Differences

This shows you the differences between two versions of the page.

--- intern:railflyer [2012/10/22 17:00] – [General problems and suggestions] svesch
+++ intern:railflyer [2025/06/11 20:35] (current) – external edit 127.0.0.1
@@ Line 1: / Line 1: @@
+====== Railflyer PCB review ======
+===== Truck PCB (2012-10-23) =====
+==== Revisited 30.11. ====
+  * No current limiting resistor on the LEDs but LED22 and LED23.
+  * If no external resistor is used there might be a problem with to high current through the LEDs which reduces lifetime or can destroys several LEDs
+  * Christopher states, that maximum of 10 LEDs per truck are switched on, but 20 mA each. This means a total current of 200 mA per truck. This is to much for the 3.3V regulator.
+Suggestions to solve this:
+  * Use the regulated 5V instead of the 3.3V here.
+  * Use current sources instead the transistor/resistor combination.
+  * Connect the LEDs in series instead of parallel: i.e. if the LEDs need less than 2.5V voltage drop.
+Further problem: +5V is not available at the moment.
+Use the free contacts 2 and 3 for this OR if each truck PCB gets its own motor driver the Motor N and P signals.
+==== General problems and suggestions ====
+  * Wires go through parts on schematic: cleaned up.
+  * Logo and text: Removed.
+  * Checked package of 2SC4617 (Q1-Q11): Ok
+  * Check the used LEDs, particularly LED9-LED16. Transistor can only drive up to 150 mA!!!
+===== Processor PCB (2012-10-22) =====
+==== General problems and suggestions ====
+  * As send by Christopher board and schematics are not consistent: That is bad :(
+  * Following steps to solve this: Renamed J2 to FRONT_TRUCK (rotated and mirrored for correct pinning), renamed J1 to REAR_TRUCK, Rotated and mirrored J7 for correct pinning (Pin 1 is low battery, P20 is 3,3V now)
+  * Layer 2 and Layer 3 in board removed as not needed.
+  * Board: Some vias are to close together: Shorts possible while soldering.
+  * Board: For current wires the distance inbetween is too low
+  * Keep the connections to the truck PCBs as short as possible to prevent EMI issues!
+  * Renamed DCC_P to DCC_R for consistence with Power PCB
+  * Resized the smallest vias to standard 0.25 mm drill
+  * Resized motor and DCC wires to 0.4 mm drill
+==== Sound chip IC3 (SGTL5000) ====
+This IC converts a I2S stream into an audio stream.
+  * The package is QFN-20 (SGTL5000XNLA3/R2), checked.
+  * I2C: Address is 0x14/0x15, up to 400 kHz are supported
+  * C3 is too far away from VDDA pin: Fixed.
+  * Ground connection is suboptimal, resized ground layer.
+  * Capacitors are 0201: Ok.
+==== Serial flash IC2 (MT29F4G01) ====
+This IC stores the wav-Files for sound generation. Altough a Flash with parallel bus access, only the SPI pins are used (as a common SPI flash/eeprom SO8).
+  * Package: Checked.
+  * Optimized routing for VCC.
+  * Missing decouple C: Added C5 (100nF).
+==== Accelerometer IC4 (MMA7361LC) ====
+This IC measures 3 axes acceleration (as analog signals).
+  * Package: LGA. Checked.
+  * 0g-Detect is open: ok.
+  * Sleep pin is open: not ok, connected to controller with pulldown (for saving battery power).
+  * g-Select pin is open: ok, internal pulldown resistor enables 1.5g operation.
+  * Three outputs (X,Y and Z) go to ADC inputs of controller: ok, decoupling/filter is installed (C1-C3).
+==== Microcontroller IC1 (MC13224V) ====
+This controller controls the locomotive and is connected via wireless or DCC to the command station.
+  * I2C Pullups are missing: Added R4 and R5.
+  * VDD and capacitor not optimal: Optimized.
+  * Antenna: Ok
+  * Pinnings: Checked
+==== BEMF circuit ====
+This circuit can be used to measure the back EMF of the connected motors. It is a simple voltage divider with hold element.
+  * Traces are ok, values must be checked in prototyping phase
+===== Power PCB (2012-10-12) =====
+==== Battery monitor IC6 (MAX17059) ====
+This part is for monitoring the battery voltage and has an I2C interface.
+  * correct: version for 2 cell monitor
+  * Not correct: C36, bypass VDD and GND with 100nF NOT Cell and GND
+  * Not correct: Connect CTG to GND!
+  * Recommendation: ALRT is open drain, pullup needed (better do not use the internal MC ones, so the board works stand alone)
+  * I2C: Address is 0x6C/0x6D, up to 400 kHz are supported
+  * Package: TDFN-8, mark pin 1 on ground plane, pinning is correct.
+  * Bypass-C is not optimal routed
+  * Routing of SDA/SCL to connector J2 is okay
+==== Battery charger IC5 (MAX1873) ====
+This part is for loading the battery.
+  * D7 can be omitted, VDCC is sourced by a bridge rectifier which prevents back source.
+  * Shunt R17 with 0.04 Ohms limits load current to 2.5 A from rails
+  * Shunt R15 with 0.08 Ohms limits load current to 2.5 A
+==== Voltage regulator IC4 (TL1963ADCQ) ====
+This part regulates voltage to 5V from input voltage which can be up to 20V DC. Current can be up to 1.5A.
+  * Input capacitance is missing (10 uf)
+  * Correct: Calculated output voltage due resistor ratio: 5.0336
+  * Output capacitance C1 is ok
+==== Voltage regulator IC3 (MIC5205-3.3) ====
+This part regulates voltage to 3.3V from input voltage 5V. Current can be up to 150 mA.
+  * No errors but the ouput current might be a show stopper
+==== H-Bridge interface and H-Bridge ====
+  * C3 is far away from H-Bridge, bring it nearer
+==== DCC Track input ====
+  * Decoupling capacitor at IC1 is missing (100nF)
+==== General suggestions ====
+  - Circuit around D9, C38 and R42 is not considered in BOM, maybe it is a smoother for VDCC voltage, leave it for now.
+  - Bad: All schematics one one Eagle sheet, better use two (for printing and reading)
+  - Bad: I would avoid using P$ as prefix for pin assignments
+  - Board: For current wires the distance inbetween is too low
+  - Board: Remove old logos as ordered