ASHP - Flow Temperature Result

nxdmsandkaskdjaqd

I ran my underflow heating for the first time yesterday.  I have attached a small temperature probe which is connected to the flow side of the heat pump as it enter the storage unit.  Below is the result.  The first peak is for DHW and the second is the flow temperature for the UFH.  The system ran for 8 hours and was demanding heat for all of that time.  I wasn't expecting to see dips, does this indicate that the heat pump is short cycling?

MattMattMattUK

It could indicate lots of things but you will need to gather more data first.

The key one is how accurate do you think the temperature probe is and have you tested it, was it Insulated against the pipe or could air flow have changed the temp it was reading?

What is the flow temperature of the unit set to and what does it report the output temperature as being? Does it also report return temperature?

It would also be worth running a test where you set the demand temp for space heating well above the set flow temp. That way the unit would only short cycle if there were issues with dispersing heat due to inadequately sized underfloor heating or radiators. 

Reed_Richards

There is cycling and short cycling.  All heat sources will cycle when the heat demand is less than the lowest heat output they can achieve.  What you don't want is for these cycles to be short, say less than 5 minutes in duration.  I suspect that short cycling is worse for boilers than for heat pumps but it is still best avoided.

FreeBear

MattMattMattUK said:

It could indicate lots of things but you will need to gather more data first.

The key one is how accurate do you think the temperature probe is and have you tested it, was it Insulated against the pipe or could air flow have changed the temp it was reading?

Three different ways of mounting a sensor -

From top to bottom - A DS18B20 clipped to a pipe - These things can be inaccurate and non-linear, especially with the cheap Chinese knock-offs (which most of mine are).

Middle one, a PTC supplied with a purpose designed pipe clip (this one should be on 15mm pipe, not 22mm as in the photo). Accuracy & linearity yet to be determined.

Bottom one - PTC set in a custom made thermal pocket so that the business end is in the centre of water flow. Should be reasonably accurate and avoid problems from draughts affecting the readings.

Rather than measuring the return temperature, you'd be be much better off using the outward flow - The peaks & dips could be a result of the UFH controls increasing/decreasing flow. Comparing the outward flow and return temperatures would confirm this (and if you can measure the flow rate, you can calculate power). To determine if the HP is short cycling, I'd suggest monitoring the electrical consumption - This would tell you if the HP is modulating the output or really short cycling.

More sensors for @nxdmsandkaskdjaqd to add to Home Assistant I'm afraid.

P.S. Internal view of the thermal pocket.

nxdmsandkaskdjaqd

MattMattMattUK said:

It could indicate lots of things but you will need to gather more data first.

The key one is how accurate do you think the temperature probe is and have you tested it, was it Insulated against the pipe or could air flow have changed the temp it was reading?

What is the flow temperature of the unit set to and what does it report the output temperature as being? Does it also report return temperature?

It would also be worth running a test where you set the demand temp for space heating well above the set flow temp. That way the unit would only short cycle if there were issues with dispersing heat due to inadequately sized underfloor heating or radiators. 

I am happy with the flow temperature probe, the flow result for DHW would support this.

Not sure what the flow temp was set to as not familiar with the temperature compensation system yet. The system reported a flow temperature after 8 hours of being on as 31C.

I will run another test today, its mild with an outside temperature of 15C.

I have been looking at the flow and return on the heating manifold.  I could only see a delta of 4.5C, which may not be enough.  I believe I should be looking for 7C.  Got to work out how to adjust the flow rates.  I do have the design spec for the flow rates.

nxdmsandkaskdjaqd

Reed_Richards said:

There is cycling and short cycling.  All heat sources will cycle when the heat demand is less than the lowest heat output they can achieve.  What you don't want is for these cycles to be short, say less than 5 minutes in duration.  I suspect that short cycling is worse for boilers than for heat pumps but it is still best avoided.

I have looked in a bit more detail at the flow temperature graph.  There are 2 points where the off cycle is less than 5 minutes. I have added an expanded plot of yesterday's 8 hour run.  Interesting that after 8.30 pm, there are no short cycles.  Does this indicate anything?

As mentioned above, I have only 4.5C delta across the mainfold, I suspect this may not be helping.

FreeBear

nxdmsandkaskdjaqd said: I have been looking at the flow and return on the heating manifold.  I could only see a delta of 4.5C, which may not be enough.  I believe I should be looking for 7C.  Got to work out how to adjust the flow rates.  I do have the design spec for the flow rates.

Which make/model of heat pump do you have ?

Some models have an interface that allows you to take temperature & flow readings from internal sensors (and also change parameters on the fly).

nxdmsandkaskdjaqd

FreeBear said:

nxdmsandkaskdjaqd said: I have been looking at the flow and return on the heating manifold.  I could only see a delta of 4.5C, which may not be enough.  I believe I should be looking for 7C.  Got to work out how to adjust the flow rates.  I do have the design spec for the flow rates.

Which make/model of heat pump do you have ?

Some models have an interface that allows you to take temperature & flow readings from internal sensors (and also change parameters on the fly).

Hitachi Yutaki (ASHP and Indoor Unit & DHW Cylinder)

Which I believe has sensors internally, but these are only available under the engineering login.

FreeBear

nxdmsandkaskdjaqd said:

FreeBear said:

nxdmsandkaskdjaqd said: I have been looking at the flow and return on the heating manifold.  I could only see a delta of 4.5C, which may not be enough.  I believe I should be looking for 7C.  Got to work out how to adjust the flow rates.  I do have the design spec for the flow rates.

Which make/model of heat pump do you have ?

Some models have an interface that allows you to take temperature & flow readings from internal sensors (and also change parameters on the fly).

Hitachi Yutaki (ASHP and Indoor Unit & DHW Cylinder)

Which I believe has sensors internally, but these are only available under the engineering login.

OK. Done a little bit of digging (some of this you probably already know). There are at least 8 different variants of the Yutaki HP depending on the final application. But they all look to be fairly similar in design, so quite likely share the same electronics board.

Hitachi uses a proprietary H-BUS to communicate between the HP and a remote controller. It goes without saying, information is virtually non-existent from Hitachi <grumble><grumble>. However, several people have been working on reverse engineering both the bus and the protocol. It appears to be a two wire serial bus powered at 12V running at 9600 baud.

A couple of links for you -

https://github.com/LenShustek/M-NET-Sniffer - Has a circuit diagram to interface directly to the H-BUS pins.

https://github.com/hankerspace/HLinkSniffer - Reverse engineered packets with (possible) variables.

https://github.com/Arnold-n/P1P2MQTT - Primarily aimed at Daikin Altherma, but also claims to work with Hitachi.

Edit - If Daikin is compatible with the Hitachi H-BUS, then this project might be of use - https://github.com/revk/ESP32-Faikin

Once you go start messing with Home Assistant (and Grafana), it becomes a very deep rabbit hole - One I've been in for several years now.

michaels

Do you have anything that monitors your household energy usage more accurately than the smart meter (half hourly)?

For example we have a battery system that gives us battery charge/discharge and grid load at 2 min intervals so is useful to see in general what different stuff is using (although this obviously does not differentiate all your other uses from the heat pump)

nxdmsandkaskdjaqd

michaels said:

Do you have anything that monitors your household energy usage more accurately than the smart meter (half hourly)?

For example we have a battery system that gives us battery charge/discharge and grid load at 2 min intervals so is useful to see in general what different stuff is using (although this obviously does not differentiate all your other uses from the heat pump)

No unfortunately I only have the smart meter, which only gives a general view on consumption.