Sunamp container bulging

[newhome]

I wonder how Sunamp factored the unit’s restriction into their calculations when they quoted for customers. I suspect that they haven’t given that there is an assumption that no generation will be lost to the grid. Surely such a restriction would have a significant affect on the model required?

 

Here is my colleague’s quote that states the assumption that there is no loss to the grid. This was used to specify the Sunamp he required. 

 

[Jeremy Harris]

47 minutes ago, Onoff said:

The conspiracist in me wonders if these issues with the product have been influenced by an outside party, with maybe a "Grand Plan".

 

Customer base loses confidence, figures go down and the product moves abroad...

 

Still no comment here from SA???

 

 

First off, I can't see Sunamp responding directly on here.  Secondly, I'm not convinced there is anything underhand going on at all; having been dealing with them since 2015 I'm absolutely convinced that their focus is primarily on the technology, and sadly not product engineering now.  That definitely wasn't always the case, though, the Sunamp PV was a very nicely engineered product, as I mentioned in this blog entry http://www.mayfly.eu/2015/10/part-forty-getting-into-hot-water-episode-two/ .   My view is that the fairly simple control system that the current eHW product has is already known to be sub-optimal for the excess PV generation use case, but that they aren't so focussed on this model now, presumably for reasons to do with the largest market segment they currently have. 

 

Putting aside the case and top insulation design, which clearly needs some further work, the only identified operating issue is the inability to keep the eHW models topped up when charged from excess PV generation, as far as we know.  I'm not aware of any issues with the other products in the range, so heating those from a boiler, ASHP etc should be fine. 

 

The difference between the models is that all except the eHW are primarily heated by hot water, very much like the original Sunamp PV.  We have speculated, with some understanding of the probable issues surrounding the use of direct electric heating of the PCM, that the rational behind the eHW model having the selectable 90%/50% depleted charge acceptance threshold as being related to the use of a direct electric heating element embedded in the PCM, as it's only that one model that seems to have this limitation. 

 

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.  A better option for us would have been to buy two more Sunamp PV heat cells to double the capacity of our old unit, whilst retaining the heated water charge system.

 

 

[scottishjohn]

25 minutes ago, JSHarris said:

 

First off, I can't see Sunamp responding directly on here.  Secondly, I'm not convinced there is anything underhand going on at all; having been dealing with them since 2015 I'm absolutely convinced that their focus is primarily on the technology, and sadly not product engineering now.  That definitely wasn't always the case, though, the Sunamp PV was a very nicely engineered product, as I mentioned in this blog entry http://www.mayfly.eu/2015/10/part-forty-getting-into-hot-water-episode-two/ .   My view is that the fairly simple control system that the current eHW product has is already known to be sub-optimal for the excess PV generation use case, but that they aren't so focussed on this model now, presumably for reasons to do with the largest market segment they currently have. 

 

Putting aside the case and top insulation design, which clearly needs some further work, the only identified operating issue is the inability to keep the eHW models topped up when charged from excess PV generation, as far as we know.  I'm not aware of any issues with the other products in the range, so heating those from a boiler, ASHP etc should be fine. 

 

The difference between the models is that all except the eHW are primarily heated by hot water, very much like the original Sunamp PV.  We have speculated, with some understanding of the probable issues surrounding the use of direct electric heating of the PCM, that the rational behind the eHW model having the selectable 90%/50% depleted charge acceptance threshold as being related to the use of a direct electric heating element embedded in the PCM, as it's only that one model that seems to have this limitation. 

 

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.  A better option for us would have been to buy two more Sunamp PV heat cells to double the capacity of our old unit, whilst retaining the heated water charge system.

 

 

yes wire them in a cascade arrangement,like solar thermal buffer tanks   triggered by the requirement for a unit to go below 50% charge before swopping to next unit ,that way you can use your PV could you not?.or maybe have a divertor valve when excess PV to make it run UFH heating to drop it into charge zone ,then divert back while its charging--alot of buggering about i know  but a solution to use what you got to best .

