我必须使用什么值电阻来限流？

通常，对于220VAC，I系列使用2x470K 1／2W。理论上，一个1M 1/2也可以工作，但是你没有那么孤立……我实际上从未使用过“真正的”ZCD模块，而只是一个标准输入引脚：但是我想电子规格应该是SAMP。PS：有时，输入引脚上的10NF电容有助于“稳定性”。



      以下为原文

    Usually, for 220VAC I used to use 2x 470K 1/2W in series. In theory one single 1M 1/2 works too, but you have less isolation then...
 
I actually never used the "true" ZCD module, rather just a standard input pin: but I suppose electrical specs should be the same.
 
PS: at times, a 10nF capacitor at the input pin helps "stability"

R＝峰值V/300 uA.SEEHTTP://MyCHIPCDeave/COM/8BIT：ZCDAND HTTP:/WW1.MICROCHIP.COM/DeLoSs/En/AppNOTES/90003138A.PDF



      以下为原文

    R= Peak V/300uA.
See http://microchipdeveloper.com/8bit:zcd and http://ww1.microchip.com/downloads/en/AppNotes/90003138A.pdf
 

啊，是的，在某些情况下，我添加了一个偏置电阻…这要看情况而定。



      以下为原文

    Ah yep, in some cases I added a Bias resistor... it depends.

很抱歉很苛刻，但是如果你要问这些问题，你就不应该靠近电源电压。



      以下为原文

    Sorry to be harsh, but if you have to ask these sort of questions, you shouldn't be going anywhere near mains voltages.
 

我计算的像（220 - 5.75 / 0.0003）＝7xx欧姆。



      以下为原文

    I was calculating like
 
(220 - 5.75 / 0.0003) = 7xx Ohms.

（220—5.75／0.0003）＝-18946.666…



      以下为原文

   
(220 - 5.75 / 0.0003) = -18946.666...

（220 - 5.75 / 0.0003）＝-18946.666…抱歉，是（220～5.75）/0.0003。



      以下为原文

   
(220 - 5.75 / 0.0003) = -18946.666...




Sorry, It was 
 
(220 - 5.75) / 0.0003

我总是用括号来消除歧义，我还没有记住优先顺序。



      以下为原文

    1and0, nice catch.
I try to always use parentheses to remove any ambiguity, I have yet to memorize the order of precedence.
 

Tryr =（220V*SqRT（2））/ 300 UA＝1.04 MOHM &编辑；听Qub。要非常小心。你在玩火！



      以下为原文

   
Try
 
R = (220V * sqrt(2)) / 300 uA = 1.04 Mohm
 
Listen to Qub. Be very careful. You're playing with fire!

也要检查电阻（S）和PCB轨道和焊盘间距的额定电压。我通常使用0805个SMD电阻器，通常额定200-300伏，因此需要串联两个。这将是一个好主意，使用两个电阻器，一个在240个VAC源的每一个腿上，大约470K，因此提供“触摸安全”的低电流限制从电源到电路板的其余部分。一般来说，小于1毫安的任何东西被认为是安全的，甚至低于大多数人的检测阈值。



      以下为原文

    Also be sure to check the voltage rating of the resistor(s) and the PCB track and pad spacing. I usually use 0805 SMD resistors which are typically rated 200-300 Volts, so two in series would be needed. It would be a good idea to use two resistors, one on each leg of the 240 VAC source, about 470K, and thus provide "touch safe" low current limiting from the mains to the rest of the board. Generally anything less than 1 mA is considered safe, and is even below the threshold of detection by most people.

