Hello!

I would greatly appreciate some more established advice on my situation.

 

I have three devices to run from UPS power during Stage 1-4 loadshedding.

D-Link DSL-2750U (12V, 1A), TP-Link Archer C20 (9V, 0.6A), Raycore CPE (Not 100% sure about model but believe the PSU is also 9V, 06A)

I have a Mecer ME-650-VU (Rated at 360W max output) UPS. It has a single 12V, 7AH battery. I do not know how much power the UPS subsystems use to monitor and power itself, but I (hopefully over-) guesstimated no more than 12 Watt.

If my calculations are correct (and my knowledge on this matter is mostly assumptive and self-tought, I would appreciate corrections!) my system's max usage at any given time should be 35W. In a zero-loss scenario the battery should provide 84AH worth of power. In a real-world scenario this is probably lower than 75AH (super optimistic assumption of only 10% efficiency loss).

Currently the power lasts between 90 minutes and 120 minutes. Unfortunately I couldn't see the exact time it went off but I'm certain of this time window.

 

My questions are as follow:

Would running my current system off a 12AH battery extend my uptime to over the 180 minute mark?

and

Would swapping out the standard 7AH battery in the ME-650-VU with a 12AH one cause any problems?

Bonus Question: Would disabling either the 5.0GHz or 2.4GHz channels on the C20 router reduce power consumption enough to make a noticeable difference and if so, which band uses more power?

 

My understanding so far is that upping the capacity would proportionally increase the charge time on the battery.

Thanks for your help and time!

Bear in mind that you will require more time to charge the larger capacity battery.

If I sound like a stuck record please insert earplugs!

This is a DC application. There is no need to install an inverter UPS, especially one of those single 7Ah battery devices. They are hopelessly inefficient. If you insist on using an inverter start with a dual battery system (24V)

But you can even do way better than that by using a mini UPS. There's no need to generate 220V so why do it?

Google Mini UPS or see: PUPS UPS

59 minutes ago, Richard Mackay said:

If I sound like a stuck record please insert earplugs!

This is a DC application. There is no need to install an inverter UPS, especially one of those single 7Ah battery devices. They are hopelessly inefficient. If you insist on using an inverter start with a dual battery system (24V)

But you can even do way better than that by using a mini UPS. There's no need to generate 220V so why do it?

Google Mini UPS or see: PUPS UPS

Thank you for the info Richard. To my best knowledge I do have an UPS and not an inverter. Sorry if I was unclear or if I'm misunderstanding.

EDIT: On inspection of your recommendation I have some more questions. I want to know how the PUPS unit achieves longer battery life than a conventional UPS. I also do not wish to buy excessively right now and would rather try and figure out a solution using the tech I have installed. The PUPS unit seems too much like a mystery with no access to a spec sheet as well as being R150 more than a line-interactive UPS of the same caliber. It was worth checking out regardless, thank you for bringing it to my attention.

 

1 hour ago, Crankshaft said:

Bear in mind that you will require more time to charge the larger capacity battery.

Thanks Crankshaft, I will keep that in mind.

Edited March 10, 20205 yr by phLOx
Adding some info

1 hour ago, Richard Mackay said:

If I sound like a stuck record please insert earplugs!

This one time I actually agree with you. He has three appliances, one that is 12VDC and two that are 9VDC. I still have my reservations about making a consumer-level product that the average idi^H^H^Hperson on the street can use, but this certainly strikes me as the kind of application where a bit of study, R100 worth of buck converters from a hobbyist place and a fuse or two... is going to solve the problem neatly and efficiently.

22 minutes ago, phLOx said:

To my best knowledge I do have an UPS and not an inverter. Sorry if I was unclear or if I'm misunderstanding.

I believe we find ourselves in terminology hell again. A UPS is an "uninterruptible power supply", which means it must contain a transfer switch, charger and inverter. Often the inverter and charger is the same component running in reverse half the time. But now there are DC backup solutions, which are essentially LiPo packs with some sort of DC output so you can power DC applications directly. People have now started calling these "DC upses"... which I suppose they kinda are... but I can also see how it confuses the heck out of anyone.

