Real world testing example of Condensing Boiler in older home

[John.g]

It means shutting in the rad return valves (a form of balancing) or to get a quick and easy feel of what can be achieved, just throttle the pump suction valve or preferably a common return valve from "all" the rads.

 

[Chuck]

What does "throttle the returns" refer to please?

Increase their impedance, i.e. restrict them by partially closing their lockshield valves.

 

[JC22024]

It means shutting in the rad return valves (a form of balancing) or to get a quick and easy feel of what can be achieved, just throttle the pump suction valve or preferably a common return valve from "all" the rads.

When downstairs is on again tomorrow I could partially close one of the ball valves on the mag filter at return line would that have same effect? Just to test temporarily. Or better to turn down the gate valve on the pump inlet instead?

I do think the boiler is oversized really. I am interested to take all readings including the hour counter measurements and compare before and after nozzle downsize.

Aka to use 2no hour counters. 0.01 hr accuracy. One connected to the circulator pump which represents the heating flow / duty time and the second to a 240v circuit on the burner. May try to measure actual nozzle firing time VS just when the boiler stat is calling.

When considering boiler cycling am I correct in thinking this is the total number of times the boiler turns on and off inside an hour of average running? Or how is boiler Cycling assessed in your opinions? For example at present the boiler must be firing up again about 16 - 20 times per hour. If a nozzle downsize reduced that to 10 or less then is that an indicator of a more suitable / efficient boiler size?

 

[John.g]

This is a screenshot of the old spreadsheet that I do the calcs on and contains more info.

You don't have permission to view attachments. Attachments are hidden.

 

[Chuck]

When considering boiler cycling am I correct in thinking this is the total number of times the boiler turns on and off inside an hour of average running? Or how is boiler Cycling assessed in your opinions? For example at present the boiler must be firing up again about 16 times per hour. If a nozzle downsize reduced that to 10 or less then is that an indicator of a more suitable / efficient boiler size?

Room thermostats for use with gas boilers are typically limited to a maximum of 6 on-off cycles per hour, for oil boilers this should be changed to 3 per hour. The boilers themselves will usually have their own short-cycle prevention controls built-in.

Obviously, the manufacturers' instructions trump these 'rule of thumb' values.

 

[JC22024]

This is a screenshot of the old spreadsheet that I do the calcs on and contains more info.

You don't have permission to view attachments. Attachments are hidden.

So the Delta temperature makes perfect sense there. It is a great influence on the heat output of the radiator. The LPM flow is calculated by the delta temp and rad size(?)

 

[John.g]

When downstairs is on again tomorrow I could partially close one of the ball valves on the mag filter at return line would that have same effect? Just to test temporarily. Or better to turn down the gate valve on the pump inlet instead?

I do think the boiler is oversized really. I am interested to take all readings including the hour counter measurements and compare before and after nozzle downsize.

Aka to use 2no hour counters. 0.01 hr accuracy. One connected to the circulator pump which represents the heating flow / duty time and the second to a 240v circuit on the burner. May try to measure actual nozzle firing time VS just when the boiler stat is calling.

When considering boiler cycling am I correct in thinking this is the total number of times the boiler turns on and off inside an hour of average running? Or how is boiler Cycling assessed in your opinions? For example at present the boiler must be firing up again about 16 - 20 times per hour. If a nozzle downsize reduced that to 10 or less then is that an indicator of a more suitable / efficient boiler size?

Use the Ball Valve.

Boiler Cycling: "Burner firing time 1min 8 secs. Off time 2 min 10 secs during this time." the boiler fired for 68secs and was off for 130sec = cycle time of 198 secs, boiler output 26*68/198, 8.9kw. cycles/hour 3600/198, 18. Oil fired boilers couldn't care less how often they cycle because this is the only way they can control their output since they cannot modulate.

A smaller nozzle may help but not a lot IMO.

 

[John.g]

So the Delta temperature makes perfect sense there. It is a great influence on the heat output of the radiator. The LPM flow is calculated by the delta temp and rad size(?)

