If you want to check the calculations yourself, there's an online heat-loss modelling tool, Heatpunk, which has a free tier where you can enter your house room dimensions, materials, windows/doors etc and it will give you a calculation of total and room heat loss, adequacy of existing radiators for different design flow temperatures etc.

@baroquebank

Heat loss calculations are based on imprecise (or incorrect) assumptions and so you should expect some spread.  The industry likes to represent them as gospel, but this is simply not the case.  They look scientific, but garbage in = garbage out.

You have two which are overall similar but different in detail.  That doesn't actually prove either is correct (I had 2 @ 16kW for a 7kW house) but does give some confidence particularly as heat geek appears to be pretty good at getting them about right.  Personally I would always advise to do a sense check against gas consumption.  If you have half hourly readings you can do a pretty robust check, but even if you only have annual consumption you can check that they are in the right ballpark.  In the latter case divide your annual consumption in kWh by 2000, and separately by 3000.  The house loss is almost certainly between the two figures, more likely nearer the latter than the former unless you allow it to cool a lot for a fair proportion of the day.  

I would also agree with @editor ,ask about the reasons and good explanations are given tend towards that installer.

Rooms share heat and you can anyway adjust, within reason, by balancing so I wouldn't worry too much about room level discrepancies between heat loss surveys.  I might however take a close look at the assumptions behind very large differences (say more than 50%) in case that reveals anything.  Don't get too hung up on percentages of the smaller numbers (IE lower loss rooms) though!

@temperature_gradient thanks for replying, but I'm not too worried about the heat loss calculations, they were near enough for me to be OK with them. It's the discrepancy in required radiator replacements that has bothered me.

@editor thanks for your response, which is very helpful. I do plan to have further discussion with the Heat Geek (whom I do feel more comfortable with) when he's back off holiday.
I believe the heat pump size discrepancy is an accident of where I fall. Octopus clearly feel the 6kW Cosy wouldn't be good enough (and I'd agree) so the next one is 9kW. Heat Geek clearly believe the 7kW Vaillant fits, which again feels OK to me.

Have either quote provided room by room heat loss calculations and radiator sizing and outputs at design temperature?

Assuming you are hapy with the overall heat loss figures, a room by room breakdown and how that heat loss will be met (or not) by each installed radiator will tell you exactly which radiators need upgrading. I would expect this level of information to be available if you request it. I understand not all installers will want to provide it to all customers as it may be overkill for more non-technical customers, but it will instantly tell you why each as recommended the radiator upgrades that they have.

Get the radiator sizing right, and you can run a fully open system with an even temperature throughout the house. Get it wrong and you will end up with rooms either too hot or too cold rooms and a less efficient system that is difficult to balance.

Posted by: @baroquebank

↑

@temperature_gradient thanks for replying, but I'm not too worried about the heat loss calculations, they were near enough for me to be OK with them. It's the discrepancy in required radiator replacements that has bothered me.

If you know the room heat losses for each of the rooms (have they both provided that, room by room basis?) and they're roughly the same, then presumably the difference must be in the estimated/calculated heat outputs of the existing radiators at the proposed flow temperatures.

Have they measured and modelled the existing radiators correctly (sizes, number of panels, convectors, or are they designed type rads?) Do you know their calculated/estimated heat outputs at the two flow temps?

Also, do you know the design room temperatures used - were they the same in both cases? Oh, and are they both using the same lowest outdoor air temperature?

I'm guessing they would only be proposing to replace radiators if the design heat-loss at the design temperature, is above their modelled heat output from the existing radiators.

Individual room heat losses matter just as much as total heat loss, indeed for most spreadsheet based heat loss calculations, the latter is just the sum of the former. At design time, the room heat losses determine the radiator sizes, and the total heat loss determines the heat pump capacity. 

It is perfectly possible to have two or more total heat loss calculations that are similar, but the underlying room heat losses are very different, and any or all of them may be wrong. 

Posted by: @baroquebank

↑

HeatGeek - we must upgrade 2 upstairs bedroom radiators, both in rooms where the calculation shows existing radiators exceed heat loss

Octopus - we must replace 2 downstairs radiators, both in rooms where I do feel current radiators are marginal. Indeed one of these, in the kitchen, Octopus are saying must be significantly upgraded (or I must provide an additional heat source) in order to qualify for the government grant.

To me, this is a bit of a red flag. If I have understood correctly, HeatGeek have said you need to downgrade two upstairs rads (because the 'existing rads exceed the heat loss') whereas Octopus have said you need to upgrade two downstairs rads. These are significantly different proposals.

These proposals can only sensibly come from satisfactory individual room heat losses set against the existing radiators outputs when running at your proposed heat pump settings. You need to ask both HeatGeek and Octopus to give you those numbers, and then compare the two, and see which (if indeed either!) makes better sense. If you can post the numbers here, we may also be able to comment one way or another.

If either installer cannot provide the the numbers, you know what to do... 

Posted by: @cathoderay

↑

To me, this is a bit of a red flag. If I have understood correctly, HeatGeek have said you need to downgrade two upstairs rads (because the 'existing rads exceed the heat loss') whereas Octopus have said you need to upgrade two downstairs rads. These are significantly different proposals.

I agree 100% with this.  I can't think of an engineering reason to reduce radiator size, just turn down the LSV.  One of mine is at least 50% over and it's not a problem to balance.

Posted by: @cathoderay

↑

It is perfectly possible to have two or more total heat loss calculations that are similar, but the underlying room heat losses are very different, and any or all of them may be wrong. 

They are almost certainly all wrong, this is not an exact science (well it is in principle, but we don't know everything we need to know to do the calculation). 

That's why I wouldn't personally get too hung up about differences provided they are within reason.  Certainly examine the calculations, but if they agree on a room by room basis to within 20% or thereabouts, you are doing well.

I too am not to bothered about the total heat loss estimates at this stage. What does bother me is that one installer wants to downgrade two rads (which I agree is an odd thing to do) while the other wants to upgrade two rads. This suggests to me the individual room heat loss assessments (and/or radiator outputs) are significantly different for at least for some rooms, maybe all rooms (and yet by some accident they both add up to similar totals). This room level assessment is what needs looking into, to understand why they are different, and perhaps shed some light on why one (or both) are wrong.

Posted by: @cathoderay

↑

I too am not to bothered about the total heat loss estimates at this stage. What does bother me is that one installer wants to downgrade two rads (which I agree is an odd thing to do) while the other wants to upgrade two rads. This suggests to me the individual room heat loss assessments (and/or radiator outputs) are significantly different for at least for some rooms, maybe all rooms (and yet by some accident they both add up to similar totals). This room level assessment is what needs looking into, to understand why they are different, and perhaps shed some light on why one (or both) are wrong.

 

A good analysis with which I agree