无法使软件UART和I2C模块同时工作

大家好！我正面临着这个难题。我已经建立了一个跨阻抗放大器来放大光电二极管的信号，该光电二极管馈入PIC12F1822上的一个引脚，由ADC以1KHz进行采样，当采样发生时，这些采样通过串行与UART一起发送到PC。允许调谐增益的电路。由于无法使软件UART和I2C模块同时工作（尽管我确实使I2C自己工作），所以用I2C协议编程了数字的电阻值。然而，反馈电阻的NGE对其他人表现出奇怪的行为。如果我将数字罐编程为在大约125k-175k之间或在270k-370k之间，ADC提供的样品总是包括巨大的电压尖峰，但是如果我使用不同的电阻值，它们就不会出现。这种效果是100%可靠的，因为它们总是出现在这些范围中的值，而不会出现在外面的值。我在取样时用示波器观察放大器的输出，而示波器没有拾取电压尖峰，所以我相信它引入了一些伪影。ADC没有问题，因为我已经试过用固定的电阻值代替数字，同样的效果一直持续到如果我选择一个电阻值在可疑的范围内，如果我使用一个电阻值在它们之外就消失了。TWALE发出的话，它不会受到电阻值的影响。我也尝试过以不同的速率进行采样，但仍然会发生，所以我不认为这是某种混叠类型的影响（但又为什么它会依赖于阻力，如果是？）我还试着用同一种型号的芯片替换，以防我用的那个有毛病，仍然不走运。我在eevblog上看到一篇帖子，上面有个家伙在PWM的同时使用ADC，结果很奇怪。UART模块？我还是不相信这一点，为什么会对放大器中的电阻产生影响呢？我真的没有主意了。我已经上传了ADC正在接收的样本的图片，这里是示意图和示波器图片的链接。任何建议或想法，将不胜感激。



      以下为原文

    Hello all! I am facing quite the puzzle. I have built a transimpedance amplifier to the amplify the signal of a photodiode which the feeds into a pin on a PIC12F1822 to be sampled by the ADC at 1KHz and these samples are then sent over serial with the UART to a PC, as the sampling is occuring.

I have added a 1Mohm digipot to the amplifier circuit to allow for tuneable gain. The resistance value of the digipot is programmed with a I2C protocol which I have implemented with software bitbashing as I was unable to get the software UART and I2C modules to work concurrently (although I did get the I2C working by itself).

The system works completely as expected for certain ranges of the feedback resistance, however exhibits strange behaviour for others. If I programme the digitpot to be between roughly 125k-175k or between 270k-370k the samples provided by the ADC always include huge voltage spikes, however if I use different resistance values, they do not appear. This effect is 100% reliable in that they always occur for values in these ranges and never occur for values outside.

I am observing the output of the amplifier with an oscilloscope while sampling, and the oscilloscope is not picking up the voltage spikes, so I believe its some artifact introduced by the ADC.

It's not a problem with the digipot itself, as I have tried replacing the digipot with a fixed resistor value and the same effects till persist if I choose a value of resistance within the dodgy ranges and disappears if I use one that has a value outside of them.

I don't believe it's a software issue as then it wouldn't be effected by the resistance value. I have also tried sampling at a different rate and it still occurs, so I don't think it's some Aliasing type effect (but again why would it depend on resistance if it was?). I have also tried replacing the chip with another one of the same model, in case the one I was using was faulty and still no luck.

I read a post on the eevblog where a guy was having weird ADC spikes from concurrently using the ADC at the same time as PWM, could my effect be due to concurrently using the UART module? I'm still not convinced by that as again, why would that effect on the resistance in the amplifier?

I am truly out of ideas. I have uploaded a picture of the samples the ADC is receiving and (as you can only upload a single picture per post) here are links to the schematic and oscilloscope pictures. Any suggestions or ideas would be greatly appreciated.

[

   Attached Image(s)