打印

今天遇到鬼了...

[复制链接]
6780|44
手机看帖
扫描二维码
随时随地手机跟帖
跳转到指定楼层
楼主
两块很小的4层PCB通过贴片接插件叠合的,追求紧凑所以4面都全部贴满了元件.硬件软件都是俺弄的.
是个电机驱动电路.之前没有发生过这样现象,现在是修改了软件中上电后电机发滴滴声后出现的,俺当然不会白痴到改成驱动短路.这个鬼是啥呢?就是不接电机时候,PCB板居然能发出和电机一样的滴滴声,很清晰.这要推广的话,以后连蜂鸣器都可以省了.板子厚1.0,大小才4*5厘米左右,先把两块板的夹心铝板撤了,还有声音,把电解电容撤了,还有声音,很多的元件撤了只剩跟电机驱动有关的,还有声音.现在板子上也就单片机,贴片电容电阻,MOS管,LDO,居然还有声音,只是音量很小了,把声音频率改1倍后它也可以发出对应的,服了,发声音就要震动,现在俺觉得都把机械上所有的条件全给他排除了,但是就是有声音,加上夹心铝板就把声音放大了.叫了12个人去听,居然没一个听的出,有的甚至开始没撤铝板前很清晰声音都听不出,服了那些Y们,听觉太不敏锐,俺3米外就听到还能分辨是哪一个板子.现在的声音也就俺一个人能听到了,背对着俺就知道他们上电没上电,上电了就有滴滴声.
实在没辙,,,光一片PCB加些贴片元件能发出预期的声音,俺去申请个专利把蜂鸣器废了.哪位老大能解释下灵异现象不?  咋抓鬼?

相关帖子

沙发
chunyang| | 2009-8-24 10:54 | 只看该作者
能发声的元件不少,特别是电感,有的电容也可以,甚至包括PCB,关键是你的电路中存在尖峰脉冲。

使用特权

评论回复
板凳
cjx2009| | 2009-8-24 11:04 | 只看该作者
嘿嘿

使用特权

评论回复
地板
huangqi412|  楼主 | 2009-8-24 11:08 | 只看该作者
电感肯定是没有一个.  电解电容之类都去了,只有普通铁片电容应该不会吧,  所有的剩下的元件俺觉得不太可能能够振动了       PCB板,有可能

CHUNYANG 给稍点拨下这方面知识.

使用特权

评论回复
5
chunyang| | 2009-8-24 11:13 | 只看该作者
帖片电容都是独石叠层结构,如果品质不佳,会遇到发声情况的,而你的PCB及其接插方式当然也是有可能的。

使用特权

评论回复
6
aweyfan| | 2009-8-24 11:15 | 只看该作者
当彩色的声音尝起来是甜的


电池也能发声,遇到过

使用特权

评论回复
7
frienz| | 2009-8-24 12:15 | 只看该作者
不稀奇,不稀奇。从LZ讲述的内容来看,贴片瓷片电容嫌疑很大。

把贴片电容都吹掉,看看声音还有没有?

使用特权

评论回复
8
frienz| | 2009-8-24 12:24 | 只看该作者
转载一篇**,看看贴片瓷片电容如何“唱歌”的(压电效应)

2006-08 Arrow Asian Times Article

KEMET Electronics Corp--PIEZOELECTRIC EFFECTS CERAMIC CHIP CAPACITORS
(Singing Capacitors)
Most dielectrics of ceramic capacitors exhibit a characteristic identified as piezoelectric effects than can cause unexpected signals in certain circuits. In some cases, the piezoelectric effect may result in the appearance of electrical noise, while in other cases; an acoustic sound may be heard, emanating from the capacitor itself.
The basic element in most MLCCs is barium-titanate, or some close derivative of this. Piezoelectric properties are common to the barium-titanate structure. If you are old enough to recall before CDs, acoustic recordings were mostly sold on vinyl records and the signals were picked up using a needle that was contained in a spiraling groove cut in the record’s surface. Within these grooves, the needle would ride on roughened surfaces that created mechanical vibrations of the needle. The needle was connected to a crystal structure (in less expensive turntables), and this crystal would generate an electrical signal that correlated with the vibration’s frequency and magnitude, which was then amplified to generate the speaker signals. What was the crystal structure of these ceramic cartridges? They were based on barium-titanate!

Piezoelectric effects can result in noise for ferroelectric ceramic chips, such as those of the middle to high dielectric constants like X5R, X7R, X8R, Y5V, Y5U, Z5U, etc. Piezoelectricity occurs in all ferroelectric dielectrics, regardless of manufacturer, and the means to reduce these effects usually requires the dielectric constant be lowered (the capacitance capability is also lowered with the lower dielectric constant) while moving to higher dielectrics (Y5V, Z5V are usually cheaper) creates a higher susceptibility to this effect. Note that
there are no measurable piezoelectric effects in Class 1 capacitors, such as C0G or NP0 - neither of which is considered ferroelectric.

Historically, the piezoelectric noise has been only an occasional issue, since it was at such a low level. It can occur as a mechanically induced electrical noise or it can occur as an electrically induced mechanical noise (this is what a speaker or buzzer does). If a capacitor is surface-mounted on a PCB, there is a direct mechanical connection between the board and the capacitor. Vibrations created on the board can create electrical signals within the capacitor. Electrical signals in the capacitor can create mechanical vibrations of the
board. Multiple chips located in a specific area can flex the board, creating a larger speaker area.