Is the real problem one of cycles --most lithium  batteries have a maximum  number  of cycle you can use them for --maybe all these little charges have same effect long term and would reduce  full capacity life span 

[Jeremy Harris]

2 hours ago, newhome said:

I wonder how Sunamp factored the unit’s restriction into their calculations when they quoted for customers. I suspect that they haven’t given that there is an assumption that no generation will be lost to the grid. Surely such a restriction would have a significant affect on the model required?

 

Here is my colleague’s quote that states the assumption that there is no loss to the grid. This was used to specify the Sunamp he required. 

 

 

That's pretty clear.  In the last week (which has only had two or three periods when we were exporting significant amounts of PV generation) we've only been able to utilise about 30% to 40% of our excess PV generation to charge the Sunamp at a guess.  Part of this was one day last week where we were exporting and I first noticed that the Sunamp wasn't accepting charge, despite knowing that ~4 kWh had been used that morning.  To be fair, when I investigated this, and sought clarification from Sunamp, I discovered that the Qontroller for our unit was set up wrongly, and wasn't accepting charge until the unit was 90% depleted.  I fixed that on Friday, so now it accepts charge from the 50% depleted condition, but it still refused to accept charge during a sunny spell on Saturday, so we were still exporting energy that I'd have much rather went to charge the Sunamp, so we could use it.

 

If I had to guess, I'd say that the best the current eHW models can manage in terms of excess PV utilisation may be around 50% to 60%.  There's no way at all that they can always accept charge when there is capacity available, so they will force grid import when it shouldn't be required.

[newhome]

9 minutes ago, scottishjohn said:

the obvious answer is send them a link to this thread and then the make their own decision

 

I’ll likely do that. I’ve already mentioned the container issue to him but given that his system is being installed by Sunamp themselves as opposed to a reseller, and given that they have stated that they know the issues and can mitigate against them happening it shouldn’t be an issue for him, should it? Nevertheless I have suggested that he asks the installer how he will ensure that it doesn’t happen. The PV restriction is more concerning of course. 

[Jeremy Harris]

2 hours ago, scottishjohn said:

yes wire them in a cascade arrangement,like solar thermal buffer tanks   triggered by the requirement for a unit to go below 50% charge before swopping to next unit ,that way you can use your PV could you not?.or maybe have a divertor valve when excess PV to make it run UFH heating to drop it into charge zone ,then divert back while its charging--alot of buggering about i know  but a solution to use what you got to best 

 

 

One or two people here have already got the doubled up Sunamp PV system, as it's easy to do.  The heater, flow sensor, control system and variable speed pump in the original unit is retained and a box with just two additional Sunamp PV cells is added and plumbed in in parallel.  Sunamp offered this as an option, and I provided a base wide enough to add a second heat battery when I originally fitted our Sunamp PV, thinking it might be useful to increase the capacity in future.

 

There's none of this 50% depleted stuff with a water heated cell, as used in the Sunamp PV.  They accept charge as soon as they have any spare charge capacity.

 

The same will apply to all the other Sunamp models other than the eHW, as it's only the direct electric heating control system that seems to have limited charge acceptance.

[scottishjohn]

is the real problem one of cycles ?.

i know that lithium -ion batteries all have a max cycle capacity   ,my impact gun is 2000,after that the charger will not charge it , and every now and then i have to totally flatten battery to make all cells equal again 

whe i asked they told me it was heat problem and at end of life it would just keep heating and not totally filling up  .

I understand its not electric,but if it was water and you kept boiling it and freezzing it would change its make up 

,maybe the pcm has same problem --so   they want to make sure  the  servicelife is not shortened  by lots of small charges??

electric cars are same --run them down to 20%  -charge up to 80/90% for best life ?

