 Notice. New forum software under development. It's going to miss a few functions and look a bit ugly for a while, but I'm working on it full time now as the old forum was too unstable. Couple days, all good. If you notice any issues, please contact me. |
Forum Index : Microcontroller and PC projects : NY Geothermal Battery Backed Power Supply
| Author | Message | ||||
| Paul_L Guru  Joined: 03/03/2016 Location: United StatesPosts: 769 |
I am going to need a battery backed power supply to power an E100 board and a 7 inch display during power failures. I have a 10kw backup generator which starts automatically after about 45 seconds of a power failure, but I don't want to loose the program thread or data. An E100 input will sense the power failure and the program will open all the high power relays and wait for the generator to start delivering power. Then the program will operate the heat pump in a reduced power mode. I was thinking of using 5 NiCd D cells in series and keep them charged with a 6v wall wart. The E100 will be powered by a separate 6v wall wart feeding an LDO 5v regulator. The backup batteries will feed the regulator through an isolation diode during power failures. Has anybody got any better ideas? Paul in NY |
||||
Grogster Admin Group  Joined: 31/12/2012 Location: New ZealandPosts: 9066 |
Here is my latest PSU design:  Up to 3A output, fully adjustable and short-circuit protected. You need to bolt this one's aluminium right-angle heatsink bracket, to a bigger heatsink depending on input-output voltage difference and load current. This one has transistor buffered mains-fail and fuse-fail lines designed to interface to MM I/O pins via the standard 10k/1k pull-up arrangement. When powered, the mains-fail is a separate circuit on the board via an optocoupler, which pulls the MM input LOW while powered. The fuse-fail is the same concept, but it uses a transistor rather then the opto, as the fuse supply has the same ground as the rest of the circuit being powered - the mains-fail does not, hence the opto isolator. If the mains fails(or the fuse blows), the artificial pull-down will vnish, and the 10k/1k pull-ups will pull that line(or lines) high, and the MM can pick up on that with an interrupt. Battery charging is via a simple diode-gate and series charging resistor chosen to TRICKLE-CHARGE a suitable gel-cel. This design DOES NOT allow for fast-charging, and does not monitor the battery. Trickle-charging is fine done like this, but naturally, it takes the battery time to charge up when done like this. Smoke makes things work. When the smoke gets out, it stops! |
||||
bigmik Guru Joined: 20/06/2011 Location: AustraliaPosts: 2870 |
Gday Paul, I use one if these XIOAMI battery packs, this particular one I have 4 of and love them.. They have the capability of supplying 2A for 5hr and whilst I haven't TESTED it for that long with that load, they do drive reasonable current for quite a while.. The great thing about these, that many similar units can't do, is deliver power at the same time as it is being charged.. There are unfortunately fakes out there so do a Google search first to identify a fake as many of these fakes have caught fire.. Of course you could also just get a basic 600VA UPS unit. Regards, Mick Mick's uMite Stuff can be found >>> HERE (Kindly hosted by Dontronics) <<< |
||||
| WhiteWizzard Guru Joined: 05/04/2013 Location: United KingdomPosts: 2794 |
I would also recommend a 'mobile charging pack' similar to what Mick has linked to. I have used these countless times and have several MMs around the house using these 'packs' as mini UPSs. The 10400mAh units are great for powering E100s however, note that some of these packs don't 'monitor' the charge level once reached to capacity. What this means is that IF you leave the 5v charging PSU plugged in all the time, AND run your circuit from the battery-pack's 5v output, then the pack will charge itself (and run your circuit), BUT when it has reached full charge, it will discharge to empty. Make sure the Battery Pack is ok to deliver power while charging. I would not recommend buying 'cheap' online as the fakes do indeed get warm and in some cases catch fire. Pop into a 'branded' local mobile phone shop (or national computer store) and get a branded 'pack'. May cost you double the online price, but is far safer  WW For everything Micromite visit micromite.org Direct Email: whitewizzard@micromite.o |
||||
| robert.rozee Guru Joined: 31/12/2012 Location: New ZealandPosts: 2290 |
