On 08 August 2016 at 2:45 PM, Weasel said:

Welcome to the forum @Farmer

What type of generator do you have. Perhaps look at some autostart options. 

Important thing is to get your baseline correct for how far you want to expand. In otherwords get your batteries and inverter right. 

Ive found too much batteries is not a mistake. You can always add more panels and other goodies but batteries not really. 

I use a 14 kw, 4cyl Lister.....is  the nu of  of panels not dependant on the size of the inverter.....and to get that right I will have to opt for the bigger inverters..3-4 kw ...and then they are usually 48 v..??

3 hours ago, Farmer said:

.....is  the nu of  of panels not dependant on the size of the inverter.....and to get that right I will have to opt for the bigger inverters..3-4 kw ...and then they are usually 48 v..??

No of panels is dependant on the load you want powered daytime.

Yes, the larger inverters are 48v.

One little thing to know, if you have a lot of panels, and less batteries, the charge current must be set lower to not overcharge the batteries.

I am sorry this thread went south again. I hope you understood what I was saying re. loads that match batteries and panels with extra panels daytime to power more.

EDIT: Inbetween all the opinions, sometimes we lose track of a post where it is said what needs to be powered like fridge and lights only on batteries, rest daytime using panels, for the lister is going to run in any event UNLESS you go all out and try to use the lister as little as possible? But that is a huge upfront capital outlay for more panels and batteries to be able to replace what the lister can do at night.

2 hours ago, Farmer said:

is  the nu of  of panels not dependant on the size of the inverter.....and to get that right I will have to opt for the bigger inverters..3-4 kw ...and then they are usually 48 v..??

not exactly, i can try explain by example. if i had a 3kw battery based inverter with a built in mppt such as hybrids or solar inverters or something like the infamous axpert then the mppt inside it is large enough to power the 3kw output.... usually anyway. That however doesn't have a bearing on the amount of panels you choose to put on it. each manufacturer will have minimum and maximum recommended wattage's for the PV input but if i chose to have only 3 x 300w panels in series for one string on the 3kw then it will do with the 900W as much as it can. On the flip-side if i choose to put 5 strings parallel like the one above on it for 4500W. it wont ever charge the batteries at that 4500W instead limit the power to the internal mppt's rating. in this example it could be like 60A into batteries. in other words the inverter wont blow up if you have too much panels, unless its crap and cant handle running on the limit the whole day.

This is one of the nice things when you consider a more modular system where the MPPT is not part of the inverter and is bought separately, It does always work out more expensive if you do but the advantages are obvious. its more "professional" and "industrial" in terms of quality, serviceability, support and so on. You can also choose to expand in only one area instead of a whole unit e.g. to increase PV only by adding another mppt. Not a massive advantage since the alternative is very cheap but its a difference that needs mentioning. one important thing to be aware of here is how well an inverter handles a generator input and with stuff like victron the support is very good, they where designed for it. i don't really know about others, needs some looking into as i said. they have some programmable relays where you can start a generator based on your conditions.

all things considered i would look at 8 x T105 RE or similar batteries, you wont get away with less.I would prefer a 3kw victron multiplus and an mppt with around 1kw pv for now. so either a 24v config with an mppt 100/50 and 4 x 250-300W panels or 48v config with 150/35 and 3 x 300 later add another 3 panels. 

cheapest would be to go with the Sunniva, it could serve all your needs, im just not sure about generator in, too lazy to go and read up right now.

 

9 hours ago, Farmer said:

Deepbass..I think this is how my system should look like yours.I would like to go for more panels to directdrive my freezer,washing  machine computer and charging during daytime and use the gennie for pumping water and power for the workers for 3 hours at night.I only need the batt for the lights and freezer during nighttime.

What area are you in? You could come and have a look if close by. I am between Magaliesburg and Koster. 

I have opted not to have auto-start. Battery voltage or SOC being a poor indicator of whether the gennie needs to run or not. An example is that you run the batteries down more than normal and the battery voltage drops more than normal. A SOC cutoff autostarts the gennie but if you look outside it is going to be a sunny day and the energy to recharge the batteries should come from the panels but now your system has started the gennie.  

A hybrid inverter with a built in MPPT allows one to expand the PV array in stages as @Weasel has pointed out. 

1 hour ago, Chris Hobson said:

... is going to be a sunny day and the energy to recharge the batteries should come from the panels but now your system has started the gennie.

I have this exact same scenario where SOC is set and you see the sun is there so system must switch back to solar long before set SOC is reached. Same as gennie that must not start based on SOC.

Found a little trick with the BMV. Just need time to have it developed that the SOC setting is ignored in favour of more power available so batteries can be charged whilst load is powered using solar as quick and early as possible each day, yet if there is insufficient sun, let it all be.

8 hours ago, Weasel said:

so either a 24v config with an mppt 100/50 and 4 x 250-300W panels or 48v config with 150/35

If Weasels suggestions are considered, would suggest only the 150v range of MPPT's. Gives more room to maneuver for the 100's are 24v max, the 150's are 12/24/48v.

