PopularFX
Home Help Search Login Register
Welcome,Guest. Please login or register.
2020-08-11, 14:12:51
News: Registration with the OUR forum is now by invitation only.

Pages: 1 [2]
Author Topic: Miscellaneous Technical Debates  (Read 49958 times)
Group: Elite
Hero Member
******

Posts: 3639
It's turtles all the way down
I'm with humbugger on this one, and this is one of the reasons careful designers often put a diode b-e in cases where there may be a negative going pulse in order to protect against the reverse breakdown failure mechanism.

Obviously a lot of folks don't bother to read data sheets or just don't understand them. Some of us did a lot of our design work before integrated circuits were invented, and had to know this stuff or we would be met with tons of returned product on our shipping dock.

Interestingly, some data sheets don't show the 10ua at 5 -6 volts breakdown, but the attached sheet for the PN2222 from Fairchild and 2N2222 from SGS do. Other data sheets, such as the attached from Philips,  refer to a maximum b-e voltage with the collector open. The Philips data sheet only specifies Vebo of 5-6 Volts. Since the forward b-e voltage must be on the order of 0.6 volts, they must be referring to the reverse voltage, and do not even specify a current.

Go h !


---------------------------
"Secrecy, secret societies and secret groups have always been repugnant to a free and open society"......John F Kennedy
   
Group: Guest
I'm with humbugger on this one, and this is one of the reasons careful designers often put a diode b-e in cases where there may be a negative going pulse in order to protect against the reverse breakdown failure mechanism.

Obviously a lot of folks don't bother to read data sheets or just don't understand them. Some of us did a lot of our design work before integrated circuits were invented, and had to know this stuff or we would be met with tons of returned product on our shipping dock.

Interestingly, some data sheets don't show the 10ua at 5 -6 volts breakdown, but the attached sheet for the PN2222 from Fairchild does.

Go h !

Thanks Poynt and ION.

I long ago agreed to accept the idea that the original argument may have been simply a misunderstanding of terms and even offered an apology to Wattsup for being so harsh in my original response.  Since then, however, there have been a bunch of straw-man arguments offered over and over again and I (being me) simply had to persist until the argument was clearly understood and addressed directly.

Poynt:  Regarding the avalanche transistor, it is true that the avalanche begins at the C-B junction but the avalanche current of interest flows through the C-E junction...just to be clear on that.

ION:  So that readers don't get confused, the 10uA figure typically used on small signal transistors to specify the B-E junction breakdown voltage isn't a hard and fast limit, per se.  It is simply the current that is allowed to flow in the avalanched junction in order to measure the breakdown voltage.  A very low current is used (and carefully limited) because significantly higher currents will break the part. 

It is interesting to note that the B-E breakdown voltage and test current are invariably listed under "Typical Characteristics" rather than "Absolute Maximums".  The reason for this?  I'd say it's because it is well assumed that nobody is going to use the device in a way wherein they would need to know how much abuse that junction can take before breaking.  Another indication that reverse-breakdown B-E junction operation is not a "normal" or expected mode that circuit designers would use.  Otherwise, absolute maximums would be specified clearly.  One more nail in the coffin of this "tempest in a teapot" debate.

Happy Friday all!

Humbugger
   
Sr. Member
****

Posts: 337
Your valiant attempts to nurse Wattsup's bruised ego are commendable, however misguided.  I do not see any reference to the bogus term "reversing threshold" in any of your references.  The Ft (transition frequency) specification and the reverse breakdown voltage for B-E junction are entirley unrelated and appear on every transistor data sheet.  Citing a diagram showing a reverse-biased transistor is not a proof that pounding a B-E junction repeatedly into reverse breakdown will not have a destructive effect, as was my point.  Note that the reverse breakdown is rated at 10 micro-amps, not 10mA as you state.  The reason is simple: If the current into the reversed B-E junction is not limited to a very low level, the junction is destroyed.  Poof!

More straw men fishing for red herrings... ;)  All you have proven is that you don't understand the specifications on transistor data sheets or how they relate (or don't relate) to each other.  Try it!  Get yourself one of those avalanche transistors and pound the B-E junction with 10mA reverse-bias beyond the breakdown voltage pulses repeatedly.  Then come back and tell me what happened.

Humbugger

P.S.  Regarding your quoting me about reverse bias, that little snippet, taken out of context and misrepresented is the classic example of a fallacious straw man argument.  Not only that, but the exact same misquote had been discussed and addressed in this thread when another poster tried the same straw-man argument.

See here: http://www.overunityresearch.com/index.php?topic=726.msg10709#msg10709

Hum,

That "another poster" was me and I certainly did NOT try to make any "straw-man argument". I did misread your post and that I explained to you
in a later post.  Using my post as an example for a straw-man argument is plain wrong, and rude.

