System check and a couple of question...

bobc · June 2013

Hi folks,

Great to see that there's a strong, supportive community here at wind-sun...

So, I, unfortunately, have to come up to speed on solar in short order and just so I don't make any blunders, can I run this system by you? And, there's a couple of questions that I can't find answers to. I'm sure they're out there but they are just buried deeper than I've had time to dig for...

For my first solar project, I'm wanting to run a water feature pump. You'll note that the water pump and inverter are 230v/50Hz. That's because I'm buying everything in the US, where it's cheap, cheap and shipping it (along with everything else I own) to New Zealand, where everything is super, super expensive. This is happening in the next couple of weeks. So, I'm trying to learn some things about solar while I'm trying to do everything else associated with a major move. That's why I haven't been able to spend boatloads of time searching.... Here's the system as I see it:

--Water pump: 230v 32w 0.14A Run time about 12 hours/day. But flexible.
--Pump will be 50' away from controller so, wire will be THWN, maybe UF, 14AWG.
--Inverter will be 12v to 230v "pure" sign wave, 300w. It will be a cheap eBay special, given the application. I also thought I'd get something with higher wattage than needed in the event that we ever lose mains power (very rare), I can run the pellet stove (~100w when running) for the evening.
--Generic charge controller 12v, 10A (I'm buying a Renogy panel and a Renogy controller comes with it). This controller appears to be re-branded many resellers (10A). http://www.ebay.com/itm/271211938333?ssPageName=STRK:MEWNX:IT&_trksid=p3984.m1497.l2649
--Inverter will be next to controller so, 14 AWG wire between the two.
--80AH 12v battery located next to controller
--14 AWG wire between battery and controller
--One panel ( Max power: 100w; VOC: 22.5v; VMP: 18.9v; IMP: 5.29A; ISC: 5.75A) This panel has a great price point and very good reviews on Amazon...
--5' of MC4 10AWG wire to get from panel to inside house. Then, 30' of generic 10AWG to get to the controller (single story house).
--Christchurch averages 5.77 hours of daylight over the year, 4.0 hours in winter.

How's it all look? When I did my battery calc's, I planned for a 60-70% discharge versus 50%.

Here's the questions I can't find answers to:

How do you physically make the connections at the controller? The terminals on this controller look like it can handle 16AWG but not anything bigger. How am I going to attach 10AWG to it (I had originally spec'ed out a 12v pump requiring 4AWG wire but it was too cost prohibitive. But again, I got to wondering how to make this connection)?

The second question is about the load disconnect. I'd like to be able to set the system so that when the battery hits 60% or 70% (or 50%) that the pump would automatically shut down. Are there reasonably priced controllers or some other unit that do this? And, how can it be that this and other controllers have the load disconnect at 11.1v? Wouldn't the battery be shot by that point?

Sorry for the long email but I wanted to lay it all out and give you all the info...

Cheers,

-bob

NorthGuy · June 2013

Re: System check and a couple of question...

If you're going to order things from the US, the voltage standards over there are different. Not 230V/50Hz, but rather 240V/60Hz, or for these smaller loads 120V/60Hz.

The panel looks a little small, both relative to the pump and batteris. You need about 400Wh for the pump (or 500Wh if you factor in Inverter consumption). With PWM controller, the panel cannot produce that much, even not counting losses. It would be good to have 10-12A of charging current for batteris. The panel can produce only half of this. If I were you, I would use 2 panels.

bobc · July 2013

Re: System check and a couple of question...

Thanks for the reply! I waited a couple of days to see if anyone else would add thoughts.

So, what did I miss that this panel is too small? Using several worksheets out on the 'net, they seemed to lead me to having just one 100w panel.

Also, answers anyone, for the two questions?

thanks again!

vtmaps · July 2013

Re: System check and a couple of question...

bobc wrote: »

So, what did I miss that this panel is too small?

Tell us what math you used, and we can tell you what you missed.

Five hours of sunlight on a 100 watt panel through a PWM controller is NOT going to produce anywhere near the 500 wattHours that you need each day for the pump... and you also need to factor in the battery inefficiency.