Buy right once on the controller, and it will serve you for many years to come.

Chris, i think the option i had with the multi is not exactly what farmer had in mind. My guess is he is willing to spend around 25 to 35? its just what i would have done and what i considder the best option. But i want to explore your reasoning behind auto start. I dont have much need being city dweller so its hypothetical to me.

Would an soc of 50% charging up to say 70% not suffice in this situation. Realise of course you want to save genny time and give solar a chance to charge and your not taking into account day and night in the start up.

Or if its set on a timer for say 5pm start, until its full. Then have the low soc and a high current draw starups as a backup but only up to an acceptable soc or until load ends.

My question is that is there some conditions in your experience that cant be solved by some logic. Do you consider it a luxury and like the manual control. Im just curious really.

Sent from my SM-G900F using Tapatalk

27 minutes ago, The Terrible Triplett said:

If Weasels suggestions are considered, would suggest only the 150v range of MPPT's. Gives more room to maneuver for the 100's are 24v max, the 150's are 12/24/48v.

Agreed absolutely, just that if you shop around you can get a 24V 3000 for a lot less than the 48, then by extension you'd get the 100/50 because you'd want the amps. By the way i wouldn't do more than 3000 on 24V. 

 

I gave quite an extensive discussion on autostart on one of the threads. Basically if you are off grid, but on the farm, it is better to choose when to use the genny, as an autostart can't anticipate what loads you are going to use, or what the weather is going to be like. Better to switch on or off based on experience and anticipated need. Autostart has no idea if it is day or night, sunny or cloudy, you are welding for 5 mins or 5 hours etc. Also if you need autostart to tell you that your batteries are flat, you'd better have another cup of coffee in the morning.. 

23 minutes ago, Weasel said:

By the way i wouldn't do more than 3000 on 24V. 

Agree 100%.

33 minutes ago, Weasel said:

... then by extension you'd get the 100/50 because you'd want the amps

This is also a very good little caveat to note, if you want to get to 500ah battery bank.

For again this comes right back to what is the load powered on batteries and for how long? It makes a HUGE impact on the rest of the equipment and the bank voltages.

Powering 5000w daytime when there is sunlight and only average of 50-100w per hour at night, get a 5000va 12v Victron Multiplus. 

 

Just a pearl of wisdom. Having bought myself in the past into corners I leaned that with controllers always get the one with all options like the 150's for maybe you smokkel an inverter one day that is 48v, needs change. Last thing you want then is to replace a controller also.

With more choices there are more "responsibilities".

The line in the sand for the 150's are the amps they can push. A 150/35 is still at a ok price for up to 350ah bank, above that, eina magtag, but then if you go above 350ah bank, carefully check your needs and wants. And do check the array limits. Moet nie 'n TTT trek met controller met te veel panele op 12v. 

So if you are a 12/24v man, and you are SURE it will not change, then the 100/50 is a very good option up to 1000ah bank, YES, the 2v bank, here we go - T&C apply off course. 

 

18 minutes ago, DeepBass9 said:

I gave quite an extensive discussion on autostart on one of the threads. Basically if you are off grid, but on the farm, it is better to choose when to use the genny, as an autostart can't anticipate what loads you are going to use, or what the weather is going to be like. Better to switch on or off based on experience and anticipated need. Autostart has no idea if it is day or night, sunny or cloudy, you are welding for 5 mins or 5 hours etc. Also if you need autostart to tell you that your batteries are flat, you'd better have another cup of coffee in the morning.. 

Deepbass, your experiences are well noted for nothin like using stuff to have experiences.

What Weasel and I, out of pure interest are saying is that with some sort of electronic intelligence, can it maybe be automated? We all have BMV's.

Autostart has everything to do with SOC of the batteries, day or night. IF the batts get to 50% SOC or below, you are breaking them, maybe it is a crisis or whatever, so then auto start the genny to protect the batts?

If you are going to weld, start the genny, if you see there is not enough sun yet, start the genny. Manual override is still there.

And once gennie runs let it power the load and recharge batteries till SOC is back up to say wot 70/80%, then auto switch genny off, for once the gennie has started, use all it can give, max use of each liter of diesel. 

Hypothetically, will this work in the real world?

PS. With auto cutt off if diesel is below X in tank or temp of gennie goes above Y - just sommer.

If the controlling software is a bit smarter and can monitor a few variables, then yes it will work. Even something as simple as a timer to manual start the genny and then turn it of automatically after X hours would be useful for me (easy enough I suppose, just haven't got round to it). You would need battery SOC, time of day, expected solar kwh for the day (i'm not sure if you can look up the weather on the net and get an estimate from somewhere), expected extra load, etc. Then diesel in the tank and expected run time at a specific load. Gets a bit complicated but could be done, would need a bit more than some baling wire, a dripping tap, a bucket and a counterweight though! Maybe someone smart can put together something based on an arduino or a raspberry pi.

