Can a silicon diode be used as an inaccurate zener?

Go To Last Post
60 posts / 0 new

Pages

Author
Message
#1
  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Hi - I was just thinking about this... can you use a normal silicon diode as an inaccurate zener diode? I mean, if so I suspect it won't be nearly as accurate as a zener, but for my application I don't care. Or when a normal silicone diode breaks down is there actual damage occurring to the part? Is it the same effect as is happening on a zener, or something different?

Thanks!

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Reverse breakdown, like at 50 volts for an 2N4001? Or forward volts like .6 volts at 20ma fwd voltage? (There is a bandgap ref in the AVRs with an a/d converter, accurate to +-10%)

Imagecraft compiler user

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

I think revers breakdown voltage is not a good controlled parameter for normal diodes and they
are not constructed to withstand that. Forward
voltage may be used, but I personally don't
like even zener-diodes and prefer a TL431 or another
"precision" regulator.

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

I don't get it! What is wrong with using a zener diode, they are cheap and reliable and work quite OK in a correctly designed cct.

Charles Darwin, Lord Kelvin & Murphy are always lurking about!
Lee -.-
Riddle me this...How did the serpent move around before the fall?

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

bobgardner wrote:
Reverse breakdown, like at 50 volts for an 2N4001? Or forward volts like .6 volts at 20ma fwd voltage? (There is a bandgap ref in the AVRs with an a/d converter, accurate to +-10%)

Reverse breakdown. Very low current so power dissipation would be a non issue. I want a very high voltage zener (500v+) but don't care about how precise it is, so this seems like a potential alternative as hv zeners don't seem to exist. Also considering using a neon for the application, as their iv curves are quite steep like a zener.

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Why not just use a few of these or similar in series?
http://www.onsemi.com/pub_link/Collateral/P6SMB6.8AT3-D.PDF

Reverse breakdown of a normal diode can break the diode.

oddbudman

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

oddbudman wrote:
Why not just use a few of these or similar in series?
http://www.onsemi.com/pub_link/Collateral/P6SMB6.8AT3-D.PDF

Reverse breakdown of a normal diode can break the diode.

oddbudman


So reverse breakdown damages a normal diode, even if you current limit it like you would a zener?

Multiple zeners is an alternative, but I'm space constrained so it is not ideal.

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

I would guess that a current-limited reverse breakdown would not damage the diode.

But the reverse breakdown is not going to be clearly defined. e.g. a IN4001 is not going to break down at its maximum reverse voltage. You are only guaranteed that it is some higher voltage.

But why should you want to use a cheap device for 'out of data sheet' usage? When you could use an equally cheap 'zener' device that is specified for this application.

Just what sort of voltage are you trying to achieve?

David.

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

You could build your own 500V Zener with a circuit like below. (T1 and T2 must be high voltage transistors, of course).

At approx. 500V the voltage drop across R2 is 0.6V - T1 gets active which then activates T2.

Regards
Sebastian

Attachment(s): 

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

S-Sohn wrote:
You could build your own 500V Zener with a circuit like below. (T1 and T2 must be high voltage transistors, of course).

At approx. 500V the voltage drop across R2 is 0.6V - T1 gets active which then activates T2.

Regards
Sebastian


Sebastian - it's an interesting approach - but I'm very space constrained - so single component solutions are pretty ideal for me. I probably will need a higher voltage rating - like 1KV or even more. Accuracy is not terribly important - and I don't mind testing a bunch of components to find a single one that works - volume for this project is quite low.

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

How about a gas discharge tube rated at 470 volts?

http://www.bourns.com/data/global/pdfs/2036.pdf

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

david.prentice wrote:
I would guess that a current-limited reverse breakdown would not damage the diode.

But the reverse breakdown is not going to be clearly defined. e.g. a IN4001 is not going to break down at its maximum reverse voltage. You are only guaranteed that it is some higher voltage.

But why should you want to use a cheap device for 'out of data sheet' usage? When you could use an equally cheap 'zener' device that is specified for this application.

Just what sort of voltage are you trying to achieve?

David.


It'll need to be at least 500 volts, but probably more. Exact voltage is unknown at this point. The problem is that, at least according to Digi-Key, Zener diodes are not made past 200V. Price is of essentially no concern in this project.

I am reading up on avalanche diodes - these may be an alternative.