That's it, the rad size is in Kw, I just used 1kw, the flowrates are then pro rata for other outputs.

Your pump is definitely oversized, a normal 7M pump would have done the job IMO, the old pump was knackered I'd say.

The ads are fabulous, did you know that ear wax can make you deaf?

 

[JC22024]

That's it, the rad size is in Kw, I just used 1kw, the flowrates are then pro rata for other outputs.

Your pump is definitely oversized, a normal 7M pump would have done the job IMO, the old pump was knackered I'd say.

The ads are fabulous, did you know that ear wax can make you deaf?

Yes original grundfos 15 50 130 pump must have also been faulty. But house never heated properly until now so do think it was undersized and agree new pump is perhaps flowing too fast even on speed 1.

May look at changing to a 180mm smart type pump again by grundfos or similar. But first will ensure rads are putting out min 11c drop each before going back to pump. Aka balance rads and slow the flow in them if they are only showing a few c drop.

This looks good

Dual Channel K Type Digital Thermocouple Thermometer 6802 II, 2 Pipe Clamp, HVAC | eBay

Selection between Celsius, Fahrenheit and Kelvin. Thermocouple sensor specifications Clamp jaw opening: 1-3/8" (3.5cm). 2 x K type Thermocouple sensors/ wires. Pipe clamp Specifications Dual temperature monitoring T1/T2.

www.ebay.co.uk

Yes the adds here are terrible I had to install an add blocker!!

 

[John.g]

You can have very small dTs by decreasing the flow temp and increasing the flow rate, this is why UFH can operate with very low temps, with radiators the same can be achieved but would need to be massively oversized X 2.5 in the example below to give similar output as a 50 deg rad.

You don't have permission to view attachments. Attachments are hidden.

 

[ShaunCorbs]

I would say the pump is sized correctly you have a few too many rads for a 15-60 balance the rads first and come back

 

[John.g]

If the system requires a 5M head which the pump is set to now then a 8M replacement necessary as the UPS3 will only service 8kw of rads at dt 6C, 13kw @ dT 10C & 19kw @ dT 15C. all at a 5M head .

Carrying out some quite interesting "throttled" tests with my 20kw Firebird + Smart pump which can be set up to give the exact flow just by looking at the power/pump curves, will post later.

 

[John.g]

Did some testing this morning, I set the pump to constant speed setting of 5M to correspond to your setting, I then shut the pump discharge valve and confirmed that the pump power was 21W at no flow, I reopened pump discharge valve and throttled the suction valve to give me 24W which gives a flowrate of 5LPM (at 5M). The (20kw) boiler cycled 6 min on and 6 min off with measured temps exactly as below, the return was 44/42C, avg 43C and the boiler flow 77/69C, avg 73C.

Carried out these tests on rads with a total output of 15kw with all the TRVs full open.
Will do some more later on this evening with all systems on (apart from HW cylinder) and fully up to temperature with suction valve still throttled in.

You don't have permission to view attachments. Attachments are hidden.

 

[Ric2013]

Interesting thread. This is the kind of investigative spirit we need to have if we're really to make meaningful improvements to heating systems.

Re the infra-red thermometer, I find they work really well. However, they are usually geared to high-emissivity surfaces (it should normally explain that in the destructions). As such, any reasonably shiny metallic surface can't really be read accurately as it is a low-emissivity surface. However, there is an easy fix in that a piece of masking tape on the pipe can be used as a reliable test point.

 

[JC22024]

Yes I have used the tape trick before on IR thermometers but it's not working this time unfortunately. The readings with IR gun are all over the place so no good.

I ordered more of the little digital thermometers on ebay as they are less than £3.50 each. And for accurate radiator balancing for dummies I ordered one of these:

You don't have permission to view attachments. Attachments are hidden.

It can display either temperature reading from the clamps and / or the actual temperature drop. I would like to make sure all rads and precisely balanced so this should do it.

