有关PIC16F1713（或类似）上的PWM和PPS系统的问题

哦，是的，当我说“或类似”时，我猜这是一个28引脚的“4位数字系列”设备，它的PPS系统很可能像PIC16F1703，如果我必须改变部件和“不做”PPS，那么我必须从头开始一个新的PCB（并添加PPS外部的等）并尝试猜测什么一个工作组的PIN可能看起来像（这已经是一个问题，我相信两次在同一个项目，所以这个前提是旧的真的很快）。此外，如果它有助于两个引脚，我试图使用PWM是RC1和RC2，如果这慢跑任何人的记忆。哦，当我说“跑得很快”时，我改变了OSCCon的速度，在任何地方，它仍然以任何速度失败，我尝试，仍然会运行良好的单步，所以GEE，这不是一个真正的时间为基础的问题，为什么我不感到惊讶。对不起，如果我看起来有点恼火，因为我只是知道很多的勘误表没有得到“公开”的记录（只是经历了一个不同的过程）。



      以下为原文

    ...oh yeah when I say "or similar" I guess that's a 28 pin "4 digit series" device with a PPS system which could probably be like the PIC16F1773, if I have to change parts and "do without" PPS then I have to start over with a new PCB (and add the equiv of the PPS externally) and try and GUESS what a working set of pins might look like (that's already been an issue I believe twice on this same project so this premise is getting old real quick). Also if it helps the two pins I'm trying to use for PWM are RC1 and RC2 if that jogs anyone's memory. Oh and when I said "running at speed" I changed the rate in OSCCON all over the place, it would still fail at any speed I tried and STILL would run fine single-step so gee it's not REALLY a time-based issue, why am I not surprised. Sorry if I seem a little annoyed because I just know a lot of the errata doesn't get "openly" doc'ed (just went through that on a DIFFERENT proc).

我被你提到的仿真弄糊涂了。你是在真正的硬件上运行全速，用调试器连接速度，还是在模拟器中全速运行？



      以下为原文

    I'm confused by your references to emulation.
Are you running this at full speed on real hardware,
full speed with a debugger connected,
or in the simulator?
 

我不希望对PPS寄存器的写入对LAT寄存器有影响。数据表中有什么让你相信会发生这样的事情吗？



      以下为原文

   
 
I would not expect a write to a PPS register to have an effect on a LAT register.  Did something in the data sheet lead you to believe that would happen?

在一个非常真实的电路板上有一个芯片。电源是通过一个“5伏3针开关LDO更换”稳压器从28伏“墙疣”切换器，这是什么桥连接。没有提供大量的能量存储（只是开关数据表所要求的），所以如果有大量的电流通过桥接而被击穿，那么它会将壁疣向下按压，因此供应暂时消失，调试器返回到复位（和右）。现在这些桥上根本没有载物了。架构是这样的，板上有四个芯片用于应用程序，但我只调试一个，因为它们可以为不同的输出运行相同的代码。桥只需要在几秒钟后完全反转，所以操作可能慢（PWM只有一个KHz），所以这个问题很难处理。



      以下为原文

    It's with a PICkit3 on a chip in a VERY REAL circuit board. Power is provided to the chip via one of those "5 volt 3 pin switcher LDO replacement" regulators down from a 28 volt "wall wart" switcher which is what the bridge is connected across. There's not a whole lot of energy storage provided (just what the switcher data sheet asked for) so if there's a substantial amount of current drawn by shoot-through via the bridge then it honks the wall wart down and so the supply goes away momentarily and the debugger returns to reset (and right now there's no loads on these bridges at all). The architecture is such that there's FOUR of these chips on the board for the application but I only debug one since they can all run the same code for different outputs. The bridge only has to fully reverse at most after a few seconds so the operation can be VERRRRY slow (and the PWM's only a couple kHz) so it's extremely frustrating that the problem is so hard to get a handle on.

在PPS输出寄存器中的一个零点意味着锁存针被引导到输出端，并且锁存针先前已经被清除，所以如果我已经分配了一个PPS输出引脚到PWM，那么我清除了寄存器，PIN应该马上变低，或者至少这是数据表将给你BEL。唉！（除非有什么毛病？）会有故意拖延，他们没有记录吗？这是我想知道的事情之一。



      以下为原文

    A zero in a PPS output register means that latch pin is directed to the output, and that latch pin has previously been cleared so if  I had assigned a PPS output pin to PWM then I cleared that register the pin should be going low immediately, or at least that's what the data sheet would have you believe. (Unless there's some kind of glitch?) Would there be some intentional delay and they didn't document it? That's one of the things I'm wondering about here.
 