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

alwelch wrote:
How about a gas discharge tube rated at 470 volts?

http://www.bourns.com/data/global/pdfs/2036.pdf


Very interesting! What is the third terminal for? Is it equivalent to the third terminal and xenon flashes?

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Quote:
but I'm very space constrained

You don't use 0.1mm track space at these voltages. I don't know the security requirements at these voltages, but I would expect they're in the range of 1mm+. I can't imagine that you can use SMD parts in these voltage areas.

Regards
Sebastian

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

S-Sohn wrote:
Quote:
but I'm very space constrained

You don't use 0.1mm track space at these voltages. I don't know the security requirements at these voltages, but I would expect they're in the range of 1mm+. I can't imagine that you can use SMD parts in these voltage areas.

Regards
Sebastian


Sebastian - your concern is very valid. In fact on a recent test board I had to mill out my ground plane to extend clearances around some high voltage traces, as it was sparking. However, the whole PCB is going to be potted - so I'm not too worried about sparking between traces!

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

I am horrified by the thought of a 1000V zener or even a 500V zener.

But in terms of producing a regulated high voltage o/p, you can use regular variable voltage chips. You sense a fraction of the o/p voltage, and your series pass element is a high voltage transistor.

This is just from memory. I am sure if you google "LM317 high voltage" or similar, you will get the appropriate app note.

As Sebastian has suggested, you will need to consider board area and track space very carefully.

If you are really considering a shunt-regulator. Which is what zeners effectively behave as.
The shunt element is a high voltage transistor, and you arrange it to maintain a constant collector voltage. The base and emitter can be at "low-voltage" potentials.

David.

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

According to "Horowitz Hill: The Art of Electronics"
you also should take care that the resistors
are specified for the high voltage.

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

nleahcim wrote:
alwelch wrote:
How about a gas discharge tube rated at 470 volts?

http://www.bourns.com/data/global/pdfs/2036.pdf


Very interesting! What is the third terminal for? Is it equivalent to the third terminal and xenon flashes?

They seem to offer them in either 2 or 3 terminal configurations and the 3 terminal one has a thermal protection that shorts the two outer legs to the center to bypass the device to ground for high current clamping. It resets when it cools.

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

ossi wrote:
According to "Horowitz Hill: The Art of Electronics"
you also should take care that the resistors
are specified for the high voltage.

Yep - I'm aware of this. Designing for high voltage is a royal PITA!

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

david.prentice wrote:
I am horrified by the thought of a 1000V zener or even a 500V zener.

But in terms of producing a regulated high voltage o/p, you can use regular variable voltage chips. You sense a fraction of the o/p voltage, and your series pass element is a high voltage transistor.

This is just from memory. I am sure if you google "LM317 high voltage" or similar, you will get the appropriate app note.

As Sebastian has suggested, you will need to consider board area and track space very carefully.

If you are really considering a shunt-regulator. Which is what zeners effectively behave as.
The shunt element is a high voltage transistor, and you arrange it to maintain a constant collector voltage. The base and emitter can be at "low-voltage" potentials.

David.


I'm not trying to make a regulator. I'm trying to get a signal as to when I have high voltage present. I'm hoping a zener sort of configuration will be more efficient than a resistive divider. This feedback signal is used to control a switching regulator that is producing the high voltage signal. The resistive divider is the fall back plan. Also, like I said, I don't care about accuracy - so if I have a signal that says "yes it's about 500V or higher" that's probably good enough.

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Quote:
so if I have a signal that says "yes it's about 500V or higher" that's probably good enough.
In this case you could go with this circuit. T1 can be any standard transistor. No high voltage requirements anymore. Signal HighVoltageAvailability goes low if the input voltage is higher than 500V.

You should have posted this information directly when you had started this thread.

Regards
Sebastian

Attachment(s): 

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Quote:

I don't know the security requirements at these voltages, but I would expect they're in the range of 1mm+. I can't imagine that you can use SMD parts in these voltage areas.

http://www.smps.us/pcbtracespaci...

Without examining the standards the link above gives a rough idea. 1.5mm to 3mm according the the chart. At those voltages PCB material >>does<< matter. Been there, done that.

Quote:

I can't imagine that you can use SMD parts in these voltage areas.

Hmmm--you should then see the dropping resistors on this product... ;)
http://www.anderson-bolds.com/Me...

You can put lipstick on a pig, but it is still a pig.

I've never met a pig I didn't like, as long as you have some salt and pepper.

