BJTs as ultra low leakage protection diodes
Ebrahim
leakage diode(<10pA) in a serious design. The reason that I do not use
JFET is that I need high breakdown(100V) voltage. So I also need
Bootstrapping I have down some measurements with different JFETs and
BJTs and I didn't find quite big differences.
Ebrahim
Sorry I forgot to say that I mean Collector-Base Diode of BJT.
Regards
Robert Baer
Phil Allison
>> Hello, I want is there anyone that has used a BJT instead of ultra low
>> leakage diode(<10pA) in a serious design.
"Ebrahim"
** You are a totally stupid pile of autistic shit, a waste of space a waster
of good people's valuable time.
Fuck off and DROP DEAD !!!
Ebrahim
Hello Robert :)
Thanks for replying to my post. The main problem with LEDs is their low
breakdown voltage. I have not measured leakage current of an LED but it
makes sense because GaAs Reverse saturation current is about 10 times
less than Si. Have you used LED as an ultra low leakage diode ? have you
measured leakage currents of LEDs?
Regards,
Ebrahim
Phil Allison
Ebrahim
> Ebrahim wrote:
Hello Robert :)
Thanks for replying to my post. I have not measured leakage current of
mi...@sushi.com
What exactly are you trying to protect? IC, fet, ??
Ebrahim
> What exactly are you trying to protect? IC, fet, ??
Hello,
I'm Trying to protect input of Ultra low bias current FET OpAmps.
Regards,
Ebrahim
Robert Baer
remaining visibility which i think is astounding, and <1nA in the 1V
(forward) region with corresponding reverse values.
Certainly not in the region you want.
Robert Baer
ionizes; 90V beforehand.
Ebrahim
> In a way i have "measured" leakage current in LEDs; low microamps for
> remaining visibility which i think is astounding, and <1nA in the 1V
> (forward) region with corresponding reverse values.
> Certainly not in the region you want.
Wow, uA and nA :-o !
then they can't be used as an ultra low leakage. Even humble 1N4148 has
lower leakage ;) .
Joerg
Nanoamps? Wow. Is that spec'd anywhere? If so then they'd be really
useful as limiters.
--
Regards, Joerg
http://www.analogconsultants.com/
"gmail" domain blocked because of excessive spam.
Use another domain or send PM.
mook johnson
"Ebrahim" <Engine...@gmail.com> wrote in message
news:h0vdqr$314$1...@news.eternal-september.org...
not cheap but
John Larkin
wrote:
Then limit the current before you try to clamp it. You're not going to
have a lot of luck finding a half-amp, 100V diode that leaks picoamps.
Why 100 volts? The opamp can't stand that.
Use a resistor or, if you can't stand the Johnson noise, a pair of
depletion fets or some such. SSRs are interesting as low-resistance
current limiters.
A high-value resistor bypassed with a small cap is interesting here
too. If the fet has internal ESD diodes, that may be all you need.
John
Ebrahim
> Then limit the current before you try to clamp it. You're not going to
> have a lot of luck finding a half-amp, 100V diode that leaks picoamps.
> Why 100 volts? The opamp can't stand that.
***
John:
100V is needed for another use(protecting a high voltage discrete
amplifier), not protecting FET opamp. I was hoping that I could find one
low leakage diode for both uses. though for the first application(HV
Discrete amp) I can live with some leakage but for protecting the FET
opamp I really need ultra low leakage and about 500mA load current.
> Use a resistor or, if you can't stand the Johnson noise, a pair of
> depletion fets or some such. SSRs are interesting as low-resistance
> current limiters.
***
The FET Opamp that I try to protect is opa627 (
http://focus.ti.com/docs/prod/folders/print/opa627.html)
resistor is out of question because of Johnson noise as you
mentioned(because voltage could rise up to 100V in some rare occasions
and it needs a really big resistor) and for the same reason depletion
FETs can't be used. SSR(Solid State Relays ?) How?