 

or is it like "bomb suprise" where you have ice cream inside someything but lots of heat on outside+ latent heat delays the heat flow  

you can ahve ice ,water and steam in same bucket if you add alot of heat quickly 

and the electric element will heat from one end not evenly through out the medium  

[Jamie998]

I suppose one potential workaround would be to have multiple small capacity units - not sure how they could be connected to deplete one before moving onto the next?  That way you will ensure it is depleted each time you have showers/bath etc and will therefore jump straight back into a charge cycle.  However without looking at costs I suspect this will raise a 'whole system' price significantly.

[Jeremy Harris]

2 minutes ago, scottishjohn said:

is the real problem one of cycles ?.

i know that lithium -ion batteries all have a max cycle capacity   ,my impact gun is 2000,after that the charger will not charge it , and every now and then i have to totally flatten battery to make all cells equal again 

whe i asked they told me it was heat problem and at end of life it would just keep heating and not totally filling up  .

I understand its not electric,but if it was water and you kept boiling it and freezzing it would change its make up 

,maybe the pcm has same problem --so   they want to make sure  the  servicelife is not shortened  by lots of small charges??

electric cars are same --run them down to 20%  -charge up to 80/90% for best life ?

 

or is it like "bomb suprise" where you have ice cream inside someything but lots of heat on outside+ latent heat delays the heat flow  

you can ahve ice ,water and steam in same bucket if you add alot of heat quickly 

and the electric element will heat from one end not evenly through out the medium  

 

 

No, not at all, what on earth gave you that idea?  This is getting to be just scaremongering now, and I think we should stick to facts, not wild speculation that is way off the mark.

 

The current accelerated cycle testing programme has run to well over 35,000 cycles with barely any detectable degradation at all.  Put into context, and assuming two cycles per day (which is pessimistic) then 35,000 cycles equates to a life of at least 48 years.

[Jeremy Harris]

2 minutes ago, Jamie998 said:

I suppose one potential workaround would be to have multiple small capacity units - not sure how they could be connected to deplete one before moving onto the next?  That way you will ensure it is depleted each time you have showers/bath etc and will therefore jump straight back into a charge cycle.  However without looking at costs I suspect this will raise a 'whole system' price significantly.

 

 

This isn't needed, as there is no problem at all with charging units with heated water, from any state of charge.  The old Sunamp PV used this charge method and worked very well, and I can't see any reason why using heated water to charge any size cell shouldn't work just as well.

 

Worth remembering that the limited charge acceptance issue is only with one Sunamp model, the direct electrically heated eHW.  None of the other models have this issue, as far as I'm aware.

[scottishjohn]

1 minute ago, JSHarris said:

 

 

No, not at all, what on earth gave you that idea?  This is getting to be just scaremongering now, and I think we should stick to facts, not wild speculation that is way off the mark.

 

The current accelerated cycle testing programme has run to well over 35,000 cycles with barely any detectable degradation at all.  Put into context, and assuming two cycles per day (which is pessimistic) then 35,000 cycles equates to a life of at least 48 years.

35000 full cycles prpobably not partial cycles ,which could alter results and I, m guessing that the tests were done with firsttype water heated unit --not electric element type unit--

will be total differnet heating path 

I,m not scaremongering suggesting possibilities for discussion

[Jamie998]

1 minute ago, JSHarris said:

 

 

This isn't needed, as there is no problem at all with charging units with heated water, from any state of charge.  The old Sunamp PV used this charge method and worked very well, and I can't see any reason why using heated water to charge any size cell shouldn't work just as well.

 

Worth remembering that the limited charge acceptance issue is only with one Sunamp model, the direct electrically heated eHW.  None of the other models have this issue, as far as I'm aware.

The problem is, that is the model I wanted.  My original broad brush plan was to max out on PV and dump it into a SunAmp directly - it would seem this is no loader optimal for efficiency so an alternate course of action is required to work around this issue.

