Geekzone: technology news, blogs, forums
Guest
Welcome Guest.
You haven't logged in yet. If you don't have an account you can register now.
View this topic in a long page with up to 500 replies per page Create new topic
1 | 2 | 3 | 4
Aredwood
3885 posts

Uber Geek


  #2267270 30-Jun-2019 11:54

@Terciops Can you add support for a second temp sensor?

As on my own hot water cylinder (300L mains pressure) I have both a lower temp sensor and an upper temp sensor that I have installed behind the upper element cover. Which are part of a homemade Arduino controller for managing solar thermal, and a (now defunct) waste oil boiler for heating the hot water cylinder.

I find that it doesn't take much hot water usage to cause the lower temp sensor to indicate completely cold water. While the upper temp sensor stays hot. Unless you have used lots of hot water.

Using an upper temp sensor would give a far better basis for control of heating using grid power. As the controller would then know if the cylinder has a reasonable amount of hot water stored Vs cylinder close to running out of hot water.





Terciops
21 posts

Geek


  #2267284 30-Jun-2019 12:18
Send private message

Aredwood: @Terciops Can you add support for a second temp sensor?

As on my own hot water cylinder (300L mains pressure) I have both a lower temp sensor and an upper temp sensor that I have installed behind the upper element cover. Which are part of a homemade Arduino controller for managing solar thermal, and a (now defunct) waste oil boiler for heating the hot water cylinder.

I find that it doesn't take much hot water usage to cause the lower temp sensor to indicate completely cold water. While the upper temp sensor stays hot. Unless you have used lots of hot water.

Using an upper temp sensor would give a far better basis for control of heating using grid power. As the controller would then know if the cylinder has a reasonable amount of hot water stored Vs cylinder close to running out of hot water.

 

 

 

Most domestic tanks only have a single opening to place a temperature probe, and a hell of a lot of experimentation went into finding an optimal solution for a single probe operation.  Yes the upper is hotter, but the inertia of the thermal mass and mixing column make it a poor analog of true tank temperature at the thermocline - which is what is needed.  Better to have very precise control of temperature just above the thermocline when you are only topping up small amounts as there will very little heat injected and the thermal plume will not get moving fast enough for the top temp to change.  In the end the top temp - even if you have one - is much slower to react.

 

 

 

The trick is to put the temperature probe above the thermocline - about a hand span above the element by design.  This avoids seeing the cold inrush of incoming water from the bottom and unless there is a rapid water use,  this lower probe works fine.  The temperature at the top of the tank will always be higher as warm water rises.

 

If the probe is set too low then you always see that cold inrush and temperature measured is erratic and unreliable.   Get that lower probe up above the thermocline and it becomes very usable.  As a guide, tank designers put the mains thermostat slightly below this point to account for the dwell on the thermostat mechanism.   So you need to be about 5cms / 2 inches above that thermostat level to be just about golden.

 

 

 

We also use a digital probe and read it every 8 seconds to get an average result.  This works well in practice and is very accurate.   The dynamics of a hot water tank are surprisingly complex.  There are some long, but interesting PhD papers on the 'net about this.

 

 

 

 

 

 

 

 

 

 


raytaylor
3856 posts

Uber Geek

Trusted

  #2267378 30-Jun-2019 16:59
Send private message

Terciops:

 

The next generation Paladin is considerably more sophisticated, and there is comms  designed in to work on the local LAN.  The whole thing around providing some form of external readout and control has been a nightmare frankly.  Bluetooth is a mess and is very spotty particularly around all the pipes and wires where Paladin is usually mounted.  So I went with the ESP32 and used that as a secondary processor / comms unit that talks to the main Paladin 'engine' on a hard serial link.

 

This has been great and it works very well without compromising the core Paladin performance.  However getting a reliable Wifi signal in a lot of houses is difficult for the same reasons as above for BT.   

 

 

 

 

I would consider ethernet LAN capability to be more important than wifi or bluetooth. Especially when selling to new houses as they have homehubs with structured cabling being installed. 

 

We are advising all our new home builds to include an ethernet cable to the hot water cylinder and the fuse board for future solar installations if they are not getting solar installed at the time of the build. 





Ray Taylor

There is no place like localhost

Spreadsheet for Comparing Electricity Plans Here


Terciops
21 posts

Geek


  #2267485 30-Jun-2019 20:44
Send private message

raytaylor:

 

Terciops:

 

The next generation Paladin is considerably more sophisticated, and there is comms  designed in to work on the local LAN.  The whole thing around providing some form of external readout and control has been a nightmare frankly.  Bluetooth is a mess and is very spotty particularly around all the pipes and wires where Paladin is usually mounted.  So I went with the ESP32 and used that as a secondary processor / comms unit that talks to the main Paladin 'engine' on a hard serial link.

