C/H "Pumping Over" - What Else Can I Do?

[Peter Hemmings]

Hi,

I have read various solutions to the above problem and (I think) I have exhausted them all. Any advice would be appreciated.

System: Baxi Barcelona open vented, with two 2 port valve circuits. The pump is situated between the air separator and the two valves in the flow of the boiler. The Boiler is situated in the garage with 28mm pipework up to the valves, air separator and pump (all in an airing cupboard on the landing. The air vent goes vertically to the small header tank in the loft. History: System plumbed in about 7 years ago and originally had this problem but though I cured it by lowering the header tank water level to 2 inches and extending the vent height. Recently replaced a large hall radiator with one about 3/4 its size. During balancing the system I found it had been pumping over again (dirty water in header tank). As the pump was getting noisy I replaced it with a Grundfoss "smart" pump. When I removed the old one I foung it had quite a lot of hard black sludge on its ports. I then realized this was probably due to the oxygen from pumping over. Had the boiler serviced and is OK! I asked the engineer for advice, and was told to clean the system as there was probably some restriction in the flow. Cleaned it with Sentinel X300 and ran boiler on high temperature with all valves open, but still the problem exists. It did clean the system slightly but I have alway had X100 in it and I did take all the old radiators off the walls and cleaned them when I plumbed in the system. Symptoms: When the boiler starts and the pump switches off at the end of priming and when the boiler stops I get about a tumbler measure pumped over. I think there is also some fluid coming into the tank from the C/H circuit feed. I tried closing a gate valve (in the feed) so it was only just open, but it obviously did not improve the pumping over. Even with the pump on its lowest speed (both auto and fixed) I still have the problem. The Baxi installation manual did not specify a need for a bypass and as I have 2 towel warmers without TRV's I assumes it would make no difference!! Well that's the sorry saga so far, I would appreciate any comments on my system and any possible ways I can cure the problem as its now exhausted my knowledge!!?

I am tending to think it is my design that may be the problem as I have always had problems with air in the system!!

Thanks & sorry for the length of this message.

[Peter Hemmings]

And one further observation is that it is not the valves closing causine the problem (as the "professional" engineer advised). They don't close on the priming cycle.

[Vortex]

Modify to a sealed system. That will fix it once and for all....and bring multiple other benefits

I did this 2 years ago. It was a very simple job....but my advice would be to make sure you have a mechanism to introduce liquid inhibitor (rather than the Gel/Mastic gun approach which IMHO is a hassle and expensive).

I suggesr reading Ed Sirretts FAQ:

[John Rumm]

When you say air separator, are you just referring to the vent pipe?

Remember the CH header tank is a Feed *and* expansion tank. CH water is supposed to flow into it as well as out. Hence you would normally expect the water in it to be as dirty as that in the rest of the system.

Are these in the controlled zones or parallel with them though?

Sealed system is the obvious answer if the boiler can be adapted to run that way. Failing that, making sure the vent pipe is not too close to the suction side of the pump can help reduce air ingress.

[Doctor Drivel]

Two solutions:

1. Convert to a sealed system. Not that difficult.

1. The easiest is to convert the system to a combined vent and feed tank system. One 22mm pipe top the bottom of the F&E tank. Your boiler has a high limit stat on it so no problems here.

Aerjec Water De-aerators

Look at the mounting positions. Just cap off the cold fee port. Take the vent in 22mm pipe to the base of the F&E tank. Make sure the tank is not too full, overwise pumping through the overflow. Bending the brass ballcock lever is the way here to reduce the water level.

Potterton/Baxi specify an air ejector where the combined Feed and Expansion pipe meets the system. Works very well. With this method extremely low heads can be attained (200mm from top of boiler).

I prefer a combied vent and feed than a sealed system.

[Doctor Drivel]

See this too:

[Peter Hemmings]

No, the installation manual said it was not necessary!, but maybe not in all cases?

No, its a purpose made "small tin can" about 4 inches diameter and one and half high. Vent goes out of top (one side) and the flow and and out to the pump are at bottom. Feed goes in at side.

OK but on my old system (similar type) I never got pump over. It was a smaller house though. Someone did advise me that it should not pump over and should be clear, I could even shut off the feed pipe once the system was bled, obviously wrongly!!!

The two warmers are in the zones, not directly across the boiler (as per bypass).

I do not think a Barcelona will work, I will have a look.

