AMB Laboratories DIY Audio

[dougigs]

That all makes sense. Battery-level indication is indeed useful on long-charge devices like this, but not mandatory, and you're right that it would be daunting. With the exiting Mini3, I can tell if the 9V needs charging because the colour of my blue front-panel power LED changes slightly. I can't remember how the voltage-discharge curves of LiPo cells look, but I suspect that this may also apply here.

[amb]

The battery protection chip will automatically shut off power to the amp if the battery voltage droops too low. This will prevent a deep-discharge which may degrade the battery. You can simply use the amp until it shuts off and then plug it in to recharge, or plug it in any time. The charge controller will determine the battery state and what to do. So there is no need (other than curiosity) for the user to see more status information. We provide test pads on the protection chip's over-voltage, under-voltage, and low-power pins for built-time debugging.

[goldkenn]

Two questions:
1) The sonical difference due to lower voltage swing compare with 9V battery?
2) As long as battery life extended, which a wall wart may no more useful even for desktop utility, for convenience, why not consider USB charge?

[MisterX]

why not consider USB charge?

USB cannot supply enough power to charge the batteries we're using at the desired charge rate.

Edit:

Where are you going to fit a USB jack in a Mini3?

[amb]

USB cannot supply enough power to charge the batteries we're using at the desired charge rate.

That's right. We are charging the battery as high as 600mA and we need 9V of voltage to charge two series-connected Li-Po cells. USB power cannot satisfy either of these needs.

[goldkenn]

why not consider USB charge?

Where are you going to fit a USB jack in a Mini3?

Simply put a mini B socket on this new board upon the original DC jack, desolder the original DC jack, it should fit the hole originally for the wall wart, for postioning for USB socket, maybe need something like s SMD to DIP opamp adpater, to make the mini B socket fit the postion of the panel hold

If for charing only, a simple 5V step up should make the USB cahrge idea work, of course the curent not as high as a wall wart, which will takes longer charging time, but in real time there should no need for rapid charge.

Long battery life is convinence for outdoor, therefore my point of view is all target to outdoor application, USB charge also convience for outdoor, in case we need to charge outdoor, or at least not at home, nowdays around the world maybe find a USB power supply will be more easy then find an AC mians for a wall wart; And in case we need to wait for charging outside, one hour or three hours waiting may not be a big difference when outdoor, since I don't think user will willing to wait for one full hour even could rapid charge by a wall wart, mostly in real life will only charge up for short terms, enough for arrive the destination will be fine.

[amb]

The battery charge controller chip will be programmed to charge at 600mA (maybe even a little higher, I am still experimenting) during the first phase of the charge cycle (called "constant current phase"). When the combined cell voltage reaches approximately 8.3V it will enter into "constant voltage" mode and the charge current will gradually descend until the chip decides that the charging is complete. If you starve the circuit with insufficient current, the circuit will not work properly (i.e., the charge controller will time out and terminate the charge cycle prematurely). It's not just a matter of making the charge time longer.

If you step up the USB voltage (e.g., with a DC-DC converter), you gain voltage but lose current capability. To get from 5V to 9V is a ratio of 9:5 drop in current capability (assuming a 100% efficient converter, which is not the case), so the 500mA current capability out of a USB port becomes 278mA maximum. Many computer USB ports are also a bit low on voltage (not quite 5V, and gets worse as the current demand goes up). So by this alone USB is totally inadequate for charging these batteries. We also don't have any room for such a step-up circuit on the board. It's FULL.

Lastly, we're retaining the DC power jack on the Mini³. There will be no jacks on the χ1 board (as can be seen on the PCB layout and 3D render pics I posted). This reduces the number of hard-to-remove parts to make upgrading an existing NiMH-powered Mini³ to Li-Po power easier.

Given all this, USB-charging is out of the question.

[Yoga Flame]

This discussion on USB got me thinking. Would it be possible to construct a custom USB adapter cable that plugs into two USB ports and outputs 9V and over 600mA? Maybe combine the two USB power lines in series, and feed it through a regulator before terminating into a DC barrel plug.

Not that this would be very practical. It's sort of the worst of both worlds. You're constrained to charging only from USB ports, and you also need to carry around a special adapter. But still, would it work?

[amb]

You can't put two non-isolated 5V sources in series. When connecting the positive of one to the negative of another you'll short it out. Even if the 5V from different USB ports are isolated, this will merely double the voltage but does not increase current capability. You could connect the two in parallel to double the current capacity -- and many external disk drives for laptop use come with a special cable with two USB-A plugs on one end -- but this does not increase voltage. IMHO paralleling two USB ports' 5V lines is an ugly hack and it uses two available ports.

[MisterX]

If you reduce the value of the fast charge current programing resistor (R1 or RPROG from the MCP7386* datasheet) and increase the value of the saftey timer programing capacitor (C1 or CTIMER from the MCP7386* datasheet) one of those ZipLinq 9V Phone Charger Booster Adapters should work.
But I'm not sure doubling the re-charge time just to add the inconvenience (IMO) of USB charging is worth the effort.



http://www.cables4computer.com/Ziplinq/ ... _BULK.html