 

This has been great and it works very well without compromising the core Paladin performance.  However getting a reliable Wifi signal in a lot of houses is difficult for the same reasons as above for BT.   

 

 

 

 

I would consider ethernet LAN capability to be more important than wifi or bluetooth. Especially when selling to new houses as they have homehubs with structured cabling being installed. 

 

We are advising all our new home builds to include an ethernet cable to the hot water cylinder and the fuse board for future solar installations if they are not getting solar installed at the time of the build. 

 


Terciops
21 posts

Geek


  #2267486 30-Jun-2019 20:45
Send private message

raytaylor:

 

Terciops:

 

The next generation Paladin is considerably more sophisticated, and there is comms  designed in to work on the local LAN.  The whole thing around providing some form of external readout and control has been a nightmare frankly.  Bluetooth is a mess and is very spotty particularly around all the pipes and wires where Paladin is usually mounted.  So I went with the ESP32 and used that as a secondary processor / comms unit that talks to the main Paladin 'engine' on a hard serial link.

 

This has been great and it works very well without compromising the core Paladin performance.  However getting a reliable Wifi signal in a lot of houses is difficult for the same reasons as above for BT.   

 

 

 

 

I would consider ethernet LAN capability to be more important than wifi or bluetooth. Especially when selling to new houses as they have homehubs with structured cabling being installed. 

 

We are advising all our new home builds to include an ethernet cable to the hot water cylinder and the fuse board for future solar installations if they are not getting solar installed at the time of the build. 

 


Terciops
21 posts

Geek


  #2267487 30-Jun-2019 20:46
Send private message

"I would consider ethernet LAN capability to be more important than wifi or bluetooth. Especially when selling to new houses as they have homehubs with structured cabling being installed. 

 

We are advising all our new home builds to include an ethernet cable to the hot water cylinder and the fuse board for future solar installations if they are not getting solar installed at the time of the build. "

 

 

 

Ethernet is not native to the ESP32 in its' present form.  You have however given me a considerable impetus to implement a new idea that has been sitting on my circuit design board as a 'maybe' for the next run of motherboards.  To explain :

 

 

 

One of the annoying / time consuming installation issues in some houses is the temperature probe cable, which needs 3 wires, and is usually extended on a CAT5 cable or similar.  But I could improve my WiFi performance considerably if I moved the the SSR away from the main box entirely (and by inference the Transfer CT as well).

 

Thinking that through laterally the obvious solution is to move the transfer CT the SSR and the Temperature probe connection to a separate box next to the cylinder and leave the main Paladin box, sans SSR and heatsink,  adjacent to the main fuse box where the mains and solar CT connect, and for visibility.

 

 

 

Now there is still a mains connection to that mains box, about 4W worth, to run the power supply for the CPU(s) and the mains signal transformer.  But all the heavy amps are gone, and with them a considerable potential 'hum' that can play merry havoc with the WiFi.  A win-win if I ever saw one. 

 

The obvious vehicle to move the electric pixies for the Transfer CT and temperature probe Data and the SSR commands is a CAT5 cable, so adding a J45 socket to the motherboard and a J45 socket to a little daughter board in a box that carries a 3.5 socket for the Transfer CT and a connection for the temp probe and the SSR would be simple and very efficient.   Just route the HW element live through that SSR and all done.  

 

A couple of indicator lights on the box to show SSR activity for diagnostics, a nice cast metal box about the size of a single plug surface box with the heat sink on the lid through connected to the SSR solves the cooling in spades. 

 

I can also mount the connection for the secondary SSR on that daughter board as well,  as the SSRs carry a common earth - there is the 8 of the CAT5.

 

 

 

Which leaves me with a nice empty Paladin case with room to mount any new goodies where the SSR / heatsink once was and no nasty heavy wire connectors either.

 

 

 

If you have to run a CAT5 anyway, or as you are telling me - a CAT5 run would be a new build standard,  then it is a no brainer to get this option built into the new motherboards that will be order this coming month for the next build sequence.

 

 

 

Good call, Ray, thanks.

 

 

 

 


raytaylor
3856 posts

Uber Geek

Trusted

  #2267506 30-Jun-2019 21:12
Send private message

Sorry just to clarify

 

Cat5/6 is for future lan connections to the solar controller and a monitoring device for home automation, cloud connectivity, SNMP monitoring and a web interface for the devices. 

