Hi Dimiter, On 11/12/2013 7:20 PM, dp wrote:> I remember some years ago I had a tiny MP3 player > which worked for a while until on a very hot day it died (while > not being in use). I opened it, the MCU clock was not starting. > My main suspect was overvoltage kill (I did not go much further > to verify that though). It had a Li-ion battery which powered > it unregulated.Actually, on a related note... monitoring the temperature of a battery (esp under charge) is usually worth the extra component cost. As batteries fail (lose the ability to take a charge), they often get a lot warmer during charging. Many charge controllers have provisions for a temperature sense input. And/or, safeguards (often simple timers) to ensure a battery doesn't keep calling for charge indefinitely! Finally, it's worth the effort to "qualify" your battery provider. Impurities in cells can lead to spectacular results! :> (not the sort your lawyer/accountant will enjoy)
Preferred LiFePO4 USB charger/regulator IC
Started by ●November 11, 2013
Reply by ●November 13, 20132013-11-13
Reply by ●November 13, 20132013-11-13
On Wednesday, November 13, 2013 6:37:57 AM UTC+2, Don Y wrote:> Hi Dimiter, > > On 11/12/2013 7:20 PM, dp wrote: > > > I remember some years ago I had a tiny MP3 player > > which worked for a while until on a very hot day it died (while > > not being in use). I opened it, the MCU clock was not starting. > > My main suspect was overvoltage kill (I did not go much further > > to verify that though). It had a Li-ion battery which powered > > it unregulated. > > Actually, on a related note... monitoring the temperature of > a battery (esp under charge) is usually worth the extra > component cost. As batteries fail (lose the ability to take > a charge), they often get a lot warmer during charging. > As batteries fail (lose the ability to take > a charge), they often get a lot warmer during charging. > Many charge controllers have provisions for a temperature > sense input. And/or, safeguards (often simple timers) to > ensure a battery doesn't keep calling for charge indefinitely!Of course a temperature sensor is good. Usually the problem that comes with it is the necessity for as many sensors as battery cells and ensuring a good temperature contact mechanically. On the Nukeman I had put a single sensor (it had 18 cells, mind you) which was just to detect overheating. But NiMH being full can still be detected by the dropping voltage while being charged (because of heating). Worse than NiCd but still usable. Then I also put a max. battery voltage limit to the regulator (which regulates for current) - this also allows the thing to run with no battery connected - and some sane time limit to the charging in case slope detection fails or if I want to set the charging to too low a current for slope change to take place. Dimiter ------------------------------------------------------ Dimiter Popoff, TGI http://www.tgi-sci.com ------------------------------------------------------ http://www.flickr.com/photos/didi_tgi/sets/72157600228621276/
Reply by ●November 13, 20132013-11-13
Hi Dimiter, On 11/13/2013 11:28 AM, dp wrote:> On Wednesday, November 13, 2013 6:37:57 AM UTC+2, Don Y wrote:>> Actually, on a related note... monitoring the temperature of >> a battery (esp under charge) is usually worth the extra >> component cost. As batteries fail (lose the ability to take >> a charge), they often get a lot warmer during charging. >> As batteries fail (lose the ability to take >> a charge), they often get a lot warmer during charging. >> Many charge controllers have provisions for a temperature >> sense input. And/or, safeguards (often simple timers) to >> ensure a battery doesn't keep calling for charge indefinitely! > > Of course a temperature sensor is good. Usually the problem that > comes with it is the necessity for as many sensors as battery cells > and ensuring a good temperature contact mechanically.I think many OEMs get away with a single sensor by expecting the cells to be from the same manufacture batch -- and expecting their behaviors to largely "track". Likewise, expecting this to continue to be the case thereafter (replace the BATTERY and not just individual cells out of it). Clifford is apparently using a single cell design so one sensor covers "all cells" :>> On the Nukeman I had put a single sensor (it had 18 cells,Yikes! Because you needed ampacity (wide) or potential (stacked)? (as well as CAPACITY)> mind you) which was just to detect overheating. But NiMH being full > can still be detected by the dropping voltage while being charged (because > of heating). Worse than NiCd but still usable. Then I also put a max. > battery voltage limit to the regulator (which regulates for > current) - this also allows the thing to run with no battery > connected - and some sane time limit to the charging > in case slope detection fails or if I want to set the charging to too > low a current for slope change to take place.Batteries are a royal PITA! They make a device infinitely more useful (depends on device) but bring so many "side effects" to the design (and use). Esp for non-replaceable designs! I can't count the number of things, here, that have lost their utility simply because they weren't used *often* enough to merit keeping them charged. And, as a result, the "batteries" have died inside. Cheaper/easier to replace the entire device than to hope to find a battery replacement! :< [Hence my comment to Clifford re: "battery holder" to endear him to DIY customers!]
