|
Blotto Skorzany posted:Alkalines don't discharge linearly with any load I can think of.
|
#
?
Sep 13, 2014 00:00
|
|
|
|
| # ? May 5, 2024 22:55 |
|
|
So just to clarify (and I'm sorry if this is repetetive), 2 AAA batteries is probably still worth doing because it'll have a vastly longer lifespan than the 3 button cells, right? Assuming the circuit continues to work fine off of 3V, I just want to make sure I'm not basically substituting power source A for power source B without a meaningful gain. I would assume based on size alone the AAAs are likely to last longer, and I'd love to cram 3 in there, but I only have room for 2. :V
|
|
#
?
Sep 13, 2014 00:17
|
|
|
Funky Bunch Wikia posted:So just to clarify (and I'm sorry if this is repetetive), 2 AAA batteries is probably still worth doing because it'll have a vastly longer lifespan than the 3 button cells, right? Assuming the circuit continues to work fine off of 3V, I just want to make sure I'm not basically substituting power source A for power source B without a meaningful gain. I would assume based on size alone the AAAs are likely to last longer, and I'd love to cram 3 in there, but I only have room for 2. :V Since it's the same battery chemistry in this case, you can roughly judge capacity by size and weight. Which in this case should be real drat obvious.
|
|
#
?
Sep 13, 2014 04:57
|
|
|
Funky Bunch Wikia posted:So just to clarify (and I'm sorry if this is repetetive), 2 AAA batteries is probably still worth doing because it'll have a vastly longer lifespan than the 3 button cells, right? Assuming the circuit continues to work fine off of 3V, I just want to make sure I'm not basically substituting power source A for power source B without a meaningful gain. I would assume based on size alone the AAAs are likely to last longer, and I'd love to cram 3 in there, but I only have room for 2. :V 2 AAA batteries will give you a larger total number of mAh, but you have less wiggle room before you hit the minimum voltage requirements for your component to run reliably. 3 button cells in series will give you a larger starting voltage which allows more voltage drop before the device stops working, but your total mAh is reduced. Which one is the limiting factor will depend on the circuit draw and the cut off voltage for the equipment. AA norms are much better tested than AAA, but here are some discharge curves for one alkaline AAA: http://data.energizer.com/PDFs/E92.pdf Honestly, while you could probably test for which one is optimal with a breadboard supply, I'd just go ahead and test both and see which one works better.
|
|
#
?
Sep 13, 2014 05:15
|
|
|
poeticoddity posted:AA norms are much better tested than AAA, but here are some discharge curves for one alkaline AAA: http://data.energizer.com/PDFs/E92.pdf If you want more than that, how about three AAAAs? Feel like disassembling a 9v?
|
|
#
?
Sep 13, 2014 05:41
|
|
|
Blotto Skorzany posted:Alkalines don't discharge linearly with any load I can think of. What if it's an intelligent variable-draw load that continuously adjusts the current to produce a constant droop in the battery voltage? Then what, hmmmmmmmm? 
|
|
#
?
Sep 13, 2014 05:49
|
|
|
^^ checkmateEffective-Disorder posted:So I'm looking at a load that draws roughly 1 A at 5 volts. I want to use a battery. I was thinking a two-cell lithium battery would work nicely. But, I want to integrate on-board USB charging. The simplest solution would seem to be something like a MAX1555, but that's designed for one cell only. Before I planned to go with USB charging, I was thinking I'd just go with a 7805. But, if I go with the MAX1555 or equivalent, I'm looking at using a switching converter to get me up from the 3.7 V provided by my sole lithium cell. Are there any go-to solutions for multi-cell USB charging? Would it be easier to build a switch IC into my board? If so, are there any ideal solutions for 1.2 A max @ 5 V that come off the top of anyone's head? I've been doing some thinking on a USB -> Li-ion charging setup the last couple days. My load is about the same as yours, ~5W. A lot of the highly-integrated USB charger ICs are limited to 1 cell, I assume because it is easiest with 5V. There are some all-in-one options that integrate a charger and SMPS. Check out these chips from Linear for example. TI has some as well. I want 24V in my project, so I think I will need to use an external switcher. Currently, I'm looking at using MAX14578 for charger detection and MAX8934 for the charger. I wanted DCP detection, ability to re-use the circuit with an actual USB device (this one is charge only), powerpath, stand-alone operation and thermal monitoring of the battery. I have no experience with any of this yet, I'm trying to get organized on a test board. taqueso fucked around with this message at 19:08 on Sep 13, 2014 |
|
#
?