Awaiting the other stuff too before I can re run the tests before and after re jetting the boiler.

Interesting that John's Firebird has a 6 minute cycle and mine is so much shorter at just over 1 min. I wonder is this a difference in thermostats? Or the heating demand? Most likely the former.

When heating is on later I will turn down the ball valve on return and see if I can reduce the return temp to 20 for the moment and / or at what point of return temp does the condensing start again. Would like to know that information at least in my oil boiler.

In a way its kind of a shame oil burners have "dumb" controls. They can't modulate down like their gas counterparts and on top of that it is difficult to find a weather compensator compatible with a standard system anymore. Correct me if I'm wrong but you could buy generic weather compensators 10 or 15 years ago but the advances in gas boiler controls must have done away with the generic "fits all" compensators.

 

[JC22024]

Interesting thread. This is the kind of investigative spirit we need to have if we're really to make meaningful improvements to heating systems.

Re the infra-red thermometer, I find they work really well. However, they are usually geared to high-emissivity surfaces (it should normally explain that in the destructions). As such, any reasonably shiny metallic surface can't really be read accurately as it is a low-emissivity surface. However, there is an easy fix in that a piece of masking tape on the pipe can be used as a reliable test point.

Thanks for the comments, I wasn't sure how this thread might be perceived but I hear and read different opinions on condensing boilers and I for one am genuinely curious on how efficient they are and what can be achieved for optimum running

 

[JC22024]

Sorry to also ask a stupid question here but just so I understand correctly:

Radiator flow and balancing to obtain the desired 11c drop:

If the flow is too low then the radiator is too cold may not even heat at all. That's simple.

If the flow is too high what would happen? Eg locksheild fully open? With too much flow radiator would be very hot, putting put alot of heat but may have only 3c drop say.

In that case then are we saying that by turning down the lockshield / reducing flow the desired 11c drop is tunable? And in this condition the radiator/ system should have optimum heat output to room? With the primary aim being to slow the hot water so it transfers its heat to the room.

 

[ShaunCorbs]

If it’s low flow you will struggle to balance it properly as chances are you increase the resistance the water might stop to that rad but that’s normally cause by two things blockage or bad / incorrectly sized pump

And yes spot on see with them fully open the water isn’t given the chance to full transfer it’s heat eg 10dc difference so you need to slow the outgoing water down to give you more temp drop you will get to a point where closing more of the lockshield won’t increase the drop

 

[JC22024]

This is getting interesting- downstairs zone now. Restricted the return so as the boiler is supplying 70 and return at or below 45c.

Burn time 2 Min 05 secs. Off time 2 min 30 secs.

Watching now to see how low the return stabilises to and if any condensate starts to be produced. As a barometer for when condensing will occur on the return temp.

You don't have permission to view attachments. Attachments are hidden.

Edit- down to 36.5c return temp now and still zero condensing action! Oil condensing boilers only condense on first warmup it appears - in real world conditions this new Firebird boiler is NOT condensing even at 36c.

So consider them as high efficiency only in this case.

 

[John.g]

I'm not sure re Firebird condensate trap but some traps, to avoid the waste pipe freezing up, fill up the trap with condensate and then blows this down and refills again so on the off chance that this is occuring keep a eye on it but I wouldn't hold my breath as firebird supplied a plastic trap originally which used to melt.

 

[JC22024]

I'm not sure re Firebird condensate trap but some traps, to avoid the waste pipe freezing up, fill up the trap with condensate and then blows this down and refills again so on the off chance that this is occuring keep a eye on it but I wouldn't hold my breath as firebird supplied a plastic trap originally which used to melt.

I can edit my post above but I suppose mybtwst conditions there were incorrect. Restricting the return means the boiler isn't actually firing anywhere near what it would normally be. The radiator temps dropped inside the house.

So when I just reopened the ball valve there whilst return temp was below 40c approx and the burner firing there was some condensate being produced albeit only about 50ml. I suppose I would need a Lower c stat in order to conduct that test properly.