> A high-value resistor bypassed with a small cap is interesting here
> too. If the fet has internal ESD diodes, that may be all you need.
***
No, Ultra low bias current opamps usually do not have internal
protection diodes.
Cheers,
Ebrahim
Phil Hobbs
It's far lower than that. I've measured ordinary red display LEDs whose
leakage was below 100 fA from -5V to +0.5V bias.
Cheers
Phil Hobbs
--
Dr Philip C D Hobbs
Principal
ElectroOptical Innovations
55 Orchard Rd
Briarcliff Manor NY 10510
845-480-2058
hobbs at electrooptical dot net
http://electrooptical.net
Phil Allison
"Ebrahim"
Phil Hobbs
What op amp have you ever seen that didn't, and how do you know? Such a
device might be quite interesting in applications, but I don't know of one.
Even femtoamp leakage devices (e.g. the venerable LMC660) have
protection diodes. You can use them for resetting integrators, by
briefly inverting the power rails (with suitable current limits). Win
Hill posted a circuit for that in this very group a few years back.
There are folks here who post stuff that's way beyond their
competence...you've been here long enough to know who some of them are.
Joining that group isn't necessarily the best way to achieve respect.
Just saying.
Ebrahim
> Ebrahim wrote:
> What op amp have you ever seen that didn't, and how do you know? Such a
> device might be quite interesting in applications, but I don't know of one.
> Even femtoamp leakage devices (e.g. the venerable LMC660) have
> protection diodes. You can use them for resetting integrators, by
> briefly inverting the power rails (with suitable current limits). Win
> Hill posted a circuit for that in this very group a few years back.
>
***
Hello Phil
Look at data sheet of OPA627, It has shown equivalent schematic. I do
not see any protection diodes in inputs.
> There are folks here who post stuff that's way beyond their
> competence...you've been here long enough to know who some of them are.
> Joining that group isn't necessarily the best way to achieve respect.
> Just saying.
I really do not understand what you are talking about.
I'm not here to join some internal groups, I'm not here to woo members.
I'm not aware of factions of this newsgroup and honestly I do not care
about them if they exist. I'm here to discuss technical issues. I never
claim something that I'm not. I ask questions and if I know something
about a topic I post my opinion. If someone provided documents and
proved that I'm wrong about something, then I learn from him and thank
him, provided that he present his reasons politely and respectfully.
I do not remember I've claimed anything here. so your post really
surprised me.
regards,
Ebrahim
John Larkin
wrote:
>Phil Hobbs wrote:
>> Ebrahim wrote:
>
>> What op amp have you ever seen that didn't, and how do you know? Such a
>> device might be quite interesting in applications, but I don't know of one.
> > Even femtoamp leakage devices (e.g. the venerable LMC660) have
> > protection diodes. You can use them for resetting integrators, by
> > briefly inverting the power rails (with suitable current limits). Win
> > Hill posted a circuit for that in this very group a few years back.
> >
>***
>Hello Phil
>
>Look at data sheet of OPA627, It has shown equivalent schematic. I do
>not see any protection diodes in inputs.
>
An ohmmeter will tell. If it has them, all you need to do is limit the
current.
If not, use the c-b junctions of a BFT25.
John
Robert Baer
bright LED EL-334-15/T2C2-1TVB:
<3nA @4V reverse (seems linear from 0 at 0V to there and then tends to
curve up near 4.5V); <400nA forward.
A bit light sensitive either way; no bright lighting on it.
Maybe for picoamps at low capacitance try various NPN and PNP
transistors; E-B and C-B junctions.
Robert Baer
conditions and test to what is reasonable like i did (4V or less in
reverse); see response posted a bit earlier.
Robert Baer
> Ebrahim wrote:
>> Robert Baer wrote:
>>> Ebrahim wrote:
>>> Do not think BJTs in general are that good; try an LED...