Failing that, making sure the vent pipe is not too close to

Is that horizontally of vertically close!?

Because I have this "tin can" separator with flow in and out at the bottom The pump is only about 4inches to the side of the vent! The pump is also only 8 inches below the separator:

| | _|____ | | | | | |__| |______| | | / | / |OUT | | | | | | _ |___ IN | Pump | |______|

To Zone Valves (2 port, CH and HW)

The distance header tank water level and top of vent loop is about 2 foot ATM. I could (with a bit of artistic plumbing!) extend it another foot.

Would a bypass across the boiler in the garage also be a solution/help!?

Thanks

[John Rumm]

Don't confuse expansion with pump over. For normal expansion (i.e. what happens as the water heats up) the flow will be up the feed pipe into the base of the cistern (i.e. under water). Pump over is water going up the vent and falling into the tank.

It should not pump over, but the water will not be clear.

You could do this - this would force all expansion to take place up the vent pipe. This is more likely to cause pump over though.

Does the boiler overrun the pump after the call for heat from the zone stats has gone away? This may lead to it pumping against a mostly blocked system except for the feed/ expansion and vent pipes.

With the addition of overheat stat, expansion chamber, and filling loop it most can be converted.

Ed may be able to give you a definitive answer.

Well either, but having the pump lower than the gizmo probably helps a little in this case.

Depends on why you get the pump over. If it is as a result of the pump pumping against closed valves, then yes a spring controlled bypass might help.

[Doctor Drivel]

at page 13 fig 12.

at page 11.

[Doctor Drivel]

I gave the answer.

[Doctor Drivel]

Look at my posts to solve.

[John Rumm]

Common sense would suggest that even on the occasions you do stumble upon the correct answer, it would still be wise to wait for conformation from someone with credibility.

[Doctor Drivel]

Which is a Chav from Essex like you? Gosh!! It is best he take noet of a pro like me, not an amateur know-it-all.

[John Rumm]

Could be, or any number of others...

Oh we do take noet of you - often out of morbid curiosity.

[Peter Hemmings]

Peter Hemmings wrote: Thanks for the posts,

I could have gone away and not said anything but.....

here I am, and I know the drawing is not good, but did anyone spot the error!!

snip

OK look at this:

have an Aerjec de-aerator, unless the design has changed, I have the flow/returns wrong!!! I think I should correct this first, and probably make it a combined single pipe as per the posts.

In reply to John Rumm's question about the boiler over-running the pump after heating is satisfied, no the pump and valve operate together. As I said previously, I checked and I get pump-over (not expansion!!) at the end of an initial 10 second (cold) or 60 second (over temp) with no no boiler heating. I get pump-over from cold at the end of the 10 seconds when the pump switches off and on (quickly) with no movement of the valve(s). I also manually held them open and I still get the problem.

[Ed Sirett]

I have read most of the posts in this thread.

With the setup you have you should not get much if any pumping over. The little you get is a transient effect which therefore is much much much less serious than if it happened all the while. Frankly I'd just put plenty of inhibitor in and leave well alone.

My commiserations are with you seeing as you have one of the dreaded Barcelona boilers, nevertheless a sealed system is possible with this model. Modifying the tin-can/aerjec unit will probably not acheive anything anyway, IMHO.

[Peter Hemmings]

Yep, I know they have problems but keeping it cleaned out and fingers crossed!!

nevertheless a sealed system is possible with this

OK Modifying the tin-can/aerjec unit will probably not acheive

OK but if I make a single pipe vent/feed it will stop the "whooshing of water" in the loft. I will need to flush it out a bit more with clean water so I have not yet put in the inhibitor and I know that doe improve it somewhat!

I will post results later next week, just for interested readers!

Thanks for all the replies

[Doctor Drivel]

Yes, Essex is Chav county UK.

[Doctor Drivel]

You don't want "any" pump over or air sucked in of any description. Convert to a combined feed and expansion and all solved.

[Doctor Drivel]

If cleaning the system use Fernox Cleaner rather than Senitel. Fernox is aggressive and does get rid of the crud in the system. Flush out WELL. Sentinel X-100 is fine as the inhibitor after.

If you have hard water and using a Combined feed and expansion pipe and de-aerator. Fill the system with 1/2 the inhibitor, vent and make sure all is fine. Then add the rest in the tank. This makes sure the tank is not just fresh water and is rich with inhibitor, which can cause scale problems around the aerator (remote I know, but just to be sure).