Converting a mains device to battery power (Li-Pol in series and charging).

[castingflame]

Hi all.

I have a mains powered 'Device' that I wish to run from battery power so that it is not tethered by cable. The specs on the back of the unit say that it is rated at 200VA from a supply of 240 volts.

200VA @240 volts (if I did it correctly) translates to just under 840mA current draw.  If I wish to bypass its built-in PSU and run the device directly off of the required 24 volts, is this just a straight equation like  VA/Supply (ie 24) still?,  ... making it just under 8.34 Amps draw from a 24 volt DC source?

Apologies if this is a naive question.


Cheers

[mariush]

It would be best to measure the average current draw and all that using a multimeter (at the very least). 
That 200VA rating could be just the peak current the device will take when it starts up (like inrush current charging capacitors inside, creating the magnetic flux inside the transformer etc) but the average power consumption could be much lower than 200VA.
The power consumption could also vary with the other things like how bright the lcd screen is (if the device has any etc)

Depending on the power supply the device has, it may be only around 85-90% efficient when converting 240v ac to 24v DC (or whatever your device uses), so if you measure 200w taken from the mains socket, the device may use only 180w or something like that, while 20w would be waste.

[castingflame]

Thank you mariush for you reply. You raise some very good points.

The unit has one of those stock looking Switch Mode PSUs.

As this is really a feasibility study and not my unit I am limited in using the unit in a fully functional way at the moment. Could I use my multimeter to measure the current between the output of the PSU and the PCB?

[mariush]

It's hard with a regular multimeter but it would be possible to get a reasonably close measurement of the current using a clampmeter (which can also work as a regular multimeter, but for the price you'd pay for a clampmeter you would get a more specialized multimeter, so think of the regular multimeter part as a bonus.

You would need a multimeter that can also measure DC current. A cheap and more than good enough clampmeter would be Uni-T UT210 :  https://amzn.to/2ISWmCE



So basically what you would do is start the clampmeter in DC current mode, then insert the +24v wire(s) between the psu and the board (or whatever) in the center of your clamp, and the meter will tell you the amount of current going through the wire.  The measurement will vary slightly with the position of the wire inside the clamp, so it's best to be consistent if you make multiple measurements and keep the wire(s) as close as possible to the center.

If the current flowing through the wire is too small, you can practically make a loop through the clamp, for every loop the meter will show double the actual current value or something like that.

That's pretty much it... you just have to separate the positive and negative wires and only insert one type through the clamp otherwise you won't get a good measurement.

[castingflame]

Great advice mariush, I have just ordered one. 

Hopefully I will be able to get the instructions to get the unit fully working soon and I will take a measurement when the clamp arrives.

I really appreciate your help mariush.


I am just doing this a a favor to my father and it is really a bit of a stretch for me. Once I have an idea of the requirements I will need to look at a suitable battery solution and charger. Not sure if a 'pack' made from Li-Poly would be suitable at this stage ...

I will update in a few days.

[castingflame]

I have used the above clamp meter to take measurements. The Unit is being supplied by a built-in SMPSU and it is providing 23.8v dc and 2.6 Amps dc while the unit is fully running. Using an online calculator, that came out at 61.88 Watts

The requirement is for an internal battery solution and charging circuit. The battery needs to last 1 hour but 2 hours would be preferable.



I was looking for a Li-Pol solution and wondered if using something like 2 x (1300mAh Lipo Battery Pack 22.2V) wired in parallel would double the current to 2600mAh. I know it is a bit shy on voltage so maybe a DC/DC converter would help there.

https://www.ebay.co.uk/itm/RC-Turnigy-Graphene-1300mAh-6S-65C-Lipo-Battery-Pack-XT60-22-2V-GRAPHENE-POWER/253180141295?hash=item3af2b65eef:g:ckkAAOSwKNJZzTSX


Also, would it be possible to use 7 x of these 3.7v 3800mAh units?

https://www.ebay.co.uk/itm/3-7V-3800mAh-AKYGA-Battery-Li-Pol-Rechargeable-Cell-for-GPS-MP3-MP4-L7836140/253456235210?hash=item3b032b3aca:g:IscAAOSwHPdalbXM


Thanks.

[james_s]

Keep in mind you need to be able to separate the wires from each other to use a clamp meter. If you put the clamp around both the V+ and return the current will cancel out and you'll read zero.

[castingflame]

Thanks James,

Yes I put it around both cables initially and not much happened. Then I RTFM'd (after printing one out) and did it correctly.

[castingflame]

Li-Pol in series and charging

More focused questions ...

I am thinking about using 7 of these 3.7v  3800mAh li-pol packs in a project wired in series to get up to the required 24 volts. Firstly, is that okay in theory as far as using li-pol cells go?


https://www.ebay.co.uk/itm/3-7V-3800mAh-AKYGA-Battery-Li-Pol-Rechargeable-Cell-for-GPS-MP3-MP4-L7836140/253456235210?hash=item3b032b3aca:g:IscAAOSwHPdalbXM



The question is this; I am aware that there are specific requirements for charging these packs in a correct and safe manner. I would much prefer an off the shelf solution for this part of the project. It would be possible to a discrete small PCB charger like the ones sold by Mikroelektronika for each of the packs, but this seems a poor way forward.

https://uk.rs-online.com/web/p/products/8828904/?grossPrice=Y&cm_mmc=UK-PLA-DS3A-_-google-_-PLA_UK_EN_Semiconductors-_-Semiconductor_Development_Kits%7CPower_Management_Development_Kits-_-PRODUCT+GROUP&matchtype=&gclid=Cj0KCQjw-JvaBRDGARIsAFjqkkrFvuKQ-s3n82o6qUc4hsEoz_0_3KW3gUTA5DUZFle8n6m3JtnJZhUaAlItEALw_wcB&gclsrc=aw.ds


So what I really need is a charger or circuit design that would enable them to be charged in series. Is this possible or even safe?


The complete power pack will be internal in the case of the device so any series/parallel/individual 'rewiring', for charging V usage, will need to be done with either by solid state components or relays/solid state relays.


Thanks for any help you can give.

[HB9EVI]

Very tight with those cells; maximal discharge current rating 1C - that's maximally 3,8A, but you have to count with at least 8,3A (likely more) on the 24V side; if I see the wiring of that cell, i don't think they even take 3,8amps without heating up.

[castingflame]

Thank you for you response. It is appreciated. I am now looking at NiMH.


Also, wheres does the 8,34 Amps come from. Sorry This area is all very new to me.

[GigaJoe]

70WH at 24V - to run one hour,

as an option to find suitable lipo block from let say cordless drill (a few batteries that easy to swap, with a charger)  and use a pre-build boost converter to get voltage up to 24V. need to look at output filtering as device can be sensitive to the noise in the power ... That basically some fast hack, build it as a separate unit,  and GX16-2 (-3)  as a power connector.

[castingflame]

Thanks GigaJoe for your help.

That connector is exactly the type of thing I was looking for. Sam as on my 3D printer.

I have done some practical testing and moving forward. Initially I am opting to use a Ni-MH pack and charger to get some real world numbers in a cost effective way. So far so good.