Interesting couple of days with the new heat pump.

[MikeSharp01]

I started playing with the new heat pump on Sunday, first day I had since commissioning. Thought it would be fun to run it and heat up the slab so just set it going, I had not done any tuning or fiddling about so this is the out of the box operation and in the context of a very cold slab! So we are setting out to raise 70T of concrete 8 or so degrees. (Spoiler alert we still are not there yet.) Here is the first day as observed by the EMON heat meter system.

 

 

Clearly a number of issues, two main ones - of course the experts I hope will see more. Firstly the delta T is huge 11-12 deg C for the most part, and really does not improve across the day. Secondly the flow rate is quite low - which may explain the high delta. So I did fiddle with the flow meters at lunch time but put them back to where they were as 5l/m was the design flow - rightly or wrongly and it didn't improve the delta T massively although, as you can see it did close to between 10-11. I checked the pump settings and it has a few of them PUMPf. PWM.OUT PE and PUMPf.PWM.IN.PE the former was at 5% and the latter at 15%. (PWM being Pulse Width Modulation and 50% being the duty cycle of the pulse so the motor could be wound up to 100% from there if needed. The room temperature is climbing and carries on doing so after turn off as the slab continues to give up heat to the space. (15.2 at turn off tops out at 16). That was Sunday, I turned it off at 23:00, remotely via the Huilian app, because I was worried about over heating the slab. COP across the day 4.3. I had the set point at 28 degrees, but as we shall see that was a very daft idea as it goes way beyond that, or so I thought!

 

So by the end of Sunday I knew I had a flow problem but was not sure where that was coming from given that the pump appeared to be running at 5% PWM, the flow meters were all, just the three of them, showing 5l/m flows and the calibrated heat meter flow was showing just around 9l/m actually flowing. Something wrong somewhere. So it could be a flow restriction or a pump setting or something I was not understanding.

 

Monday I just started it up again but this time I played with things a lot more in the AM. I spoke to the Heat Pump manufacturer just really confirmed what I already knew. I fitted a slab temperature sensor to give me some comfort I was not going to crack the slab, I had fitted the pockets at the slab pour so now I just stuffed the DS18B20 down the pipe 10m or so and connected to my data logger. (Turns out I could have connected it to the EMON heat meter system had I wished) So I now knew what my slab temperature was mid slab. (I could get the surface temperature via the IR camera.)

 

As you can see the playing did not reveal much I cleaned the filter, tiny amount of clag in it, replaced it no improvement in flow and I adjusted the set point down to try and control the max temperature, as you can see it had no effect - I was obviously not controlling the set point I thought I was. The pulses you can see in the AM are me playing, observing, stopping, making an adjustment and restarting - nothing worked so I set everything back as it was. I looked at the Carell controller settings didn't find much. My setpoint was there but was not affecting anything. The Huilian App has loads more settings, you just have to find the one you want, or think you might want!  Lunch time and I thought I would just work through all the settings on the Huilian app and see what was there. I found one which set either 'control on out' or 'control on in' it was set to 'in' I thought that cannot be right for my setup, I suppose you might do that in industrial situations but surely not here, I need to control on flow (out) as I want to limit the temperature of the water entering the slab, don't I? - so I toggled it. Now it looked like I could limit the flow temperature but as a tweaked that down all the happened was that the heat pump shut down hence the pulses after lunch. So I could control the upper limit but not close the delta T gap as if I tried it tripped.

 

At about 15:00 I tried another trick asked the machine to heat the hot water. Let's see what 28mm pipe all the way did to the flow rate. You can see the result below:

 

19l/m so no restriction in that pat of the system, if that was the problem. So I had a cup of coffee at 3:15 (how very civilised) and took stock over the last M&S chocolate digestive! I thought, OK, the flow must be restricted or the pump settings are wrong. Let's remove all restriction in the UFH system by opening the flow meters to full bore. This did have an effect on the flow rate immediately. I could just about see the outer two flow meters in the bottom of the sight glass and the middle one is maxed out at 5l/m. We now have 12.2 l/m flow and control of the setpoint so the last few runs were done tweaking down the setpoint and I got to a delta T of 6.6 deg C which I thought was OK to head home with. Tuesday (today) could wait but there were still some thoughts to be had.

 

We cannot open the flow any more, but there may be a restriction in the central loop, this is the longest at 185m, as I cannot get the flow meter float into the bottom of the glass, possibly air in there somewhere. The pump looks like it is just ticking over so I needed to know more about the pump and its control but my search on Monday had not revealed any setting for the pumps 'aggression' level! SO I decide that first thing tomorrow I would fit the pressure transducer to the UFH flow line and see just how hard the pump was actually working.