 

I suppose it could be used for control signalling but we wouldnt be advising customers to connect home infrastructure like a solar controller or dump load controller to wifi because every time we swap out the router they get a new ssid and password. 

 

Our technician walks in, swaps the router, tells the customer the new ssid and password is on the front of the router and walks away. Testing is done to the technicians cellphone or laptop only. Customers often call our helpdesk saying "your technician swapped our router and now our solar isnt reporting to the cloud" and our response is simply "too bad, we dont know about solar, we just fix your internet, contact the manufacturer and your solar installer should have hard wired your solar controller with a LAN cable" but in nicer words. 

 

Our advice for electricians fitting out a new home build is a the switch hub in the homehub-to-HotWater and homehub-to-fusebox is simply so that a solar installer can make use of a LAN cable when the solar system is installed, so they avoid the problem of wifi stopping every time the router is replaced. 
Thus the installer has no reason to be lazy and use wifi. 

 

I suspect any other ISP would be the same. 





Ray Taylor

There is no place like localhost

Spreadsheet for Comparing Electricity Plans Here


Aredwood
3885 posts

Uber Geek


  #2267533 30-Jun-2019 22:36

Terciops:

Aredwood: @Terciops Can you add support for a second temp sensor?

As on my own hot water cylinder (300L mains pressure) I have both a lower temp sensor and an upper temp sensor that I have installed behind the upper element cover. Which are part of a homemade Arduino controller for managing solar thermal, and a (now defunct) waste oil boiler for heating the hot water cylinder.

I find that it doesn't take much hot water usage to cause the lower temp sensor to indicate completely cold water. While the upper temp sensor stays hot. Unless you have used lots of hot water.

Using an upper temp sensor would give a far better basis for control of heating using grid power. As the controller would then know if the cylinder has a reasonable amount of hot water stored Vs cylinder close to running out of hot water.


 


Most domestic tanks only have a single opening to place a temperature probe, and a hell of a lot of experimentation went into finding an optimal solution for a single probe operation.  Yes the upper is hotter, but the inertia of the thermal mass and mixing column make it a poor analog of true tank temperature at the thermocline - which is what is needed.  Better to have very precise control of temperature just above the thermocline when you are only topping up small amounts as there will very little heat injected and the thermal plume will not get moving fast enough for the top temp to change.  In the end the top temp - even if you have one - is much slower to react.


 


The trick is to put the temperature probe above the thermocline - about a hand span above the element by design.  This avoids seeing the cold inrush of incoming water from the bottom and unless there is a rapid water use,  this lower probe works fine.  The temperature at the top of the tank will always be higher as warm water rises.


If the probe is set too low then you always see that cold inrush and temperature measured is erratic and unreliable.   Get that lower probe up above the thermocline and it becomes very usable.  As a guide, tank designers put the mains thermostat slightly below this point to account for the dwell on the thermostat mechanism.   So you need to be about 5cms / 2 inches above that thermostat level to be just about golden.


 


We also use a digital probe and read it every 8 seconds to get an average result.  This works well in practice and is very accurate.   The dynamics of a hot water tank are surprisingly complex.  There are some long, but interesting PhD papers on the 'net about this.


 


 


 


 


 



Let's say that you have a fully heated hot water cylinder, that is 270L or so capacity. You then use 50L to 100L of hot water in the early morning. The temp sensor at the bottom of the cylinder will say that the cylinder is now completely cold. When in reality there is plenty of hot water still stored and available for use. And it would only take 3-6KW/Hr to reheat the cylinder.

Yet the controller will then be using grid power to reheat the cylinder. But the cylinder won't then be able to accept much diverted power, as it would already be up to temp. If there is also an upper temp sensor. The controller would then know that the cylinder still has plenty of hot water available. And it would then only supply diverted power to the element.

If the upper temp reading drops, then the controller could either immediately reheat the cylinder on grid power. Or it could do a full reheat cycle overnight.

I can see why some of your customers are having to kludge on external timers to stop needless reheating on grid power. And so they can actually use the full storage capacity of their cylinder for storing diverted power.

Even better if there is also an upper element. As that could then be used for quick “boost” heating. While the lower element would solely be used for diversion heating and anti legionella heat cycles.

Guessing that the temp sensor you have used is the DS18B20. In which case, no hardware changes are needed to support multiple temp sensors. Just wire the second sensor in parallel with the first and upgrade the software.

