ABCDEFGHIJKLMNOPQRSTUVWXYZAAAB
1
Erratas and change propositons for next versions of PiDSP
2
3
Date
Version
No.FileDescriptionProposed fixSeverity[1=minor, 5=critical]Add/Remove/ChangeStatus Note
4
18.2.2017B24Buck+PM.SchDocR37 + CK67 – not neededRemove1removefixedDifferent compensation scheme was selected
5
18.2.2017B25Buck+PM.SchDocCK24,R40,TP39 – not neededRemove1removefixedDifferent compensation scheme was selected, since we measure output impedance, we dont need stability measurement
6
18.2.2017B26Buck+PM.SchDocMissing testpoint at switching node
Add testpoint to switching node + Gnd testpoint nerby
2addfixedEases measurement and design of the snubber + its possible to measure output load via measuring voltage at the inductor
7
18.2.2017B219Buck+PM.SchDocOverhatiing inductor @3A
Different Inductor - with bigger footprint, 70um PCB
4changePending
Already reduced inductor value for reduced losses, I 'm not sure how much heat is generated in the inductor and how much heat is just transfered from the SMPS ic
8
18.2.2017B220Buck+PM.SchDocMissing testpoint at compensator output
Add the testpoint + see if theres GND testpoint nearby
3addfixed
Usefull for: Transconductance measurement, can be used for measuring output current, verification of compensator transfer function
9
7.3.2017B236Buck+PM.SchDoc
R24 + R25 - connected to wrong power rail. Disabling 3.3V will shut dwon whole board. Consder connecting r24 to both 3.3 -on board and RPI aswell
pull up connected to 5v + mosfet as level shifter to 3.3V logic
5changefixed
10
01.04.2017B241Buck+PM.SchDocPB pin sensitive to touch. You can turn it off by touching
Add pullup resistor - tied to input votlage
3changefixed
11
18.2.2017B228DSP_DIO.SchDocRename I2s singal to cmply with FreeDSP namingrename it1changefixed
12
7.3.2017B239dspclk.schdocgnd testpoint is too far from clk testpointsadd/move gnd testpoints1changefixed
13
18.2.2017B224DSPControl.SchDocReset timing - test if we need it at all... If so, add discharge diode
add diode, or remove timing capacitor
2addfixedSeems that once you get your power sequencing in order, you dont need reseting RC constant
14
18.2.2017B226DSPControl.SchDoc2ic adress selection - do I really need to address multiple DSP on one bus? It messes with my layout
Think about it, possibly remove the resistors
2removefixedfixed in layout
15
02.04.2017B246DSPControl.SchDocreset pullup is too big, there is a pull down in adau.
Reduce pullup value, reclaclulater timing cap accordingly
3changefixed
16
18.2.2017B216DSPPower.SchDoc0.1R Resistors are inductive at 40MHz
Remove degeneration resistors from 0603 capacitors
4changefixedLesson learned - measure your components before you use them
17
18.2.2017B217DSPPower.SchDocDegeneration resistors are missing at 10uF MLCC
Add degeneration resistors for 1206 package
2addfixedESR of capacitor was way lower than stated in datasheet - same as above, measure before use
18
02.04.2017B245DSPPower.SchDocpvdd is missing testpoint - only sma 1changefixed
19
7.3.2017B237frontPanelConnectorOtuput filter - check for resonances 2changefixedadded another 10u cap - there is just a slight 1db peak
20
18.2.2017B23InputFilterSchDocR41,CK69,R13,CK49 – superfluous, capacitors are not big enough to effectivelly damp possible filter ringing - design errorSince there is no board estate for bigger capacitor package, lets remove it. 1removefixedDesign oversight
21
01.04.2017B242InputFilterSchDocVin testpoint is missingadd testpoint1changefixed
22
08.04.2017B247InputFilterSchDocTransil PN and library name messed upfixed
23
18.2.2017B222
L5907MFX-3.3NPB.SCHDoc
Input filter - not evaluated. Possibly not realy performing well. New chipbeads or change to indcustor might be needed - possible add damping resistorEvaluate the performance. 4changefixedCurrent filter design is more than satisfactory - 60dB attenuation at sw freqency of dual channel buck
24
18.2.2017B21PiDSP.pcbDocPrevious changes have resulted in split power plane ner I2c dsp address resistors and at the top edge of DSP ic. This results in incerased impedance of power distribution systémLayout change – reconsider need for i2c resistor.1changefixedThat what happens if you design your stuff while really tired...