Probably the most important part is to figure out for how long the genny must run, compared to what you would have got from solar. So if batteries are at 50% at dawn, there is no reason to turn on the genny as the sun will charge them, but at 50% at dusk, then time to start the genny. Or 50% at dawn, but partly cloudy weather expected, what then?

With you.

I, as you, am "off-grid". Only difference between us, is that my "gennie" is Eskom, and my rule is do not go below X SOC unless there is no Eskom and no solar. One small caveat I NEVER use Eskom ("gennie") to charge my batts. Makes it very interesting and the measuring of it all become very important.

My problem comes in during the mornings where at one point in the morning the SOC is still low, panels produce enough power to charge and power the load. I now need to switch my "gennie" off and run on solar for the batts can take longer to re-charge.

After having had a look at measuring local irradiation and pulling weather data from reputable sites and all that, it got very complicated and costly very fast, if you do it right. Benefit value did not compute.

In the end I deduced that the BMV SOC of the batteries is the one and only bottom line, with manual override - which I then also automated. If "gennie" is running to protect batts, because SOC is below X, and "gennie" stops, then switch power back to inverter and run until "gennie" comes back on or inverter switches off due to low batts, obviously reducing load before then to core use only.

Timers are easy, take i.e. a swimming pool pump timer, start gennie at X time, off at Y time, if Z condition is true, with manual override.

Z Condition is the BMV relay so if BMV relay is triggered due to SOC say below 50%, and it is between X and Y times, gennie must start, otherwise not. Will that work?

So I am going to see what I can cook up in the next few months and report back, to "start" stop my "gennie", that it is used as efficiently as possible, taking into account the solar panels and what they produce, for I THINK I have seen the pattern.

Then one day, as I do not have a Multiplus, I am going to split the loads that must be on if Eskom and solar is off, that only the very basic core loads are powered, that the batts can carry them for many hours with no strain.

2 hours ago, Weasel said:

My question is that is there some conditions in your experience that cant be solved by some logic. Do you consider it a luxury and like the manual control. Im just curious really.

I like the manual control. I have been turn on a gennie all my farming career so it is not a big deal. We have a stop/start button behind the front door so it is as easy as pressing a button. Human logic beats any system hands down. I can evaluate sunshine levels, anticipated load and timing of the load in seconds and make appropriate call on whether we need gennie power or not. It is more difficult for a urban dweller but remember I walk 30 m to work. Gennie power is expensive so not something I want to leave to logic. Autostart for hospitals and critically electronic systems by all means but on a farm? Generally there is someone who I can delegate to switch the gennie on or off if I am elsewhere on the farm. I can look at PV Output and make a call if I am away.

11 hours ago, The Terrible Triplett said:

No of panels is dependant on the load you want powered daytime.

Well yes, but that is not all. The constraints are multiple. You want to:

1. Power day time loads
2. Have enough left to recharge the batteries
3. Total power into batteries must not exceed 15% of C20 rate, and you need to remember that the inverter might trip on overload/error so this also applies during zero-load conditions.
4. Have enough battery to get through the night.

This usually means that there is little point in having more PV than the recommended charge rate of the batteries... or at least... that is what it meant in the old days, which is why I'm also going to talk about the "separate MPPT" @Weasel suggested. The constraints were rather tight, you had to size the PV array to the battery (by the 15% rule) and then you sized the PV/battery combo against your load.

But now with these hybrid inverters, you can adjust the PV power so that it dials back the MPPT if there is no load, and therefore keeps the batteries safe. That is definitely one upside to them.

 

10 hours ago, Weasel said:

This is one of the nice things when you consider a more modular system where the MPPT is not part of the inverter and is bought separately, It does always work out more expensive if you do but the advantages are obvious.

But... like Weasel I prefer them separate. I don't like Multifunction devices. Like that huge device in my home that can scan pages and send (but not receive) faxes, looks rather a lot like a laser printer but can't print a page to save its life, nor can it be repaired. Who even sends faxes anymore? And stuff do break and if they don't, it's just less flexible. But when you operate them separately, you cannot open the taps and size the array towards the day time load... you're once again back to that safe 15% of the C20 rate.

The only way I know around this at the moment is Victron's MPPTs. You can dial them back by writing a current percentage to the relevant register using the Vedirect Hex protocol. So you can open them up flat-out during the day, but if the loads are disconnected suddenly it will adjust the MPPT back to a safe level. That means you size your PV array to daytime load plus 13% of the battery capacity so you can recharge them.

In this setting, the hybrids really shine, and they are cost effective. But give the blue team some time... the CCGX software already runs on the beaglebone black, and soon it will run on the Raspberry Pi (well it sort of already does, but it is terribly buggy).

For now, just understand the constraints when you design the system. If it has the capability to adjust the MPPT to keep battery charge at a safe level AND consume the power as it comes in, then you can install more PV and match the PV to the day-time loads. If it doesn't have that capability, then you have to size it to the battery capacity, and you might end up having to increase the battery capacity to fit inside the constraints.

In one word: FUN! :-)