Groundloop.
   
Group: Guest
Just a few general comments.  Certainly when you take your electronics courses they explain to you how transistors work.   The current flow paths for NPN and PNP transistors are explained.  Heck I remember doing the lab where we put our transistors on a "transistor curve tracer" machine that had a scope-type display.  The biasing techniques for small signal amplification and the switching methods were also explained.  There are basically no options for finagling with how current flows through NPN and PNP transistors if you want to do something useful with them.

So it came as a big shock to me when I started reading the free energy forums a few years ago about people using NPN and PNP transistors in "alternative" and "think outside of the box" setups.  Transistors wired backwards just don't make sense peoples!  lol

But, for fun here is a cool clip from a cool guy:

[youtube]http://www.youtube.com/watch?v=K0A2gnlNe64[/youtube]



   
Group: Guest
Hum,

That "another poster" was me and I certainly did NOT try to make any "straw-man argument". I did misread your post and that I explained to you
in a later post.  Using my post as an example for a straw-man argument is plain wrong, and rude.

Groundloop.

I'm sorry if that offended you, Groundloop.  It wasn't a purposely-done straw-man argument, I know.  You owned up to your out-of-context quote right away when I pointed it out and it was clearly just a mistake.  It was just that Harvey then repeated almost the exact same quote but truncated it even more (leaving out the part about current flowing just as you had).  I simply pointed out that false words were being put in my mouth by quoting me without the entire sentence, just as you had done by accident.

No offense was intended.  Being misquoted or quoted out of context once in an argument is one thing.  But having a second person do exactly the same thing after the first incident was explained and clarified to everyone's satisfaction...well, I had to bring up the fact that that bit of misunderstanding had already been addressed.  I'm sorry if recalling it again offended you.  I thought you handled it quite admirably.

Humbugger
   
Sr. Member
****

Posts: 337
Humbugger,

No problem. :-)

GL.
   
Group: Guest

But, for fun here is a cool clip from a cool guy:


Where do I get a 40K ohm pot?  Radio Shack is out of stock and Digikey doesn't have any either!

Seriously...his explanation falls a bit short when he talks about it acting like a zener diode.  There is more to it than that.  Substituting a zener won't get the oscillation.  It has to latch up, like an SCR or a "zener with hysteresis" so that once it turns on, it stays on until the cap is discharged.  A plain zener won't do that.

Also, I'm pretty sure it will work with PNP parts just as well, if all the polarities are reversed accordingly.  Don't you think?  But he's right about MOSFETs...probably won't work at all.

After a little research, I'm not convinced this is the Esaki Effect at all.  I may be wrong here, but I think it is a case of "latch-up under reverse breakdown".  But far be it from me to try to start an argument!   ;D ;D ;D ;D ;D

Humbugger
   

Group: Administrator
Hero Member
*****

Posts: 2911
It's not as complicated as it may seem...
Thanks Poynt and ION.

Poynt:  Regarding the avalanche transistor, it is true that the avalanche begins at the C-B junction but the avalanche current of interest flows through the C-E junction...just to be clear on that.

Indeed, I did not specify, but hopefully implied that the output is taken off the collector, and the current path is still from C-E.

Using a transistor in this mode is very similar to the "self-activated switch" I demonstrated recently, because that is essentially what it is.

.99
   
Group: Guest
Where do I get a 40K ohm pot?  Radio Shack is out of stock and Digikey doesn't have any either!

Seriously...his explanation falls a bit short when he talks about it acting like a zener diode.  There is more to it than that.  Substituting a zener won't get the oscillation.  It has to latch up, like an SCR or a "zener with hysteresis" so that once it turns on, it stays on until the cap is discharged.  A plain zener won't do that.

Also, I'm pretty sure it will work with PNP parts just as well, if all the polarities are reversed accordingly.  Don't you think?  But he's right about MOSFETs...probably won't work at all.

After a little research, I'm not convinced this is the Esaki Effect at all.  I may be wrong here, but I think it is a case of "latch-up under reverse breakdown".  But far be it from me to try to start an argument!   ;D ;D ;D ;D ;D

Humbugger

You fight me you fight my gang!  lol

Actually I'm pretty sure this guy knows his stuff and I think he intentionally simplifies his presentations in his clips.  I think he may even "dumb down" his statements sometimes to appeal to the YouTube masses and not appear to be in an ivory tower.  For example he must know that MOSFETs are based on an entirely different technology.  So when he says he doesn't know why it doesn't work for MOSFETs he is acting like "one of the common folk."

No biting off of ears and no scratching!

MileHigh
   

Group: Elite
Hero Member
******

Posts: 1859
Transistors wired backwards just don't make sense peoples!  lol


?

I guess I'm one of those 'peoples'.

