The Mini SSTC

迷你特斯拉线圈

Page Created: 1/15/04

创建时间：1/15/04

Updated:4/30/07

最后编辑时间：4/30/07

I built this small coil in about 1 day of work.  Its small and relatively safe.  It produces up to 7" of spark, though this is limited to my small heatsinks getting too warm.  If one was to copy the design but use larger heatsinks, the power could be much greater.  Please scroll down to the end of the page to see the finalized schematic!

我用了大约一天的时间来制作这个小线圈。它很小，而且十分安全。尽管受到我发烫的小散热片限制，它还是放出了大约7英寸（约17到18厘米——译者注）的电弧。如果有人采用这个设计并使用更大的散热器，应该就能达到更大的功率。请到页面的最下方查看最终的图纸！

Here are some pictures of the original setup.  It worked well but needed a really long antenna to work properly.  See the updates for the slightly revised version.

这是早期结构的一些照片。它工作得很好，但是为了保证正常工作，需要一个相当长的天线。看看做过少许改动的最新版本。

Schematic (not final, see end of page)

原理图（不是最终版本，请看页面的最后）

Update 1/18/04

04年1月18日更新

I removed 2 primary turns for a total of just 4 turns.  This surprisingly helped with heating problems of the MOSFETs!  I also removed the toroid and now just have a wire pointing off the edge of the coil.  The coil runs much happier now.  The heatsinks have yet to get overheated.  Also, these changes brought about 7" sparks (compared with 6" previously).  Here are some snapshots of the new setup and the various sparks this little coil can create. 

我去掉了初级线圈的两圈，因为总共只需要四圈就可以了。令人惊喜的是，这个改动大大环节了MOSFET（互补金属氧化物半导体场效应晶体管——译者注）。我还去掉了圆环（指放电顶端——译者注），只留下了一根金属丝。这时，线圈运行得更加令人兴奋。散热器也过热了。然而，这些改动得到了7英寸长的电弧（和以前的6英寸比较）。这里有一些关于新的改动和这个小线圈制造的各种电弧的照片。

Here the coil is operating with a moderate pulse width and rate.

线圈被控制在适中的脉宽（应该是指灭弧器的脉宽——译者注）。

Here the coil is running at a low pulse rate and a low duty cycle.

线圈运行在更低的脉宽和更低的占空比。

Now the pulse rate is maxed but still a low on time.

脉宽调整到最高，但是曝光时间短。

Low pulse rate, but now the duty cycle is a bit higher, notice the thickness.

低脉宽，但是占空比非常大，注意一下电弧的宽度。

The highest duty cycle at a low pulse rate produces flame like discharges that do burn things easily.

最大的占空比和低脉宽制造了像火焰一样的效果，小心不要被烧伤。

How the coil works

线圈的工作原理

Here is an overview of the schematic and how the coil operates.  In the top of the schematic we see the low voltage power supply section composed of a small transformer, 4 diodes, some caps and voltage regulators.  The 7812 gives 12V at about 1A.  This regulator should be heatsinked as it supplies the power to the demanding gate driver section.  The 7805 supplies 5V for the logic ICs.  U1 is any typical 555 running at about 200khz or so (this is not critical as long as it is running relatively fast, or about 1/2 the frequency you intend the coil to run at).  Its sole purpose is to trigger the oscillation.  There is a 100K resistor on its output that makes the 555 look "weak" to the input of the 74hc14.  In fact, its *just* strong enough to trigger the 74hc14 (which is a hex inverter).  Now, also on the input of the 74hc14 we see an antenna with a series capacitor.  When the coil starts to oscillate, this antenna picks up a signal from the TC output and feeds it right back into the system. This signal looks "strong" so it overrides the 555 (U1).  Now the coil is running from its own noise and will be perfectly in tune at all times!  Also on the input are 2 1n60 germanium diodes.  Their function is to clamp the antenna's voltage to the 5V and gnd rails so that we don't fry our 5V logic chip!  The output of the 74hc14 is fed to the inputs of an inverting and non-inverting gate driver chip.  They work in opposition creating an effective 24V peak to peak voltage across the primary of the 16T transformer.  This transformer is just a small ferrite core wound by hand and is used to supply isolated gate driver signals to each of the MOSFETs.  This transformer is known as a Gate Driver Transformer (GDT).  Q1 and Q2 make up a half-bridge.  One fet is on while the other is off.  This creates a square wave across the primary of our TC of about 85V RMS (1/2 of the voltage supply, 170VDC).  The MUR860 diodes are fast diodes there to catch any reverse current that is back flowing from the tesla resonator.  D9 and C12 provide the DC for the half-bridge. The last piece of the puzzle is U5.  Its another 555, but its running at a low pulse rate.  This 555 controls the ON/OFF period of the tesla coil by enabling/disabling the gate drivers.  The adjustment of R5 and R6 allows for a wide variety of spark outputs.  You can also tailor the component values to your own liking.  This makes for a really cool and small SSTC.  With proper heatsinking on Q1/Q2 this thing should run for a long time.  Just make sure the MOSFETs are not overheating!