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Thank you Lee!
Did you program the motor guardian?
Looks nice, but the prize doesn't fit to the size of my wallet.
What SMD parts have you used there?

Regards
Sebastian

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

If you just want to detect a voltage that exceeds 500V, the job is trivial.

You use a resistive divider, and an analog comparator to the 'slider'. You can easily have a protective transorb or zener at the 'slider'. And since you are using such a large division ratio, the 'output impedance' at the slider is comfortably low.

For example a resistive divider of 10M and 47k will give you 2.33V for 500V i/p. and only 4.66V for 1000V.

The o/p impedance of 47k is perfect to drive the AVR analog comparator. The divider will draw 100uA and dissipate 100mW. You will need to construct the 10M from several discrete resistors to satisfy the resistor voltage rating.

If 100uA is too much load on your signal, you could possibly use 50M and 220k. e.g. 20uA at 1000V.

I would guess that you can live with this sort of loading. Bear in mind that your original plan of a 500V zener would take a massive current from a 1000V signal!

David.

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

david.prentice wrote:
If you just want to detect a voltage that exceeds 500V, the job is trivial.

You use a resistive divider, and an analog comparator to the 'slider'. You can easily have a protective transorb or zener at the 'slider'. And since you are using such a large division ratio, the 'output impedance' at the slider is comfortably low.

For example a resistive divider of 10M and 47k will give you 2.33V for 500V i/p. and only 4.66V for 1000V.

The o/p impedance of 47k is perfect to drive the AVR analog comparator. The divider will draw 100uA and dissipate 100mW. You will need to construct the 10M from several discrete resistors to satisfy the resistor voltage rating.

If 100uA is too much load on your signal, you could possibly use 50M and 220k. e.g. 20uA at 1000V.

I would guess that you can live with this sort of loading. Bear in mind that your original plan of a 500V zener would take a massive current from a 1000V signal!

David.


I'm well aware of the ability to use large resistors. Like I said, that is a backup solution. 100ua is more than my entire system load. My loads are more like 1-5ua. If my sensing is using up even 10ua, that means my efficiency is in the single digits. If I keep on upping the resistance I start to have all sorts of nasty issues with contaminants. I don't want to use a resistive divider.

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

S-Sohn wrote:
Quote:
so if I have a signal that says "yes it's about 500V or higher" that's probably good enough.
In this case you could go with this circuit. T1 can be any standard transistor. No high voltage requirements anymore. Signal HighVoltageAvailability goes low if the input voltage is higher than 500V.

You should have posted this information directly when you had started this thread.

Regards
Sebastian

Sebastian - a resistive divider like you are suggesting is perfectly reasonable, but I see advantages in a more Zener like system as I think I could reduce current draw.

Can anybody definitively answer my original question?

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

alwelch wrote:
How about a gas discharge tube rated at 470 volts?

http://www.bourns.com/data/global/pdfs/2036.pdf


It is unlikely that this sort of device will operate as a shunt regulator as it is intended for
surge protection - that it's impedance falls to near zip to pass amperes upon triggering.
Operating in a current limited mode may work somewhat, but the very small electrodes
and discharge volume would make it unreliable.

Victoreen manufactured the GV series of Corona Regulator Tubes that operated between
5 µA and 2 mA to regulate from 350 to 3500 V for use in radiation monitoring equipment.
While obsolete, they can still be found on ebay. Attached is the application note portion of
Victoreen's catalog with the GV3A page.

Stan

Attachment(s): 

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Quote:
Can anybody definitively answer my original question?
This is a difficult job as you come out with the details bit by bit only. If you could summarize ALL your requirements in ONE post, things would become much easier for everybody.

Regards
Sebastian

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

sbennett wrote:
alwelch wrote:
How about a gas discharge tube rated at 470 volts?

http://www.bourns.com/data/global/pdfs/2036.pdf


It is unlikely that this sort of device will operate as a shunt regulator as it is intended for
surge protection - that it's impedance falls to near zip to pass amperes upon triggering.
Operating in a current limited mode may work somewhat, but the very small electrodes
and discharge volume would make it unreliable.

Victoreen manufactured the GV series of Corona Regulator Tubes that operated between
5 µA and 2 mA to regulate from 350 to 3500 V for use in radiation monitoring equipment.
While obsolete, they can still be found on ebay. Attached is the application note portion of
Victoreen's catalog with the GV3A page.