1. NROs
2. Need to use the alpha gain vs. the beta
3. Bidirectional analog switch
4. Current mirrors

Granted, none of these are garden variety uses but 'transistors wired backwards just don't make sense' ?



---------------------------
"As far as the laws of mathematics refer to reality, they are not certain; as far as they are certain, they do not refer to reality." - Einstein

"What we observe is not nature itself, but nature exposed to our method of questioning." - Werner Heisenberg
   
Group: Guest

?

I guess I'm one of those 'peoples'.

1. NROs
2. Need to use the alpha gain vs. the beta
3. Bidirectional analog switch
4. Current mirrors

Granted, none of these are garden variety uses but 'transistors wired backwards just don't make sense' ?




I'd be curious to see your diagram for a bi-directional analog switch and a current mirror that use "backward" transistors.  That would be a new one for me...not saying it can't be done, but I've never seen it and I've used discreet transistor (usually MOSFETs in back-to-back series) analog switches and bipolar transistor current mirrors with one transistor having a shorted B-C and actingas a matched diode for the mirroring transistor .  I don't consider either of those to be "backward" or unusual.  Hip me to the backward way, man!   O0

Humbugger

http://asic.ee.cuhk.edu.hk/ele3210/2008/t7.pdf  current mirror

Aha!  I found the backward analog switch!  http://www.freepatentsonline.com/6426667.pdf  Pretty neato!  And quite backward, as you say.  And very obscure.  Thanks for the stimulus to searcha nd learn!   O0
   
Group: Guest
Perhaps further clarifying what I am saying will help.  I will only do it for the case of an NPN transistor.  I am using the standard definition of current also.

For an NPN transistor, you have a device with three terminals:  Current flows into the base terminal.  Current flows into the collector termonal.  Current flows out of the emitter terminal.

Furthermore, we can use the standard convention where current is positive when it flows into a node and negative when it flows out of a node.

So for a typical NPN transistor using rationalized units we can say:

+1 unit of current flows into the the base input.
+50 units of current flow into the collector input.
-51 units of current flow out of the emitter output.

When you add all three currents together that go into the NPN transistor you get zero.

That is the only way a transistor is supposed to work as far as I understand.  There may be some non-standard "tricks" that people do with transistors that I am not aware of.

However, any person that thinks that you are "lacking imagination" and you can set up transistors in circuits in a "backwards" fashion to to functional things for real-world applications is dead wrong.

MileHigh
   

Group: Elite
Hero Member
******

Posts: 1859
It should be known that a bipolar transistor has the same layer type or 'doping' on the collector and emitter.

The only reason it is highly inadvisable to run them backward is because they are optimized to run in one direction. If you don't really understand what to avoid you will surely destroy the transistor. There is gain in either direction. Modern types are very asymmetric between 'right way' and 'wrong way' current.

I remember a time when the military was transitioning from Nuvistor tube-based equipment to 'transistorized'. The RF end of the receivers were built around germanium bipolars that didn't care how they were connected as long as you didn't mix an emitter or collector with a base connection point.

You should be able to choose any basic, one-transistor amplifier circuit and flip the C & E. The trick is to play with the bias values to bring it into an operational window before it fries.

The reasons for my use of such 'obscure' ideas:

Almost zero signal distortion....

The logarithmic difference between input and output signals are minimized (also makes for a more accurate current mirror)....

A neat little trick is when the circuit is driven into clipping. Only at the waveform peak the transistor will break into negative resistance oscillation.....

Since reverse recovery time now has more affect upon the circuit performance, you may find the backwards transistor also acting a bit like a low-pass filter....

I think the usefulness of these types of circuits are like any other circuit. If you can't find a use - then they are useless for you.





---------------------------
"As far as the laws of mathematics refer to reality, they are not certain; as far as they are certain, they do not refer to reality." - Einstein

"What we observe is not nature itself, but nature exposed to our method of questioning." - Werner Heisenberg
   
Group: Guest
That all sounds about right to me.  I'm at a bit of a loss to understand why the superior traits you describe in terms of signal distortion and bursting into oscillation at peaks are attributed only to "backward" usage if the collectors and emitters are interchangeable, though.  I guess you must mean that the modern day ones do that but the old, symmetrical ones didn't.

Anyway, you'll get no argument from me on this one, especially since you didn't mention radiant energy, dark matter, or other kinds of magic.  I used to play with CK722s, 2n107s, 2N35s, 2N176's and all those early germanium types, but I was only 12 or 13 years old then and very careful to follow the prescribed Popular Electronics or GE Transistor Handbook circuit hookups. 

I mean, you know, I had to skip candy and cigarettes and Playboy magazine for a month to afford a single really crappy (and ultra-delicate) transistor!  Most of them, I'm proud to say, died from having leads break off right at the glass frit seal (or at the plastic bottom for the CK722).  I remember desperately carving away at the plastic to get enough of a nub back to solder a new lead to.