Reply by ●November 13, 20132013-11-13
Hi Don, On Thursday, November 14, 2013 12:36:33 AM UTC+2, Don Y wrote:> ... > > Yikes! Because you needed ampacity (wide) or potential (stacked)? > (as well as CAPACITY)Both, generally I needed many hours (was it 6 or 8...) autonomous operation (here it is: http://tgi-sci.com/tgi/pntb.htm ) . I had two groups of 9 batteries in series each tied in parallel. I was wise enough to have them discharged before making the parallel connection, obviously, and was careful enough to measure the voltage before that, but generally things worked quite OK. I had mounted each cell to be pressed against its parallel counterpart and that was all, the battery lasted for years. Used those 76mm long 17mm diameter cells, the whole pack was 75Wh. Not that much more compared to what todays laptops have, nor was it much larger. Somewhat heavier than equivalent Li-ion of course. Dimiter ------------------------------------------------------ Dimiter Popoff, TGI http://www.tgi-sci.com ------------------------------------------------------ http://www.flickr.com/photos/didi_tgi/sets/72157600228621276/
Reply by ●November 13, 20132013-11-13
Hi Dimiter, [still 30C, here! Ice on your trees, yet? :> ] On 11/13/2013 5:50 PM, dp wrote:>> Yikes! Because you needed ampacity (wide) or potential (stacked)? >> (as well as CAPACITY) > > Both, generally I needed many hours (was it 6 or 8...) autonomous > operation (here it is: http://tgi-sci.com/tgi/pntb.htm ) . > I had two groups of 9 batteries in series each tied in parallel. > I was wise enough to have them discharged before making the parallel > connection, obviously, and was careful enough to measure the voltage > before that, but generally things worked quite OK.Hmmm... I guess I would have tried to treat them as separate packs charged independently with their outputs merged through sharing "diodes" (or actively)> I had mounted > each cell to be pressed against its parallel counterpart and > that was all, the battery lasted for years. Used those 76mm long 17mm > diameter cells, the whole pack was 75Wh. Not that much more compared > to what todays laptops have, nor was it much larger. Somewhat heavier > than equivalent Li-ion of course.I'm treading carefully on my battery requirements. Trying hard to find ways to "eliminate load" so I can reduce the size and complexity of the "battery". But, I'm more constrained on space/volume than you are so I don't have much choice to "add capacity". I guess I'm hoping for a "breakthrough" before I *need* it! :> My current strategy is to make devices inexpensive enough that you can buy several and just swap them out as their batteries need recharging. Even if that means carrying a "spare" with you...
Reply by ●November 13, 20132013-11-13
On Thursday, November 14, 2013 4:38:17 AM UTC+2, Don Y wrote:> Hi Dimiter, > > [still 30C, here! Ice on your trees, yet? :> ] >No, summer was half as long as it should have been but I guess that paid for a delayed cold. Still around +15C max/5C min, fairly merciful for mid November.> > On 11/13/2013 5:50 PM, dp wrote: > >> Yikes! Because you needed ampacity (wide) or potential (stacked)? > >> (as well as CAPACITY) > > > > Both, generally I needed many hours (was it 6 or 8...) autonomous > > operation (here it is: http://tgi-sci.com/tgi/pntb.htm ) . > > I had two groups of 9 batteries in series each tied in parallel. > > I was wise enough to have them discharged before making the parallel > > connection, obviously, and was careful enough to measure the voltage > > before that, but generally things worked quite OK. > > Hmmm... I guess I would have tried to treat them as separate packs > charged independently with their outputs merged through sharing > "diodes" (or actively)These were (well, must have been) my initial thoughts back then (20 years ago when I first did it this way). But things are not that bad, really. Given that I sell a battery pack (cells soldered to a PCB) and I buy cells of the same batch, have parallel cells stuck to each other, it makes little difference to whatever they have wound inside the batteries whether it is in one or in two packages. Once you charge the whole pack and keep for a day or two on trickle charge things level out quite OK.> I guess I'm hoping for a "breakthrough" before I *need* it! :>Hah, aren't we all (for this or some other breakthrough) :D .> My current strategy is to make devices inexpensive enough that you can > buy several and just swap them out as their batteries need recharging. > Even if that means carrying a "spare" with you...That's of course a good strategy but not applicable to my low volume (sometimes one-off) designs, hence the different approach. I have yet to make a Li-ion charger though, hopefully my bold approaches don't put me in the news :D :D :D . (I think I am wise enough not to be too bold with these things but then who knows.... :D ). Dimiter ------------------------------------------------------ Dimiter Popoff Transgalactic Instruments http://www.tgi-sci.com ------------------------------------------------------ http://www.flickr.com/photos/didi_tgi/sets/72157600228621276/
Reply by ●November 14, 20132013-11-14