Stan


Stan - thanks for the information. Sounds like gas discharge tubes are out. The part you provided a datasheet for looks nice - but for my application it is much too big, unfortunately.

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Quote:

Did you program the motor guardian?
Looks nice, but the prize doesn't fit to the size of my wallet.
What SMD parts have you used there?

Yep, that's one of my babies. A full Mega32.

It is somewhat at the high end of that sort of thing for that class of motors (or similar AC monitoring). However if you looked at the feature list there is a LOT offered. Now this one is a half-full Mega8 non-isolated:
http://www.automatictiming.com/D...

The default specs on the MPA are +/- 1%. But the heart is an Analog Devices ADE7754 energy-meter billing-quality chip, that can be calibrated down to the nubbins (basically a 24-bit device). All for a few $$; cheaper than the Mega32 IIRC.

I just unfolded the three-board set, and see nothing >>but<< surface mount parts.

The voltage front end is three smallish sized SM resistors in series for each phase. That drops the (up to about 700VAC) raw voltage to ~2.5V input to the ADE7754. Lessee--looks like size 2010, and about 250k each.

There was a model that had dropping resistors in the several meg range. A new board spin had a different material and the high-voltage, high-resistor impedance allowed leakage through the board and distorted results. The "solution" was to do as in the MPA and have less impedance in the dropping circuit so the board leakage became immaterial.

On the same site we also did this whole series of pressure regulators, all with AVRs:
http://www.automatictiming.com/P...
http://www.automatictiming.com/P...

as well as many/most of the timers and counters on that site (all with AVRs of course ;) [well, there are a few AT89s in there])

You can put lipstick on a pig, but it is still a pig.

I've never met a pig I didn't like, as long as you have some salt and pepper.

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

S-Sohn wrote:
Quote:
Can anybody definitively answer my original question?
This is a difficult job as you come out with the details bit by bit only. If you could summarize ALL your requirements in ONE post, things would become much easier for everybody.

Regards
Sebastian


Hi Sebastian - I thought my question was pretty clear, but I'll restate: My question was if you can use a normal silicon diode as a zener. Specifically, if a silicon diode could be used as a constant voltage drop when reverse biased, as long as you current limit it to protect against excess current flow. My assumptions are that it would not have as steep of an IV curve as a zener, and that the zener voltage would be somewhere beyond the rated breakdown voltage of the diode.

Thanks!

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

[no; sorry; I'm not answering that question] Why fuss with all that when simple dropping will let you put the signal right into the AVR's A/D (we still haven't been told AC or DC; AC might need rectification or level shifting) or analog comparator for your 500V test.

Space requirements are the same. Component count about the same. Spacing about the same.

You can put lipstick on a pig, but it is still a pig.

I've never met a pig I didn't like, as long as you have some salt and pepper.

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

theusch wrote:
[no; sorry; I'm not answering that question] Why fuss with all that when simple dropping will let you put the signal right into the AVR's A/D (we still haven't been told AC or DC; AC might need rectification or level shifting) or analog comparator for your 500V test.

Space requirements are the same. Component count about the same. Spacing about the same.


Because that would probably be higher power. This is a very low power system.

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Quote:

Because that would probably be higher power. This is a very low power system.

But you want to do a zener-type system?!?

(I'm out. Same dropping impedance as the diagrams above...)

You can put lipstick on a pig, but it is still a pig.

I've never met a pig I didn't like, as long as you have some salt and pepper.

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

theusch wrote:
Quote:

Because that would probably be higher power. This is a very low power system.

But you want to do a zener-type system?!?

(I'm out. Same dropping impedance as the diagrams above...)


Well, the idea was that it'd be used as an overvoltage indicator. So either I have essentially no current draw from my feedback (zener not conducting), or I have just a trickle (zener conducting, current limited with a resistor) telling me to stop the switcher. Whereas with a resistive divider I always have a load, no matter what. Make sense?

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

I am confused about this zener thing. With a 400V zener and a 10M series resistor: You can obviously have the effect of 0uA up to say 400V. 10uA at 500V, and 60uA at 1000V.

You can scale up the resistances a bit, but coping with 500V to 1000V is a serious problem. And if you used a 480V zener with a 2M series resistor, your 10uA at 500V turns into 260uA at 1000V.

Regarding the reverse breakdown characteristics of regular rectifier diodes, I seem to recall a negative resistance 'knee' following breakdown. This would give you even more problems keeping the dissipation within limits.

But hey-ho it is your idea, just google for 'reverse breakdown curves' and see for yourself.