更可能是硬件或软件设计问题。既然你已经选择了对这个问题进行哲学思考，而不是展示硬件或软件或范围图片，那么我们的硬件和软件工程师很难帮助。但是我将把我的哲学帽放在上面，尽管我对社会科学不太感兴趣。我很肯定PPS硬件不知道PWM硬件是什么，所以我不相信这是你的问题。



      以下为原文

    More likely a hardware or software design problem. Since you’ve chosen to philosophize about the the problem rather than show hardware or software or scope pictures it’s difficult for us hardware and software engineers to help.

But I’ll put my philosophy hat on, though I’m not much for the social sciences. I am pretty sure the PPS hardware has no idea what the PWM hardware is up to, so I don’t believe that’s your problem.

对，在这个芯片上模拟使用的任何引脚在正常使用中的任何一个引脚都不是“文档特征”。然而，如果你尝试使用这个芯片上的RB0作为模拟输入，你将不会得到一个转换的输出，这将表明该管脚上的电压一直低于约7伏或高于约1.3伏特（这是我所说的版本2001和逻辑的GND作为参考）。我绝对肯定这个芯片错误是在内部记录的，但它只是在他们拒绝承认的问题清单上，我只是一直在寻找更多的错误。是的，我对过于简单的离散高边驱动器的性能有点怀疑，当我用尽时间把电路板弄出来时，我发现了它，但是如果它实际上是在“脉冲”上拉伸毛刺，我可以告诉你，如果没有除草剂，那根本不是问题。首先是毛病。在稳定的输出端口上不应该有故障，但是……总之，它可能是另一个“特性”。所有这些“无证的特征”，当他们知道他们在那里，但不会承认的时候，不得不去处理，真是累人！



      以下为原文

    Right, it's also not "a documented feature" that any of the pins designated for analog use on this chip are any different in normal use than any others. Yet if you try and use RB0 on this chip as an analog input you won't get a converted output that will indicate that the voltage on that pin ever goes below about .7 volts or above about 1.3 volts (this is release 2001 I'm talking about and logic to gnd as reference). I'm absolutely certain this chip error is documented SOMEWHERE internally but it just goes on the list of problems they refuse to admit, and I just keep finding more of them all the time. Yes I'm somewhat skeptical about the performance of the overly simple discrete high-side driver that I came up with when I was running out of time to get the board out, but if it's actually stretching glitches into "on pulses" I can tell you that wouldn't be an issue at all if there weren't decoding glitches in the first place. And there shouldn't be glitches on stable output ports but...all in all it's probably another "feature". And all these "undocumented features" are getting REALLY tiring to have to deal with when they know they're there but won't admit it!

我模糊地记得指出，RB0是ZCD引脚，你说这是不可能启用的。



      以下为原文

    I vaguely remember pointing out that RB0 is the ZCD pin, and you said that couldn't possibly be enabled.
 

不，我不相信我们在这里讨论过。问题是RB0也是一个12，“默认”应该是ZCD禁用。在这个电路板上没有AC，我从来没有试图启用它，甚至没有想到它会在那里。很可能这是一个DOC错误，而PIN不打算用于AN12。不管怎样，我刚才把它换成了RC4（我想起来了），所以现在我没有用它来做任何事情。我会在商店里买一些“快门挡”的帽子，因为在这一点上，我倾向于相信它是产生马刺的高边，而在我倾向于认为它是“停留”的低端或是因为我显然没有。了解文档或它们有缺陷。看到我的问题是，我只需要确保，当我希望低侧引脚低，我真的必须确保我可以用PPS，在PWM侧没有注册，实际上有一点，它被记录为“转PWM现在设置（或清除）这个位”，我假设清除PPS REGPIST的针可以做到这一点，但大家都知道什么“假设”是不是？



