Cabling for Controller to Battery and battery to Inverter

[paulcheung]

Make some one check the battery. check if they can do what they claim. Buying some non quality brand battery you are taking a big chance there.

[mapmaker]

I rang the dealer who sold me the equipment, and after I emailed him a photograph of my set up, he said the problem could be because the controller and inverter share the same battery cabling - his full comment is in one of my earlier posts above.

Any chance you could post a diagram or schematic of how your system is wired? I'm a bit uncertain what you're describing. Assuming it is wired correctly, I would guess that there may be a high resistance connection somewhere.

The battery serves as a buffer between the controller and the inverter... if the battery has too low a capacity, too high a resistance, or a bad connection, it doesn't 'buffer' the controller's output and the inverter may see too high a voltage.

You mentioned the battery voltage when the OVD occurs... how and where are you reading that number? Does your controller have battery sense wires? Do you have a digital voltmeter to measure the actual voltage at the inverter input?

--mapmaker

[Sub]

Thanks Mapmaker. I have had a look at the Midnite web pages, and find quite a few wiring diagrams and I think you refer to the diagram named "Created 12V configuration", dated 11-10-08, approved by RWG?

Again that seems to have the inverter and controller sharing the cables to the battery bank. So it does seem to be an acceptable methodology.

My concern about this originated (remember I have just had the new battery bank and MPPT controller installed, so have been hyper vigilant in how things are working) from an event a few days ago. My inverter started to beep. I have heard it beep before, but that was on my old battery bank, when it went into a LVD.

I checked the inverter manual trouble shooting page, but if there is an adverse inverter event the beeping is meant to be accompanied by one of the LED lights flashing to indicate the precise problem. That wasn't happening. I then checked the monitor on the controller. All seemed well, but while I was standing there suddenly the LED lamp, usually green for ok, started flashing red, and the message "PROMPT, OVD" came up. This only lasted less than a couple of seconds when the green light came back on.

I rang the dealer who sold me the equipment, and after I emailed him a photograph of my set up, he said the problem could be because the controller and inverter share the same battery cabling - his full comment is in one of my earlier posts above.

Yesterday, the same thing happened in mid afternoon. The OVD kept happening every 10 minutes or so, much to my flat mates dismay, as she had a special dish in the oven - it is a gas oven but has an electric fan, and the oven kept switching off. Likewise the washing machine was on and that kept switching off as well. I would have thought that an OVD event would have caused the controller to stop sending volts to the battery, but it seems it turns everything off, and the inverter beeping was telling me power had resumed.

I got things working again by pulling the PV fuse and running on battery alone for a few hours.

The same thing just happened a few minutes ago, turning off the internet modem. What concerns me about this is we are in the middle of winter here so not yet getting full sun - admittedly we are having periods of really bright sun right now but with periods of rain showers and occasional cloud cover. This morning's event happened at 11am and we had less than two hours of sun before that.

Each time I checked the controller monitor when the OVD happens the battery voltage is showing no more than 58.8V, and PV volts are showing at the most 77V. SOC is rapidly cycling between 100% and 80%.

Until now I had been concerned that my PV array of only 1280W may not be enough for my new battery bank. But if it is going into OVD regularly in the middle of winter, what will happen when we get full summer sun!

Question, could the way the system is wired together be causing this, or is the controller simply doing what it is designed to do?

Controller settings for OVD is 64V, OVR is 60V and Charging limit is set at 62V.

[mapmaker]

I am reassured by Paul's comments about my wiring set up. I have been doing a lot of Google searching trying to find an appropriate wiring schematic, but the only one I came across is on page 16 of a tutorial, and I hope this is a link to it

There are quite a few things I don't like about that schematic on page 16. I prefer the use of bus bars. Check out the wiring diagrams for the ePanel over at Midnite Solar. Note: their wiring diagrams assume an inverter/charger, rather than separate units. Also, they assume you are using a Midnite controller which has GFP built in. That means there is no bonding between battery negative and ground, because the built-in GFP provides that bonding. In your setup you should have that bonding.

The battery monitor on page 13 of the pdf link you provided is a Trimetric battery monitor. I have one and like it.

--mapmaker

[Sub]

Balanced wiring

Thanks Paul and Mapmaker.
Re-reading my first post I see I left out info about my inverter - it is a Rich Electrics 3000W pure wine wave inverter.
I am reassured by Paul's comments about my wiring set up. I have been doing a lot of Google searching trying to find an appropriate wiring schematic, but the only one I came across is on page 16 of a tutorial, and I hope this is a link to it - http://www.solarflex.co.za/PDF%20FIL...Guidelines.pdf
While not exactly like mine it has major similarities. In my first post I said that the positive battery lead goes to a 100A shunt, to which cables also run from the controller and to the inverter. I have had a closer look at the device - it is not a shunt, it is a 100A "blow fuse". There is a transparent window through which I can see a single copper wire, so I guess that corresponds to the item labelled "Circuit Breaker" on the diagram in the link. The main differences are - I do not have
1. a charger/inverter; (I have my back up 4.5kW generator connected directly to the battery bank via a 48V charger)
2. a battery monitor;
3. a shunt on the negative lead;
From Paul's reply it seems that this is probably an acceptable way to wire the inverter and controller to the battery bank. My belief it was not OK, stems from advice from the dealer/supplier that -
"The inverter and controller should not share any wiring. Large current changes from the inverter draw will affect the battery voltage readings that the controller makes"
That advice is at odds with what seems to be good practise by others, as evidenced by Paul's experience and the wiring diagram in the above link.
I would be interested in reading further thoughts/ideas about this.