Hi Dimiter, On 11/13/2013 8:52 PM, dp wrote:>>> Both, generally I needed many hours (was it 6 or 8...) autonomous >>> operation (here it is: http://tgi-sci.com/tgi/pntb.htm ) . >>> I had two groups of 9 batteries in series each tied in parallel. >>> I was wise enough to have them discharged before making the parallel >>> connection, obviously, and was careful enough to measure the voltage >>> before that, but generally things worked quite OK. >> >> Hmmm... I guess I would have tried to treat them as separate packs >> charged independently with their outputs merged through sharing >> "diodes" (or actively) > > These were (well, must have been) my initial thoughts back then (20 > years ago when I first did it this way). But things are not that bad,Ah, OK.> really. Given that I sell a battery pack (cells soldered to a PCB) and > I buy cells of the same batch, have parallel cells stuck to each other, > it makes little difference to whatever they have wound inside > the batteries whether it is in one or in two packages. Once you charge > the whole pack and keep for a day or two on trickle charge things > level out quite OK.I suspect for an instrument with the level of "investment" as yours, you probably don't encounter folks trying to replace batteries "on the cheap".>> I guess I'm hoping for a "breakthrough" before I *need* it! :> > > Hah, aren't we all (for this or some other breakthrough) :D .Sometimes we get lucky! :>>> My current strategy is to make devices inexpensive enough that you can >> buy several and just swap them out as their batteries need recharging. >> Even if that means carrying a "spare" with you... > > That's of course a good strategy but not applicable to my low volume > (sometimes one-off) designs, hence the different approach. I have yet > to make a Li-ion charger though, hopefully my bold approaches don't > put me in the news :D :D :D . (I think I am wise enough not to be > too bold with these things but then who knows.... :D ).Some of my devices are in the "coin cell" ballpark. If you're already that small, adding another cell is a big increase in volume, etc. (One I am targeting as a "large wristwatch") Of course, the advantage of that small size is that it is not unreasonable to stuff a spare in your pocket in case the battery in the first is exhausted and needs to sit on a charger for some time...
Reply by ●November 15, 20132013-11-15
On 13/11/13 10:34, mike wrote:> Is this a REAL USB device that can negotiate power levels > with the host? Or are you using the host as a dumb power supply?Initially, the latter. Using USB connections means chargers are readily available.> Is this a production device that is required to meet, and be tested to > international safety standards?Not a concern at this level - it's a ham radio device (fox transmitter and sniffer receivers). The transmitter is programmable (and the RTC is settable) but that's through the iButton port which is also used for event scoring - not through the USB port. I don't want to roll my own charger, although it's not hard to do a basic-level job (current limit into a 3.4V regulator). The devices have very low quiescent current (uA) so I'm not worried about the battery going flat on the shelf. LiPo cells are much less stable (need temp monitoring and cutout) which is why I'm choosing LiFePO4. I should only need low-voltage cutout. Thanks all for your comments. It appears no-one wants to recommend a chip for this, or perhaps has even used one. I'll be off USENET for a couple of weeks now. Any further comments would be appreciated, even if I don't respond quickly. Clifford Heath.
Reply by ●November 15, 20132013-11-15
On 11/14/2013 10:57 PM, Clifford Heath wrote:> On 13/11/13 10:34, mike wrote: >> Is this a REAL USB device that can negotiate power levels >> with the host? Or are you using the host as a dumb power supply? > > Initially, the latter. Using USB connections means chargers are readily > available. > >> Is this a production device that is required to meet, and be tested to >> international safety standards? > > Not a concern at this level - it's a ham radio device (fox transmitter > and sniffer receivers). The transmitter is programmable (and the RTC is > settable) but that's through the iButton port which is also used for > event scoring - not through the USB port.I'd be interested in learning more about the device. Back in the early '60's we had transmitter hunts on 6 meters. I was about 15 years old at the time. Went to my first hunt and watched guys driving their mobile rigs around watching the S-meter trying to find the hidden transmitter. Second hunt, I took a 6-meter converter cobbled onto the front of a transistor radio and a loop antenna. Took one direction reading. Drove 90 degrees away from it and took another reading. Then drove right to the hidden transmitter. Third hunt, nobody else showed up. Good times... As for your charger...I'd go to Best Buy. They've got a recycle bin for cellphone batteries. Talk the guy into letting you have a pile. Check out what chips they use. These guys have some battery protection devices and cells with such devices already packaged http://dx.com/p/charge-discharge-protective-circuit-board-for-rechargeable-li-ion-batteries-17-4mm-1-9mm-26112 I bought two, but never used 'em. There are protected batteries for flashlights in various sizes with chargers available at the same place. I don't think your problems are technical. Depending on how long you plan to sell and support the device, finding something that will be available in the future (next week) is a big gamble. If we have the same concept of a transmitter hunt, I think the rational choice is 3 AAA alkaline cells or thereabouts. Not everything in the world needs to be rechargeable.> > I don't want to roll my own charger, although it's not hard to do a > basic-level job (current limit into a 3.4V regulator). > > The devices have very low quiescent current (uA) so I'm not worried > about the battery going flat on the shelf. > > LiPo cells are much less stable (need temp monitoring and cutout) which > is why I'm choosing LiFePO4. I should only need low-voltage cutout. > > Thanks all for your comments. It appears no-one wants to recommend a > chip for this, or perhaps has even used one. > > I'll be off USENET for a couple of weeks now. Any further comments would > be appreciated, even if I don't respond quickly. > > Clifford Heath.
