Sunamp container bulging

[Jeremy Harris]

9 minutes ago, joe90 said:

I am surprised that SA have not come back about this, there is plenty of feedback info on this site alone?

 

So am I. 

 

I can confirm that the unit behaved the same this morning as yesterday.  Turning the Qontroller off and then on again reset it and switched the contactor on, ready to receive any charge from excess PV that might be available (looks doubtful so far today - it's cloudy and the output isn't yet covering the house base load).

[Ed Davies]

10 hours ago, Nickfromwales said:

I can promise you that that does not happen

 

Yes, the things clearly work so obviously it doesn't. But it must take some cleverness in the design to make sure. I imagine that's at least part of what their secret sauce is about.

 

10 hours ago, Nickfromwales said:

From my direct experience, the water comes out noticeably hotter at the start,

 

Yes, the last bit of energy stored is as sensible heat warming up the liquid above the melting point.

 

10 hours ago, Nickfromwales said:

The PCM doesn't start to melt until 58oC or above.

 

Yes, again. I imagine it goes over a fairly narrow range of temperatures but you'll need an extra few degrees (say an extra 7 °C → 65 °C) to get the heat flow in to work quickly enough to be useful.

 

If left on its own it'll likely just sit at 58 °C. As heat is lost (slowly) through the insulating panels and pipework, a small amount of material will solidify releasing replacement heat to keep the temperature stable.

[willbish]

2 hours ago, JSHarris said:

 

No, none at all.  The unit still delivers hot water in exactly the same way when powered off or on, as the Qontroller has no way of knowing that hot water is being drawn off, even when powered on, other than by monitoring three temperatures within the heat cell, near the bottom, centre and top of it.

That's great to hear. You don't get that with a combi boiler during a power outage

[Nickfromwales]

29 minutes ago, Ed Davies said:

If left on its own it'll likely just sit at 58 °C. As heat is lost (slowly) through the insulating panels and pipework, a small amount of material will solidify releasing replacement heat to keep the temperature stable.

If the little SA sachets are anything to go by, they cool to ambient and then lose nothing of the heat stored from the phase change. 

[Jeremy Harris]

9 minutes ago, willbish said:

That's great to hear. You don't get that with a combi boiler during a power outage

 

No, that's one of the big advantages.  When I decommissioned the old Sunamp PV I removed all the wiring, left it for a weekend to cool down and when I came back on the Monday to disconnect it I still had to run the tap for half an hour to get rid of all the heat stored in the thing, before I disconnected the plumbing, ready to swap it over with the UniQ.

[Ed Davies]

5 hours ago, Nickfromwales said:

If the little SA sachets are anything to go by, they cool to ambient and then lose nothing of the heat stored from the phase change.  

 

Like a handwarmer? But would they do that if there was any solid in there to seed the phase change? I suspect not.

 

I actually put in another sentence after the one you quoted then deleted it as a) an additional complication and b) too speculative, but let's try now:

 

My guess would be that if a unit was fully charged (all liquid) then was left to stand then it might supercool like a handwarmer; that is, it might drop well below 58 °C while still remaining liquid but when some water was drawn off so a part solidified that would seed further solidification so that whole lot rose to 58 °C where it would reach an equilibrium of solid/liquid mix which would change slowly as it lost further heat.

 

On reflection, though, I think that's silly for two reasons. The trivial one is that the temperature sensors wouldn't work for state-of-charge sensing. More seriously, if the whole cell supercooled to ambient temperature then there wouldn't be any sensible heat available to warm water flowing through (assuming the “cold” water is at ambient temperature, too) so nothing would trigger further solidification.  So I'm going to stick with my view that any charged cell is likely at around 58 °C except the last bit of the charge/first bit of the discharge where it's a bit hotter and the last bit of the discharge/first bit of the charge when it's cooler. I think they must have something which triggers solidification as soon as the temperature reaches the appropriate point, like the clicker in a hand warmer but “always on”.