Richard is our resident dissenter from the usual pattern where you first boost the voltage to 230VAC and then immediately buck it back down to 12VDC again. He says it makes more sense to use the DC directly. Today he is actually right... 😛

Edited March 10, 20205 yr by plonkster

2 minutes ago, plonkster said:

Richard is our resident dissenter from the usual pattern where you first boost the voltage to 230VAC and then immediately buck it back down to 12VDC again. He says it makes more sense to use the DC directly. Today he is actually right... 😛

Whew!

 

4 minutes ago, plonkster said:

This one time I actually agree with you. He has three appliances, one that is 12VDC and two that are 9VDC. I still have my reservations about making a consumer-level product that the average idi^H^H^Hperson on the street can use, but this certainly strikes me as the kind of application where a bit of study, R100 worth of buck converters from a hobbyist place and a fuse or two... is going to solve the problem neatly and efficiently.

I believe we find ourselves in terminology hell again. A UPS is an "uninterruptible power supply", which means it must contain a transfer switch, charger and inverter. Often the inverter and charger is the same component running in reverse half the time. But now there are DC backup solutions, which are essentially LiPo packs with some sort of DC output so you can power DC applications directly. People have now started calling these "DC upses"... which I suppose they kinda are... but I can also see how it confuses the heck out of anyone.

Richard is our resident dissenter from the usual pattern where you first boost the voltage to 230VAC and then immediately buck it back down to 12VDC again. He says it makes more sense to use the DC directly. Today he is actually right... 😛

I completely agree that its wasteful to conversion, but I am not knowledgeable enough to wire directly to battery AND have the battery on an auto-charging inverter from AC, hench the UPS, to do the work for me.

But I'm willing to learn! We are losing some power in this scenario to both the conversion and the UPS monitoring subsystem, I would assume. Going directly from the battery would be far less wasteful. Now if only I knew how to safely wire that and still keep everything "automated".
I just acquired some cheap electronic components and hardware and will solder if need be.

Today's loadshedding reported that the system stayed up and running for 130 minutes. I believe it is so much more than yesterday's 90-120 ballpark since no wireless devices was connected to the one router and only half to the other, so less traffic meaning less power draw, I assume.

15 minutes ago, phLOx said:

But I'm willing to learn!

You see, that sets you appart from the average idi... uhm person on the street 🙂

If you have a 12V battery, you can easily make 9V from it using a step-down converter. In my day we had to solder up our own using a linear regulator, but the world has gotten so much easier. You can buy one of these, and adjust the screw so that it makes 9V. The input and output are neatly marked and you can't really get the polarity screwed up if you know how to read. Put a fuse inline (1A will probably do), and find a plug that fits the socket  of your router/appliance, and you can supply it with 9V.

For the 12V side itself, there are two options. The one is to supply it directly from the battery (with a fuse in line, as usual), but I have concerns with this. My concern is that a lead acid battery, while charging, goes as high as 14.5V. Our appliance SHOULD have some margin, but we don't know that for sure. So you may want to look at a buck/boost converter, that is one that can either buck or boost the voltage. It takes a voltage of between 4V and 32V and gives an output of whatever you adjust it to (12V). So again, the 12V battery can go as high as 14.5V and as low as 10.5V (when it is almost completely flat), but you will get a flat 12V out of it.

You will need a good soldering iron, some fuses and fuse holders, black and red wire, plugs to fit the various sockets, and a multimeter. And some common sense.

Edited March 10, 20205 yr by plonkster

Most of the routers I've had that were using 9V also worked perfectly fine at 12v.

I would buy this and a 12Ah or maybe slightly bigger battery.

5 minutes ago, Gnome said:

Most of the routers I've had that were using 9V also worked perfectly fine at 12v.

I would buy this and a 12Ah or maybe slightly bigger battery.