      以下为原文

    No I don't believe we discussed it here. Problem is RB0 is also AN12 and "default" is SUPPOSED to be ZCD disabled. There's no AC even available on this board and I never attempted to enable it or even thought about it being there. It could well be this is really a doc error and that pin wasn't intended to be used for AN12. Anyway I swapped that out a while ago for RC4 (as I recall) so right now I'm not using it for anything. I'll get some "glitch stopper" caps for my too-fast driver at the store later because at this point I'm inclined to believe that it's the high side that's generating the spurs, whereas before I was inclined to think it was the low side that was "staying on" or something because apparently I didn't understand the docs or they were defective. See my problem is I just need to be sure that when I want that low side pin to be low I have to REALLY be sure I can do it with PPS, on the PWM side there's no register that actually has a bit in it that's documented as "to turn PWM off NOW set (or clear) this bit" and I ASSUMED clearing the PPS register for that pin would do it but you all know what "assume" means don't you?

在最初的帖子中，你做了一个概括的概括，即在一个完整的设备系列中声明了错误，并坚持不懈地抵制显示了即使是一个设备的代码。我们中的一些人展示了特定设备的输出，这驳斥了你的概括。我当时认为，您的问题是由于未能理解全局上拉使能的逻辑状态和独立端口/PIN上拉状态能够以不同的方式影响上拉的原因。（0为全局启用位，1为个别启用位），而您试图读取模拟输入与启用UPS。我无法证明我的信仰，因为我们从来没有看到任何代码从你。你再也不会被打扰了。现在你又在一个设备中声明了一个bug，并假设在整个“类似”的设备中必须有未记录的bug。因为你没有显示任何代码，我们无法验证或反驳它。好吧，我质疑你对“类似”设备的一般要求。我将展示一个特定设备的例子，它显示RB0（ZCD引脚）可以用作模拟输入引脚（如果只有），ZCD模块被禁用，因为Q上边-H -B已经提到（两次）。我没有PIC16F1713来播放。我最靠近的是PIC16F1718，我认为它和你能得到的相似。（参考这些数据表），我在Vdd和VSS之间连接了一个10K的容器。我把雨刷通过1K电阻器连接到RB0，并运行了一个程序，显示了ADC读数与ZCD启用和禁用。我通过在ZCDCON寄存器中设置ZCSDN位来实现1。我通过清除ZCCDN位为零来禁用它。---注释---注释---注意，如果你有PRADMA设置ZCDDIS= OFF，ZCD将被启用，而不管ZCDCON注册器中ZCD允许位的状态。如果ZCD是启用的，RB0不能用作模拟输入引脚。这里是主程序（）在我的测试程序中。输出数个值。7月14日2018在16:06:19由XC8版本1450 CPU= PIC16F1718配置设置为ZCDDIS＝on（即ZCD被禁用）。10K锅通过1K电阻器连接到RB。0、将每次通过循环切换ZCD1CONBITS.ZCCDN。注意-注释-***********如果在配置设置中没有ACDDIS= on，它将忽略ZCDEN，ZCD将永远在ON上。如果ZCD是ON，则无法读取ZCD引脚上的模拟值。真的。时期。完全停止。***********************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************1015和lt＝＝ZCD禁用1×238和lt；＝＝ZCD使0 1008和lt；＝＝＝ZCD禁用1 237和lt；＝＝＝ZCD。通过循环的前两次是在中间的某个锅中。然后通过循环将锅设置为低值几次，接下来的两个值为高值。鲁迅：RB0为这个芯片工作，如果（如果只有）ZCD是一个模拟输入。完全停止。如果你不想创建一个简单的小程序，显示RB0不能作为模拟输入，那么，就这样吧。我不会再张贴，试图说服你。在我看来，广泛推广而不支持它对社区造成了伤害。但这只是一个观点。我的观点（就像我提到的另一个线程）是试图消除其他用户对整个家庭设备中的未被证实的bug的恐惧。所谓的bug肯定不存在于我所测试的设备。关于，戴夫（Over and Out.）



      以下为原文

   
 



+1 to you for that.
 
To the original poster:
In a previous thread you made a sweeping generalization that declared bugs in an entire family of devices and adamantely resisted showing us code for even one device.  Some of us showed outputs for particular devices that refuted your sweeping generalization.  I believed then that your problem was caused by failure to understand that the logic state of the global pull-up enable and the state of indiviual  port/pin pull-up enables influence the pull-up in a different manner.  (0 for global enable bit, 1 for individual enable bits) and you were trying to read analog inputs with pull-ups enabled.  I can't prove my belief, since we never saw any code from you.  You just couldn't be bothered.

