Flow and return temperatures for optimum efficiency

[johnny2007]

You don't want a flow temperature of 75C with any type of condensing boiler. You really only need the flow temperature to be high enough to keep you comfortable, because the lower the return temperature, the more it will condense, and the more efficient it will be. A flow of 55C, or even lower, is often recommended these days.

I'll try that, although the engineer who fitted the boiler said to keep it between 60 and 65 degrees.

 

[Harry Bloomfield]

Does this apply to combi boilers?
If so, do we measure temperatures at the radiator pipes (flow and return?) Or at the boiler?

It is the boiler which condenses, or not, dependent on temperature - so obviously you measure at the boiler flow and return pipes.

My boiler runs at a flow of 75C, but only when it is called upon to heat up the cylinder, and not much can be done about that. The rest of the time it is well into condensing temperatures.

 

[JonathanM]

I'll try that, although the engineer who fitted the boiler said to keep it between 60 and 65 degrees.

As regards efficiency, the science is pretty straight forward. The lower the return, the more efficient.

But the best temperature for your needs may be different in practice, and you may have to sacrifice some efficiency.

I don't know whether you have seen one of these type of graphs before?

 

[johnny2007]

As regards efficiency, the science is pretty straight forward. The lower the return, the more efficient.

But the best temperature for your needs may be different in practice, and you may have to sacrifice some efficiency.

I don't know whether you have seen one of these type of graphs before?

View attachment 288973

So, if I have 65 degrees on the flow as it's set by the engineer, and below 54 on the return I should be ok, correct?
Or should I aim for the 20 degree drop by better balancing the radiators lockshields?

 

[JonathanM]

So, if I have 65 degrees on the flow as it's set by the engineer, and below 54 on the return I should be ok, correct?
Or should I aim for the 20 degree drop by better balancing the radiators lockshields?

It would be useful to know what the current return temperature is. Some Logic combis have a built in menu where you can check the return temperature. Do you know if your boiler is either a Logic Max or a Logic+?

For the best efficiency, you want to get the return as far below 54 as possible. At 54 there will be almost no condensing. Better balancing might reduce the return temperature. But older radiator systems can't always achieve that. Do you know how old your radiators are?

You could also try reducing the flow temperature to see whether you can cope, because that will definitely reduce the return temperature. There have been a couple of posters, however, who had Ideal Logics where at lower flow temperatures they got a lot of very quick cycling.

 

[vulcancontinental]

So, if I have 65 degrees on the flow as it's set by the engineer, and below 54 on the return I should be ok, correct?
Or should I aim for the 20 degree drop by better balancing the radiators lockshields?

The latent heat of condensation is released when the vapour condenses and it only condenses when it is is contact with a surface below its dewpoint which at the excess air levels in the boiler starts at 54.

If the return is 54 and the flow 65 very little of the overall surface in the heat exchanger will be available below the flue product dewpoint.

The flue products themselves, being over 54 will not condense in the flueway, only upon contact with a cooler surface.

Ideally the flow would be as far below the dewpoint temperature as possible, not the return. That's why I emphasize the flue temperature not the return or flow; the flue temperature is the giveaway on how successful cooling of the flue gases has been or put another way, how good the heat exchanger is, something people seem to overlook.

When looking at a boiler the figures are test bench figures, look for the flue temperature at 80/60 and 50/30 in the mi's, that tells you how good the exchanger is. You have to get rid of the sensible before the latent becomes available. Not every manufacturer will list flue temperature.

 

[vulcancontinental]

It would be useful to know what the current return temperature is. Some Logic combis have a built in menu where you can check the return temperature. Do you know if your boiler is either a Logic Max or a Logic+?

For the best efficiency, you want to get the return as far below 54 as possible. At 54 there will be almost no condensing. Better balancing might reduce the return temperature. But older radiator systems can't always achieve that. Do you know how old your radiators are?

You could also try reducing the flow temperature to see whether you can cope, because that will definitely reduce the return temperature. There have been a couple of posters, however, who had Ideal Logics where at lower flow temperatures they got a lot of very quick cycling.