25
18.2.2017B22PiDSP.pcbDocSolder mask is not removed from the edges on top layerRemove solder mask1removeclosedlooks ugly without it, I'll keep it
26
18.2.2017B214PiDSP.pcbDoc6.8u inductor footprint - not hand soldering friendlyIncrease pad size 2changePendingReflow soldering is ok
27
18.2.2017B215PiDSP.pcbDocOscillator footprint - not hand soldering friendly increase pad size 2changePendingReflow soldering is ok
28
18.2.2017B218PiDSP.pcbDocOverheating of main 5V power supply @3A
Derate power supply to 2.5A, or use 70um PCB - dunno how much this helps, not tested
4changePending
design error, SMPS is located at the edge of pcb, RPI connector on the other side, not enough cooling area. I've already reduced switching freqency
29
18.2.2017B227PiDSP.pcbDocMounting holes not complying with FreeDSP standard
See if that can be fixed, and fix it. Might not be possible due to connector placement
3changeclosed
I moved them in order to be able to screw multiple boards togeather in hammond cas. FreeDSP hole arangemend does not enable tyhis. I need to bring it up with other guys.
30
18.2.2017B229PiDSP.pcbDocWhen MLW connector with cable relief is used, J3 is coliding with SD card on RPI3Do not use cable relief1changefixednot going to be fixed.Only so much you can fit into 10x10cm..
31
18.2.2017B231PiDSP.pcbDocRPi connector missing designatoradd designator to silk screen1changefixed
32
26.3.2017B240PiDSP.pcbDocFront panel connector is missing pin description in silk screen. Programing connector dittoAdd silk screen description1changefixed
33
02.04.2017B243PiDSP.pcbDocToslink not fixed in pcb
add solder pads for mounting pins
5changefixed
34
02.04.2017B244PiDSP.pcbDocfront panel connector - wrong voltages for some GPIO pushbuttons - pullups/downs are selectable only for output. fix pinout3changefixed
35
18.2.2017B230PowerSupply.SchDocstartup sequencing - used wrong pull up voltages, start up seqence does not work as intended, test disable singal
Test functionality of disable signal. Redesign for proper function
5changefixedA little causality issue....
36
7.3.2017B238PowerSupply.SchDocrc timing reset diodes misingAdd resting diodes2changefixed
37
18.2.2017B233RPi&MCUHAT EEPROM connected to dsp 3.3V - if you disable PiDSP it has no power ...
Verify if thats an issue ? Connect to RPI 3v3
1changeclosedby default, PiDSP is on, you need to pull high to disable board. so by default there is a power, so you can read the eeprom
38
18.2.2017B27TPS54292.SchDocR69,CK33,R70,CK60 - not neededRemove1removefixedDifferent compensation scheme choosed
39
18.2.2017B28TPS54292.SchDocCK55,CK56,R61,R65,TP41,TP43 - not neededRemove1removefixedDifferent compensation scheme choosed, since we measure output impedance, we dont need stability measurement
40
18.2.2017B29TPS54292.SchDocMissing testpoint at switching nodes - dual channel buck converter
Add testpoint to switching nodes + Gnd testpoint nerby
2addfixedEases measurement and design of the snubber + its possible to measure output load via measuring voltage at the inductor
41
18.2.2017B210TPS54292.SchDocCK28 - only big mlcc in 0805 package
Change to 1206 package with value >=4.7u, prefarably somtehing that is already used on the board
2changefixedsimpler bom
42
18.2.2017B211TPS54292.SchDocNot enought footprints for R66 - I need two resistors parallelAdd parallel resistor3addfixedFor hand soldering put two atop each other
43
18.2.2017B221TPS54292.SchDocMissing testpoint at compensators outputs
Add the testpoints + see if theres GND testpoint nearby
3addfixed
Usefull for: Transconductance measurement, can be used for measuring output current, verification of compensator transfer function
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100