That unit looks perfect! If only I did my research before going out and getting the UPS. 😢

 

  

44 minutes ago, plonkster said:

You see, that sets you appart from the average idi... uhm person on the street 🙂

If you have a 12V battery, you can easily make 9V from it using a step-down converter. In my day we had to solder up our own using a linear regulator, but the world has gotten so much easier. You can buy one of these, and adjust the screw so that it makes 9V. The input and output are neatly marked and you can't really get the polarity screwed up if you know how to read. Put a fuse inline (1A will probably do), and find a plug that fits the socket  of your router/appliance, and you can supply it with 9V.

For the 12V side itself, there are two options. The one is to supply it directly from the battery (with a fuse in line, as usual), but I have concerns with this. My concern is that a lead acid battery, while charging, goes as high as 14.5V. Our appliance SHOULD have some margin, but we don't know that for sure. So you may want to look at a buck/boost converter, that is one that can either buck or boost the voltage. It takes a voltage of between 4V and 32V and gives an output of whatever you adjust it to (12V). So again, the 12V battery can go as high as 14.5V and as low as 10.5V (when it is almost completely flat), but you will get a flat 12V out of it.

You will need a good soldering iron, some fuses and fuse holders, black and red wire, plugs to fit the various sockets, and a multimeter. And some common sense.

I've got these from China. (See Attached)

The top stuff with a red border I'm still waiting for but haven't heard from the post office for obvious reasons.

 

The multimeter works. The soldering iron works. It's not good by any standard I'm allowed to judge with as I have no experience with soldering (but the headset I "fixed" works!). The reason I got cheap stuff is so I don't have to cry too much when I accidentally destroy it while practicing.

I think I have heard about the buck converters before while looking up a way to convert a battery operated device to AC wall power. Thanks for the link, Communica has great parts and good prices. I also use Yebo Electronics and DC3 Distribution.

2 hours ago, phLOx said:

Thank you for the info Richard. To my best knowledge I do have an UPS and not an inverter. Sorry if I was unclear or if I'm misunderstanding.

Your Mecer UPS has a built in inverter. The term UPS doesn't specify any particular technology but has become the name given to an inverter UPS.

To differentiate these DC UPS's (i.e. that don't generate 220V AC)  are being referred to as a 'Mini UPS'

Quote

Would swapping out the standard 7AH battery in the ME-650-VU with a 12AH one cause any problems?

I added a second 7AH battery in parallel on mine (outside the case).  Still works perfectly with no issues two years later.

6 hours ago, Richard Mackay said:

Google Mini UPS or see: PUPS UPS

I wont trust this device, tried to go to the site displayed on the device and they don't exist anymore. Dont think you can expect any backup if this unit fails. 

11 hours ago, JustinSchoeman said:

I added a second 7AH battery in parallel on mine (outside the case).  Still works perfectly with no issues two years later.

This is what I expected but wanted to know for sure. Your comment is invaluable to me and I thank you!

8 hours ago, Jaco de Jongh said:

I wont trust this device, tried to go to the site displayed on the device and they don't exist anymore. Dont think you can expect any backup if this unit fails. 

I'm thinking the branding agency printed the website name wrong since it should be .info and not .co.za. In fact I cannot find references that the .co.za domain in question ever even existed. If it did I bet they moved over to the free .info domain since its, well, free. I'm 99% sure the device is simply an imported "brandless" device which is then rebranded locally. Without a specification sheet I won't be willing to consider it though. Need to know whats inside 🙂

11 hours ago, Richard Mackay said:

Your Mecer UPS has a built in inverter. The term UPS doesn't specify any particular technology but has become the name given to an inverter UPS.

To differentiate these DC UPS's (i.e. that don't generate 220V AC)  are being referred to as a 'Mini UPS'

Thank you Richard, this makes sense and I learned a new term!