Sep 13, 2014 18:52
|
|
|
taqueso posted:I've been doing some thinking on a USB -> Li-ion charging setup the last couple days. My load is about the same as yours, ~5W. A lot of the highly-integrated USB charger ICs are limited to 1 cell, I assume because it is easiest with 5V. There are some all-in-one options that integrate a charger and SMPS. Check out these chips from Linear for example. TI has some as well. Those are some sexy ICs, however I'm deliberately eschewing any microcontroller involvement on this one. No I2C, no logic to speak of. At this point, I've started looking at using a pair of 14500 lithium cells and going with external charging. Size and mAh wise, it works. The terminal edge of the typical lithium ion discharge curve for a pair would be right around where the 7805 gives up at minimum voltage (7 V), so I think I can get away with that and not screw up some batteries. If I'm wrong, I hope someone will tell me so. (How does a 7805 behave at the border of the minimum input voltage?) I have another design that integrates an 8-bit microcontroller, and once I get around to working on that some more, I'm sure I'll start with looking at those Linear chips. Thanks!
|
|
#
?
Sep 14, 2014 01:55
|
|
|
Effective-Disorder posted:Those are some sexy ICs, however I'm deliberately eschewing any microcontroller involvement on this one. No I2C, no logic to speak of. At this point, I've started looking at using a pair of 14500 lithium cells and going with external charging. Size and mAh wise, it works. The terminal edge of the typical lithium ion discharge curve for a pair would be right around where the 7805 gives up at minimum voltage (7 V), so I think I can get away with that and not screw up some batteries. If I'm wrong, I hope someone will tell me so. (How does a 7805 behave at the border of the minimum input voltage?) I have another design that integrates an 8-bit microcontroller, and once I get around to working on that some more, I'm sure I'll start with looking at those Linear chips. Thanks! I want mine to work without a uC, too. A few of those all-in-one chips do have stand-alone variations or can work standalone without touching the I2C, but it sure is hard to sort through all the various options. There are a ton of choices. I don't know how the 7805 will operate near dropout, but if I remember right TLV1117LV33 (another linear regulator) I was using recently kept running with a lower output voltage when the input voltage dropped too low. You are right to question if you can trust the 7805 to not discharge the batteries. Definitely test or use an undervoltage lockout. Adjustable UVLO are built into some switcher modules, if you wanted to go that route. What program(s) are you using for your PCB renders? I wish Altium's 3D export included traces. taqueso fucked around with this message at 04:09 on Sep 14, 2014 |
|
#
?
Sep 14, 2014 04:06
|
|
|
Here's a kind of interesting problem that came up at work the other day: Suppose you have a DC signal somewhere in the +/-10V range with a 10Hz, 0.2V AC sine wave superimposed on top of it. What is the simplest circuit you can build with an output that gives only the DC component to within 1-2% when sampled at a random time? In other words, given a signal like V(t) = Vdc + Vac*sin(10Hz), we want to produce an output of Vout(t) = Vdc using the smallest reasonable number of (realistically valued) components. I came up with one solution that seems to work in simulation but I'd be interested to see if someone can come up with something better. I'll be happy to post mine later for comments/ridicule. e: I should mention that the value of Vdc changes in steps of ~1V over time, so the output of the circuit should follow that and settle in within ~1 second or so. PDP-1 fucked around with this message at 04:48 on Sep 14, 2014 |
|
#
?