 

Today dawned another day and I had some proper work to do for chunks of it so I left the heat pump off after fitting the pressure transducer. At Midday I started it all going and you can see the results below. First things first the pressure transducer revealed that the pump pressure was about 62  kPa (9 psi ). The pump has a max output of about 88 kPa (12.8psi) so there is some headroom in there to increase output somewhat if it wanted to, or could be forced to. 

 

 

So here, above, are the runs up to around 15:00 from 12:00 today. Looks like its making sense 2 defrosts and one 'breather' maybe not a short cycle because the pump stays running. But at the end you can see what looks like something else! here it is:

 

Two short cycles and defrost over a couple of hours - should I like this - I think not. So where to next. My plan is to pressure flush the UFH loops again, remove all the Glycol (I know I didn't tell you about that but hey you would just have been putting everything down to that if I had told you earlier - didn't want to put it in but felt I needed to for cover over the XMAS break when the system was not running.) and pressure flush the rest of the system and fill it with one of the sludge removing chemicals in case there is any residue in the UFH loops, they have been filled with the previous Glycol since 2018 after all).

 

Anyway if you have been reading this far and have any thoughts, observations, advice or even distain you might like to share I would love to hear it.

 

The playground should you want any further voyeuristic opportunity!  

 

[JohnMo]

185m loop is very long, it will dominate the system pressure drop 

 

I removed glycol a year or so ago, system flow jumped from 18 to 22L/min. Add inhibitor and biocide to system, use only one manufacturer for both - do not mix as you may have chemical incompatibility issues.

[SimonD]

Obviously your flow rates & dt need looking, but also be very patient. I commissioned a new installation at the beginning of December that has very thick stone walls and had no heating for 3 months. Initially I thought I'd messed up the installation, but it's taken until the last week or so to find its balance. This is todays:

 

 

This is what it looked like just after installation:

 

 

 

[Gus Potter]

56 minutes ago, MikeSharp01 said:

I started playing with the new heat pump on Sunday,

And that sir is the fun of it.. play away. I love this stuff. But have not taken time to really study your info. 

 

Firstly though. Your building is going to take time to stabalise.. not just the soil dumpling but the exteral walls. The moisture levels and the temperature in the dumplig below your insulation will have to settle into the new regime. Are you moving furniture about, changing rugs on the floor? Have you taken your socks off and just walked about in bare feet? 

 

1 hour ago, MikeSharp01 said:

as I want to limit the temperature of the water entering the slab, don't I?

Well technically yes. But in real life I've heated floor slabs by accident by a mile with flows in the pipes up to 60 -70C.. but the flooring was laid to compensate for this kind of accident as I have a gas boiler.. accidents do happen? and I'm after all a UFH philistine. Put it in.. lots of short loops for redundancy, make it buildable..

 

That plant room is a bit of a dream space wise.. you lucky sod!

 

From the graph I think I can see that no flows go above 40C? 

[MikeSharp01]

27 minutes ago, JohnMo said:

185m loop is very long, it will dominate the system pressure drop 

Yes, it is. I should have split it. I have not found any controls for the pump in all the places I have looked but I am hopeful that I can squeeze that up a little further and witha complete pressure flush I can get the performance I want. Getting your head around how these things work and the perverse incentives (just the laws of physics really) they seem to have built in is fun. 

 

3 minutes ago, SimonD said:

but also be very patient

I have had that thought, the whole place is cold and it has a significant decrement / increment delay as you can see from the trends. I need to get the delta T down a bit more, I have 6 or so at the ends of those runs today which is re-assuring. I am hoping its just a sediment restriction in that long loop and I can get it cleared if not I am going to have to think a bit more out of the box aren't I. I was interested today to see how the heating affected the building and the slab. Going into the heating session this afternoon we had a room temperature of 16.5, at the end of the day before we had 16.9 so a drop of 0.4 deg C in 12 hours with a delta T inside to outside of 11 deg C. 6.549 kWh of electricity raised the temperature back to 16.9 deg C now. Over the three days 33kWh (£8.91) has lifted the internal temperature 5 degrees against an average delta of 10 degrees. 

 

3 minutes ago, Gus Potter said:

And that sir is the fun of it.. play away.

Yes, rediscovering my inner child is real fun, I wish I still had the sponge brain I had then you can learn so much so fast its frightening now to think about how slow I am!

[Gus Potter]

51 minutes ago, MikeSharp01 said:

Yes, it is. I should have split it.

