The TOMO Power Supply

TomoWell, I started to make a video of this thinking it was going to be a great new uninterruptable supply. I’d listened to others, I’d read the spec (or so I thought) and this was to be the new saviour of battery supplies.

The TOMO unit comes in different sizes, I chose the 3-battery solution. I bought mine from AliExpress.  It has a USB input and 2 USB outputs and can power up to 2.1 amps in TOTAL (2.1 amps out of one output OR 1 amp out of the other – must use the same chip as everyone else).

It is nice looking and comes MINUS the batteries and hence is cheap.  You can use one battery or more – either way it still acts as a power supply.

You can connect power and disconnect to your heart’s content, it will not interrupt the output. You can let the battery die (no input power) and then plug in power and it will recover gracefully, turning on the output.

Tomo[6]So for an uninterruptable supply, apart from lack of monitoring (when running a load you cannot see the state of the battery as it is showing load current and voltage), this looks ideal.

Except for one tiny but vital detail… and I quote from the spec which I don’t think was in the advert but I might be wrong….  “when input and output at the same time, it will automatically charge the device first, then charge the battery”. I guess that’s what I get for not reading the instructions VERY carefully…

Do you see the problem? I must say it took me a little while to grasp this….   I noted when I put a meter inline with the incoming charge voltage to the unit – the current used was the same as the current used by the Raspberry Pi I had on test.

In short – when it is powering a computer like a Pi, it thinks it is CHARGING it – and hence continues to do this ad-infinitum and does NOT charge it’s own battery because the “Pi” never becomes “fully charged”.

Oh dear,  another one bites the dust. I can’t even chuck it back at them as the function is fairly clear in the paperwork.

Even as a phone charger – why on earth they don’t show the state of the batteries when discharging into a load…. is beyond me.

But it gets worse. After fully charging the unit (according to the indicators) with 3 batteries, mid-afternoon yesterday, then last night I plugged my flat phone into the unit.  This morning when I arose, the unit showed the batteries were less than half full (one being down to 25%) and yet the phone was only about 20% charged.  Something seriously wrong there – I will try again.


39 thoughts on “The TOMO Power Supply

  1. Throughout all of this I have my two Ravpower units, one providing constant power to a Raspberry pi, the other one travelling with me. Spot on, both of them.

  2. Just an update from the use of my Tomos at home … BIG PROBLEM.

    There’s 2 USB outputs on the Tomo, and I had a Raspberry Pi 2 on one of them and an Orange Pi Zero on the other. I had a problem with the Orange Pi Zero which required me to pull the USB cable feeding the Orange out of the Tomo – this rebooted the Raspberry too. This was with fully charged batteries AND with DC power in connected.

    At first I wasn’t sure what caused the Raspberry to reboot but repeating the above procedure interrupts the power to the Raspberry every time.

    This renders the 2x USB output facility of the Tomo completely pointless if a problem with one device even with the batteries charged AND with DC power input causes the other to reset. The voltage display on the Tomo briefly shows 10 (yes TEN) volts during this reset. I’ve not verified whether the 10v is true (I could connect one of my USB power meters to check this) but if it is true, it’s rather scary and could damage the 5v devices connected.

    I had hoped to power 2 devices from the Tomo, reluctantly accepting its weakness that it does not recharge the batteries whilst there is a load as the batteries should cover several power outages before requiring removal for recharge – but if I can only power one device without risking one device rebooting the other, it’s starting to run out of useful attributes.

    I can no longer recommend the Tomo to anyone – it’s just an expensive paperweight.

    1. It gets worse Darren, as the batteries start to go down, maybe halfway, the unit disconnects and reconnects frequently. On Android at least this has the effect of momentarily turning the phone on so if you are trying to sleep it wakes you up. Eventually the unit turns off completely before the batteries are anywhere near done. The company I bought mine from continues to hide behind “send us a video” and will not discuss anything else despite having been pointed to this blog and having had the problem clearly explained by email. I need to find a way to let Aliexpress know they are playing games. Yes, one to avoid. A shame as it looked promising.