Sep 14, 2014 04:45
|
|
|
PDP-1 posted:Here's a kind of interesting problem that came up at work the other day: Suppose you have a DC signal somewhere in the +/-10V range with a 10Hz, 0.2V AC sine wave superimposed on top of it. What is the simplest circuit you can build with an output that gives only the DC component to within 1-2% when sampled at a random time? The solution that comes to mind immediately is to sample at an integer multiple of the frequency you want to drop and average over an integer multiple of the frequency you want dropped. So for example, sample at 20Hz, 40Hz, etc. and average over 2, 4, etc samples. Multimeters use this technique to attenuate 60Hz/50Hz noise in DC mode, in addition to a low-pass filter.
|
|
#
?
Sep 14, 2014 05:25
|
|
|
PDP-1 posted:Here's a kind of interesting problem that came up at work the other day: Suppose you have a DC signal somewhere in the +/-10V range with a 10Hz, 0.2V AC sine wave superimposed on top of it. What is the simplest circuit you can build with an output that gives only the DC component to within 1-2% when sampled at a random time? A low pass filter. Your problem is that you don't have a lot of bandwidth to work with if you want settling time of 1s on a <10hz filter. So start by trying a single pole RC filter and seeing if you can fit it into your requirements. You may be able to. The good news is that your AC signal is small. Actually, with no filter at all you practically hit your numbers by the way. 0.2V is 1% of +/-10V. Also, if you can 1) do this digitally and 2) know the sin wave frequency accurately you can be extremely accurate with as little as a 1 cycle settling time by simply averaging over N number of cycles. But this depends heavily on 2. If you don't know your frequency you'd need to average over many samples to get adequate accuracy, which would impact settling time negatively.
|
|
#
?
Sep 14, 2014 15:31
|
|
|
PDP-1 posted:Here's a kind of interesting problem that came up at work the other day: Suppose you have a DC signal somewhere in the +/-10V range with a 10Hz, 0.2V AC sine wave superimposed on top of it. What is the simplest circuit you can build with an output that gives only the DC component to within 1-2% when sampled at a random time? So you can just use a simple filter with a 3dB point of 10Hz. A two pole filter like this will have a step rise time of 0.04 Seconds.
|
|
#
?
Sep 14, 2014 18:07
|
|
|
taqueso posted:I want mine to work without a uC, too. A few of those all-in-one chips do have stand-alone variations or can work standalone without touching the I2C, but it sure is hard to sort through all the various options. There are a ton of choices. I'm looking at this circuit or something similar for my UVLO. Lithium ion is too expensive to be messing up just because of the lazy. I use a ULP for Eagle called EagleUp that creates an importable file for SketchUp with its associated plug-in. I have yet to get it to attach the components automatically. I think it's due to problems with standardization of package names in Eagle. I pull the 3d objects for the various components off of the SketchUp online library, and sometimes import from 3d files obtained from this site, which is pretty drat handy. Those SIP resistors are something I had to spin up myself from the data sheet. They're supposed to be Bourns 4600X series, but I fudged some of the dimensions just to get the drat things done. Effective-Disorder fucked around with this message at 19:48 on Sep 14, 2014 |
|
#
?
