head loss alone will be about 3.6m (four times greater), which is more than a 15-50 can deliver. This would not someone trying to balance for an 10°C, even if they could only achieve something higher.

Thanks, I don't really know what you mean by your last paragraph? What differential should be looked for at the radiators to achieve 20C at the boiler?

I know that the system was not balanced following the change of radiators / pipework. From everyone's input I'm thinking I should ask for the following to be investigated.

1. Change pump to a higher rated one, probably 6m.
2. Check that boiler output has not been adjusted from 18kw, if it has get it changed back.
3. Carry out full balance of system.
4. Re-set bypass valve, change if required.

 

[doitmyself]

Thanks, I don't really know what you mean by your last paragraph? What differential should be looked for at the radiators to achieve 20C at the boiler?

I'm sorry for confusing you. The last sentence should read:

"This would not stop someone trying to balance for an 10°C, even if they could only achieve something higher."

In the past, before condensing boilers, the standard was to balance the system - that's boiler and rads - for an 11°C (actually 20°F, which shows how old it is) differential. This has changed to 20°C for two reasons: to reduce the head loss due to friction, particularly due to the heat exchanger which tend to have very narrow pathways; and so the return temperature is closer to the dew-point of natural gas (55°C), which is the temperature at which condensing begins (It increases as return temperature drops.).

A condensing boiler should be balanced for a drop, across boiler and rads, as close to 20°C as possible, but not any higher. You may not be able to achieve 20°C due to the way the pump interacts with the system.

You may find that balancing the system will remove the need to replace the pump. There is over 3m of head available for the index circuit.

 

[doitmyself]

I based the info above on the boiler's output of 18kW. You have only 13.5kW of rads so the boiler will modulate to that output. This means a flow rate of 580 litre/hr giving a heat exchanger head of 0.5m and an available pump head of 3.5m at speed 3. This should be more than enough.

 

[Chris Clarke]

Thanks, makes sense to me. Although is 3.5m head sufficient for the longest microbore to manifold circuit of 18m? My rough calculations would suggest not.

And it seems like that was too long in the first place but that's a different story.

 

[Ric2013]

Too late for me to read all this. Basically, a pump that is too weak for the job will NOT be opeing the bypass. My experience with differential pressure auto bypasses is that they can very easily be jammed very slightly open with a bit of crud from the system and then they let by, even when the head is not sufficient to overcome the spring pressure.

If, as I suspect, your system is old and has a lining of black stuff on the pipe bores, I wonder if a bit of this could have got in between the valve seat and seal on your ABV when it was last forced to open and this has lodged there. I once tried an experiment in my own house with an ABV going through a bypass radiator, but, after the valve got jammed slightly open a second time in the same number of weeks, I gave it up.

 

[chris watkins]

I know that the system was not balanced following the change of radiators / pipework. From everyone's input I'm thinking I should ask for the following to be investigated.

1. Change pump to a higher rated one, probably 6m.
2. Check that boiler output has not been adjusted from 18kw, if it has get it changed back.
3. Carry out full balance of system.
4. Re-set bypass valve, change if required.

Changing the pump to a 6m head one will only make the by-pass opening worse however you my need a higher head pump to over come the resistants of the micro bore system, so what is to be done.

First thing to understand is why we need a by-pass in the first place.
The main reason (& in this case) is to maintain a minimum flow rate through the boiler at all times, this allows the boiler controls to adjust the burn of fuel to try to achieve a 20deg C difference between the flow & returning water. The system is pressure dynamic, that is to say zone & TRV valves are opening & closing all the time so the work the pump has to do keeps changing. With all the valves open the pump needs move a lot of water (heat) but with only enough force to overcome the resistance in the whole circuit. As the valves close the flow rate around the circuits will decrease but the pressure generated by the pump will rise (they are directly related, see pump graphs).
You have fixed speed pump installed so the pressure will climb, this rise in pressure was used to open a auto by-pass valve at the point were the minimum flow rate required by the boiler manufacturers was not going to be achieved. This sudden opening between the flow & return should send hot water straight back to the boiler raising the return temperature which would cause the boiler to turn off on the boiler stat.
What are the options then?