I am collecting all condensate during the tests with a very advanced setup- a coke bottle with a hole cut in the side

You don't have permission to view attachments. Attachments are hidden.

 

[John.g]

Its the wrong container, you require a still water bottle.

 

[JC22024]

I am still also exploring boiler compensators. Or Boiler Energy Managers as they appear to the called. This one is also interesting:

Deeter Boiler Control

Prevent hot water being reheated unnecessarily, cutting your fuel bill by 20%. Product code: GVEdeeter1 24 - 36 month payback 'Fit and forget' 3 year warranty

www.greenvisionenergy.co.uk

I will email them and ask if all required parts are included in the price. And for a wiring diagram to see if it's compatible with my controls. This would be further down the line not in the next few weeks

 

[JC22024]

I spotted a great deal on one of these and bought it last night

Energy Management Solutions | Vector Group

You don't have permission to view attachments. Attachments are hidden.

It's brand new unused. I believe my boiler is switching on and off too much. I think this device is aimed to reduce boiler cycling by learning the boiler demand and firing cycles and reducing the latter.

This is a boiler energy manager not a Weather Compensator but I suppose you have have both if you wanted. I think it goes between the boiler stat and burner.

In any case this would only go on once I get the system balanced and boiler jetting complete. Then the simple meters will show what savings it is bringing. Reading up on it they appear to offer savings of 15 - 30% on boiler firing time.

They appear to guard how exactly it operates so I would imagine I there will be no instructions supplied and I'll have to study it and see what inputs it requires.

 

[John.g]

I am going to carry out a test on just one rad soon because as you say, the only sort of fit is if the flow temperature falls as well, see below. I'm quite happy that I had a flow rate of 5LPM both by throttling the 5M pump setting to replicate yours and then at a pump head of ~ 3m unthrottled to give me the same 5LPM. As the boiler output was fairly constant (cycling) at 45 to 50% then the only scenarios that sort of fit are the last two below, ie a flow temp of 64C & return of 35C for 50% output and 45% output at 60C/34C, I didn't see my return going below 42/43C, the only explanation I can give for a apparent low flow temp is that a oil fired boiler acts as a 20/25 litre buffer tank and with the burner off perhaps cold return water flows from the flow side until the burner "catches up" and reheats the buffer??.

You don't have permission to view attachments. Attachments are hidden.

 

[JC22024]

I am going to carry out a test on just one rad soon because as you say, the only sort of fit is if the flow temperature falls as well, see below. I'm quite happy that I had a flow rate of 5LPM both by throttling the 5M pump setting to replicate yours and then at a pump head of ~ 3m unthrottled to give me the same 5LPM. As the boiler output was fairly constant (cycling) at 45 to 50% then the only scenarios that sort of fit are the last two below, ie a flow temp of 64C & return of 35C for 50% output and 45% output at 60C/34C, I didn't see my return going below 42/43C, the only explanation I can give for a apparent low flow temp is that a oil fired boiler acts as a 20/25 litre buffer tank and with the burner off perhaps cold return water flows from the flow side until the burner "catches up" and reheats the buffer??.

You don't have permission to view attachments. Attachments are hidden.

Are you saying that on your radiator readings the return pipe was 34 / 36c but at the boiler pipe you are measuring no less than 42/43c?

I suppose the return temperature will be a mix of all of the radiators as they flow back to the boiler. IE the first rad in ciruit will likely have the hottest flow and return temp - so maybe those rads you measured are further into the house?

The thing that bugs me with my Firebird is that it is trying too hard to maintain the 70c stat. It would be better if there was a lag of sorts there so it drops back a few C first then re-starts. Which would make it run longer when it does and less cycling. This other BEM mentioned in my other post above should address that cycling issue hopefully. I will need to open it up when it arrives and see what inputs it requires. I would imagine pipe stats will be needed on the boiler and it will connect between the boiler stat and burner, therefore learing the burn times and heating cycles / demand and being smarter than the stat to better control the burner. I think so anyway but time will tell on that one.