[Nickfromwales]

8 minutes ago, Ed Davies said:

My guess would be that if a unit was fully charged (all liquid) then was left to stand then it might supercool like a handwarmer; that is, it might drop well below 58 °C while still remaining liquid but when some water was drawn off so a part solidified that would seed further solidification so that whole lot rose to 58 °C where it would reach an equilibrium of solid/liquid mix which would change slowly as it lost further heat.

I think this is accurate. That I believe is why the temp DROP is difficult to detect, but temp RISE can be ( more easily / reliably ) detected. The PCM is heated up near the immersion at the bottom first for this reason I think, otherwise they'd be stuffed IMO. 

Because the PCM holds such a steady temp throughout its depletion, I think its there that the 'crude' temp detection becomes apparent. My speculation / ramblings and no more, but its good to get it off your chest ( on the off-chance someone concurs lol ).  

[Jeremy Harris]

On 11/12/2018 at 16:17, Nickfromwales said:

I think this is accurate. That I believe is why the temp DROP is difficult to detect, but temp RISE can be ( more easily / reliably ) detected. The PCM is heated up near the immersion at the bottom first for this reason I think, otherwise they'd be stuffed IMO. 

Because the PCM holds such a steady temp throughout its depletion, I think its there that the 'crude' temp detection becomes apparent. My speculation / ramblings and no more, but its good to get it off your chest ( on the off-chance someone concurs lol ).  

 

 

That fits with what I'm observing with the Qontroller behaviour.  When powered off and then on again I'm guessing it forgets about looking for a temperature drop, assumes the unit needs a warm start (assuming the temperature is above the cold start threshold) and therefore closes the contactor to allow the unit to charge. 

 

This explains why our unit reliably goes into charge acceptance mode when turned off and then back on again.

 

All we need now is a firmware mod that automatically emulates the reset from turning the power off then on again and we'll be sorted.

[Triassic]

4 minutes ago, JSHarris said:

All we need now is a firmware mod that automatically emulates the reset from turning the power off then on again and we'll be sorted.

Or just copy across the original Sunamp PV charging algorithm.

 

[Nickfromwales]

11 hours ago, Triassic said:

Or just copy across the original Sunamp PV charging algorithm.

 

Not quite. The PV uses a pipe stat ( thermistor ) iirc on the return pipe to detect temp rise. Way more reliable in its measurement of what was going on in the PCM. That was;

turn on

stay on

heat until satisfied

turn off

monitor output temp when flow reversed

when low enough turn on

repeat. 

Not quite as straightforward with the Uniqs. 

[Ed Davies]

18 hours ago, Ed Davies said:

My guess would be that if a unit was fully charged (all liquid) then was left to stand then it might supercool like a handwarmer

 

18 hours ago, Ed Davies said:

On reflection, though, I think that's silly for two reasons.

 

A third reason that's likely silly: the advertised capacity of the SA includes, in addition to the latent heat which is the main function, the sensible heat stored when it's heated above 58 °C and that available when it cools below that to some other value I forget at the moment (40 °C?) which would obviously be lost if it supercooled.

[readiescards]

A new replacement Sunamp PCM58 arrived yesterday ?

yet to unbox though to see what is different

[Jeremy Harris]

Another  update, this time after a week of switching the unit off after morning showers, then back on again to "reset" it.  At the moment I've just been doing this manually, with the main isolator switch, as I want to see whether it's worth wiring a time switch in series with the Qontroller supply or not.  Not too much excess PV generation this week, but I believe there's a difference in utilisation.

 

In summary, this week the Sunamp Qontroller has been set to what should have been the default setting, which is to enable recharge at ~50% discharge.  After ~4 kWh worth of hot water was drawn off (except for two mornings where SWMBO was away, so only around 2 kWh were drawn off) I manually switched the whole unit off, waited around 30 seconds, then switched it back on again.  For 6 days out of the 7 this triggered the unit to go into charge acceptance mode; on 1 day it was already in charge acceptance mode so I didn't bother to switch it off and then on again.:

 

Excess PV energy generated = 12 kWh

Excess PV generated energy used to charge Sunamp = 11 kWh

Grid energy used to charge Sunamp = 16 kWh

 

Total DHW energy used for the week = 27 Kwh, of which 11 kWh came from excess PV generation, which isn't bad for this time of the year, in my view.

