如果将ADC GO_nDONE设置两次会怎样？

你好，我有一个PIC 16F18325，它必须在非常特定的时间和相当高的频率下从两个不同的ADC通道采样，我想用同一个TMR上的两个CCP模块和一个从CPCSI的脉冲产生ADC的自动触发的CLC模块来解决它。D我应该有足够的时间来做所有的事情，但是我在数据表上也找不到其他的问题，如果我（自动触发器）第一次设置ADC的GONNDONT位，在它完成之前（当位已经被设置）我（再次，自动触发器）再次设置时，会发生什么？转换会从开始重新开始，因此需要更长的时间来完成吗？将忽略它并继续转换已经进行，因此需要通常的时间完成吗？其他？也许是错误？我想第二个，但在数据表中没有一个明确的答案，所以我问你的经验，我知道这是一个非常极端的情况，在我的真实情况下它永远不会发生，我只是好奇，如果我必须考虑到它的任何特殊方式，使代码更重。谢谢！



      以下为原文

    Hello, I've a PIC 16F18325 that it has to sample from two different ADC channel at very specific times and quite high frequency, I've thought to solve it using two CCP modules on the same TMR and a CLC module that generates the auto-trigger for the ADC from the pulse of the CCPs

I've done the math and I should have enough time for everything, but I have some question I cannot find on the datasheet nor in other forums

what happens if, I (the auto-trigger) first set the GO_nDONE bit of the ADC, and before it has completed (so while the bit is already set) I (again, the auto-trigger) set it again?

• will the conversion restarts from the beginning and therefore it require a longer time to complete?
• will simply ignore it and continue the conversion already in progress and therefore takes the usual time to complete?
• other? maybe errors?
  

I suppose the second but there isn't a clear answer to this situation in the datasheet so i'm asking to your experience

I know it is a quite extreme situation, and in my real case it should never happens, i'm just curious if I have to take in account of it in any special way to make the code more reliable

Thank you!