--vtMaps

Cariboocoot · July 2013

Re: System check and a couple of question...

Well let's do some basic math.

You want to run a 32 Watt load for 12 hours. That equals 384 Watt hours. If you run this from a 12 Volt system that will require (384/12) 32 Amp hours. But you also have to include the inversion efficiency and the inverter's consumption. So you get a number closer to 432 Watt hours (depending on the exact inverter) and that means 36 Amp hours @ 12 Volts.

You also do not want to drain the battery 100%, or else it won't recharge. 25% is the standard for value over lifespan of the battery, so the battery needs to be 144 Amp hours. You've picked one that is 80. It will work, but you will be discharging it deeper (about 45%) and will need more power to recover from that discharge depth. Your 60-70% planned DOD is a sure-fire way to end up replacing batteries often. They are not designed to take that sort of DOD on a regular basis.

Now let's look at recharging. Typically setting the array to have a peak charge current of 10% of the battery capacity will bring everything in line using 25% DOD. This is a short-cut, not an absolute. In this case you'd have 14.4 Amps of current, which one 100 Watt panel is not going to provide. It isn't going to provide that 10% for the 80 Amp hour battery either. Why does this matter?

A 100 Watt panel will not actually produce 100 Watts during all hours. Perhaps not during any. With a PWM controller you can expect it to produce its Imp rating * the battery Voltage, or about 60 Watts in this case. For that output to provide the 432 Watt hours of load & inverter you'd need (432/60) 7.2 hours of equivalent good sun. No one gets that regularly anywhere on Earth.

Back to the rules-of-thumb. At 10% peak current for the 144 Amp hour battery you'd have an array of about 252 Watts. By the same calculation that should produce around 172 Watts and be able to provide the 432 Watt hours in 2.5 hours of good sun.

Now here's the thing: you won't find a battery that is exactly the number of Amp hours the calculations suggest, nor will you find panels that will exactly match the array sizing, nor is every day as sunny as the last one. So you have to have some fudge factor available if you want to make it work consistently throughout the year. Even things like what hours the pump will run in make a difference, as you can increase the efficiency of the system by making use of panel power that would otherwise go unrealized once the battery is charged.

It isn't so much a matter of "will this work?" as "how well will this work?" Cutting corners on equipment brings dissatisfaction with performance. Long-term you will not be happy with El Cheapo charge controller/inverter/panel.

bobc · July 2013

Re: System check and a couple of question...

Perfect!

This is the type of info I was looking for! Thanks guys! The online worksheets that I used only incorporated inverter inefficiencies, a daily temp value, no mention of recharge calcs, and 50% DOD. Every one of them used 50% values. So, this is great to see inefficiencies incorporated. I also appreciate knowing that 25% is a more realistic value to use.

Cariboocoot, there is one calc that I can't see how you got: "...At 10% peak current for the 144 Amp hour battery you'd have an array of about 252 Watts..." Did you include an inefficiency in this calc?

Where is there a good worksheet to use as I re-calculate and create a system to meet the needs?

I still don't understand what part of the system keeps the battery from not discharging below 25%...

Thanks again, -bob

Cariboocoot · July 2013

Re: System check and a couple of question...

bobc wrote: »

Cariboocoot, there is one calc that I can't see how you got: "...At 10% peak current for the 144 Amp hour battery you'd have an array of about 252 Watts..." Did you include an inefficiency in this calc?

14.4 Amps is 10% of 144 Amp hours. Multiply by 17.5 panel Vmp = 252 Watts. That's how you jump from charge current to panel size on a PWM type controller in one go; utilize the panel Vmp as the multiplier.

Where is there a good worksheet to use as I re-calculate and create a system to meet the needs?

Piece of paper, pencil, calculator, re-check on the forum. Every one of those spreadsheets I've seen has a problem in it somewhere. At best they can only get you in the ballpark, so you can organize the data how it makes sense to you and do the math manually. If I were to design one it would have separate cells for "rounded up to available 'X' capacity" for batteries, panels, charge controllers, et cetera.

