Pic18F2331 I2C不工作

我建议您将在您的SyrPracMaMigFig语句中的设置与如果不提供γ-语用值时使用的默认设置进行比较，这将显示PIC不与代码中的CONFIG语句一起运行的差异。您将时钟设置为外部晶体，这对您的安装是否正确？



      以下为原文

    I suggest you compare the settings in your #pragma config statements to the default settings being used if you do not provide the #pragmas, this should show you the difference(s) that result in the pic not running with the config statements in the code. You are setting the clock to be an external crystal, is that correct for your setup? 

嗨，Stu，这是我在另一个被劫持的线程中给出的响应的副本。调试I2C从代码可能是棘手的，特别是如果你不能同时控制主控器。当你在调试器中停止奴隶时，它不会实时响应主，所以大师可能会认为。从属器没有响应，并在消息12中的代码中给出配置设置，即：如果真的有一个晶体恰好连接到PIC18F2231，如果您不确定，那么它是一个更简单和更安全的选择。即使在晶振没有连接或不工作的情况下，也可以进入调试器。在上面的代码中，MCLRE的设置是错误的，我不知道在进入调试器时MPLAB是否会重写它。应该尝试：MSSP硬件所产生的中断取决于SSPCON寄存器SSPM字段中选择的模式。如果选择普通的从属模式，SSPM＝6，则当接收到匹配地址时将中断，然后当接收到每个数据字节时中断。如果模式SSPM＝14，则为：EN选择时，当观察到开始和停止条件信号时，也会有中断。迈西尔



      以下为原文

    Hi Stu,
This is a duplicate of the response I have given in the other thread that was hijacked.
 
Debugging I2C slave code may be tricky, especially if you cannot control the Master at the same time.
When you stop the slave in debugger, it will not respond to the master in real time, so master may think slave is not responding, and give up 
 
Among configuration settings in code in message #12, there is:
  #pragma config OSC = HS
This only make sense if there really is a Crystal properly connected to the PIC18F2331
If you are not sure, then
#pragma config = IRCIO    // Internal oscillator block, port function on RA6 and port function on RA7
is a much simpler and safer selection.
 
Also:
#pragma config FCMEN = ON    // Fail-Safe Clock Monitor enabled
#pragma config IESO = ON    // Internal External Switchover mode enabled 
Then it might be possible to enter Debugger, even if Crystal oscillator were not connected or not working.
 
Setting for MCLRE is wrong in code above, I don't know if MPLAB will override it when entering debugger.
Use:
#pragma config MCLRE = ON   // MCLR Pin Enable bit, must be ON for Debugging to be possible
 
Mode 36
The most common reason for I2C slave not responding to Address,
is that the address do not match.
You should try: 
    SSPADD = slave_address << 1;  
 
Interrupts generated by the MSSP hardware depend on  the mode selected in SSPCON register SSPM field.
If ordinary slave mode is selected, SSPM = 6, there will be interrupt when a matching address have been received,
and then for each data byte as they are received.
If mode SSPM = 14, have been selected, then there will also be interrupts when Start and Stop condition signals are observed.
 
   Mysil

MysI&杰克，谢谢你的回应，我在这方面做了一些进步。通过使用MPLAB对默认配置进行检查，我能够使用μ* PrimaMac配置使代码在微控制器上运行。正如你们提到的，我没有正确地理解振荡器配置，当我使用内部振荡器而不是外部晶体时，OSC＝HS是不正确的设置。我也在I2C问题上取得了一些进展。我认为这个问题（再次被建议）是我在模拟发现2协议分析器中实际设置的地址是发现2设备（即主设备）而不是从微控制器：开始，H30[H18] WRNAK，H01 NAK，STOPI设置地址字段为H18认为这是从地址，所以留下的设置，因为他们在发现2和改变奴隶SSPADD到0x30在微控制器代码，产生一个地址匹配（和中断）正确时，SSPCON被设置为0x36。谢谢你的帮助，我应该能够得到它WOR国王从这里，只是想给出一个解释的解决方案，以防其他人有类似的问题。



      以下为原文

    Mysil & Jack,
 
Thank you for the response, I have made a bit more progress with this. I was able to get the code to run on the microcontroller using the #pragma config by checking it against the default configuration from MPLAB. As you both mentioned I hadn't understood the oscillator configuration correctly and as I was using an internal oscillator and not an external crystal the OSC = HS was the incorrect setting.
 
I also made some more progress with the I2C issues. I think the problem (again as was suggested) was the address that I was actually setting in the Analog Discovery 2 protocol analyser was for the Discovery 2 device (i.e. the master) and not for the slave microcontroller:
 
Start, h30 [ h18 | WR ] NAK, h01 NAK, Stop
 

 
I was setting the Address field to be h18 thinking this was the Slave address so leaving the settings as they were in the Discovery 2 and changing the slave SSPADD to 0x30 in the microcontroller code, generated an address match (and interrupt) correctly when SSPCON was set to 0x36.
 
Thank you for the help, I should be able to get it working from here, just wanted to give an explanation of the solutions, in case anyone else has similar problems.
 
Stu