Power Selection Validation Rev 2.0

This page will go over the validation process for the rev 2 of the power select board for MSXIV.

To start off, I will validate the sequencing of having a load and power source to the board, and that all the appropriate rails are high. Note that a 3v cell has been placed on the cell holder of the board for the jump-start circuit.

The following is the setup:

Tests:

Aux Battery Input: Board outputs 12V.

DC DC Input: Does not work. LED flickers on and off, but eventually stabilizes after about 5-10 seconds. Will only start out stable if another power source was connected first and the DC DC input is connected afterwards. See current issues section.

• this video shows the scoping of TP3 and TP2

Open

When the line drops, the power supply is turned off. When the power supply is turned on, the line starts bouncing.

Power Supply Input: Board outputs 12V.

• I was able to replicate the LED flickering issue; seems to not stabilize at all when compared to DCDC input connector

Note: The jump start circuit that was added for rev2 is NOT needed! The LTC4417 has its own functionality.

Open

From the datasheet: “If VOUT is lower than 2.4V, VLDO switches to the highest valid priority input supply, V1, V2 and V3. If no input supply is valid, VLDO is connected to VBLDO, where the diode-OR selects highest input voltage input supply as the source”. i.e. as long as the board has power it should jumpstart itself since the STM32 at that point would have the enable pin at high impedance.

Current issues:

• If a new power source is connected while the bulk capacitor is not fully discharged, the output voltage will not be stable.

September 19 Update:

• as shown in the above video, the single flicker most likely occurs from an issue with the PS code

• from further investigation, found an issue with net naming:

• currently talking to @Hewitt McGaughey to see if the issue is possibly from pulling the enable pin low momentarily

  • Ideally, the enable pin will not be pulled low at all, and will just remain High-Z

  • It could be a different pin as well (apparently PB8 is not initialized)

  • after more testing, it was found that PB15 initialization is causing the brief flicker - it might be going low during initialization

  •  

• there exists another issue that when there is a controller board attached to the board, and there is no load on the board itself (like on the output) that it will start flickering if the power supply is turned off immediately after turned on (like when the cap is discharging)

• the above issue is a non-issue for a few reasons

  • there should never be a condition where there is actually no load and we expect the car to be functional

  • This issue is made worse based on the model of power supply, and what I assume are some switching transients that cause issues with the CB 3V3 buck

    • there is less flickering cases with the keysight E3631A than the

• overall it is determined the issue that gives slight concern is due to the load and the ability of the bulk capacitor to discharge → with a High-Z DC load, the flickering can occur because it only occurs if the supply is restarted anywhere from around 2-4V → with a High Z load it takes about two whole seconds to drop from 13.5V to under 2V, so in this time period, the flickering will persist → if the load is drawing at least 10mA, then there is not enough time to turn back on after turning it off to get the flickering (ie turn on in the 2-4V range) so this would be a non-issue (10mA of load is enough to discharge the cap). If there is less than that load and if there is immediate turning and on and off of the power supply (within 2 seconds) then there is this flickering phenonmenon that isnt good. However, this is not a functional case at all, and until further consideration there doesnt seem to be a case where there would be no loads for a case where power select is functional (should have both PDs on the output which impose a load)