This is by far the best option for awkward connectors. You could even remove the pins and put the new cables in by opening the crimp and soldering the new ones without any splice.
Just based on my experience and knowledge of these batteries, which is not that much really, don’t take any of this as gospel.
Constant voltage current-limited power supply is probably the easiest way to charge it, a regulator + current limited supply like you mention would be one way of achieving that, will waste a lot of the power and charging speed will be limited by how much heat the regulator is able to dump (which for a typical TO-220 regulator without heatsink is like a watt or two? So you’d have a max charge current of like 0.5-1.0 amp)
Your current approach probably also works just fine, but i’m not sure how good it is for the batteries in the long run
You can find charger for sale online though, at least modules that you can solder up to a battery holder or whatever, i would go for one of those if you plan to use these batteries for a longer period of time
I’d order online if there was a LiFePO4 charger on the market. However, in my country I’ve been unable to find one, and importing (excise & duties, paperwork) is more work than building it myself. I’ll also likely need a design that I can cheaply include on custom PCBs for manufacture (not for sale to end-users, but for internal use by maintenance technicians).
I gave it a test on a cell today and it seems to charge fine and at a reasonable rate – but in a sudden flash of brilliance, I forgot to physically connect the ADC pin to the battery, so it couldn’t shut off. Well, that’s what testing is for I guess.
Anyway after fixing this, it looks like I can call this a win and move on. If it undergoes destructive optimization, I’ll report back here with a warning to others.
There are really charging curves that should be followed to avoid damaging the cells.
You want to charge up to a certain voltage at a current (related to the battery capacity I believe) and when the cutoff voltage is hit, you switch to constant voltage (the max voltage of the cell) and then slowly drop the amperage as the battery is topped off.
I’m not sure about your cells, but some LiFe are 95-98% full at 3.45-3.5 volts, but the problem is that the voltage curve is really flat from 40-50% charged up to 95%. So you need really accurate measurement if you want to charge to 95%. The last 3-5% is when the battery ramps up to 3.65v and really is the riskiest part of the charge. It’s also the highest wear part of the battery use, if you can avoid charging it all the way up to that your cells will last much longer.
My normal lithium battery charger automatically slows its charging speed as battery voltage nears its capacity. I could set it at 1000mah and it will step down to less than 1/10 of that before charging is complete.
It’s just that it’s really tricky to charge the final bit because the middle 80% is such a flat voltage curve. They have a 1000x life when they’re taken care of.
It’s tricky to stop at the right point, because lithium iron only have a very small voltage increase between like 40% and 90% and they ramp up to full voltage right near the limit of their capacity.
What you are talking about is nothing special at all and not following a charging curve. The curve automatically looks the way it does when charging CCCV. Constant Current -> Constant Voltage.
If that’s not sensitive enough, another option is using a piezo element coupled to the case to detect vibration, with an op-amp or hex inverter to buffer + trigger the 555. However if you couple it too closely with e.g. the floor or furniture it will pick up nearby footsteps or cars. Might be good depending on the situation.
The Original Saleae logic analyzer (or one of its clones -> search for “Compatible Saleae”) or a LA104 (maybe with custom firmware) or maybe a DSlogic pro or ChronoVu. You can also browse the HackADay Archive to find an analyzer that suits your needs (e.g. build your own, or based on a pico). It all depends on the resolution/speed you need and usecase.
USB isolator, $10 saleae clone and sigrok/pulseview setup is very simple and cheap (except for the isolator).
The flash chip is a common 16 MiB SPI NOR flash. An easy way to read or write it would be to use flashrom on a single board computer like a Raspberry Pi.
Unfortunately, that router is not supported by OpenWrt or DD-WRT, so you probably won’t be able to do much with it.
From a conceptual perspective, very low quiescent current (aka idle/standby current) when unactivated is entirely achievable. What your design will need to do is assess how much each component will draw at idle, and if it’s too high, then you may need to have gates which turn off those high-draw components when idling.
From a cursory Google Search, the DFPlayer Mini datasheet shows a standby power of 20 mA, which far too high. A forum post shows that if the sleep mode is enabled using the serial interface, current drops to 0.4 uA. This is much better.
For the 555 itself, you mention an astable oscillating configuration, although I’m wondering what your intention for the 555 is. Ostensibly, the DFPlayer either needs a brief pulse to start playing (roughly “edge triggered”) or needs to be kept active for as long as the music should be playing (roughly “level triggered”). In either case, a 555 in a one-shot configuration would make sense, since an astable oscillator would imply the music would restart on its own every so often.
If you’re insisting on the 555, then you may not be able to access the sleep mode in the DFPlayer Mini. So your option might be to gate the DFPlayer so that it only gets Vcc power when the 555 supplies it, probably using a MOSFET. Alternatively, using a cheap microcontroller would let you control the DFPlayer Mini via serial. Your microcontroller could then also receive the signal from the vibration switch and come out of deep sleep to issue commands to the DDPlayer.