Most of you will be more familiar with AWG measurements than metric, so to clarify - the main battery leads are AWG 0000; all other cables are AWG 0 (or could be AWG 1, as I only had a handheld tape measure the gauge diameter).

Mapmaker, re your query about "the shunt", that is clarified above, but your question asking what sort of battery monitor I have raises another question for me. I have an analogue voltage meter, whose cable runs direct from a battery voltage sensor port on the controller. I know that these are not accurate, and intend to replace it with a good quality battery monitor. To simplify things I will post a separate thread asking for advice as to which monitor to chose.

[mapmaker]

The battery positive cable goes to a 100amp shunt, from which 8mm cables go to controller and inverter.

You might want to upgrade that to a 500 amp shunt... what battery monitor is using that shunt? Also, it is more usual to put the shunt in the negative circuit (same current either way). Putting it in the negative circuit (assuming a negative grounded system) means you don't need to fuse the sense wires from the shunt to the battery monitor. --mapmaker

[paulcheung]

My battery cable is also shared; my battery cables are connected to a disconnect breaker in a box. in the box there are some small breakers which connected to the PV panels and the charge controllers. I think that is how it done in most cases. I don't think you need to worry as long as the battery cable is large enough.

[Sub]

Strings and Parallel

Hi Mike
I may not have made it clear - there are two strings of four batteries each, thus two strings of four batteries connected in parallel, not four strings connected in parallel. Thus only one parallel connection.
I have seen the web page you refer to - Smart Gauge, in fact I have the connection to it saved as a shortcut on my desk top. Which is where I started to learn that for the inverter and controller to share one link to the batteries may not be such a good idea.
I know parallel is not recommended, but my last paralleled battery bank, described above, managed to last just over eight years. Unfortunately, where I live makes it impossible (for me) financially to install one string of batteries giving 500ahrs. Here AGM batteries from China are so much cheaper than FLAs from other parts of the world. And even to go to a single string of AGM means I would have had to go to 2V cells, and the cheapest of those at $400 NZ dollars each, way unaffordable as I would need 24.
The last bank were FLAs, and from your advice would have been less susceptible to unbalanced cabling/wiring than AGM.
So it seems my research, confirmed by your reply, does indeed mean I need to have the cabling redone!
The issue that most puzzled me, was to which battery terminals do I make the Controller connections. It seems they should not share the terminals that have the cables going to the inverter.
Currently I have the positive battery cable connected to one string, and the negative to the other string.
If I read you right, I do the same with the new cabling from the Controller. Would I connect the negative from the controller to the same string as the negative cable to the inverter, but at the opposite end of the string, and vice versa for the positive cable?
A bit peeved that I paid good money to a professional for the install, but now I have to redo it!!

[Mike90250]

Sadly, 4 parallel strings of AGM batteries are more susceptible to imbalance and early failure because of the inherent low resistance of AGM, and so the cable resistance is now the larger factor. The string with the lowest resistance cables will take on 90% of the work, till it falters, then the next string, and so on.

Parallel batteries are very difficult to do "right" with perfect balance wiring. See this article

http://www.smartgauge.co.uk/batt_con.html

which also covers charger and inverter connection points (the same terminals, charger on one side, inverter on other side of the lead battery terminal)

[Sub]

My off grid solar system was designed by an EE and professionally installed about 8 and half years ago. I bought the property 4 years ago so have no direct knowledge about the installation.

It consisted of 16x80W PV panels, a Xantrex C40 PWM controller, 16 Trojan T105 batteries in two strings of 8 in series connected in parallel, giving some 450Ahours C20 at 48V.

It supplies a small home with two adult occupants who are careful consumers of electricity. We did a power use audit when we moved in 4 years ago and calculated a daily use of 80 amps. Thus we were slightly undersized (80 x 3 = 240, thus 480 amp hours required) but compensated by only using things like vacuum cleaner, washing machine, toaster, hair dryer and similar items during the day when the sun was shining brightly.

The battery cables are shared both by the controller input to battery and out to the inverter. The cabling from the PV are 8mm thick, with both negative and positive cables running directly to the controller - the positive cable runs to the controller through a fuse. The cabling from the batteries are 15mm thick, the negative lead goes to a fuse, and then to a bus bar. 8mm cables go from the bus bar to the controller and to the inverter.

The battery positive cable goes to a 100amp shunt, from which 8mm cables go to controller and inverter. There is another fuse on the input to the inverter.

We have just replaced the battery bank and the controller after 8 years and 3 months of service.

New Batteries are eight 12V 290Ahr AGM, connected in two strings of 4 connected in parallel giving 48V and 580 Ahrs at C20.

New controller is an eTracer 60amp MPPT controller.

They were installed by a registered electrician who followed the old wiring/cable configuration.

Doing some research on other matters, I found that perhaps this is not ideal! It seems the cabling from the controller to the battery bank should be totally separate from the cabling from battery to Inverter.

I would appreciate comments from those more experienced than I in these things. If in fact I do need to change the cabling, do I connect Controller and inverter cables to the same battery terminals, or to terminals at opposite ends of the battery bank?

Thanks in advance, Sub.