System boiler flow temperature fluctuates wildly

[Johntheo5]

Yeah I'll do this tomorrow. I tried it with just the non-TRV rads on and the dial set even lower, still went up to 65C or something. It's definitely not working properly - either that or my thermometer. If I select type E thermocouples on my thermometer instead of type K the numbers look right But they are definitely stamped as type K.

It was 11 last night and 16 during the day here, and my room is now 19C inside. The forecast for the next week is for highs of 16-18C and lows of 13-14C.

I poked out a Proster Thermocouple Thermometer (never used) that somebody gave me a present of 3 or 4 years ago and heated up a pot of water. My probes are also stamped K type but you can select 4 or 5 different settings, I compared the K reading with types E & T, type E read consistently lower than K and T,
At a measured (with a rototherm ) "55C", K:54.4C. T:54.0. E:43.6C.
Pot boiling so had to be at 100C. K:99.7C. T:98.3C. E:72.8C
Based on that the K scale is the correct one to use so your K readings are also correct and not Es??.

 

[molgrips]

Yeah I just tested it and these are definitely K.
I readusted my thermocouples on the pipes, no change.
I removed the control panel and tested the potentiometer - fine.

I might have to get a refurbed replacement PCB off eBay they are only £30.

 

[Johntheo5]

Probably won't prove much but maybe get heating load (as above) greater than 12kw (max boiler output) and see what happens.

 

[molgrips]

Yeah a bit tricky when it's not that cold yet! I may have to continue my investigation when it's colder and open all the windows first.

Or, just run a bath and use the hot water cylinder to take the load. But I'm not sure how much heat my hot water coil can dump.

 

[Johntheo5]

Even if your house temperature is say 25C now, then if everything is calling for heat, the rad output/demand will still be over 70% of the rad output with a room temp of 10C, coupled with the HW coil demand of ~ 6/8kw with a cold cylinder shoule be > 12kw?. My standard coil of 0.65m2 will emit 6/7kw with a stone cold cylinder but only ~ 1.5/2kw when the coil temperature is within 5C of the HW temperature.

 

[molgrips]

That does not seem correct, can you ensure all rads are calling for heat, may have to increase roomstats SP and open TRVs fully and repeat that test with no HW request.

With the temperature knob on the second pip, and all rads open but no HW, the burner turns off at 57C and back on at 38C. So that's 19C differential which means it's 8.5C either side of set point.

That's probably acceptable for the heating. The return temps stay low, around 48C, and the house is getting toasty. The only issue is not being able to heat the HW properly with the heating set up like that.

The flame modulation isn't working either. When it hits 57 ish the flame throttle back for a couple of minutes, then the burner shuts off. Every time, it always seems to be the same duration and only two levels - high, low and off. That makes no sense to me.

 

[Johntheo5]

The second pip is 40C SP so apparently cut out at SP+17C, cut in at 38C = SP-2C,. Does the temperature rise very rapidly to its cut out at 57C. in other words what is the burner on time and the burner off time (a cycle).
When it throttles back for a few minutes. does it maintain 57C flow temperature or does it rise still further?

Is the circ pump installed internally in the boiler or externally?, if external, post make/model/setting.

 

[molgrips]

Does the temperature rise very rapidly to its cut out at 57C

No, it takes a good while. It was probably on for 10 mins and off for 10 mins although I didn't time it. There might be short cycling protection giving minimum burn/off times of 10 mins, I suppose, which would produce these symptoms.

The pump is a Range Tribune, I've got the flow set to low which is 1450 I guess litres per minute.

 

[Johntheo5]

!0 mins on & 10 mins = 50% average output say 50% of 10kw, 5kw, a rad flow/return (while firing) of 57/48 will give a rad output of ~ 58% or their rated output and less on the burner off 1/2 cycle. the circulatin rate based on 10kw boiler output with a dT of 9C is almost 16LPM which seems fine.

 

[molgrips]

I think my pump has been installed the wrong way round.

Looking at the boiler manual the pipe on the right should be the outward flow and obviously hotter. But it's the colder of the two.

This would explain my symptoms as the boiler is trying to modulate the return flow!

And maybe also why the rads upstairs get hotter far quicker than those downstairs. And why there are always bubbles in the pump!

 

[Harry Bloomfield]

That is how I would expect it to work @Johntheo5. But the flow temps seem all over the place so either it's faulty or my thermocouples are mis-reading.

Are you adding some sort of insulation, over the top of the thermocouples? If you have none, then they will be measuring a mix of pipe temperature and air temperature around the pipes. A short piece of split pipe insulation is enough and will act as a clip too.

 

[Harry Bloomfield]

Looking at the boiler manual the pipe on the right should be the outward flow and obviously hotter. But it's the colder of the two.

If the return is hotter than the flow, when the boiler is actually burning [1], then either the boiler, or the pump have been installed the wrong way.

[1] Be aware, that you can get a reverse flow, due to convection currents, when the boiler is not actually burning.

 

[molgrips]

Looks like the pump. They've installed it with the writing the right way up but the flow arrow pointing down. It has two isolating valves either side and two big screw connectors. I can shut the valves, release the pressure via the bleed screw and turn it round, I don't even have to release the pressure in the system.

I've been taking the readings when the boiler is running. It would explain the weird readings in my original post. And hopefully I should save a bit of money!

 

[molgrips]

Update:

I flipped the pump around - easy enough except that I didn't have any replacement rubber washers. The old ones were quite knackered so I greased them up and refitted anyway, and they somehow haven't leaked! The temperature controls now work as expected - you can set the flow temp to the fourth mark on the dial which would be 60C, it goes up to 62C or so quickly, then throttles back and continues to run for a good long time to about 65C with a return temp of 53C or so, then it shuts off. This is exactly what I'd expect. Sadly there is no short cycling protection on this boiler so it can end up doing that. However I've adjusted the flow rates and the position of the wireless stat so it doesn't do it - the stat gets warm enough before the short cycling happens.

But then the problems started. I could hear some dripping sounds from within the boiler - odd. Then when looking for draughts under the cupboards I saw dripping white milky water. I put two and two together: With the flow running backwards the condensing heat exchanger was hot rather than cool so the boiler had never been condensing in the 15 years we've had it except for a minute or two on start-up. So these few drops of water and 14 years of accumulated crud (yes, we didn't get it serviced!) in the bottom of the combustion chamber had completely blocked the drain hole and s-trap with hardened black crud. Now that the boiler was condensing properly the inside had filled with water and made a nasty slurry from the boiler residue. Of course, I got that all over the kitchen, but I cleared up the trap. Then I tested the rest of the pipe by squirting water in, and none ran out - another blockage. Well eventually I checked under the sink and found a layer of food-based gunge covering the end of the boiler drain pipe - there had never been any fluid running through it.

A load of mess and pipe fittings later (I had to build a temporary drain into the sink so we could have heat whilst I fixed it) it is now working fine, and I get a delta T in the boiler of 20-25C instead of 15-20C, so it's apparently running 25-33% more efficiently! I have tried to increase the flow rate to bring dT down a bit in accordance with the spec but that means less heat is being dumped by the rads so I get higher return temps.

Pump is much quieter now in general (in fact inaudible unless I open the airing cupboard door) and the bubble traps are in the right place to catch the bubbles so it's clearly been designed to run this way round and the original fitter had just put the writing on the label the right way up without checking the flow arrow stamped on the back of the pump.

 

[fixitflav]

Update:

Well done for perseverance!