I still don't understand what part of the system keeps the battery from not discharging below 25%...

You do, Bob.

By not using more Watt hours than that. It's calculated to be a daily average, so 30% one day won't ruin everything and a bad day of sun may take it closer to 50% but on the whole it averages 25%. You can keep it above 50% SOC with a programmable inverter: set the low Voltage disconnect to nominal battery Voltage.

bobc · July 2013

Re: System check and a couple of question...

Great! So, if I can indulge you in reviewing this to see if I understand what you taught me... Here's another example using the same panel and pump. Let me see if I can parrot this back:

32w pump running for 8 hours a day. 256 watt hours needed.
Inverter inefficiency is 12.5%. So, 288 watt hours are actually needed.
With a 12v system, 24 amp hours are needed.

Depth of Discharge for batteries is limited to 25%. So, a 96 amp hour battery is needed.

This panel puts out 5.29A (IMP) when the battery voltage is 12v, so the effective output is 63.5 watts (using a PWM controller).
In Christchurch, over the course of the year, the daily amount of sunshine is 5.77 hours. This one panel then puts out 366 watt hours per day. This is more than the 256 watt hours needed. In winter, the average daily sunshine is 4.0 hours. The panel puts out 254 watts. So, from this, one panel works.

But, the one panel can't keep up with recharging the battery: 10% of 96AH is 9.6A. This one panel only puts out 5.29A. Hence, two panels are needed.

However, using the 10% of peak current use rule of thumb, the array size would need to be 9.6 amps x Vmp (18.9) = 181 watts. Since one of these panels is rated 100 watts, but effectively puts out 63.5 watts, would two or three of these panels be needed (The Vmp thing is still fuzzy. I'll have to do some research on this)?

I also get the part that the amount of sunshine/output varies so, nothing is hard and fast and that varying the run time of the pump and wanting to use the power being produced when the pump isn't running and the battery is charged.

And for a belated response to your comment NorthGuy: I'll be buying the pump in NZ as I haven't found any 230v/50Hz available in the US. Surprisingly in the US, quite a few things are available at this voltage and with 50Hz, but pond pumps are not one of them. I had originally thought I'd save money by buying a 12v pump here in the US. Then, I sized the wire needed to run 12v 50' and there went the cost savings...

Cariboocoot · July 2013

Re: System check and a couple of question...

Are you familiar with the term "here's where it all goes wahoonie shaped"?

First of all, you probably aren't going to find a 96 Amp hour battery. So you have to either round up to a size you can find (preferable) or round down (not so good) and suffer the consequences of greater DOD and shorter battery life. How much so will be a gamble. For critical systems you definitely do not round down. Remember that battery capacity will decrease over time no matter what.

Now you've got your really available battery size. Let's say it's this one http://www.solar-electric.com/repoba12vo11.html at 115 Amp hours. So you calculate the recharging based on that real battery capacity.

Now if the pump is not running while the battery is recharging you can use a lower peak charge current: part of that 10% target is compensating for loads that draw at the same time. For example some nighttime-only lighting would only need the minimal 5% rate and assurance of enough hours to "put back" the Amp hours used. Unfortunately that is not linear, as batteries tend to need 20% more power back in than you can take out again. Again, that 10% peak target tends to compensate for this.

So let's try 11.5 Amps * 17.5 Volts and get 201.25 Watts. Hey, it looks like two 100 Watt panels, doesn't it?

Yes, it may technically be more than you need. At least on some days. On other days it won't be enough. That's when you think about back-up plans. One of those can be a generator, or even more panel - depending on the install.

I think most of the on-line calculators your were looking at were using the very basic end-to-end Watt hours formula for converting array size Watts into AC output Watt hours. It looks like this:
Array size in Watts (nominal) * hours of equivalent good sun * over-all system efficiency (typically 52%) = Ac Watt hours per day.

It's a bit daunting, but having to store energy in batteries and take it out again is not terribly efficient. The number can be improved by utilizing power during the day when the batteries are fully charged.