 

I'm inclined to view this as a big improvement, as it looks as if the Sunamp is now absorbing as much excess PV generation as possible.  I think the 1 kWh disparity between the excess PV generation recorded and the Sunamp PV charge may well be just cumulative error, as the PV diverter is designed to always err on the side of caution, and has a small "leak" to the grid that ensures it never accidentally imports.  Over a week I can easily see that this could account for 1 kWh going to the grid rather than the Sunamp.

 

So, in summary, it looks as if setting the Qontroller so that Option 1 is OFF, plus ensuring that the Qontroller is powered off, then on again, each morning, ensures that it will accept as much excess PV generation as possible.  Unless Sunamp come up with a fix for this issue quickly, then I'm just going to install a time switch to reset the Qontroller every day, as that seems to be the only way to make the unit function as intended.  It's a bit of a bodge, but at least it will allow maximum self-consumption of PV generated energy.

[newhome]

1 hour ago, JSHarris said:

It's a bit of a bodge, but at least it will allow maximum self-consumption of PV generated energy.

 

Is that only the case in the winter? What about in the summer when PV generated energy is much higher? 

[Jeremy Harris]

5 minutes ago, newhome said:

 

Is that only the case in the winter? What about in the summer when PV generated energy is much higher? 

 

The "switch off then on" bodge should work all year around, as the problem is that the Sunamp will not accept any charge at all, from any source, until it is around 50% discharged.  In effect this means that you can never be sure that you're going to be able to keep it topped up, so in summer, when there's often lots of excess PV generation, the Sunamp just wouldn't use any of it until such time as the Qontroller thought that the heat battery was ~50% discharged.  For us that means we could easily go a whole day with loads of excess PV generation without any of it being stored for use later to generate hot water.

[Barney12]

Thanks for the update @JSHarris So to use off peak electric for when the solar generation diverted was not enough to fully charge you'd have to:

 

1. Time Switch "ON" early AM and Sunamp set to receive power from solar diverter only.

2. Time switch "ON" for grid power at start of economy period where the Sunamps will continue to charge until full.

 

My new units have only been operating since Friday night so I'm still trying to build a picture of usage. So far I've ascertained that:

 

One shower uses (just under) 4kwh. (I guess I shower for longer than you and its also a shower with fairly high flow rates).

 

Switching the UFH on for a 1 hour period uses between 5-6kwh. This is where the room temperature is already circa 21 degrees and so is effectively 'topping up the heat'. Its all a bit of a back of fag packet at the moment but that two hours of UFH (with the manifolds at 30 degrees) increases the room temp by about 1 degree. That one degree rise seems to be lasting for a 24 hour period (Which I guess is pretty impressive and a nod to the levels of insulation etc of my MBC build).

 

Of course the amount of energy consumed is going to be driven by the existing slab temperature. Would it be true that slab temps broadly mirror room temps? (I know there is a thread on "measuring the chunk" or something like that but its WAY above my head at a techy level).

 

Thus with the UFH on for 2 hours it  can trigger the charge demand on my 24kwh set up (2 x 12's). I guess in theory if the two 3kwh immersions are firing at the same time as the UFH demand then the units would remain fairly well charged?

 

Clearly its all going to need a huge amount of trial and error to get the heating right without waking in the morning to no hot water. Much to do!

 

My summary would be; its feeling a VERY long way from an 'all electric' heating and DHW solution for the average consumer. The controllers are woefully inadequate for sensible control even of basic functionality, let alone solar PV maximising.

 

 

 

 

 

[Dudda]

Just finished reading all 12 pages. Very tired now.