Newer hot water cylinders are available with upper element ports. Both the Rheem and Rinnai 300L stainless steel cylinders come with upper element ports as standard. Fitting upper temp sensors to these cylinders is extremely easy. And if you want to add an upper element, all you have to do is purchase an element and thermostat. And installing them is the same as replacing an existing element and thermostat.





Aredwood
3885 posts

Uber Geek


  #2267537 30-Jun-2019 22:49

@kotuku4 has deliberately installed their temp sensor higher than recommended. Simply to reduce the needless reheating from grid power. An upper temp sensor along with some software updates would solve their problems.

Depending on how the code in the controller is written. You could implement quite a good “kludge” fix, just by buying another DS18B20 sensor (available cheap from Aliexpress). Install that sensor as an upper level sensor. And then using a small relay controlled by a timer to alternatively switch between the 2 sensors.

***Sigh***

If I get grid connect solar installed on my house. Looks like I will have to make my own PV diverter. Same as how I had to make my own circulation pump controller when I installed solar thermal.





Terciops
21 posts

Geek


  #2267626 1-Jul-2019 08:19
Send private message

Aredwood: @kotuku4 has deliberately installed their temp sensor higher than recommended. Simply to reduce the needless reheating from grid power. An upper temp sensor along with some software updates would solve their problems.

Depending on how the code in the controller is written. You could implement quite a good “kludge” fix, just by buying another DS18B20 sensor (available cheap from Aliexpress). Install that sensor as an upper level sensor. And then using a small relay controlled by a timer to alternatively switch between the 2 sensors.

***Sigh***

If I get grid connect solar installed on my house. Looks like I will have to make my own PV diverter. Same as how I had to make my own circulation pump controller when I installed solar thermal.

 

 

 

We could go back and forwards on this temperature sensor position / number issue for quite a while, and get nowhere definitive, since in our own ways both Aredwood and myself are correct.  I have tried using 2, even 3 tank sensors in determining the optimal point to measure water temperature in the column.  As in so many things when you take ideas off the drawing board and plonk them into the real world, there is no simple answer when there are so many variations of tank size / design, user use profiles, excess PV availability - the list goes on.

However in Paladin's defense the single temperature sensor mounted higher into the thermocline - which is not a line, more a diffuse boundary - works well for nearly all cases.  The over arching directive for the present design was to avoid having the users with standard tanks (180L) having cold showers.  Experience over the past 5 years has shown that a small reduction is top-ups is just not worth a cold shower.

 

That a one size fits all will never suit all is a given.  I have modified quite a few firmware loads to try to tailor Paladin to individual needs, but when the push came to shove we (nearly) always get back to standard.  

 

The latest version of Paladin does cater slightly more elegantly for personal use and experience by allowing the user to change the value of the minimum setting which effectively moves the position on the thermocline where Paladin starts to top up.  But, once again, experience has shown that doing this just exacerbates the Legionaire's problem overall during periods of low PV, so the gross savings are very small.  It all boils down to the amount of excess solar that is available to drive that transfer to hot water.  8-10kWh / day is lot, particularly in the Winter quarter.  In the summer - no problems for any installations above 3kW.  But that is the nature of PV. 

 

Anyway,   Aredwood,  when you get solar PV installed, PM me, and I will let you have one of my development Paladins on indefinite long loan.   As you obviously know how to drive an Arduino, I will also let you have the define file for the pins and hardware on the mother board and an copy of the current binary (to put it back to standard when you need to).  Or, as it is all modular plug and play, you could just have another Mega and plug that in to run your code.

 

You can then have a little 'codeplay'  time without the hassle of building hardware, which is a tedious process to get right as well.  That offer is not quite as generous as it sounds.  These ex R&D units have various physical mods that never made it to the factory versions,  so they can't be sold without just scavenging parts - which I loath to do since they can always find a useful home as is, and given the latest firmware have some interesting extras.

 

 If you join Paladin's facebook page, you see various expansion on this topic and many others and also see my home Paladin live on Youtube here :  https://www.youtube.com/watch?v=Dqr-sWS3IUA

 

 The graphic Paladin shows the DeltaT (rate of change of temperature over time - in this case degrees C / hour) - which is a vital key to managing this topic in point.

 

 

 

 

 

 


kotuku4
451 posts

Ultimate Geek

Lifetime subscriber

  #2267646 1-Jul-2019 08:58
Send private message

Hi Guys

 

I think the temperature sensor was possibly an installer issue, and a retro fit to a basic cylinder.  However, It was easy to mount in a different position.

 

Some cylinders have ports to install temperature probes.  The installer was used to immersun units that have multiple temperature probes. 

