Hitachi Yutaki Heat Pump: Hot Water – Broken by design

Published July 21, 2012 by Sean

As identified earlier in this post, my heat pump is not following it’s control settings for hot water heating, resulting in it using more power than expected. Coupled with the high standby power load, this results in an efficiency for water heating not much better than direct electric heating, and may well be contributing the the overall poor performance (compared with a typical COP of about 3).

The demand on my water heating is such that the pump needs to run once or twice a day to make up for the 4-5 degrees standing tank losses, and maybe once in addition to replace water used for a shower.

This is resulting in 2-3.5 kWh per day of electricity use, and I was surprised to see no significant improvement in the warmer weather. Most telling was the lack of response to changing the control settings. Regardless of the settings, for water heating I was always seeing heat pump flow temperatures of 55-60 degrees (and the heat pump cuts out when it’s running above 61 degrees for more than a minute). This is even when the settings on the control unit suggest I should be seeing an operating point of 46 degrees.

The system controller calculates a heat pump setting (SSUP) based on the water temperature demanded (P10, DSET) and adds P12, the heating offset. So far so good. This is converted into a 4-20mA control current and passed to the heat pump.

According to the service manual, The sensor offset (P31) is also added to this number (accounting for heat loss between the heat pump and the control unit, but not in a very useful way)

Finally, EP6 and EP7 are used to determine the response curve of the heat pump between 4mA and 20mA. The default settings are 4mA=20 degrees, 20mA=55 degrees. However, with my heat pump, 20mA corresponds to 60 degrees. Nice that it can operate at this point, but I don’t desire it to. Simple enough to change the engineering parameters though (hold set, sel and ‘i’).

The killer is this though;

4.6.6. Response of the Heat pump to a DHW demand
Since the heat pump only reads the control signal (mA) once every 20 minutes there can be a delay before the system responds to the DHW request. In order to improve this response time, the System Controller uses a feature of the heat pump that when the control signal is set to its highest value, it responds immediately. So if the supply setpoint has to be increased due to a DHW demand, then the supply setpoint is set to the parameter EP29 (Heat Pump maximum supply temperature) for a certain time. This time is the Heat Pump Max Time High Setpoint (EP35). When this time period expires, the Supply Setpoint is set to the normal calculated DHW supply setpoint.

Now, I suspect that disabling this is fine in the summer. The control current seems to be read whenever the heat pump is set to run so the function isn’t required (and we can now go back to investigating how I need to set up my system to ensure enough water for a shower).

The disaster of this configuration is for the winter. Typically, my idle water demand corresponds to 10 minutes running at full output temperature (or 20 minutes running at 50 degrees target output). Once the DHW target is reached, the 3-way valve will revert back to space heating, and the setpoint drops to maybe 30-35 degrees. However, the heat pump is still striving for 60 degrees. The radiators and pipes will have cooled off a little, so it can have a good attempt at this (and probably 3 times a day, almost regardless of the room stat settings).

For space heating, the pump continues to run even whilst the setpoint is 1 degree below the room temperature, so it is unlikely that the heat pump with trip out due to over-temp in this 10 minute period when it is operating with a temperature differential of almost 2x what is required.

There may be ways to work-around this bizare design behaviour, the most obvious one being to work out what the ‘setting mode’ parameter for Compensation for Water Temperature- Setting is. Setting mode is entered using DSW1-3, but none of the documentation which I’ve found so far details the settings. This would at least allow the factory work-around to be used whilst still retaining control over the highest operating point.

DHW demand in the summer still has a 3 minute delay as a result of the built-in hold off to a turn on request, but if the work-around isn’t enabled in the winter, the heat pump will be enabled, DHW switched in, and heat pumped out from the tank to the pump until the control current is re-sampled.

The proper fix is for Hitachi to upgrade the control software in the RHUE-3AVHN monoblock units to accommodate a better tracking of control demand. Maybe there are noise issues with sampling too rapidly, but every 3 minutes would be a nice thing to try and achieve. They have other control issues which would be good to address too – starting with the 1.8kWh (£6/month) idle drain and pump running whilst demand < room temp.

Find my other heat pump posts here.

Filed under Heating

Comments (5)

Comments RSS - Trackback - Write Comment

  1. Effect of Hitachi Heat Pump over-heat | Sean Houlihane says:

    […] noted earlier (see this post) that the Hitachi Yutaki Air Source heat pump has a design flaw. By only sampling the control […]

    Posted October 8, 2012 @ 10:10 pm (UK)
  2. James says:

    interesting to get some impartial data on the Yutaki, we have been selling these for 18mths and have some issues with water heating but this was due to installers buying cylinders that have small internal coil diameters (22mm).
    My comment is why do you have such high standing losses with the Gledhill cylinder? You mention that the heating can take from the bottom of the tank, have you used the bottom solar coil to form part of the heating?

    Posted November 22, 2012 @ 10:06 pm (UK)
  3. Sean says:

    My guess on the standing loss is siphoning, through the heat coil or the DHW take off. I’ll check the temperature drop with everything off if I remember. Yes, the solar coil at the bottom is plumbed in parallel with a relief valve. I think this helps, certainly it gives a better balance between input power and load.

    Posted November 22, 2012 @ 10:24 pm (UK)
  4. john cantor says:

    Very interesting – it will take a while to digest all that.

    I notice that this model has a high standing loss when simply doing nothing. This seems mostly the compressor oil heater which in my opinion could be controlled better. Another heat loss is the condenser plate heat-exchanger inside the unit – it has no insulation. I cannot guestimate the loss in watts, but surely a foam moulding around this item would cost very little, and save a significant amount. Surely also a considerable loss from the many un-insulated pipes inside.

    Posted September 13, 2013 @ 11:43 am (UK)
  5. Sean says:

    Hi John,

    Thanks for the comment. Interesting point on the compressor oil heater. I’ve been manually cycling the unit over the summer, and made some savings by only powering the unit overnight. Any ideas for a heater control regime which will be good for the compressor?

    Yes, there is a lot of un-insulated pipework. I added some fibreglass insulation around the heat-exchanger and some pipework, but with no clear improvement. Insulating the whole unit was my instinct, but not tried anything there yet.

    Posted September 16, 2013 @ 8:54 am (UK)

Write Comment

By continuing to use the site, you agree to the use of cookies. more information

The cookie settings on this website are set to "allow cookies" to give you the best browsing experience possible. If you continue to use this website without changing your cookie settings or you click "Accept" below then you are consenting to this.