1 hour ago, JSHarris said:

 

The "switch off then on" bodge should work all year around, as the problem is that the Sunamp will not accept any charge at all, from any source, until it is around 50% discharged.  In effect this means that you can never be sure that you're going to be able to keep it topped up, so in summer, when there's often lots of excess PV generation, the Sunamp just wouldn't use any of it until such time as the Qontroller thought that the heat battery was ~50% discharged.  For us that means we could easily go a whole day with loads of excess PV generation without any of it being stored for use later to generate hot water.

Thanks for all the testing and updates on increasing to 50% and then the on/off for a week. Very useful. Are you worried at all about any damage to the controller or PCM or anything else turning it on and off (essentially resetting it) every day?

[Triassic]

25 minutes ago, Barney12 said:

My summary would be; its feeling a VERY long way from an 'all electric' heating and DHW solution for the average consumer. The controllers are woefully inadequate for sensible control even of basic functionality, let alone solar PV maximising.

I’m left wondering if Sunamp have lost the plot!

 

Mind you, having owned a number of rental properties, I found that tenants wanted simple controls on their heating and hot water. In some cases as simple as a switch ON or OFF.    Maybe thus is the level Sunamp are now focusing on!

[oranjeboom]

@themods: Is it worth splitting this discussion into a separate topic? We've gone from bulging Sunamps to under-performing Sunamps and makes it a bit cumbersome to go back 12pages to see the history of either issue. May also be easier for one of the Sunamp reps to read/respond as needed (make the most of this free R&D!!!)....no replacement for Andy T?

 

Have Sunamp responded to this issue at all?

 

I've skimmed through most of the responses on here and as far as I can ascertain it's only an issue with eHw units? Or is it more to do with the controller that comes with that one (UniQ_SBC_01)? In which case that would impact eHeat and eDual also....

 

My eDuals arrive in the New year, so not looking forward to having to spend time 'fixing' this issue from day 1.

 

13 minutes ago, Barney12 said:

My new units have only been operating since Friday night so I'm still trying to build a picture of usage. So far I've ascertained that:

 

Are these eHW units also?

 

On 03/12/2018 at 08:01, scottishjohn said:

 

This information is included in the installation manual supplied with the product, but was not available in any of the pre-purchase information that I read.  Had it been made clear that the unit needed to be discharged to ~4.5 kWh remaining before it would accept any charge then I wouldn't have gone ahead with the change, as that's no better than the Sunamp PV I already had for our use case.

 

Where in the manual does it state this 'shortcoming' as I've gone through it a few times and can't see any references (I have version 2.0 of the reference manual).

 

 

[Jeremy Harris]

You're right, @Barney12, it is a LONG way from being a usable all-electric solution as things stand; the Qontroller needs to be a lot more user-friendly and be able to deal with common usage scenarios without letting the Sunamp run out of charge.

 

Our showers use pre-heated water, and run at about 10 litres/minute, so typically use around 2 kWh.  One option I considered in order to force the Qontroller to accept charge after two showers was to turn off the preheat, as that would use more energy from the Sunamp and probably lower the state of charge below the ~50% charge acceptance threshold.  However, whilst that might be fine in summer, in winter it would mean using more grid electricity, as pre-heat is provided from the ASHP, at around 1/3rd the cost of boosting the Sunamp from the grid.

 

The way I have our system set up at the moment, including the modification I'm planning to add a time switch to reset the Qontroller, is this:

 

Power to the Sunamp heating element is provided by the PV diverter (when there is excess PV generation) plus I have a time switch, in series with a boost enable switch, that bypasses the PV diverter. 

 

When the boost switch is on, the time switch supplies power to the Sunamp heater from 03:30 to 06:00 every morning, to ensure that there is enough charge in the Sunamp to supply morning showers (in theory the boost should ensure that there is at least 7.5 kWh of charge stored in the Sunamp, if it will accept it).

 

After morning showers a second time switch, powering the always-on Qontroller supply, will turn the controller off for 1 minute and then back on again, so "resetting" it so that it is ready to accept charge whenever power becomes available via either the PV diverter or the boost timer.