Sep 14, 2014 18:13
|
|
|
asdf32 posted:The good news is that your AC signal is small. Actually, with no filter at all you practically hit your numbers by the way. 0.2V is 1% of +/-10V. tonberrytoby posted:1% of 10V is 0.1V. So you only need to dampen your AC signal by 3dB. I see now that I phrased the problem pretty poorly - the intention was to filter the AC component to 1-2% of its original amplitude (i.e. -40dB at 10Hz) regardless of what the DC component happened to be, while replicating the DC component only at the output (0dB at 0Hz). The DC component can range from -10V to +10V, so it could be 0V and then the AC component is the entire input signal and we don't get to use the 0.2V/10V=2% trick. The digital sampling ideas are a possibility (and the thing Arcsech suggested is a cool trick in general), but I'd have to add an extra input to lock onto the driving frequency to really do synchronous sampling well. Not impossible, but certainly more complicated. Trying to design a multipole passive filter that drops -40dB in 10Hz leads to a funky pile of stacked supercaps at best. FWIW, my idea was to do a kind of active cancellation where I'd buffer off the input signal, run it through a differentiator/integrator pair to get a copy of just the AC signal with no DC component, and then add the negative of that signal to the original signal to null out the AC component. I figure I can do that with a quad op-amp and a handful of standard resistors/caps plus one trim pot to get the time constants of the integrator and differentiator stages roughly the same. It might not be perfect, but it could well be good enough and would be fairly simple to implement.
|
|
#
?
Sep 14, 2014 20:14
|
|
|
Ok, so I'm in a pretty weird situation. Basically, I have to fix a lot of electronics for my job. All the same model of thing, a ton of different kinds of failure. But I'm not really from an electrical engineering background, so while I know what to do to fix these, I don't really understand why that fixes it. An example is a particular transistor(?) normally gives 0 volts and when another pin gets "turned on" this voltage raises to 5 and turns on another part of the device. So sometimes this thing breaks and is either always on or always off, swap it out, it's fixed. I'll be hosed to understand what all these capacitors and poo poo on that circuit do, though. So I'll occasionally get one where that part is obviously blown up, but replacing it doesn't fix the problem. Or I'll get situations where the whole system will rhythmically turn itself off or on, or the microprocessor is getting normal Vcc or whatever but doesn't turn on, or the USB looks fine from a continuity standpoint but it comes up as "malfunctioning" when you plug it into a computer. This situation is made more difficult because of our regulatory environment -- I can't really talk about what I'm fixing or what the details are because we could lose certification. This isn't like nuclear reactors or anything where people's lives are on the line, but we have to conform to certain laws. I'm sure some goon could figure out what I'm talking about and get us hosed up. Not out of spite or whatever, but people talk. I'm pretty sure I've already posted enough that someone could put 2 and 2 together (if they already had one of the 2s, so to speak). We're a small business, so I guess that's why I have this job. If we were a bigtime operator they'd just hire someone competent. So I guess I'm posting this to ask if there is some kind of resource out there that could help me fill in the gaps? I guess some kind of book or something, I'm not in a position where I can go back to school. But trying to learn from books sucks and is boring as poo poo, you have to do to learn, imo. Goons, teach me how to teach myself. I want to understand circuits and microcontrollers. If that's asking too much, I do have a little project I'm working on. I'd like to make a little led bar graph like this, it lights up based on how many volts are coming from an analog signal. But my source isn't just a 0-5v kind of deal, it's more like -1 to 1, so how would I make it go? I'm thinking I would have two of these and one of them would be connected in some way that the sign's flipped, so when the source says "I'm at -.3 right now" the second chip would see ".3 volts" and light up 3 leds (or something, idk). Then I could just mount them rear end to rear end to make it look like one 20 led bar that has zero in the middle.
|
|
#
?
Sep 14, 2014 22:25
|
|
|
e: ^^^ read "The Art of Electronics" by Horowitz and Hilt.PDP-1 posted:I see now that I phrased the problem pretty poorly - the intention was to filter the AC component to 1-2% of its original amplitude (i.e. -40dB at 10Hz) regardless of what the DC component happened to be, while replicating the DC component only at the output (0dB at 0Hz). The DC component can range from -10V to +10V, so it could be 0V and then the AC component is the entire input signal and we don't get to use the 0.2V/10V=2% trick. An interferometric notch filter like it seems you want to build can be extremely sensitive. You probably would need much more then one poti.
|
|
#
?