[scottishjohn]

maybe the heater element should be in the water system AND pumped around a closed loop ,then heat distriibution will be same as early unit

could all be run from the pv circuit

Edited December 3, 2018 by scottishjohn

[Jeremy Harris]

3 minutes ago, scottishjohn said:

35000 full cycles prpobably not partial cycles ,which could alter results and I, m guessing that the tests were done with firsttype water heated unit --not electric element type unit--

will be total differnet heating path 

I,m not scaremongering suggesting possibilities for discussion

 

 

If not scaremongering, then you're certainly adding a lot of confusion to something that is only an issue for one specific model.

 

The accelerated testing has been rapidly heating a cell with a high power input, to full charge, within, IIRC, 7 minutes, then rapidly discharging the cell with cold water in, IIRC 5 minutes, so is very extreme.  It's all in the public domain and has been published by Sunamp. 

 

2 minutes ago, scottishjohn said:

maybe the heater element should be in the water system AND pumped around a closed loop ,then heat distriibution will be same as early unit

 

That's what we were discussing earlier in this thread; using the low power heat exchanger in the cell to charge the cell using a Willis heater, pump and some temperature sensors.

[PeterW]

1 minute ago, JSHarris said:

 

 

This isn't needed, as there is no problem at all with charging units with heated water, from any state of charge.  The old Sunamp PV used this charge method and worked very well, and I can't see any reason why using heated water to charge any size cell shouldn't work just as well.

 

Worth remembering that the limited charge acceptance issue is only with one Sunamp model, the direct electrically heated eHW.  None of the other models have this issue, as far as I'm aware.

 

Makes you wonder ....

 

Product engineering would have said that a single module type was the way to go, and then modify the heat module per “charge type” so that you reduce the variations. With this arrangement you would end up with some sort of “PV Boost Box” that could just hydraulically provide the heat into the units, and could even be provided as an add on to existing units where PV was added at a future date. 

 

It would only need a pump, valve and heating element, not exactly difficult to engineer .. 

[PeterW]

2 minutes ago, scottishjohn said:

maybe the heater element should be in the water system AND pumped around a closed loop ,then heat distriibution will be same as early unit

 

Thats pretty much what I’ve said - using a direct element is counter intuitive and has issues with localized hot spots around the resistance heater 

[scottishjohn]

how much cheaper are the basic unit with no electric element --are they much cheaper?

[Jeremy Harris]

3 minutes ago, PeterW said:

 

Makes you wonder ....

 

Product engineering would have said that a single module type was the way to go, and then modify the heat module per “charge type” so that you reduce the variations. With this arrangement you would end up with some sort of “PV Boost Box” that could just hydraulically provide the heat into the units, and could even be provided as an add on to existing units where PV was added at a future date. 

 

It would only need a pump, valve and heating element, not exactly difficult to engineer .. 

 

Making a charge unit that included a pump, Willis heater, some temperature sensors and a control unit wouldn't be hard.  My inclination would be to not mimic the Sunamp PV system, which used a variable speed pump, flow sensor and fixed output heater to control the temperature of the charge circuit, but to control the heating element to do this.

[scottishjohn]

1 minute ago, JSHarris said:

 

Making a charge unit that included a pump, Willis heater, some temperature sensors and a control unit wouldn't be hard.  My inclination would be to not mimic the Sunamp PV system, which used a variable speed pump, flow sensor and fixed output heater to control the temperature of the charge circuit, but to control the heating element to do this.

its called an inline boiler --"amptec " unit

[scottishjohn]

using a variable speed pump would make you think that too much heat is a bad thing ,they are trying to lift thetemperture of  whole mass evenly ,so localised boiling must be a problem  and that will be why they limit the electric type to make sure plenty of volume around element at a  low temp when it starts up with a thump

[Jeremy Harris]

On 03/12/2018 at 08:50, Jamie998 said:

The problem is, that is the model I wanted.  My original broad brush plan was to max out on PV and dump it into a SunAmp directly - it would seem this is no loader optimal for efficiency so an alternate course of action is required to work around this issue.

 

Sounds like your requirement is exactly the same as ours.  We originally purchased the Sunamp PV specifically because it did this, although it's capacity was the main limit, at just 4.5 kWh.  In practice this didn't matter too much, as it would accept any charge that was available, so tended to top up pretty quickly.