And with high modulating boilers, those over 1:10 the ignition fan speed (power input) on a relight is quite high and with the system demand so low the boiler immediately shuts down again, micro-firing. Installers have been told by manufacturers tech people this is normal operation. A lot depends on the heat exchanger.

Look for less than 8 degrees flue to return as an indicator, some are in excess of 20 degrees.

It's all dynamic, no perfect plateau.

Stuck at home yesterday I was experimenting with OpenTherm disconnected; solely weather comp. To get an idea of an operating cycle look at this video, couldn't be any fresher!

40kW combi, 30kW to heating servicing a heating load of around 8kW at -3 degrees. Boiler is 1:8 modulation. I watched the cycle five times with hardly any variation at all.

 

[JonathanM]

To get an idea of an operating cycle look at this video, couldn't be any fresher!

Thanks. It's so interesting to see everything that's going on in a cycle. I'm going to watch it again, and time the cycle. I understand now what you mentioned before about how the curve can be shifted up and down, without actually changing it. Also, I hadn't realised boilers would report the flue temperature.

I still have the FS Baxi 80,000 BTU, so my cycles are one and a half minutes on and four minutes off.

Stuck at home yesterday I was experimenting with OpenTherm disconnected; solely weather comp

I hadn't realised you could run OT and WC together. I'd assumed they would conflict in some way?

The phrase "sensible heat" makes me think of a Monty Python sketch.

 

[Agile]

I could add weather comp to any boiler with modulation.

Just use a photocell across the boiler's NTC temp sensor.

Then feed the photocell from an LED.

 

[Johntheo5]

The latent heat of condensation is released when the vapour condenses and it only condenses when it is is contact with a surface below its dewpoint which at the excess air levels in the boiler starts at 54.

If the return is 54 and the flow 65 very little of the overall surface in the heat exchanger will be available below the flue product dewpoint.

The flue products themselves, being over 54 will not condense in the flueway, only upon contact with a cooler surface.

Ideally the flow would be as far below the dewpoint temperature as possible, not the return. That's why I emphasize the flue temperature not the return or flow; the flue temperature is the giveaway on how successful cooling of the flue gases has been or put another way, how good the heat exchanger is, something people seem to overlook.

When looking at a boiler the figures are test bench figures, look for the flue temperature at 80/60 and 50/30 in the mi's, that tells you how good the exchanger is. You have to get rid of the sensible before the latent becomes available. Not every manufacturer will list flue temperature.

All very interesting, if you take flow/return of 65/54C above then the rad heat output, assuming a room temp of 20C will be 73.6% of its rated output, but/and you will get exactly the same output of 73.6% if you increase the flow temperature to 80c and reduce the return temperature to 39C, I would think that the condensing effect would be far? greater with the latter.
I run my boiler at 70C and very regularly see boiler return temps of 40/45C due to the TRVs throttling down.
I would agree that reducing flow temps in general is "good" as long as you use a very small dT (high circulation flow) to achieve the highest rad outputs with the lowest flow temps, heat pumps like this as it results in lower compressor power and higher COPs.

 

[vulcancontinental]

Thanks. It's so interesting to see everything that's going on in a cycle. I'm going to watch it again, and time the cycle. I understand now what you mentioned before about how the curve can be shifted up and down, without actually changing it. Also, I hadn't realised boilers would report the flue temperature.

I still have the FS Baxi 80,000 BTU, so my cycles are one and a half minutes on and four minutes off.



I hadn't realised you could run OT and WC together. I'd assumed they would conflict in some way?

The phrase "sensible heat" makes me think of a Monty Python sketch.

I think it ran out at just over 7 minutes.

Well not the curve exactly but the target nominal temperature

Over the evening as the temperature dropped so did the internal temperature so I adjusted the curve to 2 which was too high so this morning dropped it to 1.8.

WC and OT together are fine if they're designed to work together otherwise I don't know what would be the outcome.

