magnaclean filter and low loss header

[Onetap]

The reason I am losing 50% of the boiler flow is because the header gate valve (by-pass) is doing its job and reducing the harmful 30°C delta-T from the cold radiators down to 15°C across the boilers.

The Celsius 25 installation instructions on P11 has a diagram showing an additional system pump, which requires a permanently open by-pass (LLH, short circuit).

http://www.keston.co.uk/downloads/manuals/Celsius-25-manual.pdf


The 2 boiler reverse-return arrangement, with the ABV, shown on the link below, is intended for direct connection to the distribution (secondary) system.

http://www.keston.co.uk/downloads/support_info/multiple/2xcelsius-v2.pdf

This design assumes that "the system resistance is within the spare capacity of the boiler pumps. Otherwise consideration should be given to the use of an additional System pump."

So, the boiler pumps can be used for the secondary distribution, if capable, and an ABV is required. If an additional system pump is required, you then need an open by-pass/LLH.

The LLH is preferable since otherwise the secondary flow will vary enormously depending on whether one or two boilers/pumps are running.

The single boiler header diagram shows 1 ½” pipework was recommended for one Celsius 25.

http://www.keston.co.uk/downloads/support_info/multiple/1xblr-h.pdf


I still don't think the undersized LLH pipework is a major problem. There is a CIBSE guide which I believe recommends the pipework for a LLH should be selected so that the pressure differential at the secondary F&R connections is no more than 40 Pa, i.e., negligible pressure loss. It would be easy to look up the pressure loss per m for the LLH pipe to see what the pressure difference is at the two primary flow rates.

The differential pressure (across F&R) could cause nuisance flow in a secondary circuit when you don't want it and the pump was off. Since both the CH and DHW circuits on this installation have zone valves, it will not cause a problem.

 

[Onetap]

What happens if you replace the ABV with a fully open gate valve as in the OP's pic.

I know exactly what happens if you put in a fully open gate valve, DIA. I also know what happens when you shut said gate valve. Therein lies a tale. How long have you got?

 

[doitall]

What happens if you replace the ABV with a fully open gate valve as in the OP's pic.

I know exactly what happens if you put in a fully open gate valve, DIA. I also know what happens when you shut said gate valve. Therein lies a tale. How long have you got?

Should be good for a few more years yet. want to start another thread.

And you've never fitted check valves on smaller duty pumps, because the larger head pumps stalls them.

 

[ajrobb]

The controls will modulate the burner to deliver the set flow temperature (as set by the knob on the boiler?) AND will also alter the pumps' speeds to maintain the dT across the boiler at 15 degC, or as close as practical.

It sounds like as long as the fan control unit doesn't just use each fan speed over a fixed range of boiler output then I guess there shouldn't be a problem with boiler pumps stalling.

Once again, thanks for all the help and info guys. I find it very interesting (I suspect some of you do too )[/img]

 

[Onetap]

And you've never fitted check valves on smaller duty pumps, because the larger head pumps stalls them.

I have. I'm glad you reminded me, I'd forgotten some of the details of that job.

There were 1" (or 1 1/2"?) duty and stand-by pumps serving a DHW storage calorifier, adjacent to a big (4 or 5") pair of district heating pumps, both drawing from and returning to the same 8" F&R headers. The pumps were installed long before I saw them and the system had separate flow and return headers.

The small DHW pumps did stall. In fact, turning them on generated exactly zero flow. They were usually left turned off with all the isolating valves open. Water flowed backwards through both pumps (no NRVs) and the DHW calorifier took all night to heat up. The 3-port control valve had seized and the stored hot water got dangerously hot. The hot water was used by the catering staff at lunch time and was cold after that.

I had the boiler room repiped and connected the F&R headers, changing them into a LLH. I also got new DHW pumps with NRVs, a new control valve and a new tube bundle for the DHW calorifier; the data plate showed that the original bundle had been sized for MTHW. The plant had changed to LTHW when the Pressure System Regulations were introduced.

It worked. The DHW pumps did not stall; that's what the LLH is intended to do. The DHW was heated to 60 degC and stayed hot. In fact, I had fitted a 3-port modulating control valve; with hindsight, on/off control of the pumps would have been adequate.

 

[Onetap]

Once again, thanks for all the help and info guys. I find it very interesting (I suspect some of you do too )

You're taking the Michael now.

You asked for an opinion on a simple domestic heating system and you got 300 contradictary replies, all from self-proclaimed experts, mixed up with moronic posts from the usual trolls and unemployable geeks having a short break from surfing porn, further confused by your inability to distinguish the morons from the experts.

Isn't the internet wonderful, what would you do without it?

May I wish you the very best of luck, sir; you'll need it.

