badgerloop-software/solar_car_hardware

MPPT Rev 2 Updates

Opened this issue · 12 comments

dawiggleman commented

List of bugs/improvements to make for MPPT Rev 2

dawiggleman commented

Output Electrolytic Caps:

Problem:
The hole sizes for component UCY2G101MHD6 are too small to insert the pins of the capacitors into. Make holes in this footprint larger and update schematics/layout.

Temporary Solution:
File down leads to fit holes, or cut the leads short, put thin wire through the holes and solder to the capacitors. Hot glue capacitors down.

dawiggleman commented

Board size:

Consider placing LV components on the bottom side of the board to get under the 30 sq. inch requirement for advanced circuits to get $66 four layer boards.

dawiggleman commented

Overvoltage protection:

Add an overvoltage protection circuit to prevent switching when over a certain voltage. Would consist of a zener diode, schmitt trigger, and any additional logic to turn the gate drivers off.

dawiggleman commented

Addition of pi filter to solar inputs:

Adding a pi filter configuration could reduce the current ripple on the input of the boost strings. This would allow for better operation at the exact maximum power point over the course of a switching cycle, instead of having variation about the mpp while current is being drawn from the array.

dawiggleman commented

Replace CAN driver with ISO1050DWR:

ISO1050DWR is available at an upcharge on resaler's sites. Ordered 11 for future use.

On Rev 1 board: use this chip, but need to deadbug TX/RX pins to be correct, but every other pin on the footprint should be compatible.

dawiggleman commented

The 1uF 100V capacitors used were PN/s that required you to order in quantities of 2000+. Replace in schematics with https://www.digikey.com/en/products/detail/kemet/C1206C105K1RAC7800/5119820. This part has already been added to inventree, and was the part that was ordered

dawiggleman commented

R22 (Feedback resistor to LDO 249K needs to be replaced with 31k). Replaced w/ 30k on current board.

dawiggleman commented

Put power schottky diode in parallel with Sync FETS on boost converters to increase efficiency. Schottky diodes will remove the reverse recovery losses felt by the boost FET. QRR losses are about half our losses (11W), with the diode, the primary loss will be V*I losses, and this will be about half (6W) of average losses due to the high side switch.

Component options:
https://www.digikey.com/en/products/detail/taiwan-semiconductor-corporation/TSP10H200S-S1G/7360635
https://www.mouser.com/new/stmicroelectronics/stm-stps4s200-rectifiers/

dawiggleman commented

Add discharge resistors across input capacitors to 12V buck converter

Potentially use a depletion mode FET instead of resistor to lessen useless losses

dawiggleman commented

Add dead time circuit:
Turns out the MAX5063DASA does not have a dead time circuit built in, the MAX5064 does. Add a dead time circuit to avoid shoot through on FETs.

Not immediately necessary if first choice for conversion is rectifier diodes, but will be good to have this option for future experiments.

Use this guide:http://www.modularcircuits.com/blog/articles/h-bridge-secrets/h-bridge-control/