 

This means that the Sunamp "should" always be ready to accept charge after limited discharge, which "should" both maximise the utilisation of excess PV generation and ensure that it always accepts the top-up boost charge in the morning.  Without "resetting" the Qontroller there is a risk that it could only have around 4.5 kWh of charge available for morning showers, barely enough to deal with all scenarios and leaving us without hot water for the rest of the day and evening, despite the unit having a nominal 9 kWh capacity.

 

13 hours ago, Dudda said:

Just finished reading all 12 pages. Very tired now.

Thanks for all the testing and updates on increasing to 50% and then the on/off for a week. Very useful. Are you worried at all about any damage to the controller or PCM or anything else turning it on and off (essentially resetting it) every day?

 

No, I'm not worried about causing damage, because I'm pretty certain that there is no way that the PCM can be damaged just by resetting the Qontroller.  If it could, then the risk would be there if there was a power cut that did the same thing. 

 

I'm also pretty sure that the problem is wholly related to the way that the Qontroller tries to assess the state-of-charge of the heat cell.  It seems that it can easily detect when it's fully charged, but struggles to be able to detect partially discharged states.  I can understand why this may be, and suspect it's related to only being able to measure temperature at three vertical locations and then use that to try to guess how much of the PCM is solid and how much is liquid (which is effectively the state-of-charge). 

 

Resetting the controller puts it back to the cold start condition, where it uses the measured temperatures to determine whether to go into cold start mode, where it pulses the contactor on and off (to prevent local overheating of the PCM) or whether it goes into warm start mode, where it just switches the contactor on.  The protection for the PCM is still there, as when it senses that the heat cell is fully charged it will still shut off OK.

 

13 hours ago, oranjeboom said:

@themods: Is it worth splitting this discussion into a separate topic? We've gone from bulging Sunamps to under-performing Sunamps and makes it a bit cumbersome to go back 12pages to see the history of either issue. May also be easier for one of the Sunamp reps to read/respond as needed (make the most of this free R&D!!!)....no replacement for Andy T?

 

Good idea to split this out, IMHO.

 

13 hours ago, oranjeboom said:

Have Sunamp responded to this issue at all?

 

 

Not as far as I know.

 

13 hours ago, oranjeboom said:

I've skimmed through most of the responses on here and as far as I can ascertain it's only an issue with eHw units? Or is it more to do with the controller that comes with that one (UniQ_SBC_01)? In which case that would impact eHeat and eDual also....

 

My eDuals arrive in the New year, so not looking forward to having to spend time 'fixing' this issue from day 1.

 

 

Are these eHW units also?

 

My unit is an eHW, and I believe the issue may impact any unit with the 2.8 kW electric heating element and the UniQ_SBC_01 Qontroller, but that is only my guess.

 

13 hours ago, oranjeboom said:

Where in the manual does it state this 'shortcoming' as I've gone through it a few times and can't see any references (I have version 2.0 of the reference manual).

 

 

It's not explicitly mentioned in the manual, or at least not very clearly.  In the section for the eHW and UniQ_SBC_01 Qontroller there is reference to setting Option 1, which is defined as Option 1 ON = 90% discharge before the unit will accept any charge and Option 1 OFF = 50% discharge before the unit will accept any charge.

 

One thing that occurred to me over the weekend is that if someone wanted to do as we did, and increase the storage capacity to deal with days without excess PV generation, then the better option would have been to buy multiple smaller units, perhaps.  Might means some cunning plumbing to discharge them one at a time, but with a multi-output PV diverter such a system would then be better able to utilise all the available storage capacity.  Having said that, adding a cheap time switch looks like it will do the job on a bigger unit, anyway.

[oranjeboom]

 

1 hour ago, JSHarris said:

Our showers use pre-heated water,

? Stupid  question alert! Pre-heated water from your ASHP --> Sunamp -->showers?