 

At the weekend we had an event, and the shower got used alot, water temp was as low as I have seen it and still had water heat to max by 1pm on fine frosty day in Blenheim.

 

It has taken me awhile to understand that the system is working well.  Had Electrician/solar installer look at my power accounts for last few months to review possible battery install.

 

I am satisfied that we are doing all that we can to use our solar power, Paladin is working well, battery storage cannot be justified, don't need any additional PV panels for our use.

 

I don't need to upgrade the Paladin unit that I have just yet.

 

Will be watching used EV car market.

 

Replaced used deep cycle battery in my solar shed for charging by Ebike, Emower, battery tools, etc.

 

 

 

 





:)


Aredwood
3885 posts

Uber Geek


  #2268123 2-Jul-2019 03:49

Terciops:

 


We could go back and forwards on this temperature sensor position / number issue for quite a while, and get nowhere definitive, since in our own ways both Aredwood and myself are correct.  I have tried using 2, even 3 tank sensors in determining the optimal point to measure water temperature in the column.  As in so many things when you take ideas off the drawing board and plonk them into the real world, there is no simple answer when there are so many variations of tank size / design, user use profiles, excess PV availability - the list goes on.

However in Paladin's defense the single temperature sensor mounted higher into the thermocline - which is not a line, more a diffuse boundary - works well for nearly all cases.  The over arching directive for the present design was to avoid having the users with standard tanks (180L) having cold showers.  Experience over the past 5 years has shown that a small reduction is top-ups is just not worth a cold shower.


That a one size fits all will never suit all is a given.  I have modified quite a few firmware loads to try to tailor Paladin to individual needs, but when the push came to shove we (nearly) always get back to standard.  


The latest version of Paladin does cater slightly more elegantly for personal use and experience by allowing the user to change the value of the minimum setting which effectively moves the position on the thermocline where Paladin starts to top up.  But, once again, experience has shown that doing this just exacerbates the Legionaire's problem overall during periods of low PV, so the gross savings are very small.  It all boils down to the amount of excess solar that is available to drive that transfer to hot water.  8-10kWh / day is lot, particularly in the Winter quarter.  In the summer - no problems for any installations above 3kW.  But that is the nature of PV. 


Anyway,   Aredwood,  when you get solar PV installed, PM me, and I will let you have one of my development Paladins on indefinite long loan.   As you obviously know how to drive an Arduino, I will also let you have the define file for the pins and hardware on the mother board and an copy of the current binary (to put it back to standard when you need to).  Or, as it is all modular plug and play, you could just have another Mega and plug that in to run your code.


You can then have a little 'codeplay'  time without the hassle of building hardware, which is a tedious process to get right as well.  That offer is not quite as generous as it sounds.  These ex R&D units have various physical mods that never made it to the factory versions,  so they can't be sold without just scavenging parts - which I loath to do since they can always find a useful home as is, and given the latest firmware have some interesting extras.


 If you join Paladin's facebook page, you see various expansion on this topic and many others and also see my home Paladin live on Youtube here :  https://www.youtube.com/watch?v=Dqr-sWS3IUA


 The graphic Paladin shows the DeltaT (rate of change of temperature over time - in this case degrees C / hour) - which is a vital key to managing this topic in point.


 


 


 



Thank you for your kind offer.

At this stage, I dont know when I will be getting solar PV installed. As I would need to get the switchboard and mains cable upgraded to my house first. And as part of that, I also want to get 3 phase power connected. (mainly for future charging of multiple EVs). Existing mains cable is undersized, so still needs upgrading even if I dont get 3 phase. So it might be quite awhile until I finally get solar PV installed.

Last time I did the sums on solar PV, was during the Flick Electric Heyday. Meaning I was getting really cheap off peak power and solar just couldn't compete. Yet I'm now back on an anytime pricing plan, and there now appears to be far more winter sunshine hours (personal observation). And I now have heatpumps (therefore summer aircon). So it might make far more sense for me to get solar PV installed soon.

You should consider releasing a 2 phase and 3 phase capable diverter. Especially as someone on this thread has already asked about a 3 phase diverter. You could probably just add an extra 2 SSR, and current sensors. Although you would then need a 3 phase element in the cylinder.

And also consider a diverter for off grid systems. Maybe by taking mains power from the inverter and sending it to the cylinder. As that would allow you to use most of your original hardware. And avoid the need to install a 24V or 48V element in the cylinder.


edit to add.