>>
>> Hello Robert :)
>> Thanks for replying to my post. I have not measured leakage current of
>> an LED but it makes sense because GaAs Reverse saturation current is
>> about 10 times less than Si. Have you used LED as an ultra low leakage
>> diode ? have you measured leakage currents of LEDs?
>>
>> Regards,
>> Ebrahim
>
> It's far lower than that. I've measured ordinary red display LEDs whose
> leakage was below 100 fA from -5V to +0.5V bias.
>
> Cheers
>
> Phil Hobbs
>
That shows red LEDs to be far superior to white ones (where i got
<3nA @ 4V reverse); the "cost" prolly means that it must be protected
from all ambient light.
Robert Baer
posting stuff from investment newsletters...
Phil Hobbs
White LEDs are made of GaN--not the world's greatest semiconductor.
Cheers
Phil Hobbs
Robert Baer
> Phil Hobbs wrote:
>> Ebrahim wrote:
>
>> What op amp have you ever seen that didn't, and how do you know? Such
>> a device might be quite interesting in applications, but I don't know
>> of one.
> > Even femtoamp leakage devices (e.g. the venerable LMC660) have
> > protection diodes. You can use them for resetting integrators, by
> > briefly inverting the power rails (with suitable current limits). Win
> > Hill posted a circuit for that in this very group a few years back.
> >
> ***
> Hello Phil
>
> Look at data sheet of OPA627, It has shown equivalent schematic. I do
> not see any protection diodes in inputs.
in many cases will imply possible operation in regions that just cannot
be achieved with a real device.
Almost all analog and digital devices have a reverse polarity
protection diode "built in" under the bonding pad and many have the same
at the outputs; this appears to be standard practice and is ASS-u-ME-ed
known and therefore not alluded to by most people in the industry.
In a number of cases, an additional diode may be added to the supply
rail and/or the doping of the pad-diode may be adjusted for a (say)
zener breakdown of 5-8V.
Cannot say about that as i have not tried to measure this presumption
- and again industry is fairly quiet..
Ebrahim
> * "Equivalent schematics" AFAIK have never shown protection diodes, and
> in many cases will imply possible operation in regions that just cannot
> be achieved with a real device.
> Almost all analog and digital devices have a reverse polarity
> protection diode "built in" under the bonding pad and many have the same
> at the outputs; this appears to be standard practice and is ASS-u-ME-ed
> known and therefore not alluded to by most people in the industry.
> In a number of cases, an additional diode may be added to the supply
> rail and/or the doping of the pad-diode may be adjusted for a (say)
> zener breakdown of 5-8V.
> Cannot say about that as i have not tried to measure this presumption
> - and again industry is fairly quiet..
***
Robert:
They do look at Op27
datasheet(http://www.analog.com/en/other/militaryaerospace/op27/products/product.html)
it also has shown equivalent schematics. I can show you many BJT opamp
that have shown protection diodes and also can show you many ultra low
bias current opamps without protection.
If it is assumed to be known there should be some references to show
that can prove this. I couldn't have find one. I'm not expecting you to
show me a reference but I expect people that accuse me of "going beyond
competence" just because I posted something that I assume is true to
provide references to back their claims. I have provided my references:
2 datasheets op27 and opa627(I can post many more).
Regards,
Ebrahim
Robert Baer
mi...@sushi.com
Getting back to the basic question, breakdown voltages are only
guaranteed at a minimum. Suppose you get a device that has exceedingly
good breakdown?
Ebrahim
> Getting back to the basic question, breakdown voltages are only
> guaranteed at a minimum. Suppose you get a device that has exceedingly
> good breakdown?
I did not understand what is your mean. will you explain more please ?
Regards,
Ebrahim
Phil Allison
"Ebrahim"
>
> I did not understand what is your mean. will you explain more please ?
Winfield Hill
Ahem. Well, in 32 posts, has anyone answered Ebrahim's
question? Which was, does a C-B junction make a good
low-leakage high-voltage diode, and my answer is yes.