Now you again declare a bug in a device and postulate that there must be undocumented bugs in an entire family of "similar" devices.  Since you haven't shown any code, we can't verify it or refute it.

Well, I dispute your general claim about "similar" devices.  I will show an example for a particular device that shows RB0 (the ZCD pin) can be used as an analog input pin if (and only if) the ZCD module is disabled, as q upside-down-h b has mentioned (twice).

I don't have a PIC16F1713 to play with.  The closest I have is a PIC16F1718, which is, I think about as closely "similar" as you can get.  (Refer to the data sheets for these.)

I connected a 10K pot between Vdd and Vss. I connected the wiper through a 1K resistor to RB0, and ran a program that showed ADC readings with the ZCD enabled and disabled.  I enabled it by setting the ZCDEN bit in the ZCDCON register to 1.  I disabled it by clearing the ZCDEN bit to zero.

        ---NOTE---NOTE---NOTE---
Note that if you have the pragma to set ZCDDIS = OFF, the ZCD will be enabled regardless of the state of ZCD enable bit in the ZCDCON regiater.  If ZCD is enabled, RB0 can not be used as an analog input pin.

Here's main() in my test program.
// PIC16F1718 Configuration Bit Settings

// 'C' source line config statements

// CONFIG1
#pragma config FOSC = INTOSC    // Oscillator Selection Bits (INTOSC oscillator: I/O function on CLKIN pin)
#pragma config WDTE = OFF       // Watchdog Timer Enable (WDT disabled)
#pragma config PWRTE = ON       // Power-up Timer Enable (PWRT enabled)
#pragma config MCLRE = ON       // MCLR Pin Function Select (MCLR/VPP pin function is MCLR)
#pragma config CP = OFF         // Flash Program Memory Code Protection (Program memory code protection is disabled)
#pragma config BOREN = ON       // Brown-out Reset Enable (Brown-out Reset enabled)
#pragma config CLKOUTEN = OFF   // Clock Out Enable (CLKOUT function is disabled. I/O or oscillator function on the CLKOUT pin)
#pragma config IESO = ON        // Internal/External Switchover Mode (Internal/External Switchover Mode is enabled)
#pragma config FCMEN = ON       // Fail-Safe Clock Monitor Enable (Fail-Safe Clock Monitor is enabled)

// CONFIG2
#pragma config WRT = OFF        // Flash Memory Self-Write Protection (Write protection off)
#pragma config PPS1WAY = OFF    // Peripheral Pin Select one-way control (The PPSLOCK bit can be set and cleared repeatedly by software)
#pragma config ZCDDIS = ON      // Zero-cross detect disable (Zero-cross detect circuit is disabled at POR and can be enabled with ZCDSEN bit.)
#pragma config PLLEN = OFF      // Phase Lock Loop enable (4x PLL is enabled when software sets the SPLLEN bit)
#pragma config STVREN = ON      // Stack Overflow/Underflow Reset Enable (Stack Overflow or Underflow will cause a Reset)
#pragma config BORV = LO        // Brown-out Reset Voltage Selection (Brown-out Reset Voltage (Vbor), low trip point selected.)
#pragma config LPBOR = OFF      // Low-Power Brown Out Reset (Low-Power BOR is disabled)
#pragma config LVP = ON         // Low-Voltage Programming Enable (Low-voltage programming enabled)

// #pragma config statements should precede project file includes.
// Use project enums instead of #define for ON and OFF.