Sep 14, 2014 22:26
|
|
|
PDP-1 posted:e: I should mention that the value of Vdc changes in steps of ~1V over time, so the output of the circuit should follow that and settle in within ~1 second or so.
|
|
#
?
Sep 15, 2014 04:23
|
|
|
 I guess the ~1 sec settling time gives you a starting point for your pass band though.
|
|
#
?
Sep 15, 2014 05:29
|
|
|
I'm looking to use a 2A USB wall wort to power an arduino project- can I just put the arduino and all my servos in parallel on the 5V in? Turns out 9g servos love amperage when under high load.
|
|
#
?
Sep 15, 2014 14:04
|
|
|
Here's a dumb question. I have a self-contained audio circuit that runs off 3 1.5V coin cells. I was trying to break out the power supply to it so I could change it easily, since coin cells suck for any kind of longevity. I removed the coin cells and the holders from the circuit, soldered some new wires on, and started looking at a new power source. I have 12v N batteries laying around (and no, I'd love to just use AA batteries or whatever, but there's just no room in the device for them) that I wanted to use, and a 5v DC-DC converter like this thing which was the easiest way I had to get down to around the right ballpark for a 4.5v circuit. When I hooked the straight 5V input up to the circuit, it would only make popping sounds, so I kinda thought maybe I had to knock it down that .5v for it to work. I grabbed a silicone diode and put it in, and got the voltage to around 4.6V by doing that. Now, the circuit will play a couple seconds of each audio clip before going quiet, instead of the whole clip. To make sure I didn't kill the circuit, I hooked it back up to the original coin cells - works perfectly. I'm a simplistic moron with this stuff. I'm sure the diode is knocking the current too low or something, but I don't know what I'm doing. What's the best way of going about getting this circuit a stable 4.5V?
|
|
#
?
Sep 15, 2014 15:44
|
|
|
ANIME AKBAR posted:Then your problem statement is simply incoherent. If a signal is changing over time, it's not simply DC. You need to define a finite bandwidth or response time above which the higher frequency AC part is rejected, and use that to design a filter, analog or digital. But your output will, by definition, have an AC component. Ideal DC =/= real world DC. No DC voltage will ever really be DC, given edge cases (ie, flipping a power switch, for instance). There are a bunch of ways to deal with a stepped DC Voltage, and yes, it's not really incorrect to treat this as pure DC interrupted by step changes, and solve for each independently. Granted, treating this as a signal with some low frequency AC isn't unreasonable either, but there are weird nonlinear and digital methods for doing this kind of thing without treating it as such.
|
|
#
?
Sep 15, 2014 16:20
|
|
|
Cheekio posted:I'm looking to use a 2A USB wall wort to power an arduino project- can I just put the arduino and all my servos in parallel on the 5V in? You CAN, but with the motor loads you may find that your 5v rail gets pulled down when the servos pulse. I would put a bulk and filter capacitance as close as possible to your arduino to ensure a clean power supply. Or maybe that stuff is already built into your dev board. Not sure, I use bare pics usually.
|
|
#
?
Sep 15, 2014 17:06
|
|
|
Sagebrush posted:What if it's an intelligent variable-draw load that continuously adjusts the current to produce a constant droop in the battery voltage? Then what, hmmmmmmmm? I know this isn't serious, but I don't think a load can be smart enough for this. Won't you cap out against the C rating and make a constant droop in voltage impossible to maintain? Or can you pick the starting draw cleverly enough to keep the maximum draw necessary under that limit?
|
|
#
?
