[OpenTRV-interest] Weather compensation

Simon Hobson linux at thehobsons.co.uk
Thu Dec 22 22:46:28 GMT 2016


Rob May <rob at themayfamily.me.uk> wrote:

> I may be mis-remembering, and may be showing my lack of knowledge, but I'm sure that I read somewhere that many condensing installations were not efficient as the return flow was always too high for them to actually do the condensing bit.

Very often correct, see below ...

Tim Small <tim at buttersideup.com> wrote:

> Whilst set point is often talked about, what isn't is the % of max
> output which the boiler is operating at (modulation) - it's usually just
> left up to the boiler to determine based on distance from set point and
> flow/return variation over time.  It does have quite a large effect on
> efficiency tho' (with some boilers more than others):
> 
> e.g.
> 
> http://hpac.com/site-files/hpac.com/files/archive/hpac.com/fastrack/f1.gif
> 
> http://www.diynot.com/diy/media/boiler-efficiency.27041/full
> 
> https://www.automaticheating.com.au/wp-content/uploads/2016/11/Figure-1.jpg
> 
> http://www.greenshootscontrols.net/?p=153
> 
> ... so I think it's probably worth considering too.

I think you need to read the graphs with a little care. Notice how there's a fairly steep knee at around 55˚C in some of them at lower power outputs which is less pronounced at higher outputs - I'm thinking this is indicative of an undersized heat exchanger so that at higher power levels, there's insufficient contact time for heat transfer.
Adding a flue gas heat recovery unit in the exhaust would deal with that - at the cost of more hardware.

Part of the problem is that boiler manufacturers are still making boilers for the mid 20th century ! Read the specs for any boiler, even the brand new designs, and the design requires sufficient flow for something like 20˚C delta at full output - with no allowance for reducing the flow. That's fundamentally incompatible with TRVs - hence my suggestion that the boiler manufacturers are 50 years behind the times in their designs.
For "sensible" operation with TRVs, the system needs a modulating pump and variable flow rate - at low demand levels, the flow rate would be quite low. Contact any manufacturer* and ask them about running the system in this mode and you'll get a resounding "no chance" from them. As a result, systems are designed as a massive compromise and have to incorporate a bypass - these days commonly an automatic bypass which is just a pressure relief valve between flow and return (after the pump) so that when the TRVs are shut, the bypass opens and lets the flow rate through the boiler stay up.
This means that the return temperature behaves in what at first glance is counter-intuitive - being a maximum when the TRVs are all shut and all the flow goes direct back to the boiler via the bypass. That's fine if the flow temp is below about 55˚ - I'm finding our system works at 45˚.

That's why I suggest that controlling the flow temp based on feedback from TRV actuator positions would be a good idea. Avoid flow via the bypass, and you minimise the return temp - and thus maximise the efficiency of heat transfer.
But there's another tradeoff - lower the flow temp too far, the TRVs will all open up more, the flow rate will rise, the return temp from the rads will rise a bit, but the transit time through the heat exchanger will reduce.

As I said in an earlier message, it's something that would be interesting to experiment with.

* Exception, I've read that Atmos have some models with a solid aluminium heat exchanger and claim that they can maintain control down to very low flow rates. But when I looked at a spec sheet, they still had this 20˚C delta-T design spec.


But, in answer to Rob above ...
A lot of older systems had small radiators designed to work at higher temperatures. A key part of this is that a non-condensing boiler has a **minimum** operating temperature - if it's run too cool then condensation forms and causes all sorts of problems (leaks into the electrics, rots the internals, ...). If you just replace the boiler with a condensing one, but don't update the rest of the system, then the boiler will spend a lot of time in non-condensing mode because of the temperature it's running at.



More information about the OpenTRV-interest mailing list