#include
#include "System.h"

// CHS bits for ADCON0: Register 21-1 in the data sheet
#define channel_RB0 0b01100 /* AN12/RB0 */

void main()
{
    uint16_t rb0_value;
    
    // Initialize clock, I/O, uart, adc...
    init_system();
    printf("nCompiled on %s at %s by XC8 version %unn",
            __DATE__, __TIME__, __XC8_VERSION);
    printf("              CPU = PIC16F1718n");
    printf("Configuration setting is ZCDDIS = ON (i.e. ZCD is disabled).n");
    printf("10K Pot is connected through a 1K resistor to RB0.n");
    printf("Will toggle ZCD1CONbits.ZCDEN each time through the loop.nn");
    printf("********************NOTE--NOTE--NOTE********************nn");
    printf("If you do not have ACDDIS = ON in the configurationn");
    printf("settings it will ignore ZCDEN, and ZCD will always be ON.n");
    printf("If ZCD is ON, you can't read an analog value on the ZCD pin.n");
    printf("It's that simple.  Really.  Period.  Full stop.nn");
    printf("********************************************************nn");
    printf("ZCDEN    RB0n");
    printf("------------n");
    while (1){
        ZCD1CONbits.ZCD1EN = 0; // Disabled
        rb0_value = get_adc_value(channel_RB0); //
        printf("  %u     %4u   <=== ZCD disabledn", ZCD1CONbits.ZCD1EN, rb0_value);
        ZCD1CONbits.ZCD1EN = 1; // Enabled
        __delay_ms(100); // Give it a little time to settle down
        rb0_value = get_adc_value(channel_RB0);
        printf("  %u     %4u   <=== ZCD enabledn", ZCD1CONbits.ZCD1EN, rb0_value);
        printf("n");
        LED1 = ~LED1;
        __delay_ms(5000);
    }
}



Output for a few values.
Compiled on Jul 14 2018 at 16:06:19 by XC8 version 1450

              CPU = PIC16F1718
Configuration setting is ZCDDIS = ON (i.e. ZCD is disabled).
10K Pot is connected through a 1K resistor to RB0.
Will toggle ZCD1CONbits.ZCDEN each time through the loop.

********************NOTE--NOTE--NOTE********************

If you do not have ACDDIS = ON in the configuration
settings it will ignore ZCDEN, and ZCD will always be ON.
If ZCD is ON, you can't read an analog value on the ZCD pin.
It's that simple.  Really.  Period.  Full stop.

********************************************************

ZCDEN    RB0
------------
  0      543   <=== ZCD disabled
  1      231   <=== ZCD enabled

  0      543   <=== ZCD disabled
  1      231   <=== ZCD enabled

  0       36   <=== ZCD disabled
  1      232   <=== ZCD enabled

  0       38   <=== ZCD disabled
  1      232   <=== ZCD enabled

  0     1015   <=== ZCD disabled
  1      238   <=== ZCD enabled

  0     1008   <=== ZCD disabled
  1      237   <=== ZCD enabled

The first two times through the loop are for the Pot somewhere in the middle.
Then I set the pot to a low value for a couple of times through the loop, and the next two are for a high value.
 
Conclusion: RB0 works for this chip as an analog input if (and only if) the ZCD is disabled.

Period.  Full stop.

If you can't be bothered to create a simple small program that shows RB0 does not work as an analog input, well, so be it.  I won't post again to try to convince you.  In my opinion, making a broad generalization without backing it up is doing a disservice to the community.  But that's only an opinion.

My point here (as it was in the other thread I referred to) is to try to dispell fear that other users might have about your alleged undocumented bug in an entire family of devices.  That alleged bug certainly does not exist for the device I tested.


Regards,

Dave
(Over and out.)

先生，我对ZCD是否工作、AN12工作或是一些完全无关的部分称为“1718工程”或在施加电源时在附近产生一个巨大黑洞没有兴趣。你喜欢读这篇文章的头衔吗？我试图找出PWMS与输出PPS系统在1713之间的关系是因为它在数据表中不是完全和仔细记录的，因为与这两个系统相关的事情可能与我所面临的问题有关，并且因为它不那么不光彩。如果我认为这都是由于我对文件丢失的误解，而不是我建议一个普通的端口引脚随机地发射大的故障，并且我试图保持大功率晶体管完全在程序控制下的尝试。另一位先生只是评论，因为我碰巧在这个芯片上使用了一个已知错误文档的例子，作为前提，如果文档中已经存在“已知的问题”，那么更有可能存在更多的问题。请试着把你的评论和例子放在手边的话题上，避免这些不必要的广告，因为它们完全不合适。（顺便说一下，到目前为止，我估计的每一个16F+ 4位数字至少有一个严重的未记录的“bug”，所以肯定不是很少见。）