David.

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

david.prentice wrote:
I am confused about this zener thing. With a 400V zener and a 10M series resistor: You can obviously have the effect of 0uA up to say 400V. 10uA at 500V, and 60uA at 1000V.

You can scale up the resistances a bit, but coping with 500V to 1000V is a serious problem. And if you used a 480V zener with a 2M series resistor, your 10uA at 500V turns into 260uA at 1000V.

Regarding the reverse breakdown characteristics of regular rectifier diodes, I seem to recall a negative resistance 'knee' following breakdown. This would give you even more problems keeping the dissipation within limits.

But hey-ho it is your idea, just google for 'reverse breakdown curves' and see for yourself.

David.


David - you're exactly right about how the zener would work. But I just want to use it for overvoltage detection, not for an analog feedback signal, so I'd want the zener voltage to be my overvoltage limit. Thus there'd be no increasing current as the voltage increased - the zener would turn on, the controller circuitry would stop the power supply, and the power supply would slowly fall till it hit its intended voltage. Make sense?

I will try googling your search term.

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Would something like this work? It would turn on a resistive divider at whatever interval you want, instead of always being on. Currant usage would be the transistor leakage current, otherwise.

Note that I have little experience, so quadruple checking everything I say is required. :P

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Yoikes, that would fry the ADC when signal is low... move that connection to the other side of the cap!

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Quote:
However, the whole PCB is going to be potted - so I'm not too worried about sparking between traces!

What potting solution has a higher break down voltage then air?

It seems the OP has no idea what he wants as the requirements
keep on changing.
Space is critical, costs are no objective, voltages could be very high, can't have any load on HV etc. etc.

Come up with a firm specification, then look at if and how it can be implemented. Proper engineering dictates that components should operate according to their repeatable characteristics and not on hit & miss principles.

Charles Darwin, Lord Kelvin & Murphy are always lurking about!
Lee -.-
Riddle me this...How did the serpent move around before the fall?

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

tlucas wrote:
Would something like this work? It would turn on a resistive divider at whatever interval you want, instead of always being on. Currant usage would be the transistor leakage current, otherwise.

Note that I have little experience, so quadruple checking everything I say is required. :P

Hi - indeed such a system would work. I don't think you'd want the capacitor though. If you're looking to low pass the signal you'd want it in parallel with R2. Additionally, I think it'd be better with a FET (otherwise you'd have to be careful that the BE current didn't give you a false HV positive signal). It's a reasonable alternative - though it does require a bit of intelligence in the controller and two high voltage components. High voltage components are crazy large...

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

LDEVRIES wrote:
Quote:
However, the whole PCB is going to be potted - so I'm not too worried about sparking between traces!

What potting solution has a higher break down voltage then air?

It seems the OP has no idea what he wants as the requirements
keep on changing.
Space is critical, costs are no objective, voltages could be very high, can't have any load on HV etc. etc.

Come up with a firm specification, then look at if and how it can be implemented. Proper engineering dictates that components should operate according to their repeatable characteristics and not on hit & miss principles.


I believe most potting solutions have much higher breakdown voltages than air. All the epoxies I've used have had much higher breakdown voltages than air. A quick Google suggests that Epoxy has about 5x the breakdown voltage of air.

I shouldn't have to give any specifications for my problem - I really am just looking for an answer to my original question.

Also - what specification have I wavered on? I don't know what these changing specifications are that you're referring to. My spec is pretty darn firm - I just don't know my operating voltage. It's at least 500V, could be as high as 4KV. Possibly even higher, but unlikely.

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

The cap was to ensure the source voltage didn't creep up, making it impossible to turn on the transistor with a <5V signal.. as for the type, yes, a FET, of course :) . I just grabbed the first symbol on hand.

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

You can not rely for any kind of protection from a semiconductor in avalanche mode.
Essentially when a PN junction breaks down You do not achieve anything like zener knee voltage.
In fact I used to use a BC109 transistor in avalanche mode to generate a fast pulse which then supplied current to a laser diode.
The transistor collector was connected to an RC network connected to a 200+ V dc supply.
The capacitor voltage would rise to a point where the transistor would break down and the energy stored in the capacitor wepuld then be dumped into the laser diode. The charging path for the capacitor was sufficiently high resistance to allow the transistor to recover and the cycle would repeat.

SO TO DEFINITIVELY ANSWER YOUR QUESTION.. NO.