 

 

Another 2 sessions of loadshedding has now passed and for some reason the system held on for the entire duration. It could be that the session was cut short since the power went on 20 minutes earlier than scheduled each time, which means the system lasted 130 minutes each time. Right now I am thinking that perhaps a cheaper upgrade to the 9AH battery should suffice? The 12AH is about R200 more than the 9AH, and I'm finding it hard to justify the difference if the results will be the same. On the other hand, perhaps since the 12AH wont be fully depleted during the loadshedding it would extend the battery's life? That's probably the only reason then to fork out an extra R200. Worth it?

1 hour ago, phLOx said:

since the 12AH wont be fully depleted during the loadshedding it would extend the battery's life? That's probably the only reason then to fork out an extra R200. Worth it?

I would say it is worth it. Just the cost of your time (to go out and buy a new battery when it fails) will possibly exceed that R200 easily 🙂

 

I decided to do some measurements of an inverter UPS to get a benchmark. I found an old 600VA UPS's and did a few tests to measure the current it draws from its battery when under load.

I connected a Huawei ONT and a D-Link Wi-Fi to the UPS as the load.

The DC current that the UPS draws from its battery with this load is 1.5A 

The current that the load draws direct from a 12V supply is 0.5A

So the current fed to the load is 1/3 of what the UPS requires to drive that same load.

Edited March 12, 20205 yr by Richard Mackay

11 hours ago, Richard Mackay said:

So the current fed to the load is 1/3 of what the UPS requires to drive that same load.

So you have a 33% efficiency.

9 minutes ago, plonkster said:

So you have a 33% efficiency.

This applies only for this particular (light) load.

As you drive a greater load it gets a lot worse!

55 minutes ago, Richard Mackay said:

This applies only for this particular (light) load.

As you drive a greater load it gets a lot worse!

It's not quite that simple. The UPS will have a standing consumption - power it uses even when nothing is plugged in, this is likely where most of the power lost here goes. Let's call it 1A / 12W. You are using only about 6W. I don't think that for any load higher than 6W you will ever go lower in terms of efficiency (within the design limits of the UPS, which is not 600VA)

21 minutes ago, P1000 said:

It's not quite that simple. The UPS will have a standing consumption - power it uses even when nothing is plugged in, this is likely where most of the power lost here goes. Let's call it 1A / 12W. You are using only about 6W. I don't think that for any load higher than 6W you will ever go lower in terms of efficiency (within the design limits of the UPS, which is not 600VA)

I would agree with this.

 

I have  about 3 devices ranging from 12v to 9v, plugged into  a 24v UPS (transforming to 220V), and my power consumption on the kill-a-watt reads between 20 and 30Wh.

 

-G-

Edited March 13, 20205 yr by gallderhen

I am specifically referring to the 600VA single battery (12V) UPS

20 minutes ago, Richard Mackay said:

I am specifically referring to the 600VA single battery (12V) UPS

Me too.

2 hours ago, Richard Mackay said:

As you drive a greater load it gets a lot worse!

Actually it gets better. The self-consumption of the unit is the fly in the ointment. As the load goes up, this becomes proportionally less compared to the whole, but it remains fairly horrible over the entire range.

The same thing happens with other inverters, even better ones. A 3kVA Victron Multiplus-II, for example, has a self-consumption of 12W. If you put a single 5W LED on it, you have 5/(5+12) ~= 29% efficiency 🙂

 

Yes. The efficiency does improve with a bigger load.

I connected a 120W load on the AC output and the DC amps draw was 15A. (i.e. 180W)

This is for a 20% load of the rated capacity of the UPS. I don't know what the maximum current is of those 7Ah batteries but this must be pretty close.. 

10 minutes ago, Richard Mackay said:

Yes. The efficiency does improve with a bigger load.

I connected a 120W load on the AC output and the DC amps draw was 15A. (i.e. 180W)

This is for a 20% load of the rated capacity of the UPS. I don't know what the maximum current is of those 7Ah batteries but this must be pretty close.. 

Did you measure it with a scope, because being modified sine, the current won't be constant and a normal multimeter will probably give a pessimistic reading.

The Yuasa NP45-12 that usually ship with those UPSs can do 45W/cell - 45*6 = 270W (for 10 minutes) it can peak higher than that.