Excess solar to heat water

[viper_za]

Okay so let's start with what I have.

225ah 48v battery bank, 1800w PV (6x 300W), BMV702, Axpert 5kva wired into my DB for all loads except the stove and geyser.

My average load per hour during the day for the fridge and freezer sits around 102-150w the highest I have seen recorded so lets leave this at a constant 200w draw to make it easy.

The plan add another string of panels pushing the PV up to 2700w  (should be enough for my the iron even then)

Change the 200l geyser element from the standard 3000w to a 1500w ac element http://www.geyserwise.com/products/elements/1500w-ac-element

Then I'm still deciding on what to use as the timer for this maybe one of these http://www.geyserwise.com/products/geyserwise/geyserwise-max

 

Now the trick comes in when this needs to be turned on.

What I have come up with so far is the following

Use the relay in the BMV to control this but not based on SOC but rather on the voltages using the following options.

18. Low voltage relay

When the battery voltage falls below this value for more than 10 seconds the relay will close.

19. Clear low voltage relay

When the battery voltage rises above this value, the relay will open (after a delay, depending on setting 14 and/or 15). This value needs to be greater than or equal to the previous parameter.

My float voltage sits at 52.8V so if I set the low voltage point on 52.6V and the clear low voltage on 53v the following should happen

My battery bank gets reaches absorb stage at around 10:30 after being drained to about 80% SOC during the previous night.

I set the timer to start the geyser element at around 10:25 also this means the that battery voltage will take a dip the moment it starts up as the element has a inrush current of  ≤15 Amps

The time the voltage will dip will be less than the 10seconds the BMV allows for the low voltage as the MPPT controller should be able to ramp up the PV production faster than the 10seconds if it is available, if not the relay will close and the geyser wont run.

 

Now from the relay to the geyser I need some help in how to control this maybe with a contactor (need advice here)

What I want is to be able to turn on the geyser at night\mornings also if needed (some sort of bypass for the relay)

 

Got my hard hat on please shoot down what you think won't work etc

Thanks

Thys

 

PS I did not proofread this

 

 

[Guest]

I want to add: Set the Geyserwise to only switch element on between the hours of 11am and 3pm every day.

Then you use the battery SOC for the relay?

[viper_za]

13 minutes ago, The Terrible Triplett said:

I want to add: Set the Geyserwise to only switch element on between the hours of 11am and 3pm every day.

Then you use the battery SOC for the relay?

Yes the off time I forgot

The reason I don't want to use SOC is I really don't want to drain my batts with 1500w load in bad weather.

[Guest]

And that Sir Viper, with the tablet, is why you use the SOC. 

If you set the SOC relay to change back if SOC is say 95%, or 99 or 90, you decide, it switches off the feed to the geyser.

Once it reaches 100%, or 95, or 98, again, you decide, it switches geyser on again, provided the times are met as per the Geyserwise settings.

[viper_za]

Yes but then what if I run on batts at night then I need to go change that SOC again, automate you always say 

[Guest]

No man, I thought you had a cheap 2nd hand BMV600 there to use for the geyser!? 

They are cheaper than tablets. 

 

[Guest]

In my mind there are 2 x BMV's. One for the master system and 2nd one for the geyser element activation (using the SOC of the batteries to set the relay on/off), if the Geyserwise time schedule requires it to be on.

[Chris Hobson]

My installation is not complete and switching manually at the moment. The loft is just too hot to spend any length of time up there.

I have gone with one of these

http://www.ebay.com/itm/Multi-function-Voltage-control-relay-timer-delay-switch-voltage-protection-12V-/271832274268?hash=item3f4a77555c

If there is enough power and battery voltage is rising the relay switches on and if there is not enough power battery voltage falls and it switches off. No SOC light intensity monitor or anything like that needed.

Here is usage for today. I switched on a bit late and one can see what was happening before I installed the geyser. PV production is starting to drop already due to the batteries approaching 100% SOC. 

[Guest]

Chris, correct me if I am wrong, you are using increasing volts to determine when the switch on?

