Router Battery Backup

Thanks everyone for the advice, just tested and all is running on all networks as usual.

So my total costs were:
12AH Battery - R600
1.5A Perel intelligent charger - R600

R1200 = peace of mind for a few hours. Can easily upgrade battery size if needed.

If Geewiz timing is anything to go by this should be 16 hours of use. So this is sufficient for my purposes.

View attachment 635744
Rather expensive solution just for that router. The mini UPS they sell will power that router for about 10 hours. At a cost of just over R600.
 
Rather expensive solution just for that router. The mini UPS they sell will power that router for about 10 hours. At a cost of just over R600.

Not to phased about the cost, have no idea where we are going with the load shedding would rather have a little more juice and happy to pay that for the charger to use for other purposes if needed.
 
Not to phased about the cost, have no idea where we are going with the load shedding would rather have a little more juice and happy to pay that for the charger to use for other purposes if needed.
Ja true. Even thinking now of getting a generator. Let's see what the weeks ahead of us brings. Damn tomorrow is going to be a bad day for me if they persist with stage 4.
 
Ja true. Even thinking now of getting a generator. Let's see what the weeks ahead of us brings. Damn tomorrow is going to be a bad day for me if they persist with stage 4.

Thats a route I want to avoid as long as possible but should maybe look so long :(
 
Except that most of the thread earlier said by then you've discharged the battery basically 100%.
.
Nope not safe to so. Does so at the expense of battery life.. It is expedient because it artificially inflates standby time.
You guys read the Web which is good, but fail in engineering interpretation, so you are wrong and others repeat this rubbish.

The figure you are quoting it is a relationship between state of discharge vs battery voltage on the open circuit. It is meassured after disconnecting load and let the battery rest for half an hour or more (I don't remember exact figure), as depends on the load the battery can get hot. This is very accurate if temperature of the battery taken into account as well.

On the other side there is no way to determine state of discharge during discharging, unless the battery current and other parameters are constantly meassured and evaluated. It is not happening in any UPS you can buy.

You can also see occassionaly a figure SoC vs. voltage during discharging, but it only applies to a brand new battery of a given make and capacity. This figure is given as an example for developers, but not for referals. It is critically dependent on internal resistance of the battery and this parameter very depends on the age of the battery, it is also very unstable during breaking sulfation.

In summary, developers set cut-out voltage based on the battery capacity, chemistry and make. It is made experimentally between 10V and 10.7V for the battery they sell with UPS, so it is safe. When you replace internal battery with a much larger one, it is obvious that this assumption breaks and a cut-off is delayed during large loads due to the smaller internal resistance of the large battery. The small UPS will likely trigger cut-off first due to the other condtion, like a heat generated inside the transformer.
 
Can someone confirm if I'm being dumb here and missing something, I want to power a router and a modem, I don't know how much power they use but I'd say it's safe to assume 10W each, if I get a UPS with a 9Ah battery will that then power them for 4.5 hours?
 
Can someone confirm if I'm being dumb here and missing something, I want to power a router and a modem, I don't know how much power they use but I'd say it's safe to assume 10W each, if I get a UPS with a 9Ah battery will that then power them for 4.5 hours?

UPS I'm not sure but routers directly connected to the battery plenty even a few hours to spare.

I haven't gotten the time to get my backup setup sorted from communica so I was using my laptop battery as a test.

Laptop batter is 11.1V and 31 Wh which according to a converter is 2.8Ah battery. I powered a Huawei B618 + Huawei fibre ONT for 2.5 hours and when I plugged it back into the laptop it still had like 40% battery life left after 2.5 hours.
 
Last edited:

Currently saving for an inverter but will try this in the meantime.

The power inverter I bought has this connectionmotoquip.jpg

Can I use this adapter and plug in an extension cable? Just for the CPE and router.

plug.jpg

When CPE and router are unplugged is it safe to charge my phone and tablet?

Sorry for going on and on. Appreciate the help.

@bwana hope you don't mind me using your pic.
 
.
You guys read the Web which is good, but fail in engineering interpretation, so you are wrong and others repeat this rubbish.

The figure you are quoting it is a relationship between state of discharge vs battery voltage on the open circuit. It is meassured after disconnecting load and let the battery rest for half an hour or more (I don't remember exact figure), as depends on the load the battery can get hot. This is very accurate if temperature of the battery taken into account as well.

On the other side there is no way to determine state of discharge during discharging, unless the battery current and other parameters are constantly meassured and evaluated. It is not happening in any UPS you can buy.

You can also see occassionaly a figure SoC vs. voltage during discharging, but it only applies to a brand new battery of a given make and capacity. This figure is given as an example for developers, but not for referals. It is critically dependent on internal resistance of the battery and this parameter very depends on the age of the battery, it is also very unstable during breaking sulfation.

In summary, developers set cut-out voltage based on the battery capacity, chemistry and make. It is made experimentally between 10V and 10.7V for the battery they sell with UPS, so it is safe. When you replace internal battery with a much larger one, it is obvious that this assumption breaks and a cut-off is delayed during large loads due to the smaller internal resistance of the large battery. The small UPS will likely trigger cut-off first due to the other condtion, like a heat generated inside the transformer.