Thread where someone needs an off grid hot water system.
https://www.geekzone.co.nz/forums.asp?forumid=141&topicid=251480





Terciops
21 posts

Geek


  #2268132 2-Jul-2019 07:02
Send private message

Interesting points about 3 phase, off grid and PV increase.   I would be typing all day to do them all justice, but briefly  :

 

3 Phase for a diverter is, IMHO, just awkward and largely impractical,  as a 3 phase element is not really a domestic 'thing'.  The more interesting question is why spread your solar onto three phases anyway, when what you really want to do is consume as much of it as possible.   If you have or need 3 phase for say, EV charging (but that is another very linked topic), then you really only need those 3 extra 3 phase amps for a single high load.  The rest of the house is quite adequately catered for on a single phase.  So as I see it, you put all your solar on the same phase as your domestic loads.  But I don't have 3 phase, so I could well be wrong (again).

 

What I see happening is that people are adding up all the possible loads in their house and seeing that number above the nominal 60 amps (or so)  think they have to get into 3 phase to keep the pixies dancing.  With that logic I need 3 phase too.  If I switch on everything - 5 heater / conditioners,  electric oven, 4 induction hobs,  EV charger,  couple of assorted resistive heaters, hot water heater,  and let's not get into the lights, computers, a small arc welder,  3 kW battery charger for the house pack, 3 fridge / freezers, washing machine, tumble drier, my workshop Paladin test rig (and that has more amps hanging off of it than I care to think about :) )  - well then I am well over 100 amps.   I have 4 years of 2 minute interval grid data off Paladin and I have never seen more than 8kW (40 amps) and rarely go above 6kW (30 amps).  It just doesn't happen.  Just think about use patterns.

 

More of a worry is being on 3 phase with a 32 amp limit per leg.  This has come up quite a few times with people on a split dwelling with a 32 amp total limit and when the water heater went on when they were cooking say,  everything went dark.  Of course on a standard thermostat setup this is pretty much electric roulette.  The fix for Paladin was was actually incredibly simple.  Paladin can 'see' the grid output load and also adjust the water heater current very accurately - so it was just a matter of throttling any top-ups to keep the grid input below 5400W (think of a number - I can't even remember why 5400 exactly, it was just a number that kept enough headroom below the 32 amps to cater for say a kettle going on as well.  Paladin is fast, but kettles and induction hobs are real surge makers).   This is now a standard feature in the Paladin core and can be activated as required.

 

 

 

Off Grid ...  Oh now you have really pressed my buttons :)

 

I spent my first 16 working years in the military (RAF).  My principle secondary duty for most of that time (pilots don't fly all the time so you always get other jobs to justify the Queen's shilling) was as an NBCW/Survival Instructor.  You don't spend all that time teaching this rather exotic stuff without gaining a healthy respect for the frailty of our species and civilization.   So I might be considered a bit of a 'closet prepper' by some.  Call it as you see it, but it's a worry :)  At least according to my better half.  The mantra goes - Protection / Location / Water / Food - and without power all 4 go very rapidly down the swanny in most domestic situations.  What is generally not well understood is that without grid power all your solar is just an expensive roof ornament. 

 

Thanks to the 'islanding' rules hard baked into all but battery connected 'outback type' inverters,  when the grid goes out we (PV owners) are in the same paddle-less canoe, up the same smelly creek, as everyone else.  Suffice to say that this was one of the principle directives when I originally started thinking about Paladin - and it's modular design is has a lot to do with a solution to the 'grid down' problem.  The how, is almost certainly not practical (or even legal) to discuss on a public forum, but if you think about what Paladin does, you should see how to balance your PV load without a grid connection.  All you need to do is ..........  'Nuff said.

 

 

 

Off Grid Water Heating....

 

Now this gets really interesting and useful.  Let's dismiss a couple of things straight away.  Off grid, the most valuable things are all the electric pixies stored in the battery, so you would not want to waste these on a hot shower.  That said, once the battery is full it would be foolish not to put the excess PV into the hot water.  The problem is how???

 

Working on the basic premise that DC and AC do not play well together and segregating them using switch gear at high amps is expensive and tricky (same thing really).  You either need a dual element water heater that takes your roof DC (at 200-300V)  on one leg and AC on the other, or a straight AC solution. The biggest headache with this is, funnily enough,  controlling the temperature.  Switching DC at 3000 odd Watts is serious work, unlike AC when the 20ms (50Hz) zero crossing allows a simple switch thermostat without worrying about arcing or flame quenching.  This problem has puzzled me for quite a while and although I found solutions to both the AC and DC methods, none were anything close to practical enough to leave the work bench. 