So if a bank of clouds comes past, geyser stops heating?

[viper_za]

4 minutes ago, The Terrible Triplett said:

In my mind there are 2 x BMV's. One for the master system and 2nd one for the geyser element activation (using the SOC of the batteries to set the relay on/off), if the Geyserwise time schedule requires it to be on.

See TTT my problem with this is you will be exceeding the battries C20 rate even if it is only for 5-10minutes 1500w + my 200w load, thats +-35A on a 48v bank

We looking at +-70 for your 24v bank

If I can avoid that Im sure I can extend my battery life

[viper_za]

2 minutes ago, The Terrible Triplett said:

Chris, correct me if I am wrong, you are using increasing volts to determine when the switch on?

So if a bank of clouds comes past, geyser stops heating?

The moment the cloud passes it turns back on again, I don't see a problem with that.

I think the way I'm looking at it will do the same

[Chris Hobson]

10 minutes ago, The Terrible Triplett said:

Chris, correct me if I am wrong, you are using increasing volts to determine when the switch on?

So if a bank of clouds comes past, geyser stops heating?

Exactly -  relay just over 200 bucks BMV just over R2k

 

[Guest]

5 minutes ago, viper_za said:

The moment the cloud passes it turns back on again, I don't see a problem with that.

I think the way I'm looking at it will do the same

Jip, I agree with that. Your logic makes perfect sense.

Why I prefer the SOC method is that the batts always get preference. Once SOC is reached, then I add the other loads, for 9/10 times my setup just provides more power, with batts staying on 100% SOC.

[Guest]

10 minutes ago, viper_za said:

See TTT my problem with this is you will be exceeding the battries C20 rate even if it is only for 5-10minutes 1500w + my 200w load, thats +-35A on a 48v bank

We looking at +-70 for your 24v bank

If I can avoid that Im sure I can extend my battery life

Good point!!!

My thinking is that my loads are 600w off 930w panels. So I add 1500w more panels for geyser. 

Batt SOC is 100%, so when it is 11am, the geyser comes on and the power hop skips and jumps from panels over batteries to inverter to geyser.

If SOC is say 98-99%, then switch off the power to the geyser. 

[Guest]

Something is bothering me about that video. On that day I have a spesmaas he was draining his batteries very slowly.

16 amps in, 20 amps out, as he said he is borrowing from the batteries.

Now the system switched every few seconds based on volts, the 16 amps coming in cannot possibly put the equal amount back in, that was taken out, due to inefficiencies.
 

[Chris Hobson]

With LA batteries there is much more "movement"  in terms of voltage. Voltages above 53.3V switch on and voltages below 52.9 switch off (for my settings). As soon as you are not charging voltages fall to 52.5V even at 100% SOC. So you do not lose any gains made in terms of battery charging.

[Guest]

Sjoe, I must be careful now to not come across as an complete idiot, my wife knows that already, just hiding it well here. 

Thinking of my system, on 24v the margins are much closer than on 48v, so the effect can be seen much faster between 26.45v and 26.65 with fully charged batteries as 26.1v, if I switch off all the loads ... it gets very interesting.

I have never seen my system on 26.1v ever, unless there are no loads (never happens) and as you say, the batts are full.

During the day I am looking at +-19amps at +-32v coming in with +-600w shown on inverter going out. At the perfect time of the day I will see zero amps going into or coming out of the battery. As the day moves on I see a slow increase of amps going out of the batts. (Figures not exact.)

Starting at less than 100%, even at full solar panel power during the day, I  won't recoup the losses I started with, if I power the full load.

And that brings me to the conclusion: If you run your system using every ounce of power coming in, you cannot use volts ever to add a load in case there is spare. In using SOC, you can try, If there is spare.

But with light loads during the day you can use the volts to use switch loads on and off ... and THAT reminds me again why I am always hammering the point of SOC of the batteries to determine what you can power when.