      以下为原文

    Sir, I have ZERO INTEREST in whether ZCD works, AN12 works or whether some completely irrelevant part called a 1718 works or creates a gigantic black hole in the immediate vicinity when power is applied. Did you care to read the title of the post? I'm trying to find out what the relationship between the PWMs and the output PPS system on a 1713 is because it's not completely and carefully documented in the data sheet and because things relating to these two systems COULD be related to problems that I'm having, and because it's less dishonoring to this community if I suggest that this is all due to my misunderstanding of missing documentation than for me to suggest that an ordinary PORT pin is randomly emitting large glitches and defying my attempts to keep large power transistors entirely under programmatic control. The other gentleman was merely commenting because I happened to use an example of known faulty documentation ON THIS VERY CHIP as an entree to the premise that if there's already "known problems" in the docs then it's somewhat more likely that more exist. Please try and restrain your comments and examples to the topics at hand and avoid these unnecessary ad hominem screeds as they are entirely out of place here. (And by the way every 16F + 4 digit part I've evaluated so far has at least one serious undocumented "bug" so it's certainly not as if they're rare.)

我喜欢黑洞，尤其是在地球上。



      以下为原文

    I love black holes, especially on earth

Lawton先生，你在后8章中讨论了RB0的部分内容。然后KW7X发布了一个精确的描述，示例代码显示了您在POST 8中的语句是如何不被告知的。你随后的帖子（10和12）看起来相当敌对，似乎是一个老练的行为的例子。这就是你想要的吗？



      以下为原文

   
Mr. Lawton,
You made the configuration of RB0 part of this discussion in post #8. Then Mr. KW7X posted an accurate description with example code that show just how uninformed your statements in post #8 are. Your subsequent posts (#10 and #12) do seem quite hostile and appear to be an example of trollish behavior. Is this what you intend?

他的“例子”甚至不使用相同的芯片！我并没有对ZCD感兴趣，显然它在芯片上的存在，我使用降级AN12到不可用性的程度，但是他的“示例”没有显示出我正在努力实现的目标。如果你想用他不相关的例子来“证明”某件事，那么你的先生已经把“无知”和“呆板的行为”提高到了新的高度。天哪先生，你连英语都看不懂吗？？



      以下为原文

    His "example" doesn't even use the same chip! I'm not and never was interested in using ZCD, and apparently its presence on the chip I'm using degrades AN12 to the point of unusability but his "example" demonstrtates NOTHING towards the goal I'm trying to achieve. If you're attempting to use his irrelevant example to "prove" something then you sir have raised both "uninformed" and "trollish behavior" to new heights. My god sir can't you even read english??

我不知道你对“尽快”的想法是什么，当你单步执行的时候是可以的，但是当你运行时它可能不好，这可能是一个编码问题WRT计时（当然你不会给我们提供任何细节）。你有那个端口的SLRCON寄存器设置为，快还是慢？



      以下为原文

   
I don't know what your idea of "as soon as" is, but if it's ok when you single step but not ok when you run it's likely a coding issue wrt timing (that of course you won't provide us any details on).
 
What do you have the SLRCON register for that port set to, fast or slow?

你去吧，粗鲁的人总是让人感到很难受。从各种帖子的内容中，我只能得出结论，您从未编写过一行程序代码，否则您将发布示例代码来说明您的问题/索赔。你似乎并没有真正想要帮助任何真正的问题。PSOT是一个完整的例子，说明你提出的问题，并证明我错了。



      以下为原文

   
There you go, being rude has always gotten people so far.  From the content of your various posts I can only conclude that you have never actually written a single line of program code otherwise you would post example code that illustrates your issues/claims. You do not appear to actually desire help with any real issue. Psot a complete example that illustrates the issues you claim and prove me wrong.