小心，伙计们，你真的需要尊重230VAC电，否则它会咬你！微芯片的APP注释也有点混淆…标称电源是230VAC，具有10%的容差，最大值为254V RMS，大约为+/-360V峰值。在这个基础上，你需要1.2MOHM来限制峰值电流到正负300 U。标准的做法是使用两个（或多个）串联电阻，每个额定高电压，所以如果一个故障，电路（和你触摸它）将幸存，因为电流仍然有限。我一直使用。这种应用的特殊电阻器VISHY VR25或VR27或更近的波恩CHV表面贴装部件（800 V为1206尺寸）。实际产品的标准预期在电源上至少有2.5kV尖峰，这将基于实际情况，所以它不是RMS或峰值交流电压的问题。在输入到芯片上的某种形式的电压限制器将是明智的——MOV、电视甚至Zener diode将停止尖峰-但是检查它不会影响过零点性能。这对于多年使用1M电阻器的爱好者来说是太大了，而且从来没有任何麻烦，但是请意识到风险。它负责现场输入——这使得中性点连接到你的电路，而它经常靠近地面，这是不能保证的。在大多数情况下，仅仅通过电阻连接现场将工作，因为有足够的返回路径通过你的电路的0V，这可能是在地电位。



      以下为原文

     
Careful, guys, you really need to treat 230Vac electricity with respect or it can bite you!
Microchip's app notes are a bit confusing as well...
Nominal mains is 230Vac with a 10% tolerance so 254V rms maximum which is about +/- 360V peak. On that basis you would need 1.2MOhm to limit the peak current to plus and minus 300uA.
Standard practice is to use two (or more) resistors in series, each rated for high voltage, so if one fails, the circuit (and you if you touch it) will survive because current is still limited.
I have always used special resistors for this application - Vishay VR25 or VR27 or more recently Bourns CHV surface mount parts (800V for 1206 size). The standards for real products expect at least 2.5kV spikes on the mains and this will be based on real-life situations so it isn't a matter of just the rms or peak AC voltages. Some form of voltage clipper at the input to the chip would be wise - an MOV, TVS or even Zener diode will stop the spikes - but check that it doesn't affect the zero-crossing performance.
 
This may be too much for hobbyists who have used a 1M resistor for years and never had any bother but please be aware of the risks.
 
That takes care of the Live input - which leaves the Neutral to be connected to your circuit, and while it is often close to Ground, that isn't guaranteed. In most cases just connecting the Live via the resistors will work as there is an adequate return path via the 0V of your circuit which is probably at Ground potential.
 
 
 

也请阅读段落22.5.1。这提到使用电容器代替电阻器进行零交叉检测。这应该提供某种程度的当前隔离，但我可能仍然使用X型高压盖来增加安全性。这些帽子不像现成的陶瓷那么便宜。



      以下为原文

    Also read paragraph 22.5.1.  This mentions using a capacitor instead of a resistor for zero cross detection.  This should provide some level of current isolation but I would probably still use a X style HV cap for added security. These caps are not as cheap as off the shelf ceramic. 

电容器的问题是，它们对电力线的高频分量（如尖峰和快速上升时间谐波）提出了非常低的阻抗。我建立了一个设备，使用这种电容在220 VAC线，它经历了频繁的烧毁齐纳二极管故障。我想我们发现客户使用的是一个具有高频PWM的逆变器，即使有一个附加的串联电阻，高电流浪涌也是可能的。我使用电容器是因为我需要从交流线路获得直流电源，所以即使用10毫安电流，电阻器的功耗也会超过2瓦，这是一个小封装。对于300 UA，功率小于100毫瓦，因此不需要使用电容器来限制功率。



      以下为原文

    The problem with capacitors is that they present a very low impedance to high frequency components of the power line, such as spikes and fast rise time harmonics. I built a device that used such capacitors on a 220 VAC line and it experienced frequent failure from burned-out zener diodes. I think we found out that the customer was using an inverter, which has high frequency PWM.
 
Even with an additional series resistor, high current surges are possible. I used capacitors because I needed to obtain DC power from the AC line, so even with 10 mA current the power dissipation of resistors would be more than 2 watts, and this was in a small enclosed package. For 300 uA, the power is less than 100 mW, so there is no real need to use capacitors to limit power.

脚踏点：在标称230V 50Hz的世界（即北美洲以外的大多数地方），没有“每条腿”；有中立的和230V的距离是活跃的。只有在奇怪的世界120赫兹60Hz，看到一个240V的供应与中心抽头绑中立。



      以下为原文

   


Pedant point:
In the nominal-230V 50Hz world (i.e. most places outside North America), there is no "each leg"; there is Neutral and 230V away there is Active. It's only in the Strange world of 120V 60Hz that one sees a 240V supply with centre tap tied to Neutral.

