pierremolinaro
can.begin returns ok but can.tryToSend leaves SPI lanes untouched
Hi, I have a possible Bug (or if not, please enlighten me): Hardware: ESP32 S3 N16R8, MCP2517FD
I can Configure the MCP2517FD with SPI (errorCode 0x00). Configuration has a visible SPI signal. However, when trying to transmit. The SPI lanes remain ide. Thus, the chip cannot transmit any messages, since it receives none via spi.
#include <ACAN2517FD.h>
#include <SPI.h>
#include "config.h"
ACAN2517FD can(MCP2517_CS, SPI, MCP2517_INT);
void canSetup()
{
ACAN2517FDSettings settings(ACAN2517FDSettings::OSC_20MHz, 125 * 1000, DataBitRateFactor::x1);
//--- For version < 2.1.0
// ACAN2517FDSettings settings (ACAN2517FDSettings::OSC_4MHz10xPLL, 125 * 1000, ACAN2517FDSettings::DATA_BITRATE_x1) ;
// settings.mRequestedMode = ACAN2517FDSettings::InternalLoopBack; // Select loopback mode
settings.mRequestedMode = ACAN2517FDSettings::NormalFD; // Select loopback mode
// -- -Configure regular transmit chain(used when message.idx == 0, default)
SPI.begin(MCP2517_SCK, MCP2517_MISO, MCP2517_MOSI);
//--- Configure ACAN2517FD
delay(2000);
while (1)
{
const uint32_t errorCode = can.begin(settings, []
{ can.isr(); });
if (errorCode == 0)
{
Serial.print("Bit Rate prescaler: ");
Serial.println(settings.mBitRatePrescaler);
Serial.print("Arbitration Phase segment 1: ");
Serial.println(settings.mArbitrationPhaseSegment1);
Serial.print("Arbitration Phase segment 2: ");
Serial.println(settings.mArbitrationPhaseSegment2);
Serial.print("Arbitration SJW:");
Serial.println(settings.mArbitrationSJW);
Serial.print("Actual Arbitration Bit Rate: ");
Serial.print(settings.actualArbitrationBitRate());
Serial.println(" bit/s");
Serial.print("Exact Arbitration Bit Rate ? ");
Serial.println(settings.exactArbitrationBitRate() ? "yes" : "no");
Serial.print("Arbitration Sample point: ");
Serial.print(settings.arbitrationSamplePointFromBitStart());
Serial.println("%");
Serial.println("Configuration seems successful");
return;
}
else
{
Serial.print("Configuring went wrong with errorcode: ");
Serial.print(errorCode);
Serial.print("sizeof (ACAN2517FDSettings): ");
Serial.print(sizeof(ACAN2517FDSettings));
Serial.println(" bytes");
Serial.println("Configure ACAN2517FD");
//--- For version >= 2.1.0
Serial.print("MCP2517FD RAM Usage: ");
Serial.print(settings.ramUsage());
Serial.println(" bytes");
}
}
}
void setup()
{
Serial.begin(115200);
//--- Wait for serial (blink led at 10 Hz during waiting)
while (!Serial)
{
delay(50);
digitalWrite(LED_BUILTIN, !digitalRead(LED_BUILTIN));
}
canSetup();
}
static uint32_t gBlinkLedDate = 0;
static uint32_t gReceivedFrameCount = 0;
static uint32_t gSentFrameCount = 0;
void loop()
{
if (gBlinkLedDate < millis())
{
gBlinkLedDate = millis() + 500; // Update this to send every 2 seconds
CANFDMessage frame; // This needs to be defined here, or outside loop()
frame.id = micros() & 0x7FE;
frame.len = 64; // Be aware of payload size settings for the FIFO
bool messageOK = frame.isValid();
Serial.print("is valid: ");
Serial.println(messageOK ? "yes" : "no");
const bool ok = can.tryToSend(frame);
if (ok)
{
gSentFrameCount += 1;
Serial.print("Sent: ");
Serial.println(gSentFrameCount);
}
else
{
Serial.println("Send failure");
// Add a check for error counters if you frequently get send failure
// Serial.print("Error counters: ");
// Serial.println(can.errorCounters()); // From section 22.5
}
}
// Keep the receive part commented out for now to isolate TX issues.