 

 

 

Along with Richard Feynman,  my other guiding light is Occam's razor.  In short, no matter how 'cludgy' and complex the first working solution is,  the very fact that you now know it is possible allows a subtle reworking of your thought patterns, and given enough time the understanding will come - and eventually you get that bath time moment and find the simple answer.  Well I got there with the water heating off grid a month or so ago.  It is still on the bench, but it is testing well and looks very promising.  Making it generic enough and practical enough to be commercial is just going to be plain old hard work.  And yes, it is just standard Paladin hardware with a small change to the daughter board to 'see' battery voltage.  That modular design again ...

 

 

 

Yes the PV has been a bit variable recently.  My records show 50-100 kWh / month variation year on year is not that unusual.  But this April saw 200+ more than last year.  629 v 415 (on a 6kW array).  February was almost as big a swing at 815 v 625.   The rest of the months, were pretty flat.  Too little data to call a trend here, but worth keeping a eye on for sure.  Anyone else notice this ?

 

 


cthombor
72 posts

Master Geek

Lifetime subscriber

  #3153476 29-Oct-2023 22:05
Send private message

As at October 2023, my iBoost+ has been running for 5.5 years.  I'm quite happy with it, but I can't give it a clean "bill of health" because I had one (zero-cost) repair.   A couple of years ago, I replaced its axial fan after an error code appeared on its front panel.  I couldn't find a replacement fan on AliExpress; so I contacted the NZ distributor (easily found: Reid Solar) -- who was very helpful over the phone.  When I told him I'd be happy with a used part (which was dead-easy for me to install) he sent me one at no charge.   Well done Reid Solar!!!

My summative review: the iBoost+ is now a rather dated product, somewhat underdesigned and overpriced compared to what's now available on AliExpress... but ... is there yet any decent alternative to the iBoost+ that's sold in NZ?   If you purchase from AliExpress you'll almost-certainly not get an SDOC and won't have much recourse if the unit fails after you first accept it.

Detailed notes:

I think the fan is a weak point in the design of the iBoost+, as it's pretty noisy.  Runs full bore whenever the HWC is supplied with even a dribble of power from my wimpy 6-panel 1.6kW PV array.  Very occasionally, in summer months, the unit heats up to the point that it shuts itself off for a while -- but that's quite appropriate IMHO because power electronics age much more rapidly with every 10 degree rise in their temperature.  Even so it seems under-designed... it really "should" be able to supply 3kW continuously for 2.5 hours, for that's (roughly) how long it takes to fully heat a 180l HWC.   But hey, nothing's perfect, and it rarely overheats but instead is doing almost-always exactly what it's supposed to do, i.e. to boost my %PV utilisation far far above what I could possibly hope to achieve without it.  Even in summer, my wimpy PV array rarely produces even half of the 3kW draw of my HWC... and it's already big enough to cover my normal household baseload "on average" except in the darkest winter months.

I'm not optimistic I would have gotten much joy on the repair of my iBoost+, if I had tried to work through the firm which installed it.  It was out of warranty... so I think it'd be best to work with the importer (Reid Solar), rather than through whatever solar installation company had onsold the iBoost+ to you.

<grumble> The website of the firm which installed my PV panels, inverter, and iBoost+ still claims that they are "Solar Leaders in NZ since 1986".  However when I contacted them recently after discovering I had 6x270W panels (from Trina) rather than the 6x300W panels (from Peimar) specified on my contract, they informed me (nicely, but firmly) that the new owners of the business do not take any responsibility for contracts signed by the previous owner!   A search through the Companies Office shows ownership changes every four or five years.  Sigh.  And even if they had been willing to honour the contracts signed by their predecessor, Marlec warranties its iBoost+ for only 2 years.  </grumble>

If you have a largish PV array (producing at least 2kW during winter), are on-grid, have a standard 180l electric HWC in reasonably good nick, and only rarely use more than 150l of hot water per day, then maybe it'd be better to go "dumber" than an iBoost+.  Here I'm thinking of a timer on a switch in your breaker-board which will fully-power your HWC for 3 hours at mid-day.  That'll leave you at some risk of breeding Legionella when (not if!) your HWC calcifies to the point that it doesn't reliably heat up to at least 60 degrees C a couple of times a week.  And you may also have to reprogram (or at least inspect the settings) on that timer after every power outage... which is also BTW very important if you have an iBoost+.  At my location, power outages are not infrequent (roughly 3x per year, very occasionally for a day or three) and sometimes very "dirty" -- to the point that my iBoost+ and other "clever" electronics can have their settings scrambled.  Fortunately my solar inverter is pretty good about "remembering" its settings through such dirty or prolonged outages!