      1. Hi Pete,
        I did several battery drain load tests (with DC power in switched off to emulate a power cut) and tested 2x Tomo device and neither exhibited the power disconnect/reconnect issue that you suffered. I left a crib job running that appended timestamps to a text file every minute and there were no breaks and no evidence of reboot in any Linux log file. I wonder if the unit is fussy with its choice of 18650 battery? I shall not waste too much time guessing or speculating as there are too many negative points and question marks regarding the Tomo now to justify further investigation.

        Anyway…I thought you were supposed to be taking time out to enjoy Spanish sunshine 😉 ….

  3. Why not an 6V motorcycle battery, a charger and a good LDO from 6 to 5V 2.5A ?
    Charger can be controlled by the Pi when to charge the battery or not.

    1. Why not indeed – in fact, until recently I had 4 12v alarm lead-acid batteries charging off spare capacity of my solar cell here then running 5v via a switched-mode supply – all of which worked well until I discovered that even similar looking lead-acids can have differing charging requirements as two of them blew up like balloons. The only downside to the 6v motorcycle solution compared to a lithium solution is of course the potential cost of the motorcycle battery. I know nothing about motorcycle batteries but a quick look through Ebay suggests they need occasional acid refill which seems a little messy and seem to be around 6AH?? My 12v 7AH batteries seem cheaper, less messy and more powerful but then I could be looking entirely in the wrong place.

  4. The TOMO supplier is trying the language barrier.. “Please send a video” – I’ve explained that a video would cost more in time to produce than it is worth – and that clearly the product does not work – back came the response “please send a video”….

    I think Aliexpress tell them to say this as I’ve seen this time and time again. I’m guessing “not fit for purpose” is a phrase that would be alien to these guys.

    Ok, so be it…. DON’T BUY THIS PRODUCT

    1. i’ve one of them begging for 3 days to withdraw a dispute… i said to resend me the item missing and give me proof of that, and only then i’ll cancel dispute… every morning the same phrase… no way i’ll cancel… they’re like shark teeth, 1 down, an other on the back will take his place… no mercy…

    2. Hi Pete,
      It’s such a shame that your Tomo is faulty as I used a USB current/capacity measure thingy to record the discharge of the cells on my RPi2 test and it gave a promising indication of capacity. I’m inclined to do another full discharge test on it to see if it behaves as well for me as it did the first time. I imagine the choice and quality of 18650 cells used makes a difference to the results too.
      Your struggles with getting a refund, I can sympathise with – I’ve had hit/miss results before and nearly fell out with ‘Banggood’ a little while back but they came good in the end. Quite the opposite experience to most UK retailers but then again so is the pricing! I hope you get it sorted.

      Regarding Tindie – I’ve found most offerings on there rather overpriced but then I’ve probably been conditioned by the stack ’em high and sell ’em cheap chinese sellers….

  5. I use this:

    fed from a crappy 12V/2A mains power supply to charge 4 paralleled recycled ICR18560 and then two MT3608 paralleled (I know, pretty dirty hack, but I have had them lying around) to boost the voltage from the buck converter and/or battery up to the 5.1V.
    The buck regulator both charges the battery to 4.1V and provides 4.1V for the boost regulators (the boost regulator works fine down to less than 3V).
    It can handle my OPi PC along with a USB HDD and was rather cheap to build. Wihout the HDD, probably only one MT3608 module would be sufficient.

    1. Try as I might – I cannot log into Tindie as they’ve screwed up Facebook access..

      This begs exactly the same questions as others – and an additional one – how does he get charging at 2.5 amps when nearly all the chips out there are 1 amp or less.

      Does it support simultaneous use and charging
      Does it recover from totally flat battery without disconnecting the load – i.e. will it recover, charge the battery and supply the load

      I’m getting REALLY jaded now as the amount of “solutions” I’ve tried that simply DO NOT do the job is staggering.