 

[John.g]

What I am saying is that the (theoretically) hottest return with a 72C flow temp is 37C, with perfect distribution through all the (6 rads), the lower the flow through any of the rads will result in a still lower return temp from that rad or rads which shuld result in a still lower boiler return temp.
I'm beginning to wonder if you think about TRVs and how they operate, they just keep throttling the flow until they get their set temperature but looking at the calcs below and your boiler/rad required output of 8.9kw based on a on/off times of 68secs/130secs then the TRV would have to shut off completely for some periods to allow the whole rad to cool down in one instance to allow a average flow temp of 60C as the return temp cannot be lower than the sensing/room temperature say a minimum of 24C, see below. The only way to achieve that measured return temp of ~ 41C is to have a average flow temp of 43C so perhaps a TRV is just opening/closing to achieve this and maybe this where evohome and others got their ideas from, where they use constant on/off boiler firing times to control the mean rad temperature necessary for any particular heat demand?.. The only other way is to have the flow temperature set to 43C with a very high flow rate, this, basically is how outside temperature compensation works I would think, also UFH.

That's why I a going to carry out a few tests soon on one rad and then maybe two/three to see what is actually going on. I can also use the TRVs as all my rads have them installed, all working perfectly.,

You don't have permission to view attachments. Attachments are hidden.

 

[John.g]

Carried out the TRV test on one X 1.7kw rad and as expected, the TRV was just copening/closing but keeping fairly tight control of the mean rad temp at ~ 42/44C which gave the required rad output of 0.6kw to maintain a room temperature of 20/22C. The boiler had very long cycling times of around 2.5 mins on and 35 minutes or more off. I then progressively introduced the remaining rads (all TRVd) and their mean rad temps reflected the outputs required, the combined return temp didn't vary hugely ~ 39/45C.

 

[JC22024]

Carried out the TRV test on one X 1.7kw rad and as expected, the TRV was just copening/closing but keeping fairly tight control of the mean rad temp at ~ 42/44C which gave the required rad output of 0.6kw to maintain a room temperature of 20/22C. The boiler had very long cycling times of around 2.5 mins on and 35 minutes or more off. I then progressively introduced the remaining rads (all TRVd) and their mean rad temps reflected the outputs required, the combined return temp didn't vary hugely ~ 39/45C.

That's interesting on the off time of your boiler too- what return temp drop do you get before it fires up again? Eg - 4.5c drop before it re-fires

 

[John.g]

Its very difficult to get a good reading in the heat pac due to restricted space but initially it actually seems to rise a little but I suspect this is because the boiler flow pipe turns down immediately where as the boiler return goes out the side of the boiler and then very gradually rises so I think the boiler contents are inclined to "circulate" back through the return. The cycling time matches ~ the rad heat demand but its not that simple as the boiler once it cuts back in when the stat reaches ~ 60/65C has to reheat say 20/25 litres of boiler water from ~ 40c to its cut out of 70C so requires ~ 0.9kwh which takes ~ 2.7 minutes from a 20kw boiler. In general, once other rads are brought back on line then the % firing time does reflect the heat demand.

 

[JC22024]

Its very difficult to get a good reading in the heat pac due to restricted space but initially it actually seems to rise a little but I suspect this is because the boiler flow pipe turns down immediately where as the boiler return goes out the side of the boiler and then very gradually rises so I think the boiler contents are inclined to "circulate" back through the return. The cycling time matches ~ the rad heat demand but its not that simple as the boiler once it cuts back in when the stat reaches ~ 60/65C has to reheat say 20/25 litres of boiler water from ~ 40c to its cut out of 70C so requires ~ 0.9kwh which takes ~ 2.7 minutes from a 20kw boiler. In general, once other rads are brought back on line then the % firing time does reflect the heat demand.

Thanks for that! I just can't help thinking my boiler Is cycling alot and probably wasting alot of heat up the flue whilst doing so. It's following the stat too closely. Hence why I believe a Boiler Energy Manager will greatly help my system efficiency. They monitor the flow and return temps and calculate the load on the system, controlling the burner.