 

1 hour ago, JSHarris said:

It seems that it can easily detect when it's fully charged, but struggles to be able to detect partially discharged states.

 

Could another reason for the inability for the unit to top up at above 50% charge be to do with there being some issue with over-charging the PCM? A bit along the lines of overcharging nickel-cadmium batteries and that it was always best to run them down all the way before re-charging (in order to prolong their life). I know PCM is a totally different medium but whether there is an issue when re-charging at above 50% 'full' status?  I'm not stating that it's going to explode or result in people showering under scolding PCM (don't want to initiate any such rumours!) but just stating that there must be a legitimate reason as to why the damn thing is designed not to recharge above 50%. To guarantee the 30k(+) cycles perhaps? If purely an oversight, then hopefully it's something that Sunamp can add to their 'to improve' list. If not, I better buy one of those timers from @Nickfromwales which he no doubt has a monopoly on by now:

 

On 10/12/2018 at 15:06, Nickfromwales said:

I’ve just pre ordered 30,000 of these. 

 

 

I’ll be rich beyond my wildest dreams. 

Just don’t tell anyone ok   

 

No doubt he'll have a "Compliant with all Sunamp units" sticker on the packaging...

 

2 hours ago, JSHarris said:

Not as far as I know.

 

Could be another stupid question, but has anyone approached them with this issue? @Nickfromwales You appear to be one of their (unofficial) reps these days!! ?

 

 

[Jeremy Harris]

@oranjeboom,

 

We have a system where in winter our ASHP heats a buffer tank to 40 deg C and water from that is then pumped (on DHW demand, with a flow switch) through a plate heat exchanger to preheat the cold water that feeds the Sunamp.  This means we get a bit of heat into the water, typically it is raised to between 20 and 35 deg C before it is passed to the Sunamp, reducing the amount of energy we need to draw from the Sunamp itself.

 

There's no problem with overcharge, as that can't happen as long as the PCM is kept below its breakdown temperature of ~125 deg C.  The sensors limit the PCM temperature to well below that, no more than about 70 deg C I believe, so it's well within the allowable limit.

 

I'd pretty sure the problem isn't a safety-critical issue, just one of dealing with a challenging sensing and control problem.  I'm not convinced that the behaviour of the Qontroller has been completely and thoroughly modelled as a part of the product testing process, and that the situation of a usage pattern where less than 50% of the capacity is used before a key recharge opportunity seems to have been missed from either the specification of the test process.  It so happens that our usage pattern is exactly this, so the failing, and the consequence (running out of hot water for showers) may well  not have been considered during the design process (my personal view is that it was overlooked, and they concentrated on users who use more than half the capacity all the time).

 

I know that at least one person, other than @Nickfromwales has informed them in writing of the issue, as we aren't the only ones to have discovered the problem.

[Barney12]

4 hours ago, oranjeboom said:

 

Are these eHW units also?

 

Yes.

[Barney12]

4 hours ago, Triassic said:

I’m left wondering if Sunamp have lost the plot!

 

Mind you, having owned a number of rental properties, I found that tenants wanted simple controls on their heating and hot water. In some cases as simple as a switch ON or OFF.    Maybe thus is the level Sunamp are now focusing on!

 

I get your point but I don’t think so.

 

For standard heating requirements in an averagely insulated home even the 12kwh units aren’t man enough for the job. Plus even something as simple as doubling up means you have significant logistical issues, they take up a large amount of space and have to be on the floor. But more significantly the things are bloody heavy, in fact the 12kwh units are so heavy and cumbersome that they are almost logistically impossible to get into a house even at ground floor level. On top of all that the cost of even one unit is more expensive than an average combi boiler. So for the majority of mains gas connected properties they’re not going to tick the box of the average landlord. 

[Eileen]

2 hours ago, oranjeboom said:

Could be another stupid question, but has anyone approached them with this issue? 

 

We asked this along with several other questions a week ago but no reply. We’ve found that their communication is pretty dreadful.