Sep 15, 2014 22:01
|
|
|
Yep, as stated, the art of electronics is what you want. It's a common sense, easy (comparatively) to read guidebook covering the entire field of electronics. Be prepared for frustration because it's obviously a deep subject, but chip away and you'll start building an understanding of electronics as you go along. PDP-1 posted:I see now that I phrased the problem pretty poorly - the intention was to filter the AC component to 1-2% of its original amplitude (i.e. -40dB at 10Hz) regardless of what the DC component happened to be, while replicating the DC component only at the output (0dB at 0Hz). The DC component can range from -10V to +10V, so it could be 0V and then the AC component is the entire input signal and we don't get to use the 0.2V/10V=2% trick. Supercaps? 100k and 10uF get you in the ballpark of 1hz. Your idea sounds tricky at best, especially compared to well established filtering techniques. asdf32 fucked around with this message at 00:20 on Sep 16, 2014 |
|
#
?
Sep 16, 2014 00:12
|
|
|
I am putting together this circuit: http://www.doc-diy.net/photo/sound_trigger/ But the electret mic I have on hand has three wires rather than two. Do I hook up ground and VCC, and then run the wire labeled "Audio" into the point between R4 and C3? Do I need to go pick up a two wire mic instead? Edit: Turns out I also had a 2 wire electret mic in my random parts, so I used that. Still curious though. armorer fucked around with this message at 17:48 on Sep 16, 2014 |
|
#
?
Sep 16, 2014 00:40
|
|
|
I have a lithium battery outputting 4.2V (nominally 3.7, but it's freshly-charged) and a 3.3v DC-DC converter. When I hook up a 9V battery to the converter, it puts out a nice stable 3.3V. When I hook up the lithium battery, it puts out nothing. What's going on here? Is it just a quirk of lithium batteries or something? I don't understand what's happening.
|
|
#
?
Sep 16, 2014 17:34
|
|
|
Funky Bunch Wikia posted:I have a lithium battery outputting 4.2V (nominally 3.7, but it's freshly-charged) and a 3.3v DC-DC converter. When I hook up a 9V battery to the converter, it puts out a nice stable 3.3V. When I hook up the lithium battery, it puts out nothing. What's going on here? Is it just a quirk of lithium batteries or something? I don't understand what's happening. 4.2V is lower than the minimum input in that datasheet.
|
|
#
?
Sep 16, 2014 17:40
|
|
|
That would do it! Balls. Any easy way to get the lithium battery to put out the 3.3v I need a bit more safely? Is throwing a diode in the line to eat like half a volt likely to work? I guess I could probably go get a 3.3v linear regulator from somewhere, but I have no local electronics stores and I'm trying to finish this today.
|
|
#
?
Sep 16, 2014 17:50
|
|
|
You'll need to find a LDO linear regulator with a very low dropout voltage, there should be ones that take ~3.6v as the minimum out there, but the lower dropout you want, the more expensive they tend to be. Parametric searches are going to be your friend here. e: The TI TPS75633 series seems to be one of your best bets, although TI have a few other options in the TPS8*33 group that'll do the job, Mouser have some in stock, but they're a bit pricey. SybilVimes fucked around with this message at 18:54 on Sep 16, 2014 |
|
#
?
Sep 16, 2014 18:47
|
|
|
SybilVimes posted:You'll need to find a LDO linear regulator with a very low dropout voltage, there should be ones that take ~3.6v as the minimum out there, but the lower dropout you want, the more expensive they tend to be. Parametric searches are going to be your friend here. If you're looking to do parametric part searches, I highly recommend the greasemonkey script for DigiKey. http://eewiki.net/display/Motley/advancedsearch+Greasemonkey+Userscript+for+Digikey.com
|
|
#
?
Sep 16, 2014 19:09
|
|
|
Slanderer posted:Ideal DC =/= real world DC. No DC voltage will ever really be DC, given edge cases (ie, flipping a power switch, for instance). There are a bunch of ways to deal with a stepped DC Voltage, and yes, it's not really incorrect to treat this as pure DC interrupted by step changes, and solve for each independently. Like consider if I put a light sensor in a doorway to detect if a door is open or closed. There are two output levels, and even though it swings between the two it's reasonable to say it's a "changing DC" signal because the vast majority of the time it just sits at either output level. Now consider if I put the light sensor outside and point it at the sky. Its output will change slowly but continuously over time, and will never really reach a steady state over the course of hours and days. You can't call that a DC signal, no matter how slow it varies.
|
|
#
?