 

As an example, this is the data for the past week, the first week of energy monitoring that has included measuring the energy used to charge the Sunamp from the grid, the energy used to charge the Sunamp from excess PV generation and the PV generated energy we have exported to grid:

 

Energy exported to the grid = 18 kWh/week ( mean ~ 2.6 kWh/day, but very patchy)

Excess PV generated energy used to charge Sunamp = 6 kWh/week (mean ~ 0.86 kWh/day, but very patchy)

Grid energy used to charge Sunamp = 28 kWh (mean ~ 4 kWh/day)

 

The maximum energy exported in any single day last week was 7 kWh.

 

Total DHW energy used for the week = 34 Kwh, of which 6 kWh came from excess PV generation.

 

~ 18 kWh of PV generation was "wasted" by being exported to the grid rather then used to charge the Sunamp.

 

In terms of cost saving by using self-generation to heat our hot water, it looks like the Sunamp has only managed to use around 25% of our available self-generation, which is pretty crap.  There may have been one day when it couldn't have absorbed all our self-generation, last Friday, when it would probably only have been able to use around 5 kWh of the 7 kWh we exported (it didn't utilise any self-generation at all on that day).

 

Overall I think we could have utilised around 22 kwh out of the 24 kwh of excess generation had the Sunamp been able to do what it's supposed to do.  This would have reduced our grid import for DHW  last week from 28 kWh to about 12 kWh at a guess.

 

[Jeremy Harris]

13 minutes ago, scottishjohn said:

its called an inline boiler --"amptec " unit

 

 

Cheaper simpler and more controllable to use a WIllis and pump, as controls need to be custom built to control the flow temperature.

[Jeremy Harris]

6 minutes ago, scottishjohn said:

using a variable speed pump would make you think that too much heat is a bad thing ,they are trying to lift thetemperture of  whole mass evenly ,so localised boiling must be a problem  and that will be why they limit the electric type to make sure plenty of volume around element at a  low temp when it starts up with a thump

 

No, it wasn't over-heating control that was the primary reason for this.

 

The variable speed pump and ultrasonic flow sensor arrangement in the Sunamp PV was designed to do the opposite, allow the flow temperature to rise to the phase transition temperature when less than full power was applied to the heater from an excess PV diverter.  PCM needs to be heated above it's transition temperature in order to change phase from solid to liquid and store the additional latent heat of that phase change. 

 

It would need some work, but I believe that a simpler control system could be designed.  Using a Willis heater would allow the temperature in the "pot" to be maintained above the phase transition temperature, by switching the pump on and off.  Preventing overheating could be achieved by modulating the power delivered to the heating element.

 

 

[Guest Alphonsox]

22 minutes ago, JSHarris said:

The variable speed pump and ultrasonic flow sensor arrangement in the Sunamp PV was designed to do the opposite, allow the flow temperature to rise to the phase transition temperature when less than full power was applied to the heater from an excess PV diverter.  PCM needs to be heated above it's transition temperature in order to change phase from solid to liquid and store the additional latent heat of that phase change.

 

What is the flow detectors role in this ? I has assumed it was just there to detect cold water flow during the charging cycle and hence switch charging off. Is it performing a more complex function during charging ? Looking at its position in the water loop I assume it must be.

[scottishjohn]

8 minutes ago, Alphonsox said:

 

What is the flow detectors role in this ? I has assumed it was just there to detect cold water flow during the charging cycle and hence switch charging off. Is it performing a more complex function during charging ? Looking at its position in the water loop I assume it must be.

the "amptec" boiler already has temp controls etc

simple loop with divertor valve+ pump and it will just deliver water at temp you set. 

I used one as my inline back up for solar thermal UFH  system found one very cheap compared to normal proce  --brand new on ebay 

or use an electric shower unit maybe would work the same .

from what i can see a willis heater needs a tank to fit it into  and then you back to thermal losses