Electronics > Projects, Designs, and Technical Stuff
Bluetooth DC energy monitor
(1/1)
flangefrog:
Hi All. I'm developing a tiny BLE energy monitor designed especially for DIY ebikes (and escooters/eskate). I expect it to be useful for many other applications such as RC models, DIY renewable energy, battery charging, DC-DC converter testing, and some auto/marine applications. You can think of it as an upgrade to the common "GT Power watt meter / power analyzer" products that are all over eBay. This will eventually be released as an open source design. For some reason I haven't been able to find anything like it on the market, although I wouldn't be surprised if it already exists. The closest I can find is the Medusa Oracle which is not produced any more.
While I have been playing with electronics since I was very young I haven't had any formal training or industry experience in electronics design. I need some help with some things that most would consider to be very basic stuff. The first issue regards the boards power supply.
Initially I was looking at adding an OLED screen to my design, however I'm now considering an IPS TFT display. The problem is that this will be very hard on the linear regulator I've currently specced. The TFT will take 20mA compared to around 4mA for the OLED. When powered by 120V that's a lot of waste heat. Even with only 72V it's still too much. There are ways to manage this such as limiting the display on time, but I think it would be better to go for a buck converter. The problem is that I have very little real estate on my PCB. Without getting fancy with stacking PCBs etc can anyone suggest a solution that might fit on my board while still being relatively cost effective? One of the candidates I've found is the LTC3639 however it is very expensive and I'm not sure how much area the supporting components would take. Most other ICs don't support 120V - maybe a discrete solution is the answer. 120V support is not a hard requirement (that would be 65V) but it would be nice. One of the other requirements is a low quiescent current (10s or 100s of uA).
Here is a WIP render of my current design. Note that this is my first ever PCB I've designed and I'm still laying out components while developing the schematic. The PCB will be four layer. The space I have to work with is essentially the space above the yellow connector on the right. That SC-59 fet between the large pins is for reverse voltage protection. That's an MSOP-8 and 1812(4532) cap next to it. For reference the board size is 36*15.5mm.
flangefrog:
I think I'll try upgrading the MIC5281-YMME regulator to an MIC5283YML for the first prototype and do some real world testing. It's a smaller package with a slightly lower thermal resistance. Using 50% PWM for the backlight at 120V should keep it within the limits.
flangefrog:
Does anyone here have any insight on using a 4 terminal vs 2 terminal shunt? I would have thought 4 terminal shunts like the Vishay WSL2726 series would have a much better overall TCR but the best TCR is 75ppm. Other much cheaper 3921 2 terminal shunts are available with TCRs of 50ppm.
Is this because the TCR is measured without taking the PCB into account and once the shunt is mounted to the PCB, the effective TCR of the 2 terminal shunt would get worse (even with a kelvin connection) while the 4 terminal one would stay similar to the provided value? I did ask Vishay about this but didn't receive any reply.
Navigation
[0] Message Index
Go to full version