I took a short break from updating the interest list as I have been on vacation, and now work travel. Also I was hoping the listbot would pick the job up but it appears to not be working. I will comb the thread over the next few days and get caught up. I apologize for any inconvenience or burden this has caused folks.
That disappoints me, I think this is the last thing that needs a FET instead of well managed linear drive. Either one, or two, or even three banks of 7135s. Even just one bank with PWM. Not many 7135s needed to achieve desired lumens.
Which would be more consistent throughout the cell discharge, assuming suitable LED Vf, as well as much more efficient (run-time).
ISTR that the prototype uses the Q8 driver with the FET disconnected and a NanjG 7135 driver connected to the gate pin instead, as a second 7135 bank.
Also not sure how it fits in with the two different sets of colour temp LEDs and mixing. I don’t see how a FET has any purpose in that arrangement, just two banks of e.g. 7135s, one for the lower colour temp LEDs, the other for the higher.
Perhaps a FET could just override any of that subtlety and give “Max Power” for those who insist (maybe they shouldn’t be buying a practical lantern).
The only point of a FET is to (over) drive the LEDs to the absolute max. that the particular combination of cell max. current and ESR, state of charge, LED transfer characteristic etc. can deliver. In short blasts. Inefficiently. Sure, wind it down with fast PWM but that will make no difference to the inefficiency (short run time and inconsistent output as the cells discharge).
To summarise: why does this lantern need a FET driver ?
But I am not on the interest list, just an observer, so crack on please.
I had originally planned on the 7135s, but left the driver & firmware for that part in the lands of Lexel & Toykeeper as they know much more than i do about driver design & circuits. Maybe chat with them about your thoughts on it.
You have misunderstood.
The point is to have linear driver. But instead of a bank of 7135s it should be based on a FET driven in linear mode.
This won’t be a hot rod driven as far as batteries allow but a nicely regulated light.
I am a real fan of minimizing power losses (e.g., most lumins for the longest time). I don’t care what color the light is. I am one of those that view this as an emergency lamp. If I understand what is being debated, FET + PWM yields the best use of battery resources vs a whole bunch of analog devices. And reduces extraneous heat generation.
I remain a: No holes in the light, no charging of external devices, no charging of internal batteries. It is a light. It has batteries. Stop there. That is a lamp.
To build one prototype driver
about 17$ Oshpark with slow shipping for 3 boards
Order parts I do not have stocked 20$ With shipping costs
5$ parts I have stocked
6-7$ assembly time
5$ testing time
DHL express to china 35$
But if they make it in China its cheaper and faster to make one
When the light goes in production we can talk about free samples or a payment for the design, at the moment funds for me building a prototype would not speed it up
Table 4. Charging Parameter Default Setting
DEFAULT MODE bq25895
Charging Voltage 4.208 V
Charging Current 2.048 A
Pre-charge Current 128 mA
Termination Current 256 mA
Temperature Profile Cold/Hot
Safety Timer 12 hour
USB type is mechanical of the light not driver
max. USB out of the chip is 2.4A but this gets reduced by a resistor on DELs layout as I understood to 1.5A
Batteries get charged with 2A as we do not use the chips internal bus
the chip supports up to 5A, but likely the Powerbank chip needs to be programmed differently not sure how that works
Lantern can be used while charging/discharging,
there is an enable pin from MCU not sure how this will be configured if output is enabled or not when the lantern is off
Yeah, I’m sure they would. But any light capable of lighting up a room well enough to see will be handy, TBH. This storm likely won’t care if you have DBSAR lanterns or Q8’s with a Styrofoam cup on top as a makeshift diffuser.
EDIT: Oh yeah, if you’re at the coast, you should really evacuate if you can. Sadly, some people can’t or won’t. I’m pretty far inland at Salisbury, but it’s probably going to get interesting here too, if Florence goes as predicted.