The ATTiny uC and MSP430 uC families can draw as low as microamps or even nanoamps in some low-power modes. So that neatly addresses the standby current.
What you’ll also have to assess is the active current, or how much the music player draws when it runs for however long. This should give you an idea of the total lifetime for your application on a single battery charge.
These LEDs are not shift registers. In fact, if you look closely, there’s no input and output pin, only one data line. That’s because each LED in this strand is pre-addressed from 0 to 100. Whenever it receives the NeoPixel data, it picks out the n’th color data (matching to its pre-address) and displays that. You cannot change the pre-address, its fixed permanently - or at least, we have no idea how to re-address it.
Yes, I read that bit. I’m trying to find out more.
They are obviously made in bulk, somewhere in china, since they are quite common, and cheap. Unfortunately, I’ve only found them in pre-built products, and none actually identify the chip model.
They obviously can be programmed (at least once) since you can buy arrays of at least 400. There is no way a factory is making 400 different chips, given the end price.
I haven’t done a course in electromagnetism yet, but as far as I understand, the ferrite core is just a piece of metal with no magnetic field, so moving it doesn’t induce a voltage
I think it would, however, change the inductance, just like the iron core in a transformer does
If your coil was oscillating, then perhaps an iron core moving through it would cause perturbations which are detectable. But that would require extra logic to compare the expected oscillation frequency with what the coil is actually oscillating at.
Since you say that tilt switches are not an option – for reasons I’m not entirely sure I understand – another option is to have a linear Hall effect sensor mounted nearby a small magnet. If the magnet moves relative to the sensor, then that is a change which can be acted upon. A linear sensor makes it possible to use a trim pot to tune the sensitivity.
i failed to mention the most important aspect of the project: (near) zero power consumption when idle.
it is a vacuum (that kinda looks like r2d2) that i want to play some r2d2 noises when used.
it is hauled around on a construction site on the scaffolding and wont be in an upright position while idle or in usage, so tilt switches didnt make sense to me.
i plan to run it from a coincell or maybe 2 aa, but it should only draw power when the sensor is triggered so the power should last years if i understood the 555 timer correctly. that means accelerometers and linear hall effect sensors are out too due to microcontrollers and thus power draw.
somebody else mentioned vibration sensors (SW-18010) that look promising.
No, moving a ferrite core through a coil won’t generate a voltage. You would need to move a magnet to generate a voltage.
Look for a vibration switch like one of these. If you want more control, you could use an accelerometer and a microcontroller to trigger it from a specific amount of movement.
Are we talking a handheld flashlight? Or is it something a bit more hefty?
Reason I’m asking is the bearings in the fan and motors. A handheld flashlight is going to take a beating, and the bearings can easily be knocked out of alignment.
Soooo turns out mouser had a classical unit conversion error so they wrote 28 times as much airflow than there actually is in the fan specs so im not going to use fans in my flashlight as 2 watts of cooling matters so little that if i just use convection it would cool more.
These seem good and mouser has ip58 ones which are perfect for me. Maybe ill use two of them for example. But im ordering a few because if they dont work im gonna abandon the fan idea. Still cool fans. As for heatsinks the whole flashlight is one. Its a really high power, small flashlight(hotrod). It uses two 14500 lithium cells(basically aa size) to power a boost converter with more than 70 watts. Even tho the boost converter is really efficient the leds produce so much heat that the light overheats in seconds. So thats why i want to test if fans help.
Yeah but its still months from completion. I want to machine it from aluminium but im explorimg all possibilites before i start ordering parts/writing software. When i learn to use github im gonna upload all the files with a permissive license because i want to contribute to the flashlight community.
In seconds? Wow. I think you’re right, you might need more than a small fan!
It might be worth exploring heat pipes or peltier effect coolers. The latter makes the problem worse (they are inefficient and generate a lot of heat) but your LED can be locally cooler if you can e.g. move all that extra heat into a big heatsink (also condensation can be problematic).
One cheap source of heat pipes for testing could be old graphics cards – they often outperform simple copper heat sinks. Use thermal epoxy to stick your LED to it and see if the performance is acceptable. On the exotic end of things, you could also water/oil cool it, or (carefully) make your own thermal grease from industrial diamond powder for a small boost in thermal conductivity.
Also even at 95% efficiency, it sounds like your boost converter has some heat to dump too!
Yeah the problem is the light makes so much heat(the boost as well) that i cant dump it into the air with high enough efficiency. The bodys going to be aluminium and the pcb copper.
Hm, that reminds me! If you’re designing your own PCB, some manufacturers will make the PCB out of aluminum for you instead of FR4. This is commonly used for high-intensity LED lights to help keep them cool.
Here’s some random info about them so you can see what I mean:
I was already planning to use a copper core pcb. This is pretty common among insanely powerfull lights. The flashlight community has some great examples. But most of these lights use resistor based voltage regulators which waste a lot of energy in the form of heat so im trying to improve on the traditional design.
askelectronics
Newest
This magazine is from a federated server and may be incomplete. Browse more on the original instance.