}
SPI Signal:
First successful configuration and then idle lanes:
Hello,
The library should work with an ESP32S3.
What is your board ? What are the pin settings (MCP2517_CS, MCP2517_INT, MCP2517_SCK, MCP2517_MISO, MCP2517_MOSI) ?
Is the MCP2517FD alone on the bus ?
Note the MCP2517FD is quite buggy, it is better to use an MCP2518FD.
Pierre
Le 26 juin 2025 à 16:34, Jan Dominik Franz @.***> a écrit :
JDFranz created an issue (pierremolinaro/acan2517FD#58) https://github.com/pierremolinaro/acan2517FD/issues/58 Hi, I have a possible Bug (or if not, please enlighten me): Hardware: ESP32 S3 N16R8, MCP2517FD
I can Configure the MCP2517FD with SPI (errorCode 0x00). Configuration has a visible SPI signal. However, when trying to transmit. The SPI lanes remain ide. Thus, the chip cannot transmit any messages, since it receives none via spi.
`#include <ACAN2517FD.h> #include <SPI.h> #include "config.h"
ACAN2517FD can(MCP2517_CS, SPI, MCP2517_INT);
void canSetup() {
ACAN2517FDSettings settings(ACAN2517FDSettings::OSC_20MHz, 125 * 1000, DataBitRateFactor::x1); //--- For version < 2.1.0 // ACAN2517FDSettings settings (ACAN2517FDSettings::OSC_4MHz10xPLL, 125 * 1000, ACAN2517FDSettings::DATA_BITRATE_x1) ; // settings.mRequestedMode = ACAN2517FDSettings::InternalLoopBack; // Select loopback mode settings.mRequestedMode = ACAN2517FDSettings::NormalFD; // Select loopback mode // -- -Configure regular transmit chain(used when message.idx == 0, default)
SPI.begin(MCP2517_SCK, MCP2517_MISO, MCP2517_MOSI);
//--- Configure ACAN2517FD delay(2000);
while (1) {
const uint32_t errorCode = can.begin(settings, [] { can.isr(); }); if (errorCode == 0) { Serial.print("Bit Rate prescaler: "); Serial.println(settings.mBitRatePrescaler); Serial.print("Arbitration Phase segment 1: "); Serial.println(settings.mArbitrationPhaseSegment1); Serial.print("Arbitration Phase segment 2: "); Serial.println(settings.mArbitrationPhaseSegment2); Serial.print("Arbitration SJW:"); Serial.println(settings.mArbitrationSJW); Serial.print("Actual Arbitration Bit Rate: "); Serial.print(settings.actualArbitrationBitRate()); Serial.println(" bit/s"); Serial.print("Exact Arbitration Bit Rate ? "); Serial.println(settings.exactArbitrationBitRate() ? "yes" : "no"); Serial.print("Arbitration Sample point: "); Serial.print(settings.arbitrationSamplePointFromBitStart()); Serial.println("%"); Serial.println("Configuration seems successful"); return; } else { Serial.print("Configuring went wrong with errorcode: "); Serial.print(errorCode); Serial.print("sizeof (ACAN2517FDSettings): "); Serial.print(sizeof(ACAN2517FDSettings)); Serial.println(" bytes"); Serial.println("Configure ACAN2517FD"); //--- For version >= 2.1.0 Serial.print("MCP2517FD RAM Usage: "); Serial.print(settings.ramUsage()); Serial.println(" bytes"); }} }
void setup() { Serial.begin(115200); //--- Wait for serial (blink led at 10 Hz during waiting) while (!Serial) { delay(50); digitalWrite(LED_BUILTIN, !digitalRead(LED_BUILTIN)); }
canSetup(); }
static uint32_t gBlinkLedDate = 0; static uint32_t gReceivedFrameCount = 0; static uint32_t gSentFrameCount = 0;
void loop() {
if (gBlinkLedDate < millis()) { gBlinkLedDate = millis() + 500; // Update this to send every 2 seconds CANFDMessage frame; // This needs to be defined here, or outside loop() frame.id = micros() & 0x7FE; frame.len = 64; // Be aware of payload size settings for the FIFO bool messageOK = frame.isValid(); Serial.print("is valid: "); Serial.println(messageOK ? "yes" : "no"); const bool ok = can.tryToSend(frame);