Reply by ●November 15, 20132013-11-15
On 15/11/13 19:30, mike wrote:> On 11/14/2013 10:57 PM, Clifford Heath wrote: >> Not a concern at this level - it's a ham radio device (fox transmitter >> and sniffer receivers).> I'd be interested in learning more about the device. > Back in the early '60's we had transmitter hunts on 6 meters.Great fun, yeah? I learnt to hunt with the Melbourne crew (VK3's), who have 30+ years experience and an amazing range of talents. Google for VK3VT (now VK3FOX), VK3MZ, VK3YNG, for some of the most prolific constructors. Almost all the gear is home-built, and these days, the most sophisticated crews have fully integrated in-car systems with motorized rooftop antenna arrays (including in-car control of polarisation in one case), 2m dopplers, Watson-Watt receivers, etc, all feeding into custom-made SDR systems with real-time digital spectrum display (brick-wall filtering), automated bearing extraction from the returned antenna lobe display, drawing vectors on a moving map display, etc. The annual hunt with the VK5's at Mt Gambier in June has a big five-leg night hunt, usually spread across 3 or 4 bands. Regular hunts are on 80m, 10m, 6m, 2m, 70cm and 22cm. Foot events are on 2m mostly. When we run ARDF events (see below) they're 80m or 2m. 6m hunts are interesting in the pine plantations around Mt Gambier. Young pines are planted on a 3m grid, then thinned to a 6m grid. If you watch the vector display on my team's Watson-Watt as the car creeps forward, the bearing does a full 360 swing for every 6 metres you drive forward. Very hard to DF! Anyhow, I'm in Sydney now, and the barrier to building up such systems is very high, so I'm going back to basics, building 80m gear for ARDF and especially sprint ARDF, events held for runners on foot. ARDF events use five transmitters on the same frequency, transmitting one at a time in a five minute cycle (or for sprint, a one minute cycle, twelve seconds each). My new 80m fox is about the size of a cigarette packet, and pumps out up to a watt. An MSP430 will lock the VCO frequency to its 32KHz clock, while also doing Morse code modulation, managing the ARDF cycles, keeping real time, and running the iButton scoring that I built years ago. that's the plan anyhow. The receiver front end has been designed to feed an Si4735 digital radio chip, for which we have an SSB software patch. It will use a loop or loopstick and a sense wire, phase shifted and mixed to produce a cardioid - hunting on the null. In future I'm keen to try SDR directly in an ARM7 CortexM4 after down-converting to a 45KHz IF, a broadband DF antenna for HF to build and try, and maybe a hand-held Watson-Watt with a colour cell-phone LCD display.> As for your charger...I'd go to Best Buy.I'm not concerned with cells, and don't want 3.6 or 3.7 volt Li-ions. I'm concerned with USB charging for LiFePO4, which charge to 3.4V, and don't tend to explode at random. They're used in portable drills etc, which have a higher robustness and safety requirement than your typical phone or MP3 player.> These guys have some battery protection devices and cells with such > devices already packaged > http://dx.com/p/charge-discharge-protective-circuit-board-for-rechargeable-li-ion-batteries-17-4mm-1-9mm-26112Nice little device. I might have another use for them.> If we have the same concept of a transmitter hunt, I think > the rational choice is 3 AAA alkaline cells or thereabouts. > Not everything in the world needs to be rechargeable.Been there, done that. Alkaline is fine for sniffers, but you burn a lot of cells transmitting one watt or more... and there are six or eleven transmitters for an event (including the homing beacon). They need to be rechargable, and if I do it for the foxen, why not the receivers too? Clifford Heath.