Use ZENER DIODE thats why it was invented.f You think the zener might not have sufficient power dissipation capability buffer it with a power transistor bu connecting the anode of the zener to the base of an NPN transistor. Connect Collector to the zener cathode.
And ofcourse emiter to the point where anode of zener would usually live. You are now a proud owner of a high power zener diode.

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Quote:
I shouldn't have to give any specifications for my problem -

Read. I am too lazy to give you all the details.!

Quote:
I really am just looking for an answer to my original question.

Read. My time is more important then yours!

There were three original questions

Answers are NO, YES & NO

Charles Darwin, Lord Kelvin & Murphy are always lurking about!
Lee -.-
Riddle me this...How did the serpent move around before the fall?

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

LDEVRIES wrote:
Quote:
I shouldn't have to give any specifications for my problem -

Read. I am too lazy to give you all the details.!

Quote:
I really am just looking for an answer to my original question.

Read. My time is more important then yours!
QED

No - I gave every detail that was necessary to answer my question.

I don't think my time is more important than any of your all's - but if you feel that way than please feel free to ignore my threads in the future so that you aren't offended by your misinterpretations of my posts.

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Am I missing something here?.

C1 will charge to the applied voltage (400 < 1000 volts) and then your ideal transistor will be switched on and be expected to short the 400 volt charged cap ... sounds like a one off device = fuse.

Ross McKenzie ValuSoft Melbourne Australia

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Ross,

I think you have already realised that the OP is just after winding up the forum. If he is genuine, I advise that you keep well away from any of his products.

David.

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

david.prentice wrote:
Ross,

I think you have already realised that the OP is just after winding up the forum. If he is genuine, I advise that you keep well away from any of his products.

David.


I thought the members of this forum were above personal attacks. I guess I was mistaken.

I asked a simple question, and have been repeatedly insulted. I really thought better of the Freaks than this.

edit: to be clear - only a select few individuals have been on the attack - most of you have been great as always. But it's always disappointing to have a couple dead pixels on a nice monitor, so to speak...

Last Edited: Fri. Apr 16, 2010 - 02:46 PM
  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

valusoft wrote:
Am I missing something here?.

C1 will charge to the applied voltage (400 < 1000 volts) and then your ideal transistor will be switched on and be expected to short the 400 volt charged cap ... sounds like a one off device = fuse.


Ross - I agree.

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

ignoramus wrote:
You can not rely for any kind of protection from a semiconductor in avalanche mode.
Essentially when a PN junction breaks down you do not achieve anything like zener knee voltage.
In fact I used to use a BC109 transistor in avalanche mode to generate a fast pulse which then supplied current to a laser diode.
The transistor collector was connected to an RC network connected to a 200+ V dc supply.
The capacitor voltage would rise to a point where the transistor would break down and the energy stored in the capacitor wepuld then be dumped into the laser diode. The charging path for the capacitor was sufficiently high resistance to allow the transistor to recover and the cycle would repeat.

SO TO DEFINITIVELY ANSWER YOUR QUESTION.. NO.

Use ZENER DIODE thats why it was invented.f You think the zener might not have sufficient power dissipation capability buffer it with a power transistor bu connecting the anode of the zener to the base of an NPN transistor. Connect Collector to the zener cathode.
And ofcourse emiter to the point where anode of zener would usually live. You are now a proud owner of a high power zener diode.


Hi - thanks for sharing your experience - it is very interesting.

I agree that a zener would be the best way to get a zener-like IV curve - but the problem is that zener diodes at the voltages that I want are very uncommon. Well, they don't exist. I can only find avalanche diodes. My understanding is that avalanche diodes have IV curves that very closely resemble that of zeners. (but is that right? I am a little confused on that point...) The other problem is that the only avalanche diodes I can find at my desired voltage are quite large. Too large to fit, in fact. Also note that I'm not looking for a high power zener - I'm looking for a high voltage zener. Power dissipation will be very, very low.

So it looks like the BC109 is a 20V part. Was the base tied to the emitter? When the part broke down, did it maintain a constant voltage drop across it, or would VCE plunge to something well below the breakdown voltage? It sounds like the latter? So there is some sort of hysteresis involved with this breakdown, yes? Depending on the amount of hysteresis - that could be problematic for me. But the transistor was not getting damaged by this breakdown, right?

Also - a transistor is a significantly different beast than a diode - do you think the effect might be different?

Pages