Great story, you better stick to it! As you say no UPS that you can buy measures all these parameters so, an experiment is done and then the developers set the cut out voltage. So what! Over and over again in all the threads and fora dealing with backup power and how to look after batteries it is shown that batteries should NOT be discharged below 20% of capacity. It is then that the internal resistance becomes more significant and leads to even more heat generation which leads to more damage to the batteries. There absolutely no need to push a battery to that limit, far better to stay well above the 20% value for long life. How do you think I can make an alarm system battery last for 10 years and a gate motor battery last for 7 years? Precisely because I err on the safe side. So fine with all the theory. This is about practical power backup, not about the intracies of the chemistry of batteries and artificially pushing the product spec so that the UPS looks good on a spec sheet..

The ONLY way one can under normal working conditions approach the theoretical limits is to use a BMS system, which does measure all those parameters. No one is going to in a practical application sit there and let a battery discharge to 10 V, wait for the battery to cool down, then measure the Open Circuit Voltage to see if it is still above the 20% discharge point. And then say oh okay I have not damaged a battery.

This is about practical advice in a real world, not in a lab!
 
Last edited:
Can someone confirm if I'm being dumb here and missing something, I want to power a router and a modem, I don't know how much power they use but I'd say it's safe to assume 10W each, if I get a UPS with a 9Ah battery will that then power them for 4.5 hours?
B618 uses 6.8 watts, I can test more if you like.
 
B618 uses 6.8 watts
I don't have a kill-a-watt or something similar, it's a Technicolor TG589vn v3 which Google says is 12v 1A and a Tenda AC9 which is 12v 2A. The modem probably barely uses any power because everything except the modem part is turned off and the router I just used 10 Watt as average, so I just took 10 Watt each to be safe so that means a Mecer 850VA which has a 9Ah battery should, in theory, run both of them for 4.5 hours, unless I'm missing something.
 
I don't have a kill-a-watt or something similar, it's a Technicolor TG589vn v3 which Google says is 12v 1A and a Tenda AC9 which is 12v 2A. The modem probably barely uses any power because everything except the modem part is turned off and the router I just used 10 Watt as average, so I just took 10 Watt each to be safe so that means a Mecer 850VA which has a 9Ah battery should, in theory, run both of them for 4.5 hours, unless I'm missing something.
That sounds about right ... measured mine with only one client.
 
Currently saving for an inverter but will try this in the meantime.

The power inverter I bought has this connectionView attachment 635782

Can I use this adapter and plug in an extension cable? Just for the CPE and router.

View attachment 635788

When CPE and router are unplugged is it safe to charge my phone and tablet?

Sorry for going on and on. Appreciate the help.

@bwana hope you don't mind me using your pic.
Probably will be okay as that "universal socket" is supposed to be able to handle a range of standards.
Personally, I, after lots of bad experiences in Africa with those sockets, will not buy a device with that socket on it. Especially in SA, where we have a round pin standard.
 
Probably will be okay as that "universal socket" is supposed to be able to handle a range of standards.
Personally, I, after lots of bad experiences in Africa with those sockets, will not buy a device with that socket on it. Especially in SA, where we have a round pin standard.

Ooh gonas, thank you @Geoff.D :)
 
Can someone confirm if I'm being dumb here and missing something, I want to power a router and a modem, I don't know how much power they use but I'd say it's safe to assume 10W each, if I get a UPS with a 9Ah battery will that then power them for 4.5 hours?

A very rough calculation, not taking into account temperature effects, efficiency effects etc shows: Assuming the voltage required by the router is 9V

Amps = 20W/9V = 2.2 Amps
Stby time (maximum) = 9Ah/2.2 = 4.1 hours.
Now the less current drawn by the router the longer the stby time will be.
So this is border line if your requirement is for 4.5 hours. The only way to get a more accurate idea is to measure the actual power drawn. Or do an experiment as already suggested.

12V:
20/12 = 1.7 Amps
9Ah/1.7 = 5.3 hours.
Safe even allowing for ineffciencies and temperature effects.
 
Last edited:
A very rough calculation, not taking into account temperature effects, effciciency effects etc shows: Assuming the voltage required by the router is 9V

Amps = 20W/9V = 2.2 Amps
Stby time (maximum) = 9Ah/2.2 = 4.1 hours.
Now the less current drawn by the router the longer the stby time will be.
So this is border line if you requirement is for 4.5 hours. The only way to get a more accurate idea is to measure the actual power drawn. Or do an experiment as already suggested.

They are both 12V and I only require 2.5 hours, maybe 3.
 
@bwana hope you don't mind me using your pic.
Of course not. At the end of the day though I only needed to by a battery. My only objective was to spend as little money as possible to keep the VDSL going for a few hours, and it worked out fine for me.

BTW - I just wired a euro plug into a SA three pin because that's also what I had going. On the plus side I can also run a couple 240v LEDs on the other side of the plug.
 
Top
Sign up to the MyBroadband newsletter
X