I am trying to power the SparkCore with a LiPo battery and I am trying to use a single USB port that would allow for charging the LiPo as well as connect to the SparkCore. As an example, Adafruit has a LiPo charger backpack for it’s Trinket (https://www.adafruit.com/products/2124) that connects to a battery input pin (which would be Vin on the SparkCore), a ground pin (Gnd, obviously), and a “Bus” pin, which allows for the 5V from the USB to go directly to the charging circuit. Is there a way to get the 5V from the USB directly from the SparkCore, or is there another way to get a 5V output from the SparkCore to integrate the charging circuit with the onboard USB connector?
Battery shield by spark sounds good, but I think you might also be able to make that Adafruit LiPo Backpack work. If you hook it up like this:
LIPO BACKPACK ..... SPARK CORE
5V ..... solder to anode of large diode next to USB conn
GND ..... GND
BAT -|>|- VIN
That’s an external schottky diode between the BAT output (essentially the LiPo) and the VIN pin on the Core. Maybe a little complicated, but if you need it to be compact this will work nicely… especially if your battery pack has a protection circuit in it
Honestly, I didn’t even see that they had their own backpack for the SparkCore. After seeing it though… It’s rather bulky, to say the least. I need this project to entirely fit within a 1.5" diameter cylinder, so the size of the existing shield would definitely exceed that. Kennethlimcp and BDub, I wasn’t aware of the hardware layout, since I have only seen some of the documentation. Are Eagle or drawing schematics available?
Connecting to the anode of the Schotkky Diode would give me the access I require, so that may be the easiest way to do this. I can connect to the anode, so the forward drop isn’t really an issue. My only concern with this method whether both devices will still be protected. We can assume the USB will only provide 5V, so the Adafruit component’s 5V input side is fine, as long as I solder to the anode on the diode. Ground is ground. That’s fine. The only question is connecting the battery lead (BAT on the charging board) to the Vin on the SparkCore, which has an LDO anyway, so again, no problem. It seems like it should be fine, but I just wanted to verify.
Also, out of curiosity, can the WiFI antenna transmit through a metal honeycomb structure? Also relevant is whether or not inductive charging (looking at this right now: https://www.adafruit.com/products/1407) will work through a thin piece of stainless steel (304/304L). The real question is whether or not the voltage output will still be 5V on the inductive charger, since I need it to be pretty close to the 5V (noise and everything) that a USB provides. I think you can see where I am going with this.
Everything will be fine if you make sure to also add the diode between BAT on the lipo board and VIN on the spark core. This ensures that power from USB and BAT are "diode OR'ed" to VIN. Whichever voltage is higher will be the one that forward biases the diode and provides current to VIN. If possible, use the same diode as it on the Spark Core, or one with the same forward voltage drop. It's this one: SS3P3-M3/84A Vishay General Semiconductor - Diodes Division | Discrete Semiconductor Products | DigiKey If you sit down and do some calculations you can change it up, but without doing that this is a safe way to proceed.
If the holes are big enough, yes. How big is related to the wavelength of the frequency you are concerned with, 2.4GHz in our case. The smaller the holes, the more attenuation of signal will occur.
That inductive charger will not work through any kind of ferrous or conductive material. It effectively will act as a short circuit in air-core transformer that is created when you bring the two coils together.
If you need to put the Spark Core in a metal enclosure, it's advisable to use a u.FL core with external antenna.
Thank you! That's exactly what I was looking for. Sorry I was unclear about which component I meant.
I'm not sure I understand what you mean by
"diode OR'ed" to VIN.
Having just looked at the schematic, It seems to me that having a diode will prevent the LiPo from getting 5V (minus the forward drop of the Schottky) applied directly to it, which would most likely destroy the cell. That is, if Vin (as listed on the board) is actually RAW, as listed on the circuit schematic.
Aside from reverse protection for the cell, I think I am starting to understand what you meant. If UVCC (or VBUS from USB) minus the Schottky drop is greater than the LiPo voltage (minus the Schottky drop), then the USB will be powering the SparkCore. If the reverse is true, then the LiPo will be powering the device. It's also nice that the LiPo voltage should never be able to reach the USB voltage. That should avoid any issues with the two being very close and switching quickly.
These holes are fairly large. Given that the wavelength for a 2.4GHz emission is nearly 6" and the hole size is (at most) .5", I think there might be some attenuation. Would I be able to use the case itself as an antenna? A core design feature of this project is that there are no external features (such as charging ports or antennae), so having anything sticking out wouldn't work.
Do you know of any that will work? Also, is this related to material thickness, or another factor I might be able to control? If I understand why these losses occur, it is because currents are induced within the metal that create a magnetic field opposed to the external field, which greatly reduces the transmission of the external field through the metal. We are primarily interested in the field within the coil, so all I need to do is make sure there isn't a conductive loop pathway within the coil. Now that I think of it, what I am suggesting is an awful lot like laminated cores. I am just reducing the maximum area a conductive pathway can enclose.
Anyway, what do you think? I really appreciate all your feedback so far!
I think I what I will do for the inductive charging problem is simply have a cutout in the stainless where the charger is. I will try testing the effect of a stainless sheet separating the coils, and if it works out fine, I might stick with it as a solid piece, but otherwise a cutout is fine.
WiFi (or maybe Bluetooth) is a slightly bigger issue. I’m not exactly sure what kind of communication I want, what I want these things to communicate with, or why I want them to even communicate (well, I kind of know that one). Essentially though, I need to know if I can use the stainless itself as an antenna, or if it’s not (easily) possible to completely surround an antenna with a metal cage and still have some kind of working antenna.
This is awesome! Sorry to revive an old post, but I was wondering if there was a way to do this without soldering to the diode on the Spark? Something that can switch between charging/discharging the battery using just the Vin pin? I’m a hardware newbie Thanks!
Thanks!