Kindly asking for Advice on DIY inverter setup

[wazzerweezel]

Hi everyone. The owner of the business I work for along with myself have created a DIY inverter setup that is giving us some issues. We bought all the equipment new about a year ago and have been using the system in the office to supply uninterrupted power to a few devices. The setup is made up of the following devices:

1) 50 Amp 12 Volt Automatic 7 Stage Smart charger with 4 settings (Switching Power Supply, GEL, AGM and WET)
2) 3000 Watt, Pure Sign Wave 12 Volt to 220-240 Volt, Inverter
3) 2 x 200 Amp/Hour, 12 Volt, Gel Batteries

The charger is connected to grid power witch supplies it with 220 Volt AC. The output cables of the charger is connected to the battery poles. The batteries are connected in parallel with the idea being a supply of 12 Volt at 400 Amp/Hour total. There are a two loads connected to directly to the battery poles that draw approximately 30 Watts 12 Volt DC directly from the batteries. The inverter is connected to the battery poles drawing the 12 Volt DC towards it. All connections on the poles are made directly by using lugs with nothing separating them. Meaning all the connections are touching each on their respective positive and negative poles.

The inverter connects to an internal wiring network supplying certain wall plugs with what we call "Inverter Power". The built in Watt Meter on the Inverter indicates we draw an average on 450 Watts constant power when all the items connected to the inverter power is running. I have double checked this with Plug in Watt meter and also noted the max spike for the startup of the devices was 510 Watts.

According to all the information I gathered on various sites on the internet I worked on the assumption that we never want to exceed a DOD of 50% on the batteries. Taking this into account and doing some highly questionable math I figured we should have at least 5 hours until the batteries will need to be charged again. I got to that number as follows: 400ah / 2 = 200ah. 200ah x 12Volt = 2400Watt/hour. 2400Watt/hour / 450Watts (average constant usage) = 5.33 hours. To account for efficiency startup spikes etc I rounded down to 4 hours that should be guaranteed.

The system worked perfectly for the past year. Since we were permanently using the "Inverter Power" for our devices (Basically a bunch of computers, their monitors and one ink jet printer), we never had issues when the power went off giving us a UPS like experience but with much longer battery life. By using a connected Volt meter we monitored the system to ensure we never got too close to 50% DOD. When the grid power came back on the charger would start up automatically and start charging the batteries while also supplying the inverter. (This is all assuming my understanding was is correct that the 50amps from the charger would supply the loads imposed on the battery by the devices connected directly and the inverter. with any excess going towards charging the battery.)

Recently the setup will only last about an hour whereby the batteries would be at roughly 12.5 Volts (Measured with the loads active) then suddenly fall to something crazy like 11.6 Volts which would trigger the built in safety on the inverter switching it off.

Has anyone got advice on why this could have happened and what we may be doing incorrectly?

 

[DPG]

The obvious thing that springs to mind is that the batteries may have become past their best?

 

[wazzerweezel]

The obvious thing that springs to mind is that the batteries may have become past their best?

Hi there DPG.

Thank you for your reply. The store we bought the batteries at gave us a 10 year guarantee. Working on the assumption that any form of discharge, regardless of how much the batteries are discharged before being charged full again, counts as one cycle I would say we are at roughly 500 cycles. I don't know if this is allot for they type of batteries we have but I feel that, assuming we have not damaged the batteries in some way, they should still have some life left in them or am I being completely rediculous? Please note I am in now way or form trained in the field, what I know I have read on the internet and tried to verify by cross checking with what looked like reputable sites.

 

[newfutile]

You can get a Lead acid battery tester quite cheap which will give you a rough amp hour reading , I think it’s based on the internal resistance. The other way is to fix a load across a battery and note it’s run down time

 

[Lucien Nunes]

It does sound like your batteries have lost capacity, which would be the first thing to check. The capacity analysers that use Peukert's exponent to extrapolate from a short, light discharge to predict the performance on a prolonged one, are reasonably accurate but not anywhere near as accurate as a full discharge under working conditions. In fact you are doing that with your normal operational cycles, the only thing missing is a true measurement of the current consumption of the inverter.

Your calculations are not far out but you omitted to account for the inverter efficiency. If you take that as 80% then the load on the batteries would be 450W / 0.8 = 562W, for which the corresponding load current at 12V is 562/12 = 47A. An inline ammeter to confirm that would be helpful.

If the batteries are damaged then the question is why, given one year's use. Overcharge is a common problem, although it sounds like your charger should be capable of correctly maintaining them on cyclic duty. However, a smart charger is only as good as its configuration and if the wrong voltage has been dialed in, or the temperature sensor is not giving a reading representative of the battery temp, then it might be giving too high a voltage and very slowly gassing them dry. It would be good to know what the terminal voltage is once the charge cycle is complete and the unit falls back to continuous float.

You mention 500 cycles, however with lead-acids shallow cycles impose very little wear and tear and I would not be counting anything less than 20% DOD. Even 30-40% DOD shouldn't count as a full cycle, because there is less disturbance of the paste within the plates and less lead sulphate formed.

One final thought; the voltage that you are reading at the inverter is subject to drop in the cables. If these are too long and thin or there are high-resistance connections, the inverter will trip on low volts too soon. It might be that when the batteries were very new and lively, the extra voltage achieved compensated for the wiring drop, but now they are run-in and achieving average voltage, that is starting to manifest as a problem. At nearly 50A, high-resistance connections will probably have shown themselves up by overheating, but there's value in eliminating excessive drop as the cause, by comparing the battery terminal voltage and inverter terminal voltage on-load.