# Operating semiconductors outside of Voltage specifications?

If I were to employ the schematic below will the transistors survive.

I ask this because I was able to operate 3 low voltage(I'm guessing 1.5v), run of the mill junk LED's in series at a voltage of 340V @ 113uA using a 3M resistor with no ill result.

Is the same true for other semiconductor devices?

I'm building an off line power supply and making a linear regulator to power the primary control ICs seems like a terribly inefficient use of power.

The schematic works in simulation using LTSpice, but will the oscillator work on the breadboard or am I setting myself up for dead transistors?

EDIT The original reason for this question was because my reasoning was that .25W of power has the same charge (number of electrons) no matter the voltage. I = dq/dt, so that is essentially .25 coulombs per second at .25W.

if I=.25 and V=1 then P = .25 (P=VI) and since Q = 1 Ampere second or 1 Ampere is 1 Coulomb per second, therefore 1V at .25W should be .25Cu/sec and even if the voltage is 340V @ .25W, it is sill only .25Cu/sec. Transistors require sufficient current to operate correctly and even MOSFETs that are voltage controlled devices wont fully turn on unless they get sufficient charge (I know from personal experience), so that is why I think they may work in simulation.

I'm still a student in AC Theory, so I'm jumping ahead a little, but I figured in reality I would see an expectable voltage drop, like I did over the LEDs. There was a 335V (5V remaining) drop over just 3 little crap LEDs running at 340V in the uA range.

In a search for a totally different topic I came across this post where in the answer the user states:

The problem is that you are expecting a model to have more functionality than it has. Most models are simply parameters in an equation. They are designed to approximate the actual device behavior within specific operating conditions. For example, a diode is modeled using the following equation:

$\Large I_f = I_S(e^{\frac{Vf}{NV_t}} - 1)$ Where Vf is the applied voltage and If is the forward current. Notice that there's nothing stopping me from putting 50V as the applied forward voltage, and I will definitely get an answer from the simulator. It will be completely nonsense, but the model is assuming that the behavior of the diode is always described by that equation.

Which combined with the answers below provides me with a complete understanding of why the 2N2222 worked in simulation but WILL NOT work in a "real world" application.

Thank you all for your assistance. I is well appreciated.

• Now to the second paragraph.. How is it related to the first one? what do you mean "If I were to use a supply .... with 470k resistors" - use how? "...would that kill a transistor.." - which transistor? This is where you can show us a schematic of what you have in mind. The other paragraphs are unclear as well, but let's do one by one... – Eugene Sh. Sep 6 '16 at 17:32
• Well that's today's story about thinking and somewhere in here is a question. – Andy aka Sep 6 '16 at 18:10
• 340VDC - on a breadboard? – Bruce Abbott Sep 6 '16 at 18:19
• It is but your line of thought is very inconcise, you jump from subject to subject, make strange assumptions. You write many sentence-in-a-sentence. Making it hard to follow and unclear what you're actually asking. Try to make short concise paragraphs and use short sentences. – Bimpelrekkie Sep 6 '16 at 19:12
• And don't use 340 V on a breadboard, they're not suited for that. Also switched converters do not work well on breadbords either because of the high contact resistance. – Bimpelrekkie Sep 6 '16 at 19:13

## 2 Answers

LEDs don't switch, so there if there is a series resistance there is never 340V across them, it's all dropped in the resistor. When a transistor switches off, effectively the whole voltage will be across it.

Having overlooked voltage ratings of a 2N2222 and 2N2907 recently, across a 100V+ DC supply, I recently got a dramatic explosion-

Why Did My NPN/PNP Voltage Regulator Experiment Explode?

• I had a similar experience with a voltage regulator that I built, where LTSpice gave me inaccurate current output on a darlington configuration of some MJE13003's. Under a low current load worked fine, I added a more demanding load and needless to say, there was a violent destruction of the transistors and load. The MJE13003's are rated for up to 700V, so if I swapped the 2222's for those, do you think they would work in the multivibrator? I figure I can add a TVS to clamp the Voltage on the oscillation output so I don't blow the NMOS gate. – iuppiter Sep 29 '16 at 22:23

works in simulation (LTSpice)

But is the simulator accurate? The 2N2222 is rated for 40V between Collector and Emitter absolute maximium, but LTSpice thinks it can handle 340V!

but will it work on the breadboard or am I setting myself up for dead transistors?

Your transistors are mostly protected by high value resistors, except for the MOSFET whose Gate could be exposed to high voltage and current (relying on built-in protection diodes to prevent breakdown is not a good idea). Also your bootstrap circuit could be putting 340V onto Q3 and Q4 with no resistor to limit the current from C5. I bet they will blow up.

EDIT:

So now you only want to know whether the oscillator will work. I don't think so, because the transistors will break down at some voltage above 40V, and the diodes will break down at >100V. Even below these critical voltages their leakage currents could be excessive. Also the transistors have a reverse Base-Emitter breakdown voltage of ~7V, so even it 'worked' the timing would be off.

In any case, operating devices outside their specifications (particularly their absolute maximum ratings) is bad engineering. Don't do it!