这里叙述原理图和线圈的工作过程。在图片的上方，可以看到由小变压器构成的低压电源，有四个二极管，几个电容和几个线性稳压器。7812可以提供12V的电压和1A电流（散热工作搞好的话可以1.5A——译者注）。由于要给门极驱动电路供电，这个线性稳压器需要一个散热片。而7805可以为逻辑门芯片提供5V的电压。U1是一个普通的运行在大约200kHz（这不是关键所在，它不需要运行得非常快，可以把频率降低一倍）的555。它只是被用来引发振荡。这个100k的电阻用来使555的输出对74HC14的输入端来说显得更“弱”（意思是防止555的输出过于强劲而损坏逻辑芯片，实际上只要输入电压不超过5V就不会有问题——译者注）。事实上，这已经足够触发74HC14[这个敏感的反相器（即非门——译者注）]了。我们还能看到，反馈天线和74HC14的输入端之间串联了一个电容器。当线圈开始工作时，天线将会接收来自次级线圈的信号（就是电磁波——译者注）然后反馈到电路中。这个信号的强度已经足够无视掉555的输出。现在，线圈将会因为它本身的反馈，完美地谐振！值得一提的是，输入端有两个1N60锗二极管。这个结构用来将会损坏5V逻辑芯片的电压限制住。74HC14的输出接到一个反相和一个不反相的门驱动芯片（指UCC芯片——译者注）。他们将会交替在16匝的变压器的初级线圈产生一个Vpp为24V的电压。这个变压器只是一个很小的铁氧体磁环，我们可以用手绕出它，它用来将隔离的门极驱动信号送到每一个MOSFET。这个变压器称为门驱动变压器[GDT(哥蛋疼？哈哈哈。——译者注)]。Q1和Q2构成了一个半桥变换器，它们是交替导通的。这将能够在特斯拉线圈的初级线圈产生一个有效值为85V（电源电压直流170V的一半）的方波电压。两个MUR860快恢复二极管用来阻拦由谐振电路产生的反向电流。D9和C12为半桥提供直流电源。最后，你可能对U5的作用产生疑问。这是另外一个555，但是会产生更低的脉宽。这个555通过控制门驱动芯片来控制线圈的开关周期，R5和R6的调整可以在很大范围内调整输出的电弧。你还可以根据自己的喜好调整元件的参数。上述的这些内容造就了一个体积很小但效果很棒的特斯拉线圈。如果给Q1和Q2安装足够大的散热器，这个线圈将能工作很长时间。一定要防止MOSFET过热！

（理解这段话叙述的原理即可，因为这段话叙述的是不够完善的版本，这个版本要比最终版本复杂。这个版本使用了一个555来引发振荡，会和反馈天线的信号冲突，可能造成不好的影响——译者注）

More Updates:

更多的更新：

I'm adding an additional schematic that shows the latest revisions.  The largest change (and by far the coolest!) is the sub points added into the schematic around U5.  By allowing these 2 values to be changed an amazing variety of spark appearances can be had ranging from spark gap coils to a big flaming torch.  Observe the new changes

我给最新版本增加了一些附加的部分。最大的修改（这是到目前为止最酷的！）就是在子点加入了U5。改变了这两个值的结果便是，它产生了像SGTC一样的大火（实际上是等离子火焰——译者注），就像燃烧的火把。请查看最新的改动。

Revised Schematic (not final, see end of page)