That length of loop is bonkers! But if in the floor it is what it is.

 

Question is.. is that loop near an external wall or not and if so.. by how much and is the cold wall on the upside of the loop or the cooler side? Next is does this matter?

 

I know many are not keen on my rough and ready approach to UFH. But they will realise 5  - 10 years down the road when they are spending hand over fist to keep it working! or maybe want to sell! 

 

Now if you look at many posts on BH there is my view on a high reliance on software.. but pretty much none on BH know how that works! they blindly trust.. they certainly don't know that FE models are and often widely wrong. As an SE Fe is subject to lots of scrutiny as it often down right dangerous! The IStructE has many papers on this.. it's a great tool if treated with caution, a very handy aid!

 

51 minutes ago, MikeSharp01 said:

I wish I still had the sponge brain I had then you can learn so much so fast its frightening now to think about how slow I am!

Ah but there is an upside. The first thing is to recognise that we have not lost the skill to be able to teach ourselves. When I went to uni at 40 was in tears on my first day.. I clocked that my educators were actually teaching me how to teach myself! You can only appreciate that when you go to higher education in later life. 

 

Yes we are a bit slower.. but also faster as we have the life experience to be able to identify what we need to learn in the context of the problems we face. 

 

Mike: Is the black line your actual room temperature? If so that does not reflect real life? What room was that in? 

 

 

 

Edited 18 hours ago by Gus Potter

[MikeSharp01]

8 hours ago, Gus Potter said:

That plant room is a bit of a dream space wise.. you lucky sod!

Thanks Gus but the space is not all it seems. I have a space 245mm deep along the wall in front and 125mm on the wall to the right as there are kitchen panels, all openable, along the facing wall and cabinets, all of which will slide out for access, along the right hand wall.

[dpmiller]

Can you pop another pump is series to help the flow in that loop? The internal pump on our IVT9 is a bit marginal so I've an external Wilo to support it. Either will *just* satisfy the flow switch so with the externl pump ticking over it allows the Carel to modulate the internal pump nicely

[JohnMo]

35 minutes ago, dpmiller said:

Can you pop another pump is series to help the flow in that loop? The internal pump on our IVT9 is a bit marginal so I've an external Wilo to support it. Either will *just* satisfy the flow switch so with the externl pump ticking over it allows the Carel to modulate the internal pump nicely

I did similar with my super long DHW heating loop. Think it pulled an extra 15 to 20W

[MikeSharp01]

10 hours ago, Gus Potter said:

Is the black line your actual room temperature? If so that does not reflect real life? What room was that in? 

Yes it is. It is in the open space on the ground floor. It is the wireless  one that comes with EMON system from open energy. How does it not reflect reality?

[Dillsue]

6 hours ago, JohnMo said:

I did similar with my super long DHW heating loop. Think it pulled an extra 15 to 20W

Me to with an older section of the house having 8mm micropore drops down the walls to the rads. Set on the slowest speed it pulls a steady indicated 10 watts. An easy way to retrofit a HP to small bore pipework:)

[MikeSharp01]

8 hours ago, JohnMo said:

did similar with my super long DHW heating loop. Think it pulled an extra 15 to 20W

 

 

How do you control it

[MikeSharp01]

8 hours ago, dpmiller said:

an you pop another pump is series to help the flow in that loop? The internal pump on our IVT9 is a bit marginal so I've an external Wilo to support it. Either will *just* satisfy the flow switch so with the externl pump ticking over it allows the Carel to modulate the internal pump nicely

How did you control it. 

[JohnMo]

21 minutes ago, MikeSharp01 said:

 

 

How do you control it

I powered mine from the 3 port valve as it was only needed during DHW.  But you may have a zone 1 pump control in your wiring centre? So control as you would if there was a buffer in the system, then call or no call for heat, the main and secondary circulator do the same thing

[Nickfromwales]

8 hours ago, JohnMo said:

I did similar with my super long DHW heating loop. Think it pulled an extra 15 to 20W

Extra pumps and complexity???

 

You???

 

Surely not??? :swoons:
 

  

[Michael_S]

Do we know how much of an impact a larger delta T has on efficiency - is it just that to have the same average T the flow temp needs to be higher (eg av flow temp of 35C could be achieved with 40 flow, 30 return or 37 flow, 33 return) or do heat pumps actively dislike a large delta T?

 

Edit:  Google suggests a 1C decrease in flow T could result in a 2% increase in efficiency so in my example above the 3C lower low rate would save 6% of energy use

Edited 57 minutes ago by Michael_S