BB. · July 2013

Re: System check and a couple of question...

bobc wrote: »

This panel puts out 5.29A (IMP) when the battery voltage is 12v, so the effective output is 63.5 watts (using a PWM controller).
In Christchurch, over the course of the year, the daily amount of sunshine is 5.77 hours. This one panel then puts out 366 watt hours per day. This is more than the 256 watt hours needed. In winter, the average daily sunshine is 4.0 hours. The panel puts out 254 watts. So, from this, one panel works.

But, the one panel can't keep up with recharging the battery: 10% of 96AH is 9.6A. This one panel only puts out 5.29A. Hence, two panels are needed.

I like to do it this way (just a consistent set of math that works pretty well):

100 watt panel * 0.77 controller+panel deratings * 5.77 hours of sun = 444 Watt*Hours into the battery bank
100 watt panel * 0.52 end to end system eff * 5.77 hours of sun = 300 Watt*Hours of available "solar" power on sunny summer day

However, using the 10% of peak current use rule of thumb, the array size would need to be 9.6 amps x Vmp (18.9) = 181 watts. Since one of these panels is rated 100 watts, but effectively puts out 63.5 watts, would two or three of these panels be needed (The Vmp thing is still fuzzy. I'll have to do some research on this)?

Assuming 5-13% Rate of charge, with 10% being a healthy nominal. The array size for a 96 AH @ 12 volt battery bank should be around:

96 AH * 14.5 volts charging * 1/0.77 panel+controller derating * 0.10 rate of charge = 181 watt array "nominal"

Number came out "identical"--What are the chances of that? :cool:

I also get the part that the amount of sunshine/output varies so, nothing is hard and fast and that varying the run time of the pump and wanting to use the power being produced when the pump isn't running and the battery is charged.

Automating a solar power system is not cheap or easy. The big problem is the "battery bank". You take the battery down below 20% state of charge--And/or don't the bank >90% SOC once or twice a week--The batteries will not be long for this world.

You can set a battery cutoff at ~11.5 volts -- But that should be temperature compensated, have some hysteresis (i.e, cutoff at 11.5 volts and cut on at >13.0 volts), maybe some time limit (i.e., cutoff below 11.5 volts for >5minutes), etc...

There are some battery monitors that have a programmable output (i.e., alarm on at 50% SOC and off at 80% SOC) that you could hook up to an external relay or an inverter's inhibit input (if it has one), etc... (Victron has models with alarm outputs--possibly others do too).

If you can setup a system that only pumps when the sun is up--That can really reduce the complexity and cost of a system (and the cost of a battery bank getting murdered).

What would be really nice is a VFD (variable frequency drive) that can take a solar input and output to your three wire or three phase solar pump. It is certainly possible, but "Solar" VFD companies (at least in the US) are rare to find and may not keep the product in production very long.

Some folks have experimented and figured out they they could get a standard" VFD and connect to a solar array directly and get everything to work. But that is experimentation and a bit of "hardware risk" on your part (would be nice to have a working system fully documented here on the forum). A few links:

BB. wrote: »

Some discussions about VFD (Variable Frequency Drives)... Basically a variable frequency inverter with (typically) three phase output. Used to soft start motors (handy for 3 phase well pumps, or pumps with well head starting capacitor) and can also turn an AC motor into a variable speed motor (very handy for pumping applications).

WELL PUMP and Inverter QUESTION
Wind/solar for large scale pumping etc (out of my depth!)
could use knowledge - using Gould jet pump - transfering from 230vAC to ? DC (new link/thread 10/27/2012)
Help required to design off grid system (information on possibilities to connect "standard VFDs direct to solar panels) (new link 1/13/2013)

For DC powered pumps, there are Linear Current Boosters that do a nice job of MPPT type conversion from the solar array to the needs of a DC pump--But DC PM pumps with electronics are expensive (and so are LCB's), and DC pumps with brushes need work every few months to a year (brush replacement/commutator turning/etc.) if run 24x7.

-Bill

System check and a couple of question...

Comments

Categories