      That TOMO is the biggest heap of rubbish I’ve come across – even for charging phones – two different phones and it half-charges them and then turns off. I wrote to the Chinese suppliers and I’m now at the frighteningly familiar “will you send us a video of the problem” stage. Like I’m going to send an hour long video waiting for it to turn off! I feel a complaint coming on.

    2. Right- finally got some login info from Tindie and I’ve asked for an early sample – there are several power supply projects on there now – looking at the power output I think there’s a lot of “lack of thought” and some pretty optimistic pricing… this one however looks like it might be worth a second glance. I’ve asked hopefully all the right questions and will let you know if anything comes out of it…

    1. I am MUCH more wary than I’ve ever been after the last episode but I have to say this DOES look good – anyone up for giving it a go?

    1. Good – but discharging at 2 amps – while charging at 1 amp won’t last very long, Glen 🙂

  6. Hi, I played around with the 4 cell version tomo v8-4,
    besides the “feature” that it priories the load instead the internal battery I had 2 more issues with it.
    1. the internal 5V to 2,5V LDO was a wrong one, it drops out at 3,5V of the cells which is far to early, I exchanged it with one which drops out at 3V.
    2. I planed to use this with cells I got from older Notebook batteries. This Cells does not have any protection circuits. The tomo charged them infinitely and the cells were getting really hot. With other newer cells (also without protection circuit) it worked fine, so it could also be something with my cells.

    however, it is so sad that they don’t make real use of the dotmatrix display, the µC has all infos about voltage, current of each cell would be nice to see it on the display.
    (more details of my review (in german) of the tomo are here:

    1. Oh DEAR ME Roland – so in addition to the other issues, what you’re saying here is that if we use old laptop batteries – which people are likely to do – the unit could overheat. Good grief – did we pick a winner here… I hope the designers are reading this – I’ll do my best to make sure they do.

    2. Hi Roland – thanks for sharing your experiences.

      I wasn’t brave enough to consider unprotected cells as my requirement is for a backup power supply to be connected 24/7/365 at a remote location unattended for several weeks at a time. My Canwelum protected cells are not even getting slightly warm when left connected to DC power in (without load) … nor are my other set in my other Tomo connected to my RPi 2. Both have been running for over a week now.

      I’d have hoped that anyone building a backup power supply system that needs to be relied upon (i.e. not just a cheap phone booster) would choose brand new cells rather than unknown old laptop batteries. Laptop batteries get a hard life and are usually pretty disappointing/unreliable when they get old.

      I thought I’d just share my “no overheating here with new, protected cells” information before anyone panics too much but I’m extra glad now I paid extra for protected cells. I just didn’t trust the idea of cheap, unknown chinese electronics charging a battery pack in a kitchen cupboard 50 miles from home out of sight …

  7. Imagine Pete, your inglish, we get mislead even easier by the semi inglish/chinese translations. 😁

  8. Ahh but Pete it does still offer a UPS like facility for a Pi as loaded with 4 cells, it’s allows a lot of power out protection before the batteries run flat.

    I’ve had a USB current meter on the input whilst leaving one charging over the weekend and into this week and I can confirm that once the cells are fully charged, it does stop charging them and doesn’t keep adding current… plus no overheating.

    I think for the price, this is a good enough option for many applications but it is a shame it doesn’t charge whilst under load. I’d kinda noticed that too but equally just thought it was trickle charging and had a lot of work to do to charge 4 flat 186500s.

    If one needed even more mAh of protection before the cells run out of juice one could even cascade multiple Tomos charging out of each other. Ok…now I’m going a bit crazy…

    1. problem is, if you’re out of home and this thing does not charge till it thinks the PI is charged (so, never!), you run out of their juice in just a single run, even with power never gone away…

      1. I installed pre-charged 186500 cells in mine and applied DC power in. This may account for my unfortunate overlooking of the fact the cells don’t recharge in parallel with the Pi although I did quote the line about it favouring the load device until charged (I just didn’t digest the full meaning of that).