是的，但是除非你使用一个极化插头，并且可以确定哪一个连接是“热”到地上，最好在两者上使用电阻器。由于220/380个系统中的热线具有320伏特的正常峰值电压，因此最好在每个支路中使用两个电阻器。



      以下为原文

    Yes, but unless you use a polarized plug and can be sure which connection is "hot" to ground, it may be better to use a resistor on both. Since the hot line in 220/380 systems has normal peak voltage of 320 volts or so, it would be best to use two resistors in each leg.

你可以做的是把一个电阻和电容器连接到一个光耦合器上。然后你可以读取光耦合器，如果你快速读取它，就可以验证电源和电源的频率（因为它会随着电源频率的变化而变假），同时光耦合器也会保护你的低压C。我把它放进了Git Hub中，因为我把它写到胶乳中（因为它有数学）。http://Github.com／RoBi4…TiVyMixSnIsIds2.PDF



      以下为原文

    What you can do is connect a resistor and capacitor to an opto coupler.
You can then read the opto coupler, and if you read it fast enough
verify the mains and the mains frequency (because it will go TRUE and FALSE
in time with the mains frequency).
Also the opto coupler will protect your low voltage circuits from the mains.
 
I put this in git hub because I wrote it in LaTeX (because it has maths).
https://github.com/robin4...tive_mains_inputs2.pdf
 
 

带有电阻的电容器将引入一些相移，这对于真正的过零检测器来说可能是一个问题。此外，在大多数光电耦合器中的LED需要至少2毫安的可靠和快速的操作，这是大约500兆瓦的电阻器。您还需要一个高电压串联二极管（它使用1/2的电源，但具有更高的电压降），或二极管跨越LED，以避免反向偏置。串联二极管可能不适用于串联电容器。我为SCR触发电路做了一个过零点检测器，其中精确的过零检测是重要的。它创建自己的电源，并使用晶体管来驱动施密特触发器光耦合器，它提供了一个干净和快速方波到PIC。缺点是它的功率大约为10毫安或2瓦。



      以下为原文

    A capacitor with a resistor will introduce some phase shift, which may be a problem for a true zero-crossing detector. Also, the LED in most opto-couplers requires at least 2 mA for reliable and fast operation, which is about 500 mW for a resistor. You also need either a high voltage series diode (which uses 1/2 the power but with higher voltage drop), or a diode across the LED, to avoid reverse biasing it. The series diode may not work with a series capacitor.
 
I made a zero crossing detector for an SCR trigger circuit where accurate zero crossing detection is important. It creates its own power supply and uses a transistor to drive a Schmitt trigger opto-coupler which provides a clean and fast square wave to the PIC. Disadvantage is that it draws about 10 mA or 2 watts.
 

代替电阻，你可以使用一个电流限制电路，它能更快地接通和断开。下面的模拟结果表明，相对于一个电阻和二极管，你只能得到大约4度的相位延迟。对于120 VAC 60 Hz的输入，同一电路对于有源限幅电路具有8.5度的相位延迟，而对于电阻器和二极管则只有24度。



      以下为原文

    Instead of a resistor, you could use a current limiting circuit which turns on and off faster. The following simulation shows that you can get a phase delay of only about 4 degrees, compared to 12 degrees for just a resistor and diode.
 

For a 120 VAC 60 Hz input, the same circuit has a phase delay of 8.5 degrees for the active limit circuit and 24 degrees for just the resistor and diode.

你可以通过增加*R5来减少相位延迟（至少根据LTSpice）。例如，增加到10K使1.43DEG（~66秒下降到0.8V）延迟到60Hz的5P负载。



      以下为原文

    You can reduce the phase delay by *increasing* R5 (at least according to LTSpice).
E.g. increasing to 10k gives ~1.43Deg. (~66us to fall to 0.8V) delay into 5p load at 60Hz.
 
Edit:- Just to add that I much prefer the Opto soultion, for my own and others safety :-)