Is Bipoler designed to withstand negative voltage and current?

[EngIntoHW]

I'm talking about commonly used BJTs, as 2N2222.
Can they withstand negative VCE and ICE? (for npn)

I know that MOSFETs can withstand these.

Thanks.

 

[Jony130]

Long time ago I measure reverse-bias breakdown voltage from emitter to base and emitter to collector in general purpose BJT

For PNP

BC556
Vbe=10V; Uce=10.65V

BD140
Ube=12.55V ; Uce=13.02V


And for NPN
BC337-40
Veb=8.2V; Vec=6.7V; I=5.5mA

BC549B
Veb=8.3V; Vec=7.2V; I=5.5mA

BD139-16
Veb=8.5V; Vec=6.7V; I=5.5mA

BC639
Vbe=7.7V; Vec=6.3V; I=500uA

BC337
Veb=7.9V; Vec=6,4V; I=500uA

2SC945
Veb=8.1V; Vec=7,5V; I=500uA

 

[EngIntoHW]

Hi Jony, Thanks a lot for sharing!

Were they able to still operate normally under negative current and voltage?

 

[audioguru]

Were they able to still operate normally under negative current and voltage?

Transistors have absolute maximum voltage spec's on their datasheets. The reverse-biased emitter-base junction is usually spec'd at 5V or 6V. A higher voltage (as Jony showed) causes the junction to have avalanche breakdown (like a zener diode). The avalanche breakdown causes hot spots on the emitter-base junction and since it is not designed to dissipate heat like the collector is then the transistor is slowly damaged.

It should be obvious that a transistor connected with backwards polarity works very poorly when the voltage is low enough to avoid avalanche breakdown. If its current gain is normally 200 then it is about 5 when connected backwards.

A Mosfet has a "body diode" as part of its structure that conducts when its drain-source voltage is backwards.

 

[MrAl]

Hi,

As audio guru pointed out, the transistor can easily get damaged with an out of spec reverse voltage.

This is interesting because what happens is you might not even notice the damage to the transistor after a brief reverse Vbe that is higher than its data sheet rating. The effect can be noticed sometimes in the decrease in gain but the transistor still appears to work normally with a casual test. It's only when you go to try to get the transistor working in a circuit that requires the full spec's of the transistor (like gain and frequency response) that you can notice the difference.
For example, using a transistor in an RF oscillator circuit it might work just fine, but after a brief reverse Vbe voltage a little higher than the data sheet spec and then try to use that transistor in that same RF oscillator circuit and it probably wont work. This only takes a brief reverse voltage too, maybe less than a quarter second or even less.

So what happens is the transistor does get damaged but it doesnt look like sometimes it until you give it a full test of its capabilities. This happens with LEDs too sometimes. They still work for a while after a higher than normal reverse voltage but then one day you go to turn it on and it doesnt work at all, not even dim but completely dead.

Unfortunately the reverse Vbe can happen so easily, just by mixing up the transistor base and emitter leads and connecting it to a circuit. Suddenly it sees more than 5v reverse voltage across the base and emitter. Then after realizing the 'small' problem and fixing the leads, the transistor doesnt work as well anymore and has to be completely replaced.

This is also one of the reasons that in higher power drive circuits the drive transistors are changed when the main power transistor blows out even though the drive circuit transistors dont appear to be damaged. The damage isnt immediately apparent.

 

[EngIntoHW]

Hi,
Thank you very much MrAl and Audioguro!

I learned much from your posts

Thanks!

 

[cachehiker]

Wasn't that negative Vce you were asking about?

The 2N2222 transistor will still be an NPN transistor with the collector and emitter reversed. It will have lower gain, and much lower power handling.

 

[MrAl]

Hi cachehiker,

Perhaps, but it will be a transistor with a "collector" (actually emitter now) to base voltage rating of only 6 volts. Used normally this rating is something like 75 volts. The 2N2222 can be used up to something like 40v, but with leads reversed only 6v. Quite a difference.

 

[cachehiker]

I shouldn't be posting while talking with friends at the bar.

I kinda meant that with the power handling. Most of my 2N2222 applications have been 5V or 12V. I work in the consumer appliance field where reliability isn't exactly the first priority. As a result we are often testing at the limits of a rating and sometimes beyond it. Issues with components found being pushed too hard get addressed but time to market is often such a high priority that some reliabiity issues are often missed until after the first production run.

OK, I ramble a bit but I'm always surprised at how long some parts last when being pushed way past their ratings. I'm also continually surprised at how fast some parts fail when being pushed beyond certain ratings.

In a 12V hobbyist application with base current limited to 5mA / assumed hfe of 25 = 200uA, I wouldn't be shocked if it lasted quite a while.

 

[cachehiker]

Energizing a relay coil in reverse would not be a good idea.

 

[Roff]

Hi,

As audio guru pointed out, the transistor can easily get damaged with an out of spec reverse voltage.

This is interesting because what happens is you might not even notice the damage to the transistor after a brief reverse Vbe that is higher than its data sheet rating. The effect can be noticed sometimes in the decrease in gain but the transistor still appears to work normally with a casual test. It's only when you go to try to get the transistor working in a circuit that requires the full spec's of the transistor (like gain and frequency response) that you can notice the difference.
For example, using a transistor in an RF oscillator circuit it might work just fine, but after a brief reverse Vbe voltage a little higher than the data sheet spec and then try to use that transistor in that same RF oscillator circuit and it probably wont work. This only takes a brief reverse voltage too, maybe less than a quarter second or even less.