These created signals can be problematic enough, but the translation from electrical to mechanical, then mechanical back to electrical both involve delays and can create a slight echo or distortion effect in the circuit. The peak response of these capacitors is within and slightly above the audio frequency range. Many attempts to use these SMT devices in final stages of audio amplifiers have left many designers scrambling for alternatives, yet they can be used in the front-end stages because the signal levels are lower.

More recently, power integrity circuits involved with microprocessor decoupling have been experiencing this effect in ceramic capacitors. The microprocessor ‘sleep’ mode involves removing the voltage from the power rails on an intermittent basis, and periodically checking for any request for activity. The typical frequency for this sleep mode is at 1 kHz, and designers are experiencing a 1 kHz tone emanating from the processor region
of the computer when the processor goes into its sleep mode.

This problem is becoming more noticeable because the MLC capacitors are evolving and their applications are expanding. This effect is related to the signal strength, which is related to the electrical stress within the capacitor. As the dielectrics are shrinking to enable higher capacitance with these packages, the critical stress levels are being reached at lower voltage levels. Solutions to these noise problems may involve alternative types of capacitors (e.g., tantalum, aluminum, tapolymer, al-polymer, film, as this effect is unique to ceramic), leaded or standoff capacitors (using leadframes to eliminate the mechanical tie to the PCB eliminates a unified or cumulative response of multiple ceramic capacitors), using higher voltage ratings (lowers the stress), and others. It should be pointed out that this effect is lost at frequencies well above 30 kHz because the body cannot respond fast enough to the changing stress levels. The peak response region and the noise attributes dictates that these capacitors should be used very carefully in audio circuits, as well as high-gain circuits; and, be careful that an audio application may not easily appear to be an audio circuit (sleep mode conditions of microprocessors).

John D. Prymak – KEMET Applications Manager

使用特权

评论回复
9
gx_huang| | 2009-8-24 12:33 | 只看该作者
换成钽电容试试。

使用特权

评论回复
10
pigjiang| | 2009-8-24 12:37 | 只看该作者
我的电路板经常发出声音,同事听见了跑过来问我怎么回事,我居然没听见,死活没听见。
难道是我的耳朵频带范围太窄了?

使用特权

评论回复
11
sinanjj| | 2009-8-24 12:39 | 只看该作者
顶ls的。。。


电工就是需要想像力啊

使用特权

评论回复
12
vwwj| | 2009-8-24 12:41 | 只看该作者
15000Hz

声音.jpg (399.54 KB )

声音.jpg

使用特权

评论回复
13
maychang| | 2009-8-24 12:49 | 只看该作者
不知道楼主和你的同事都是多大岁数。
我在20岁时测过,可以听到18kHz,一个朋友可以听到17kHz,而他的女友可以听到20kHz。
现在我连10kHz也听不到。

使用特权

评论回复
14
boy4477| | 2009-8-24 12:57 | 只看该作者
上高中可以听到电视机的行频。
现在听不到了。

使用特权

评论回复
15
chunyang| | 2009-8-24 12:59 | 只看该作者
10KHz都听不到说明老maychang的耳朵提前老化了,应该70岁以上才到这个地步啊,老t都还没到这个岁数。
想测测自己的听音范围,可以找个软件信号发生器,基于声卡的“虚拟仪器”,然后用品质良好的音箱或耳机听,哪天俺也试试。

使用特权

评论回复
16
PowerAnts| | 2009-8-24 13:13 | 只看该作者
俺现在的听力上限是12.7KHz ,15625Hz的行频早在十年前就听不到了, 小时候倒是对黑白电视机行输出的叫声十分的敏感

使用特权

评论回复
17
PowerAnts| | 2009-8-24 13:32 | 只看该作者
想必,因为大电流的原因,铜箔比较宽,层间电容比较大,当施加大幅度的方波电压时,静电力场引起的振动,含有丰富的谐波成份,声音听起来就比较悦耳

使用特权

评论回复
18
ic2ic| | 2009-8-24 14:06 | 只看该作者
支持了~~

使用特权

评论回复
19
sinanjj| | 2009-8-24 14:30 | 只看该作者
用耳朵测频。。。。。

使用特权

评论回复
20
maychang| | 2009-8-24 14:30 | 只看该作者
回16楼chunyang:
多谢关心!
不止是听力的频率范围减退,现在左右两耳灵敏度也相差很远,接电话时左耳听到的声音明显比右耳小。
更糟糕的是视力。今年春节过后做了白内障手术,换了人工晶体。没做手术之前过马路都困难,看不清是否有汽车正在驶过来。计算机的屏幕戴上眼镜勉强可以看清。现在看屏幕也要戴眼镜。
人老了,各器官衰老有先后,各人不同。有的人四十多就心梗,有的人四十多已经秃顶,有的人三十多牙就全掉光了。我还算好,心脏**脏肝脏都没有什么病,主要是眼睛不行。
老tyw好像身体比较好。

使用特权

评论回复
发新帖 我要提问
您需要登录后才可以回帖 登录 | 注册

本版积分规则

622

主题

24960

帖子

18

粉丝