Now I must get back to the porn.

 

[doitall]

Onetap.

So would you say that given the common return has a higher head of water.

If both boiler pumps turned down to 1, and the system pump 15/55, on full head could over power the boiler pumps, and take the majority of the circulating water from the system side. see sketch I did earlier.

edit, remember the pipework is also undersized. in my opinion.

If I didn't make that clear let me know.


View media item 29614

 

[Onetap]

remember the pipework is also undersized. in my opinion.

And I agree with you. If I were trying to fix this, I'd do some calculations to prove, or disprove, my belief that it is not a serious problem.

On your diagram, the primary circuit (yellow) is connected to the secondary circuit (red) at two points; call these points A (LH) and B (RH). Call the pressures Pa and Pb.

The normal flow direction in the common pipework is from A to B, left to right (primary flow > secondary flow). The majority of the primary flow is drawn into the secondary pump, the remainder flows from A to B.

If the boiler pumps are at low speed, near to stalling, and the secondary pump is on high speed, then the flow direction is reversed, B to A (primary flow< secondary flow). So far, I think we agree.

You say that the adverse pressure gradient generated by the secondary pump, exceeds the pressure produced by the primary pump, stopping flow. When there is no flow ( and so no pressure losses due to friction in the primary circuit), the pressure across the primary pump is Pa at the outlet and Pb at the inlet. Pa must be greater than Pb for this to happen. The primary pumps spin, but produce no flow.

If you look at the secondary circuit, the flow direction must be from high pressure to low pressure. Secondary flow is from B to A, so Pb is greater than Pa; the primary pumps do not stall. In fact, the pressure generated by the secondary pump assists flow through the primary pumps.

A problem will probably appear when the primary flow is less than the minimum specified by Keston, and not at zero primary flow.

I think you could get the primary pumps to stall the secondary pumps (edit to add; no you can't do that either). The LLH/hydraulic separator is meant to stop any such interaction between the two circuits. This one isn't a very good LLH, but I don't think it is that bad. I'd suspect (but don't know) that the heat exchanger crack happened when someone closed the gate valve.

I could put some numbers into this and work out exactly what will happen under various circumstances but don't have time now. It's just the sort of juicy problem that I like to crack.

 

[doitall]

It would be interesting.

I think the secondary will have a major impact on the boiler pumps under certain condition.

Points to consider.

The header is under sized.

The boiler pumps are also on the return, so the resistance will also be greater.

The system return has a greater head of water.

The secondary pump will take the easy route from the system return, red line, because of the above facts.

 

[doitall]

The only reason it's working at the moment, is because of the open circuit through the gate valve.

There's not enough circulation via the system even with all three pumps running full blast.

You are wasting 50% of the boiler energy to keep the delta T from going off the scale.

 

[Biggles..]

As much as I champion kestons,these pair have had their day. Re piping it with them still attached would be like polishing a turd.

 

[doitall]

As much as I champion kestons,these pair have had their day. Re piping it with them still attached would be like polishing a turd.

Thank you Lee

 

[ajrobb]

I get the impression that the best thing I can do now is to start digging for a new soak away in anticipation of a new boiler in the spring.

 

[ajrobb]

The only reason it's working at the moment, is because of the open circuit through the gate valve.

If the gate valve was an open pipe then it would better correspond to Keston's secondary pump scheme, so I don't see that as a problem.

There's not enough circulation via the system even with all three pumps running full blast.

Ignoring DHW, the secondary circulation is fixed by the radiator balance and not the boiler(s). I agree the boilers are too big for the radiators.

You are wasting 50% of the boiler energy to keep the delta T from going off the scale.

Apart from extra pumping, no energy is wasted; half the flow twice the delta-T = same energy.

I suspect that most balanced modern domestic CH will use a by-pass to increase boiler flow once the TRVs kick in and increase delta-T above the safe limit for the boiler. This should be normal when the heating is cycling on the room thermostat and all the TRVs should be at their set points. The boiler pump would need to be infinitely variable to maintain the optimum boiler delta-T at the set flow temperature (unlike Keston's 3-speed pump) but it still needs flow through the by-pass to protect the boiler. Also most CH+DHW boilers will be oversized for CH alone and will need by-pass flow to limit delta-T when heating radiators from cold.

OK, many systems are not balanced, so the total flow through the radiators can be more than the design flow for the boiler and the by-pass need not open. Even when the TRVs reach their set points, the delta-T from the radiators might still be lower than the limit for the boiler. But you wouldn't leave radiators not balanced, would you?

 

[doitall]

Interesting last question.

Yes is the answer, other than a bit of tweaking, if the pipework is sized correctly, all the valves should be fully open. increasing efficiency and reducing system noise.