Likely as low as 10pA, but I'd want to check specific parts.
It's the same question one asks before using a transistor at
extremely low collector currents. OK, not too many people
use BJT transistors at 10pA, for one thing they're very slow,
but it's surprising how well they work down in that territory.
Often when doing this trick, one wants very low capacitance,
and I've been disappointed with the high capacitance of the
usual candidates, e.g., 5pF. This is what steers people
back to the old PAD-1 beauties, spec'd at 0.5pF and 0.3pA
When one is in this pA territory, PCB leakage can be a big
problem, and teflon posts and in-air wiring are often used.
Phil Allison
"Winfield Hill"
"Phil Allison"
"Ebrahim"
>
>> I did not understand what is your mean. will you explain more please ?
>
> ** You are a totally stupid pile of autistic shit, a waste of space a
> waster
> of good people's valuable time.
>
> Fuck off and DROP DEAD !!!
Ahem. Well, in 32 posts, has anyone answered Ebrahim's
question? Which was, does a C-B junction make a good
low-leakage high-voltage diode, and my answer is yes.
Likely as low as 10pA, but I'd want to check specific parts.
** Since when is C-B junction leakage at high voltages anything like that
low ??
You are bulshitting wildly - as per fucking usual.
It's the same question one asks before using a transistor at
extremely low collector currents.
** What absolute bollocks.
The two matters are totally independent issues.
Winfield has long be a posturing, fucking PITA, pseudo-academic pommy prig.
Now, his rampant congenital autism has rendered him totally useless.
I know he will not comprehend a single word of this.
The mentally diseased never do.
.... Phil
Ebrahim
> Ahem. Well, in 32 posts, has anyone answered Ebrahim's
> question? Which was, does a C-B junction make a good
> low-leakage high-voltage diode, and my answer is yes.
> Likely as low as 10pA, but I'd want to check specific parts.
>
> It's the same question one asks before using a transistor at
> extremely low collector currents. OK, not too many people
> use BJT transistors at 10pA, for one thing they're very slow,
> but it's surprising how well they work down in that territory.
>
> Often when doing this trick, one wants very low capacitance,
> and I've been disappointed with the high capacitance of the
> usual candidates, e.g., 5pF. This is what steers people
> back to the old PAD-1 beauties, spec'd at 0.5pF and 0.3pA
>
> When one is in this pA territory, PCB leakage can be a big
> problem, and teflon posts and in-air wiring are often used.
***Hello Winfield :) :
Thank you very much for your technical and non-political answer.
So it makes sense to look for such a candidate among high frequency
transistors that should have low capacitance.
Thank you again.
Best wishes for you,
Ebrahim
Jamie
You're correct Phil, the mentally diseased normally don't understand
each other!
Touch�!
Phil Hobbs
The part of this that makes zero sense is the 100V part. Almost all op
amps, in fact AFAIK every single one ever sold, has PN junctions between
its inputs and its supply pins.(*) That's why the Absolute Maximum
Ratings section of datasheets specifies that you mustn't take their
inputs more than 0.3V outside the supplies.
Not everything on an IC maps well onto a schematic--e.g. schematics
don't show the parasitic SCR inherent in junction-isolated CMOS
processes. It's there, though, as you'll find if you try dumping 20 mA
into an input.
If you want the same sort of function provided by the input protection
diodes, without a big leakage problem, you can use something like this:
0 VDD
|
_
A
|
*---Ri------*--------Rf-----------*
| | |
GGG _ |
A |
| |
| |\ |
0-------------RRRRR---*---RRRRR---|+\ |
| \ |
| >---*-*---0
*---|- / |
| | / |
| |/ |
| |
*---Ri------*---Rf------*
|
GGG
(Noninverting shown because inverting is reasonably obvious, and
negative polarity left as an exercise for the reader.) You have to
scale the resistances to fit your problem. The idea is that the first
diode has nearly no voltage across it, and hence no leakage or
capacitive current. This isn't a complete solution, because it costs
you SNR eventually, but it's a lot better than nothing.