修改后的原理图（不是最终版本，请看页面最下方）

Here are some substitutions I have tried so far

这是最近我刚刚试过的一些替代方案

Capacitor 电容	Resistor 电阻	Spark Appearance 电弧的表现
2.2uf	0-10k	As seen in the pictures above (this is the original configuration) 像上述的图片一样（最初的配置）
2.2uf	10Meg	Very slow rep rate allows the study of individual spark formations （没看明白这句话——译者注）
100uf	0	Allows for a pulsing flame spark, very interesting and extremely hot (I received an RF burn from touching the output for just one pulse) 一个简短的火花，非常有意思，而且极其烫[我仅仅摸了一下输出，便被烧伤了(应该不是严重烧伤，手上会出现一个黑点——译者注)]
.1uF	0-10k	Screaming tendrils of purple arcs. Noticeable ozone production.  Very similar to a small spark gap coil running at high bps.  Excellent banjo effect! 尖叫的紫色电弧。明显产生了臭氧。像一个小SGTC工作在高BPS（Burst Per Second，衡量每秒灭弧的次数的量——译者注）。非常棒的效果！
.1uF	10Meg	A nice snapping spark ranging in thickness from extremely thin to moderately bright. Similar to a single shot SGTC. 非常细的电弧。很像SGTC打火一次的效果。
1000pF	100k	Arcs similar to a flyback.  Short in length and dim until you draw the spark.  电弧很像反激变换器（开关电源的一种形式，能产生高电压——译者注）的效果。减小长度和宽度直到你找到电弧为止。

There are just so many awesome effects possible from the interrupter.  If one does not experiment, he is missing out!  I will try to get pics of each configuration later on. 

这只是灭弧器产生的奇特效果的一小部分。如果你不尝试，就会失去机会！稍后我将会得到修改配置后的照片。

1/19/04

04年1月19日

I made another change to the circuit that improves some things.  It seems the coil will work fine without U1 (the 555) even installed!  Basically, when the driver chips are enabled, the inverting chip will send a short pulse to the MOSFETs causing the oscillation to start without any problem.  The fact that there is only 1 signal going to the 74hc14 means there is no room for conflict here.  This ALSO means that the feedback is stronger... shorter antenna!  Now the coil will happily run WITH the toroid and produce up to 8" sparks now.  Here is the revised schematic:

我对电路做了一些修改，优化了某些东西。看起来，线圈即使没有U1（555）也是没有问题的！实际上，当驱动芯片被触发时，反相芯片将给MOSFET发动一个短脉冲，这个脉冲将会引发振荡二部引起任何问题。事实上，只有一个信号输入到74HC14将会避免信号之间的冲突（两个信号同时输入可能造成不大好的影响——译者注）。非常值得一提的是，这意味着我们可以使用一个更短的天线！现在，这个线圈的顶端将会令人欣喜地产生8英寸（约20到21厘米——译者注）长的电弧。这是修改后的原理图：

Revision 3 (not final, see end of page)

第三次修改（不是最终版本，请看页面最下方）

Here we add some wetted salt to the breakout point:

我们给电弧喷出的点涂抹一些盐溶液（焰色反应——译者注）：

4/30/07

07年4月30日

After having this project out for a few years I've received quite a response from many other amateurs who have copied the design successfully.  But, I very often get the same questions over and over.  In particularly "why did you series up all 6 inverter gates?".  At the time I built this coil I was still pretty new to electronics, so I didn't see the harm in it, and it was easy to wire, but now I realize the extra gates add unwanted delays, so this last version of the schematic has fixed that part.  I also reduced the value of the gate resistors... I'm tempted to say leave them out all together, but 5 ohms should be safe.  Its also possible to completely eliminate the 74HC14 and feed the antenna directly to the UCC37321/2 gate driver inputs, but this might not work as reliably for some people, so experiment at your own risk.

后来的几年里，我收到了很多仿制成功的爱好者的回复。但是，我总是被问到同一个问题：“你为什么要把6个非门全都串联起来？”当我制作这个线圈时，我还是一个新的电子爱好者，所以我没有发现它的危害，而且这是很容易就能连在一起的，但是我现在意识到，额外的门会带来不必要的延时，所以最终版本解决了这个问题。我还减小了MOSFET门极上的电阻值……我不得不说，我想把它们全部省掉，但是有了5Ω电阻应该比较安全。将74HC14省掉，直接将反馈输入到UCC37321/2似乎也是可以的，但是这样有时可能不会很可靠，所以你恐怕要冒着风险来测试这个方案。

Final Schematic (use this one to build it!)

最终的原理图（请用这一个原理图来制造线圈！）

（英语水平有限，这里面可能会出现一些小错误，但是整体应该是正确的——译者注）