I can't find any information on that Vector Group item I purchased from fleabay. From what I read they cost about £1600 - £2k from the company but someone on ebay with good feedback etc was selling a new unused one for £30 Inc postage

I'll open it when it arrives and see if anything is labeled and what sensors I would need. But may not be able to use it myself if nothing is labeled inside and its only meant to be fitted by them so looking at other BEM systems for Domestic boilers at present.

Plan is:
Balance all rads precisely and see if I can get return temps down

Meter the burner etc at 26kw setup and run for a few days.

Then downsize nozzle and re-run the same test.

Then fit a BEM (and / or try the Vector super unit) and re test to see what reduction in consumption they may offer if any.

 

[JC22024]

Tonight on upstairs zone only,

72 seconds on time
129 off time
Flow temp goes from 68.3 - 62.3.
Return temp seems to stay around 60c odd.

You don't have permission to view attachments. Attachments are hidden.

 

[John.g]

Based on cycle time, boiler output is 9.3kw so below might show why 60C return is valid but assumes ~ 11kw of upstairs rads.

You don't have permission to view attachments. Attachments are hidden.

 

[JC22024]

Today I got the twin digital thermometers in the post. Which I wanted for accurate easy comparison of pipe temps.

You don't have permission to view attachments. Attachments are hidden.

The first few I checked only had a few C drop.

I tried to guess the order of the rads and it wasn't working out. The first few would either die or have only a couple c difference. Couldn't get them to rise above that.

I did have luck adjusting one though which proves it is possible

You don't have permission to view attachments. Attachments are hidden.

So I should carry out the proper procedure. Open all rads, restart check order they heat. Cool again then start balancing in order from start. Will try to do that this Saturday all being well it will likely take ages!!

But I want to get it as balanced as possible. I'll aim for 9 or above I think the 11 or 12 may be out of reach but I'll try. The trick is to watch the flow temperature 🌡 so it doesn't start to drop whilst giving time for the difference to show. Aka easy to go too far and be on a false readout. It'll take me a while to get used to. I hope it's not working at present because I'm not following proper procedure!

The big kitchen rad in particular near boiler seems to want locksheild almost full open before it will heat up. Then drops dead if you turn below that. So some may not want to balance at all it is what it is.

 

[John.g]

What are the dimensions of the kitchen rad and is it a double or single?.

 

[ShaunCorbs]

Try with the pump speed on two instead of one

 

[JC22024]

What are the dimensions of the kitchen rad and is it a double or single?.

The kitchen rad is a double myson 160cm wide x 50cm high panel. The pipe configuration isn't ideal on this one as it was a replacement bigger rad for an older one. Need to sink into floor still, may be able to make good in future:

You don't have permission to view attachments. Attachments are hidden.

Try with the pump speed on two instead of one

Thanks I'll try that on Saturday. I'm thinking that since I'm not doing it correctly in order and all other locksheilds open that's not helping. So going by proper process maybe it will balance better.

Regarding the balancing just leave boiler stat at 70 or put up to 80 for the balancing procedure? And leave TRV heads on but full open / max temp?

 

[JC22024]

Also it appears some of the TRVs are installed on the return side. This might make obtaining the desired 11c harder on those particular rads?

They are modern bi directional TRVs but still mixed up.

If thought to be a problem I could drain down and rectify that throughout house.

 

[ShaunCorbs]

Tbh doesn’t really matter temp as that’s not a limiting factor

And tbh if your trying to balance it needs to be on the return as your throttling flow else

 

[JC22024]

Tbh doesn’t really matter temp as that’s not a limiting factor

And tbh if your trying to balance it needs to be on the return as your throttling flow else

OK so yes boiler temp doesn't really matter.

Do you mean the locksheilds should be on the return pipe yes? If so I can rectify that throughout house where it occurs before I attempt a balance.