Sep 16, 2014 21:43
|
|
|
ANIME AKBAR posted:Even if a signal has some transient component, you can only really use the term DC if it asymptotically approaches one or more stable values and stays there for a while. We'll DC means direct current. That technically applies even when the voltage changes so long as the polarity stays the same. Conversationally I'd say it's quite reasonable to say simply "DC" if a signal is changing sufficiently slowly, even if you know it's changing, and especially when the voltage polarity is constant.
|
|
#
?
Sep 16, 2014 22:29
|
|
|
In this case I am applying a DC bias plus a small AC signal to the input of a system, waiting a few seconds for all transients to settle out, and then measuring the resulting DC and AC components at the output. The "DC" component moves in discrete steps but I only collect data after the system has stabilized after each change. Anyway, mea culpa time: Those of you who were saying "just use an RC filter you dummy" were correct. I apparently hosed up my original RC time constant calculation by a few orders of magnitude because I was getting results suggesting components way, way bigger than they actually need to be. Re-did it again today and a two stage RC filter with reasonably valued components works just fine. Oops.
|
|
#
?
Sep 17, 2014 03:17
|
|
|
I'm trying to build a 12v battery pack out of the several dozen lithium ion cells I have. What would be the best way to go about this? Should I gang up multiple 14.4v BMS boards in parallel (expensive) with a simple 14.4-12v regulator, or use one 48v BMS board (or a higher voltage, even) with a big ol' 48v-12v stepdown regulator like this? or should I just do it the pain-in-the-rear end RC hobbyist way and wire up all the cells and get an expensive balancing charger and a stepdown regulator
|
|
#
?
Sep 17, 2014 23:58
|
|
|
atomicthumbs posted:I'm trying to build a 12v battery pack out of the several dozen lithium ion cells I have. Wait is this related to those semi-legal 11,000mAh packs you were selling in a shady fashion in SA-Mart? 
|
|
#
?
Sep 18, 2014 09:22
|
|
|
SoundMonkey posted:Wait is this related to those semi-legal 11,000mAh packs you were selling in a shady fashion in SA-Mart? some of them had broken USB ports or displays, and I can't very well sell those!
|
|
#
?
Sep 18, 2014 12:53
|
|
|
atomicthumbs posted:some of them had broken USB ports or displays, and I can't very well sell those! i'll trade you a broken one (with functioning battery) for a fistful of assorted ICs.
|
|
#
?
Sep 18, 2014 22:23
|
|
|
SoundMonkey posted:i'll trade you a broken one (with functioning battery) for a fistful of assorted ICs. Nice ESD handling precautions 
|
|
#
?
Sep 18, 2014 23:04
|
|
|
|
| # ? May 5, 2024 22:55 |
|
|
atomicthumbs posted:I'm trying to build a 12v battery pack out of the several dozen lithium ion cells I have. My first recommendation would be to series them all with the 48v BMS. It's been my experience that it's the superior option, provided you can charge and regulate it. Another good option would be paralleling individual cells, and then seriesing them into a single 14.4 BMS. This assumes the individual cells are of similar vintage and evenly worn. If they're decently well matched the individual cells roughly equalize. It's inexpensive and works quite well. I've found that it's harder to get a full charge than a single series though. There's also some (easy) prep work needed. http://www.batterypoweronline.com/conferences/wp-content/uploads/2012/09/Elithion.pdf You don't really want to make a bunch of individual cell and BMS strings though. That pdf also goes into why. The RC way gets you every last bit of capacity and performance. Nice under those circumstances of course or if you can reuse that equipment. Otherwise it depends on what you're doing if it's worthwhile.
|
|
#
?
Sep 19, 2014 02:49
|
|