好的，我承认我没有使用我想要使用的电源设备的所有知识。我只是想用我买的一些帽子来“驱除”这些司机，效果是不可测量的。我以前做了一些“关断定时”测量与27伏电源和35欧姆功率电阻器负载（这在桥与P沟道逻辑电平栅MOSFET的高侧和大致互补的N沟道的低侧）。事情是门门槛越低，任何电力设备关闭的时间越长，而我没有使用“有源钳位”型驱动器，因为我不认为我需要速度，并且在没有“刚性”负载的情况下，电源设备就像一个“采样和保持”，您可以是WAI。停在永远的旁边，设备就不会自行关闭！有一些很好的隔离栅极驱动器IC，但那些似乎没有被调用（他们不给你的“有源钳位”关断），另一方面，不隔离的设备，钳位最多顶出约12伏电源（我现在可能可以与现在我采取的更长的视野）。我把测试负载放在桥上，它会关闭，而且我以前的代码工作得像预期一样，但是我不能总是像一个巨大的负载那样加热这个地方。必须设计一个箝位栅极驱动器，无论是在27伏电源或12的ICS离散，我的问题将消失。很抱歉把你们这些人带到这里，但这只是我最近的经验，MCU一直表现出“无证行为”，但这次不是。



      以下为原文

    OK I'm gonna admit I went into this one without having all the requisite knowledge of the power devices I was trying to use. I had just tried to "deglitch" these drivers with some caps I bought and the effect wasn't measurable. I had previously done some "shutoff timing" measurements with a 27 volt supply and 35 ohm power resistor load (this on a bridge with P channel logic level gate MOSFETS for the high side and roughly complementary N channel for the low side). Thing is the lower the gate threshold the longer it takes for any power device to shut down, and I wasn't using an "active clamp" type driver because I didn't figure I needed the speed, and in the absence of a "stiff" load the power device is just about like a "sample and hold" and you could be waiting next to forever and the device just won't shut down on its own! There's some nice ISOLATED gate driver ICs but those didn't seem to be called for (and they don't give you the "active clamp" for shutoff either), on the other hand the non-isolated devices that DO clamp mostly top out at about a 12 volt supply (which I could probably live with now that I take the longer view). I put the test load across the bridge and it would shut down fine and my previous code worked as expected but I can't just have a big load like that heating the place up all the time. Gotta design a clamping gate driver whether it's with discretes at a 27 volt source or ICs at 12 and my problem will go away. Sorry to bring the rest of you into this but it's just in my recent experience the MCUs have been exhibiting "undocumented behavior" a lot but just not THIS time.

我不必张贴任何东西，阅读我的另一个答复，一旦我彻底诊断出问题，我的原始代码工作完全一样的预期！（除了它是专有的），我开始在这里发怒了，我更喜欢在汇编中写作，因为我不喜欢探索MCHP的编译器错误，但是如果我不为C读者写下所有的东西，那是因为你太过新手了，不知道如何用C写。这是泔水，我已经编写和记录了许多C项目到最高安全水平由DO-178C和控制发动机的最新战斗机。（现在，如果你写了一些5位数的16F设备，你必须使用C，因为有部分指令集甚至不工作，而且编译器的版本是“噱头”来修复它-这就是为什么我拒绝使用它们。）我敢打赌没有人告诉你，但他们几乎不做。在开始一个项目之前我做了大量的研究，但是你不能接受，但别介意我离开这里！



      以下为原文

    I don't have to post ANYTHING, read my other reply, once I thoroughly diagnosed the problem my original code worked exactly as expected! (Besides it's proprietary.)
 
I'm starting to get REALLY annoyed posting here, I prefer to write in assembly because I don't enjoy exploring MCHP's compiler bugs but if I don't write everything in C for the benefit of THIS audience "it's because you're too much of a newbie to know how to write in C". That's hogwash and I've written and documented many C projects to the highest safety levels provided by DO-178C and for control of the engines in the latest fighter aircraft as well. (Now if you write for some of the "5 digit" 16F devices you HAVE to use C because there are parts of the instruction set that don't even work - and that version of the compiler is "gimmicked" to fix it - which is why I refuse to use them.) I'll bet nobody told you that but they don't do nearly the amount of research I do before I start a project but YOU can;t accept that but never mind I'm out of here!

我打算建议引脚上可能有电容，这将在一段相当长的时间内保持MOSFET栅极。没有显示一个示意图或代码，就不可能分析问题并提供可能的解决方案。MOSFET几乎总是需要由某种栅极驱动器或等效电路驱动。



      以下为原文

    I was going to propose that there might be capacitance on the pin, which would hold a MOSFET gate ON for a considerable period of time. Not having been shown a schematic or code makes it impossible to analyze the problem and provide possible solutions. MOSFETs almost always need to be driven by some sort of gate driver or equivalent circuit.