So what happens is the transistor does get damaged but it doesnt look like sometimes it until you give it a full test of its capabilities. This happens with LEDs too sometimes. They still work for a while after a higher than normal reverse voltage but then one day you go to turn it on and it doesnt work at all, not even dim but completely dead.

Unfortunately the reverse Vbe can happen so easily, just by mixing up the transistor base and emitter leads and connecting it to a circuit. Suddenly it sees more than 5v reverse voltage across the base and emitter. Then after realizing the 'small' problem and fixing the leads, the transistor doesnt work as well anymore and has to be completely replaced.

This is also one of the reasons that in higher power drive circuits the drive transistors are changed when the main power transistor blows out even though the drive circuit transistors dont appear to be damaged. The damage isnt immediately apparent.

A circuit where this happens repeatedly is an astable multivibrator where Vcc>Veb(br). Most transistor models do not include breakdown. I have created a transistor model that has Vbe breakdown modeled. I'm attaching a sim which has these transistors in it. Not only do the base-emitter junctions break down repeatedly, the current can be hundreds of milliamps each time.

Here is the subcircuit of the transistor:

*
.subckt 2N3904bd 1 2 3
Q1 1 2 3 0 2N3904
D1 3 2 IdealZener6V
.model D D
.lib C:\Program Files\LTC\SwCADIII\lib\cmp\standard.dio
.model NPN NPN
.model PNP PNP
.lib C:\Program Files\LTC\SwCADIII\lib\cmp\standard.bjt
.model IdealZener6V D (Ron=10 Roff=1e12 Vfwd=6)
.ends

Try running this sim with a standard 2N3904 model. You will not see high base currents. The frequency will also be higher in hardware than in the simulation.

 

[alec_t]

I work in the consumer appliance field where reliability isn't exactly the first priority. As a result we are often testing at the limits of a rating and sometimes beyond it. Issues with components found being pushed too hard get addressed but time to market is often such a high priority that some reliabiity issues are often missed until after the first production run.

Ah, so that's why all my appliances tend to fail the day after the warranty expires .

 

[alec_t]

@Roff
Thanks for sharing that spice model. I posted a query some months ago regarding modelling junction breakdown and didn't get very far with it.

 

[MrAl]

Hi Ron,

There was an LED boost circuit that had this same problem. Cant remember what circuit it was now but it was similar to the Jewel Thief. We put a reverse diode across the base emitter to clamp negative voltages relative to the emitter to -0.7 volts. This sometimes messes up the timing a little but that can be fixed, or use a zener.

Yes it often surprises people the first time they hear that many base emitter reverse voltage breakdowns are near 5 or 6 volts. And yeah, this isnt modeled in many spice models so you can use a reverse bias of 10000 volts and get a reverse base current of 10 nanoamperes if you dont know what you are doing

 

[Jony130]

The most NPN BJT that I measured has base emitter reverse voltage breakdowns near 8V or higher. And collector emitter breakdown voltage was smaller then Vbe reverse breakdown voltage.
And in NPN when Vce reverse voltage breakdown occurs the junction has a negative resistance region.

 

[audioguru]

There is a horrible inverter where the emitter-base of the transistors continue to have avalanche breakdown. The base current is so high during breakdown that the capacitors explode. People made it using electrolytic capacitors instead of tantalum and they still exploded. Originally the polarity of the capacitors was backwards but now that is corrected and they still explode.
The output power is about 30W with a low voltage.

 

[MrAl]

Hi Jony,

Sorry i could not open your .asc file are you sure it is an asc file from LT spice or something else?

It's quite a coincidence that you were able to measure the breakdown voltage of four transistors in different modes of operation and all four started to breakdown at exactly 5.5ma each. I guess it's possible but i would have expected at least some difference.
Could it be that the breakdown voltages are really a little lower than that?
Also, did you check the frequency response and gain of each transistor before and after the breakdown tests?

 

[ericgibbs]

Hi Jony,

Sorry i could not open your .asc file are you sure it is an asc file from LT spice or something else?

It's quite a coincidence that you were able to measure the breakdown voltage of four transistors in different modes of operation and all four started to breakdown at exactly 5.5ma each. I guess it's possible but i would have expected at least some difference.
Could it be that the breakdown voltages are really a little lower than that?
Also, did you check the frequency response and gain of each transistor before and after the breakdown tests?

hi Al,
I can run it OK in LTS.

 

[Roff]

@Roff
Thanks for sharing that spice model. I posted a query some months ago regarding modelling junction breakdown and didn't get very far with it.

As you can see, it's a very simplistic model. It may not model everything that happens when the BE junction breaks down, but it suffices for some situations, such as the sim I posted.

 

[unclejed613]

"consumer appliance field where reliability isn't exactly the first priority"....... SAY AGAIN???? i have a washer/dryer combo that's been plugging along for 10 years.... just recently replaced the dryer heating element, and am replacing the motor because the shaft bushings are worn out.... sorry, but major appliances aren't something that can be replaced every 2 years or so like other consumer electronics. my mother's washer/dryer kept cranking away for close to 25 years. if it were a boom box or mp3 player i could understand an attitude of "it only needs to last for the warranty period", but not with major appliances.