i'm going to be controversial and suggest you might wish to look at selecting your backup batteries based upon their projected life: 1. lithium ion (and variants thereof) have a poor history of limited shelf life. when working in product development, we used to reckon on a 2 year service life beyond date of manufacture. this may well have improved significantly, but i would still be reluctant to rely upon any lithium technology having a service life of over 5 years. Li-Ion is ideal for cellphones, that have a short lifespan anyway, and tesla seems to be also getting good performance and lifespan. 2. Ni-Cads have an enormous instantaneous available current, ideal for power tools (though their use is regarded as un-PC these days), but perform poorly on continuous charge. 3. lead-acid gel cells are happy to be float-charged, but still outgas to some degree and may stretch as far as a 10 years service life. before presenting the last option, a question: how often do you envisage having power cuts? on a per-year basis, what mA/h drain would you estimate when powering an E100 and associated accessories? 250mA x 45 seconds x n power outages per year. now consider... 4. long-life alkeline primary cells. these can be had with a 10 year shelf life. they are cheap. it is relatively easy to create a battery replacement alarm built into the E100 that uses the number of power cuts and a maximum shelf life counter. the argument is similar to the one against using rechargable batteries in the torch that sits in your car glovebox or beside your bed for in case of emergencies. sometimes rechargable batteries are not the best choice. cheers, rob :-) |
||||
| Phil23 Guru Joined: 27/03/2016 Location: AustraliaPosts: 1664 |
Rob, You didn't mention NiMh, which probably aren't applicable here. But as a photographer, (seen you in those forums), you'd well be aware of the performance of Sanyo's Eneloop NiMh's & their long self/low self discharge rate. As far as SLA is concerned, that would be my choice, but it would be important to choose the correct ones. Panasonic seems to have about 7 different types for a given pack size & capacity in some cases. They quote life span as 1 to 10 years across the types & temperature ranges. This is one of the data sheets I received with recent stocks. Haven't looked up full specs on the LC-Q type, but it's got the best service life if it's suitable. Cheers Phil  |
||||
| Phil23 Guru Joined: 27/03/2016 Location: AustraliaPosts: 1664 |
There's a couple of data sources available, Panasonic VRLA Guide and Panasonic SLA Technical Handbook 1999 Phil. |
||||
| bigmik Guru Joined: 20/06/2011 Location: AustraliaPosts: 2870 |
GDay Paul, All, Have a read of the specs on the XIOAMI link I posted.. They claim 9 levels of protection including over charge, over voltage and over current.. I can heartily recommend this pack.. But as I mentioned BEWARE of the fakes.. (Do a search for fake XIOAMI packs... These are easy enough to pick).. I have a design I did that has 2 x RS485 TX/RX pairs and runs an LCD and emulates one of the betting terminals at work (software graciously written by Peter Carnegie). The ide being that we have these terminals running upto 1.5km away from the host in `paddocks and car parks' and when we do the wiring there is often no power there so we cannot run our terminals or have to lug around heavy UPS units.. Note this is temporary wiring... I last used this tester 12 months ago and then tossed it and the XIOAMI in the back of my van.. This year (Nov) I pulled it out again and it still had full charge (at least 75% +) showing and after several hours of driving my MicroMite board it finally dropped down to show 50-75% .... So this is after 12 months sitting in my work van and maybe 4 hours continuous use in 2015 and similar 2016. It makes powering `mite based projects easy.. Just feed it with an iPhone USB power supply and it will effectively never need to be touched again. Anyway there are several choices for you to act on now. OH.. The XIOAMI comes in 5200 and 16000mAh as well Mick Mick's uMite Stuff can be found >>> HERE (Kindly hosted by Dontronics) <<< |
||||
| Paul_L Guru Joined: 03/03/2016 Location: United StatesPosts: 769 |
@Grogs, @bigmick, @WW, @Phil23 .... Thanks guys but I think what you have proposed is a little on the overkill side. @rob .... I like the way you think. LiFe - short life, NiCd - don't work well with continuous charging forming a memory. SLA - hydrogen out gassing. Alkaline - cheap long life. The key lies in the fact that I have a fully automatic backup generator hooked up to a natural gas pipe which starts and accepts the whole house load in less than a minute. We have an average of 10 power failures a year here, mostly due to cars destroying power poles or ice and wind shorting or pulling down the 33 KV lines. In all instances the generator will be on line in about a minute. So, the batteries have to power an E100 plus a display drawing maybe 700 ma at 5 volts for maybe 10 minutes a year. Alkaline D cells have a capacity of 10,000 to 20,000 msHr. A series array of 5 cells feeding 7.5 volts to a regulator circuit should run the E100 and its touch display for at least 10 hours which amounts to an average of 60 years with our normal power failure rate! The only problem would be to limit the possibility that the regulator could apply a continuous current drain. I think I can stop this with a diode back biased by the normal power supply for the E100. Can anyone suggest a specific circuit to do this? Paul in NY |
||||
 Print this page |