        However, it isn’t as bad as you suggest DrFragle and it won’t discharge itself “even with the power never gone away”. For as long as there is DC power in, the Pi will be powered from the DC power in. It will only consume the batteries (and eventually ‘run out of juice’) if the DC power in goes away for a total of 7 hours or however long it was on my test that it took to consume the entire 10500mAh battery capacity that only starts being consumed when the DC power goes away. If DC power in is restored part way through the discharge of the batteries, the Pi will power back up again….but admittedly the batteries will remain at their part-discharged state and NOT replenish themselves. I foolishly just thought it was charging REALLY slowly (4 completely flat 18650 cells plus a Pi to feed all from a single DC input would require quite a lot of input current now we think about it more thoroughly).

        I can’t apologise enough for the confusion – I was still testing my Tomo devices at home myself before deploying them at my holiday home (I wanted to test for overcharging/overheating – so far so good on that).

        Sorry for the confusion and not noticing that

        1. No need for apologies – we read the advert as well and we too totally ignored the statement which clearly says it won’t charge both itself and the load at the same time.

          The answer is to put this to one side having warned others and keep looking.

          1. Hi Pete,

            There are several ready made “Pi UPS” options on the market (and a few PCB kits) – from the usual UK Raspberry Pi aftermarket dealers – but most are overpriced and aimed non-technical users IMO… which, if any, have you tried? I think we’ll end up having to build something from scratch to get all boxes ticked but it would be handy to know if you have tried and ruled out some (non-chinese) RPi UPS boards so that readers don’t buy any of those and face disappointment.

            1. I’ve tried several – and in each case when they failed to live up to their claims – the supplier has refunded me – I’ve yet to find a satisfactory solution. It seems that one issue is that the chips out there for charging LI-ION batteries are usually 1 amp max – while circuits can take up to 2 amps. I was talking about this with Aidan yesterday and if the only solution is to charge at a lower rate than the output might sustain – the only solution would be to diode-or the input with the battery so that either can supply the power output circuit… that way the input can be feeding the charger circuit and the power output stage at the same time. It did occur to me that then if you were electronically switching the output on and off – you’d have to run any control/display circuitry off the varying voltage battery – not good – OR have ANOTHER lower power switcher output supply to generate 3v3 for the control chip/display. Going back to my original idea of a Nano+SSD1306, that would be needed. None of these cheap circuits offer signals to warn a processor however… Aidan and I are kind of half-seriously thinking of putting something together – PCBs from China are cheap enough – but as for making something available for others, soldering even the simple 328 chip SMT version could be much of a challenge for some.

    2. Well, for me, the lack of charging under load renders the unit unusable. Imagine a situation where you are away… the power fails – maybe earth leakage or some spike – the power goes off. Eventually the batteries run down – and you managed to get someone to flick the power on for you – this unit is now running with flat batteries…. there is no way around it – you have to disconnect the load until the batteries fully charge. For many applications this renders the unit unusable.

      Currently the only working solution (aside from purchasing a far more expensive unit) I’ve come across is at board level – and I discussed this in the blog some time ago -but even that only has 1 amp output. I spent some time yesterday looking for chip level solutions – all the nice and easy 6 pin chips are able to handle 1 amp output only – OR are terribly inefficient at under 3v. There is one Texas chip which does the job and has a logic enable input, but it is no-where near as simple and even then that is only the power output side – I’ve yet to find a simple charging chip that can push 2 amps into the Lithium batteries. Of course – it’s a big world out there and I’m hoping someone is going to write in and prove me wrong.

  9. I have a suggestion, send it to BigClive and see if he can reverse engineer it enough to allow changing the logic around or even disabling it? Or have I missed something.

    1. Not a bad idea really… though I’m probably able to check it out myself… in the meanwhile I have a working solution that’s a bit messy as discussed in a separate blog entry. I was hoping this unit could be passed to the wife to charge her tablet with – but see updated version of the blog this morning.

Comments are closed.