Alternatively, you might go "smarter" ... if you can find somebody who will sell and install a home energy management system that's "solar smart" for your HWC and your EV (if you have one) and your home storage battery.  Ideally the HWC controller would have a temperature sensor, so that it can do the sort of optimisations that the Paladin v6 could do (see https://www.paladin.nz/) ... when Paladin was still actively manufacturing and selling in NZ. 

There are at least a few solar diverters of Chinese design being sold on AliExpress; some having been imported into Australia some years ago with localised branding... but AFAIK all of those importers have moved on to other lines of business.   (Phoenix companies are formed in every line of business, but seem especially common in solar.  https://provenenergy.com.au/dont-get-burnt-by-solar-phoenix-companies/)  

All to say that the question on this thread is still open AFAIK.


Terciops
21 posts

Geek


  #3153532 30-Oct-2023 02:41
Send private message

Paladin (paladin.nz)  is still here, manufacturing in NZ and selling Paladins as required.

We no longer bulk manufacture units though.  I was rather underwhelmed with the quality control of our units out of the factory and post Covid I moved the manufacturing to my workshop.  This better suited the ongoing volume of sales and has subsequently made a huge improvement to the build quality.  The other upside is that the latest Paladin firmware now incorporates a rigorous  test and tuning sequence that allows each individual unit to be 'tweeked' to provide much more accurate readings of grid, solar and transfer activity.  Paladin has always been accurate enough to do the diversion portion of its' task extremely well, but the daily totals could drift outside the nominal +/-5% I use as an acceptable limit.  

For the standard Paladin user this is never an issue, but as Paladin expanded to do more sophisticated tasks via a LoRa broadcast for 2nd tank,  remote temperature probe,  Air Con DRED control etc for specialised applications, I wanted to tighten this up.  Until the V69x firmware any tuning adjustments from standard had to be via a recompile and effectively a dedicated firmware load to match each main board and CPU combination.  Now this can be done on the standard firmware and each unit that leaves the test rig has its own tuning values stored internally.  This is total overkill for the standard unit, but as it is now part of the testing - there is no extra work and it allows a high confidence that each unit is as good as it can be when it leaves the 'shed'.

I use the Paladin FB page as a sort of ongoing tech blog (https://www.facebook.com/groups/1108024512629017) and for those who like following this sort of thing - it is worth a read.  

T


1 | 2 | 3 | 4
View this topic in a long page with up to 500 replies per page Create new topic



News and reviews »

Synology Introduces BeeStation
Posted 23-Feb-2024 14:14


New One UI 6.1 Update Brings Galaxy AI to More Galaxy Devices
Posted 23-Feb-2024 10:50


Amazon Echo Hub Available in New Zealand
Posted 23-Feb-2024 10:40


InternetNZ Releases Internet Insights 2023
Posted 20-Feb-2024 10:31


Seagate Adds 24TB IronWolf Pro Hard Drives for Multi-user Commercial and Enterprise RAID Storage Solutions
Posted 19-Feb-2024 16:54


Seagate Skyhawk AI 24TB Elevates Edge Security Capacity and Performance
Posted 9-Feb-2024 17:18


GoPro Releases Quik Desktop App for macOS and Introduces Premium+ Subscription Tier
Posted 9-Feb-2024 17:14


Ring Introduces New Ring Battery Video Doorbell Pro
Posted 9-Feb-2024 16:51


Galaxy AI Transforms the new Galaxy S24 Series
Posted 18-Jan-2024 07:00


D-Link launches AI-Powered Aquila Pro M30 Wi-Fi 6 Mesh Systems
Posted 17-Jan-2024 20:02


Newest LG 4K Lifestyle Projector Doubles as Art Objet
Posted 9-Jan-2024 15:50


More LG Smart TV Owners Set To Enjoy the Latest webOS Upgrade
Posted 9-Jan-2024 15:45


Panasonic Announces the Z95A and Z93A With Fire TV Built In
Posted 9-Jan-2024 15:30


Amazon Echo Pop Review
Posted 8-Jan-2024 14:22


Samsung Tab S9 FE Review
Posted 17-Dec-2023 08:26



Geekzone Live »

Try automatic live updates from Geekzone directly in your browser, without refreshing the page, with Geekzone Live now.



Are you subscribed to our RSS feed? You can download the latest headlines and summaries from our stories directly to your computer or smartphone by using a feed reader.