Next I'm reconnecting the OT room control but disabling it's room input so it becomes a remote boiler/time control. The reason is I notice the OT curve is curved, as it should be, but the boiler curve is linear which probably will make not a jot of difference but I have to see.

At home again tomorrow so more experiments.

1 1/2 minute burn times isn't great and you'll be losing a few percentage points but not more than 3 or 5 I'd guess. Air temp is unaffected by downtime it's just the warmth in the flue products on post and pre-purge. 3 minutes and it's 1-1.5%.

You've looked at parallel coil heat exchangers, serial coil heat exchangers some with retarder some without, large lumps of aluminium cast exchangers, the same with tubes running through them, cross tubular exchangers, some with twice as many tubes as another and larger yet the same efficiencies. Little wonder flue temperatures are not always made available.

Making the boiler condense is making the best of what the design is capable of, making the design the best it can be is another matter altogethe4r

 

[vulcancontinental]

All very interesting, if you take flow/return of 65/54C above then the rad heat output, assuming a room temp of 20C will be 73.6% of its rated output, but/and you will get exactly the same output of 73.6% if you increase the flow temperature to 80c and reduce the return temperature to 39C, I would think that the condensing effect would be far? greater with the latter.
I run my boiler at 70C and very regularly see boiler return temps of 40/45C due to the TRVs throttling down.
I would agree that reducing flow temps in general is "good" as long as you use a very small dT (high circulation flow) to achieve the highest rad outputs with the lowest flow temps, heat pumps like this as it results in lower compressor power and higher COPs.

I associate restricted flow (TRV) with narrow delta T not wider. If theres a bypass or one rad open even if boiler modulates heat returns faster without being given to the property.

To acheive 25 to 30 delta T is difficult in domestic properties if the boiler has been firing a while.

A surface temperature of 80 degrees will reduce condensing and narrow flue product to exchanger delta T.

Whether more heat would end up in system water would depend on flow rates.

 

[johnny2007]

It would be useful to know what the current return temperature is. Some Logic combis have a built in menu where you can check the return temperature. Do you know if your boiler is either a Logic Max or a Logic+?

For the best efficiency, you want to get the return as far below 54 as possible. At 54 there will be almost no condensing. Better balancing might reduce the return temperature. But older radiator systems can't always achieve that. Do you know how old your radiators are?

You could also try reducing the flow temperature to see whether you can cope, because that will definitely reduce the return temperature. There have been a couple of posters, however, who had Ideal Logics where at lower flow temperatures they got a lot of very quick cycling.

Ok, the return temperature i recorded on the return pipe on the boiler is 51 degrees, so 14 less than flow.
I have a brand new Worcester bosch 32 cdi compact and all new radiators.
I don't think my boiler has a menu to check all of the data in the video from vulcancontinental.

 

[old duffer]

It turns it off

so it does not modulate the boiler then

 

[old duffer]

I associate restricted flow (TRV) with narrow delta T not wider. If theres a bypass or one rad open even if boiler modulates heat returns faster without being given to the property.

To acheive 25 to 30 delta T is difficult in domestic properties if the boiler has been firing a while.

A surface temperature of 80 degrees will reduce condensing and narrow flue product to exchanger delta T.

Whether more heat would end up in system water would depend on flow rates.

water behaviour is predictable to a great extent but more often than not the properties of flow fluctuate outside of the "expected". Speaking with an actual qualified engineer once (MEng) he said that "water flows from one point to another. The how and why is the bit we can't determine with any accuracy".
I experienced a situation across a radiator that I could not induce flow. Measuring flow rates I recorded 1.5bar at inlet and outlet. 4 rads from the same manifold worked just fine yet this one would not work. Was like this for weeks and then one day it started to work. Anyone with extensive experience of complex heating system designs will have experienced these nuances
The properties of water are very similar to electrical current and that is a subject that is still not taught correctly today with the prevailing belief that electricity flows from one point to another.
"Heat in the system" is a very simple mathematical calculation of which flow rate is but one variable