I am always measuring against my system, running a very tight margin with panels specced to, after losses, power X items in winter, and Y items in summer, to get the max ROI, also had to do with budgets at the time, provided me with a very steep learning curve on wants versus needs. 

In your case Chris, you are using your spare to power your geyser. I NOW understand your logic even better.

[viper_za]

See what I'm planning on doing is going to have almost no effect on the batts volts except for the moment the geyser starts and the MPPT ramps up the PV production to match the load.
Since I will be setting the geyserwise to start almost just before my batts hit absorb stage they wont be needing the full 20A I have been pushing into them from early mornings.

You will have almost no voltage movement downwards (more than 10seconds anyway) on my method except if the PV cant provide the load and then the BMV relay will close and turn off the geyser

Remember I said I will be adding another 900w of panels ontop of the existing 1800w, leaves plenty to spare to even keep charging the batts on a good day with the geyser running. On a bad day I will just be charging batts and heating water with Eskom at night.

Think you should loosen the margin with regards to panels they give you more on bad days and even more devices to run on good days. (Like that AC of yours)
Wait till you hear what I'm planning on Phase 4 of mine 

[Chris Hobson]

After a bit of monitoring I find that on a sunny day there is enough power to supply the geyser and for 6-9A to go to battery. So  I just have an incredibly long bulk phase which means my batteries are only fully charged at mid afternoon. You want to start you geyser once the panels have met your max charging rate. I have 3kW on panels and I draw down my 260 Ah batteries to about 80% by morning. So come 9 am we are clicking along nicely and then the PV production starts dropping as once starts going into absorb. Look at the graph above where PV production drops from 9:30 to 10 when the geyser kick in (manually). Immediately you can see PV production picks up and carries on following a standard bell shape curve. With a smaller element time is not on you side you want to start heating water with the first available excess watt.

 

One advantage is I have a piss-willy charging rate and seems funnily enough to enhance battery balancing and we enter true float a little sooner. The BMV seems to think we have some distance to go but the battery charge rate tell me different story.

 

In your case Chris, you are using your spare to power your geyser. I NOW understand your logic even better. Sjoe you sure you want to credit a farmer with logic ?

[viper_za]

Yes Chris I will definitely play around with times to start it, a learning curve like everything with the solar but in the end I'm sure I can figure it out. Might adjust the voltages and times to start heating etc

I'm sure with a 2700w array I should be able to manage it after a little playing around.

Now I just need to decide on what I need to purchase to start the geyser with the help of the relay

[viper_za]

Okay so the spec for the relay on the BMV 702 states the following.

60V / 1A normally open (function can be inverted)

Does anyone know if that is just the Max Voltage it can do or will I be okay if I use it in conjunction with something like this?

http://za.rs-online.com/web/p/non-latching-relays/7892708/ only 24V coil

[Chris Hobson]

That would do it I think. You can either power it from the batteries if you have 24V system of if you have a 48V battery bank then buy a doorbell transformer. You have to have a full bridge rectifer as it still AC and that is how they get the doorbell to buzz.

[Guest]

I used these 220v AC relays, that are triggered by the BMV's relay.

 

[Chris Hobson]

33 minutes ago, The Terrible Triplett said:

I used these 220v AC relays, that are triggered by the BMV's relay.

I was wondering about that from earlier posts. The data sheet says 60V 1A. What you telling me is all the extra bits and pieces are unnecessary. Despite what the datasheet says the BMV can handle 220V AC? Interesting.

 

Nice and neat install by the way.

[superdiy]

1 hour ago, The Terrible Triplett said:

I used these 220v AC relays, that are triggered by the BMV's relay.

The thing is, it might work, for a while, and then one day the BMV's relay's contacts will fail and then who are you going to ask to repair it? And what will it cost if you have it repaired by a Victron repair agency?

The 1A 60V is the relay's contact rating and should not be exceeded.  Running higher than recommended voltages through the BMV is also dangerous, because it might arc over between adjacent tracks / pads on the pc board inside the BMV.

Rather use a relay with a lower coil voltage e.g. 48V.