if (ok) { gSentFrameCount += 1; Serial.print("Sent: "); Serial.println(gSentFrameCount); } else { Serial.println("Send failure"); // Add a check for error counters if you frequently get send failure // Serial.print("Error counters: "); // Serial.println(can.errorCounters()); // From section 22.5 }} // Keep the receive part commented out for now to isolate TX issues. }`
SPI Signal: First successful configuration and then idle lanes: image.png (view on web) https://github.com/user-attachments/assets/0388fde5-63c4-40e7-bdb7-d6dbd04be1b8 — Reply to this email directly, view it on GitHub https://github.com/pierremolinaro/acan2517FD/issues/58, or unsubscribe https://github.com/notifications/unsubscribe-auth/AEWKZVE7B5SRAZN6YEONFH33FQAHTAVCNFSM6AAAAACAGNNLKSVHI2DSMVQWIX3LMV43ASLTON2WKOZTGE3TSMZUGY3DANA. You are receiving this because you are subscribed to this thread.
Hello,
Thank you for your fast reply.
What is your board ? > ESP32-S3 DevKitC-1 N16R8 Module
Pin Settings: // MCP2517FD CAN Controller Pins #define MCP2517_CS 45 // CS input of MCP2517FD // #define MCP2517_MISO 35 // SCK input of MCP2517FD // #define MCP2517_MOSI 36 // SDI input of MCP2517FD // #define MCP2517_SCK 37 // SDO output of MCP2517FD //
Is the MCP2517FD alone on the bus ? > Yes
Note the MCP2517FD is quite buggy, it is better to use an MCP2518FD. > Did do that, and we've had some progress with that:
The library communicates exactly one message over the SPI after each can.begin()
Thanks in advance,
JDF
Ok, if the MCP2517FD is alone on the bus, you should select loopback mode.
In normal mode, a CAN controller sends repeatedly the first frame, as it is not acknowledged by a receiver.
Pierre
Le 27 juin 2025 à 11:21, Jan Dominik Franz @.***> a écrit :
JDFranz left a comment (pierremolinaro/acan2517FD#58) https://github.com/pierremolinaro/acan2517FD/issues/58#issuecomment-3012322504 Hello,
Thank you for your fast reply.
What is your board ? > ESP32-S3 DevKitC-1 N16R8 Module https://www.amazon.de/dp/B0DBHGL73T?ref=ppx_yo2ov_dt_b_fed_asin_title&th=1 Pin Settings: // MCP2517FD CAN Controller Pins #define MCP2517_CS 45 // CS input of MCP2517FD // #define MCP2517_MISO 35 // SCK input of MCP2517FD // #define MCP2517_MOSI 36 // SDI input of MCP2517FD // #define MCP2517_SCK 37 // SDO output of MCP2517FD //
Is the MCP2517FD alone on the bus ? > Yes
Note the MCP2517FD is quite buggy, it is better to use an MCP2518FD. > Did do that, and we've had some progress with that:
The library communicates exactly one message over the SPI after each can.begin()
Thanks in advance,
JDF
— Reply to this email directly, view it on GitHub https://github.com/pierremolinaro/acan2517FD/issues/58#issuecomment-3012322504, or unsubscribe https://github.com/notifications/unsubscribe-auth/AEWKZVGUIZOI7IVZAW5WSKD3FUEJNAVCNFSM6AAAAACAGNNLKSVHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMZTAMJSGMZDENJQGQ. You are receiving this because you commented.
Ok it was a Problem with our schematic. Turns out CANFD hardware is vrey easily disturbed when having inconsistent setups. » Always use a PCB or a professional breakoutboard ( Do not use a Breadboard like we did) Here's our breakout board for the MCP2518FD and ATA6263, that seems to have worked:
Does anyone else see significant differences to their schematic?