 

[Chris Clarke]

Changing the pump to a 6m head one will only make the by-pass opening worse however you my need a higher head pump to over come the resistants of the micro bore system, so what is to be done.

First thing to understand is why we need a by-pass in the first place.
The main reason (& in this case) is to maintain a minimum flow rate through the boiler at all times, this allows the boiler controls to adjust the burn of fuel to try to achieve a 20deg C difference between the flow & returning water. The system is pressure dynamic, that is to say zone & TRV valves are opening & closing all the time so the work the pump has to do keeps changing. With all the valves open the pump needs move a lot of water (heat) but with only enough force to overcome the resistance in the whole circuit. As the valves close the flow rate around the circuits will decrease but the pressure generated by the pump will rise (they are directly related, see pump graphs).
You have fixed speed pump installed so the pressure will climb, this rise in pressure was used to open a auto by-pass valve at the point were the minimum flow rate required by the boiler manufacturers was not going to be achieved. This sudden opening between the flow & return should send hot water straight back to the boiler raising the return temperature which would cause the boiler to turn off on the boiler stat.
What are the options then?

Fit a Grundfos pump with proportional pressure function? God knows, that's why I'm asking! Someone else must have come across this before?

 

[Chris Clarke]

Low loss header?

 

[chris watkins]

If I am looking at your boilers M.Is then 4.3 - "a variable duty pump should be set to give a temp diff of no greater than 20deg C"
"Important F&R pipe sizes - up to 18kw = 22mm, >18kw =28mm"
The flow rate required to provide 18kw @ 20 degC would main that your pump would loose approx 1M head just pushing that amount of water through the heat exchanger if you needed it full 30kw you would loose over 2.5Metres.

Your existing pump is a 5M head but that is the max it can push with no or little flow. Moving approx 0.215L/S it would only develop a pressure of approx 3.7M head, less the boiler 1M, would leave approx 2.4 M for the rest of the system. Not likely to be enough to overcome the resistance you have put in by re-piping in 10mm.

 

[Chris Clarke]

Yes so pump size needs to go up but this will play further havoc with the auto bypass? Although it does say variable duty?

 

[chris watkins]

Fit a Grundfos pump with proportional pressure function? God knows, that's why I'm asking! Someone else must have come across this before?

Yes Chris thats why they were invented.
If you fit one would you need a by-pass ??
I don't think you do & have had my one at home closed for couple of years now with an Alpha 2 installed.
But if you do want to provide the minimum flow then why not use close coupled tees instead of LLH (to expensive) to provide hydraulic seperation. You are going to buy a new pump anyway so give it a go!!

 

[Chris Clarke]

Don't you need the bypass to prevent the pump getting dead headed on the pump overrun?

 

[chris watkins]

Effectively that is what your by-pass is providing at the moment you just don't have a pump on the other side of it to push water around the heat emitters.

 

[chris watkins]

Don't you need the bypass to prevent the pump getting dead headed on the pump overrun?

Pump reads the increase in pressure and just slows down to 5Watts.
But if boiler does need to clear away heat then yes.
Forgot my one always has open circuit as I have WC.

 

[Chris Clarke]

Something like this?

 

[chris watkins]

Yes just like that.
Good old Andrew M he pointed the way for me, also have a look at Caleffi - section 5 Page 27
https://www.caleffi.com/sites/default/files/coll_attach_file/idronics_15_na.pdf

 

[doitmyself]

Thanks, makes sense to me. Although is 3.5m head sufficient for the longest microbore to manifold circuit of 18m? My rough calculations would suggest not.

It's not just the length of the pipe which is important. You have to take into account the flow rate, which depends on the size of radiator and the temperature differential. For example:

Resistance of 18m of 10mm pipe connected to a 2kW rad with a 20C differential is approx 0.58m, with a 10C differential it is approx 1.94m. The flow rate has doubled so the resistance increases by four times.

Upload your rough calculations and I can take a look at them.

 

[Chris Clarke]

So presumably the two pumps are wired to the same point and both overrun? The intelligent pump shuts down itself when the valves close?