Humbugger
   
Group: Guest
I am going to defer to Wavewatcher also because it sounds like he really knows what he is talking about.  Both of you guys have much more experience than me in this regard for sure.  I can at least see that since an NPN transistor is basically two back-to-back diodes with a common base, that it stands to reason that if you make the the base-collector diode forward biased it indeed may work "backwards."

That doesn't necessarily invalidate the statement that in the vast majority of cases you want to use a transistor the way it is intended to be used.  You analog guys are amazing.   I worked in a telecom company once and saw analog design engineers do their thing and had the utmost respect for them.

I was more of a 741 op-amp guy!  lol  I would not want to be a VHDL guy...

MileHigh
   

Group: Elite
Hero Member
******

Posts: 1859
I doubt the more symmetrical ones would exhibit any more than really bad gain. There are a few in my junk box. I may try to work with them later.

The germanium types aren't the only ones good for these circuits. Try any small signal type. I've heard electric guitar junkies use backwards transistors to develop unique sounds. Maybe some research will turn something up.. My idea of music doesn't fit with most others  C.C

Trying to solder a new lead to the nub.  ;)  I have done that lately while waiting for replacement parts >:(



---------------------------
"As far as the laws of mathematics refer to reality, they are not certain; as far as they are certain, they do not refer to reality." - Einstein

"What we observe is not nature itself, but nature exposed to our method of questioning." - Werner Heisenberg
   
Group: Guest
http://www.edn.com/article/474784-The_ultimate_zener_diode_reference.php
"Early-IC designers achieved zener characteristics using reverse-biased, npn-transistor base-emitter junctions."

http://www.freepatentsonline.com/5786972.pdf
"FIG. 1, transistors 10, 15, 20, and 25 are configured as base-emitter zener diodes"

http://leyenda2.demon.co.uk/mike/Electronics/datasheets/motorola/lm339.pdf
"Experimental data has shown that any of National's process 21 transsistors which have been selected for low reverse beta . . . can be used quite satisfactororily as a zero T.C. Zener. When connected as shown in Figure 37, the T.C. of the base-emitter Zener voltage is exactly cancelled by the T.C. of the forward biased base-collector junction if biased at 1.5mA."

http://wenku.baidu.com/view/bb809b2acfc789eb172dc885.html
See page 348

http://www.elib.edu.et/bitstream/123456789/20480/2/1001265.pdf
Page 454 section 5

http://ecee.colorado.edu/~bart/book/book/chapter4/ch4_5.htm
"Two mechanisms can cause breakdown, namely avalanche multiplication and quantum mechanical tunneling of carriers through the bandgap. Neither of the two breakdown mechanisms is destructive. "

Sorry guys, but I think Wattsup is teaching you a couple things ;)

   
Group: Guest
Harvey,

You are as persistent and tenacious as I am, or moreso.  I was wrong when I said that reverse-breakdown of B-E junctions was not commonly used except in some rare noise generator circuits.  I was right when I said they are quite fragile and easily destroyed.  I was right when I said there is no such bipolar transistor data sheet term "reversing threshold".  I was right when I said that reverse breakdown characteristics are not frequency dependent.

Regarding fragility...from your own source quotes below cut and pasted...

Remember, this whole little debate started when I challenged wattsup to show me the words "reversing threshold" on any transistor data sheet and to show me where any such reverse-bias breakdown was "frequency sensitive".  Then it evolved into a broader discussion regarding the fragility of reverse-breakdown operation in garden-variety transistors like the 2N2222.  Then it became a debate on whether such features as reverse breakdown operation were common or rare.  

In the early stages, I apologized to Wattsup for jumping on him and said we might have just misunderstood each other.  I admitted even to being a jerk and an asshole.  He's been prickly (read that PRICK-FACED) toward anything and everything I've had to say ever since and made rude comments, dismissing me as ignorant and not worth wasting time with.  Others obviously disagree. [EDIT...but I'm sure my outright arrogance and pushy, sarcastic argumentitive personality will eventually piss off everyone here] >:-)

I stand corrected on the usage of the reverse breakdown B-E junction in modern circuits (many of which I have used myself without even noticing...like in precision references, etc.).  You got me there, and Milehigh, too, I think.

I stand pat on the rest of it.  And yes, I have learned from all of it.  So there!  Take that!   >:-)

I do appreciate your mediation and your patience and the time you've spent googling references.  Thank you.

Humbugger



« Last Edit: 2011-02-19, 18:58:13 by humbugger »
   
Pages: 1 [2]
« previous next »


 

Home Help Search Login Register
Theme © PopularFX | Based on PFX Ideas! | Scripts from iScript4u 2020-08-11, 14:12:51