The diode needs almost no breakdown voltage at all--certainly not 100V.
Something like a BFT25 B-C junction (about 0.3 pF) will work great.
(WDNNS PAD-1!)
Cheers
Phil Hobbs
(*) Some chips, e.g. old CMOS->TTL converters use Zeners instead,
because their inputs are intended to be overdriven by volts. Also, some
rail-to-rail input op amps use charge pumps to allow them to run their
input stages on more than VDD-VSS.
John Larkin
<hi...@rowland.org> wrote:
>On Jun 14, 12:32�am, "Phil Allison" <phi...@tpg.com.au> wrote:
>> "Ebrahim"
>>
>>> I did not understand what is your mean. will you explain more please ?
>>
>> ** You are a totally stupid pile of autistic shit, a waste of space a waster
>> of good people's valuable time.
>>
>> � �Fuck off and DROP DEAD �!!!
>
> Ahem. Well, in 32 posts, has anyone answered Ebrahim's
> question?
Yes, I did.
Which was, does a C-B junction make a good
> low-leakage high-voltage diode, and my answer is yes.
> Likely as low as 10pA, but I'd want to check specific parts.
ftp://jjlarkin.lmi.net/BFT25.JPG
The c-b capacitance of the BFT25 is around 0.5 pF.
The forward and reverse curves should converge around zero volts; I
think I may have had 20 or 30 fA of offset in my test rig.
I've used 2N4402's below 1 pA... not sure how much below.
>
> It's the same question one asks before using a transistor at
> extremely low collector currents. OK, not too many people
> use BJT transistors at 10pA, for one thing they're very slow,
> but it's surprising how well they work down in that territory.
>
> Often when doing this trick, one wants very low capacitance,
> and I've been disappointed with the high capacitance of the
> usual candidates, e.g., 5pF. This is what steers people
> back to the old PAD-1 beauties, spec'd at 0.5pF and 0.3pA
The PAD parts are tiny jfets, so have high channel resistances, a few
K typically. So they're not very hard clamps.
>
> When one is in this pA territory, PCB leakage can be a big
> problem, and teflon posts and in-air wiring are often used.
Did you see my fA tester box?
ftp://jjlarkin.lmi.net/99S260A.JPG
ftp://jjlarkin.lmi.net/99A260A3.JPG
ftp://jjlarkin.lmi.net/99A260A1.JPG
ftp://jjlarkin.lmi.net/PAD5.JPG
John
Ebrahim
> The part of this that makes zero sense is the 100V part. Almost all op
> amps, in fact AFAIK every single one ever sold, has PN junctions between
> its inputs and its supply pins.(*) That's why the Absolute Maximum
> Ratings section of datasheets specifies that you mustn't take their
> inputs more than 0.3V outside the supplies.
> Not everything on an IC maps well onto a schematic--e.g. schematics
> don't show the parasitic SCR inherent in junction-isolated CMOS
> processes. It's there, though, as you'll find if you try dumping 20mA
> into an input.
***
Hello Phil:
I think I explained 100V in an earlier post in this thread I quote it again:
"100V is needed for another use(protecting a high voltage discrete
amplifier), not protecting FET opamp. I was hoping that I could find one
low leakage diode for both uses. though for the first application(HV
Discrete amp) I can live with some leakage"
I have also a High voltage discrete op amp and it has nothing to do with
opa627. it is another part of a system. input voltage of up to 100V is
allowed in there. It has a bootstrapped FET input stage. I may want to
ask a question about it later so maybe I post the schematics of it but
first I have to learn ASCII schematics ;) .
About reason of why input voltage shouldn't be more than 0.3 outside the
supplies, well I did not know that and your reason seems reasonable. I
will investigate about that. Thanks.
> If you want the same sort of function provided by the input protection
> diodes, without a big leakage problem, you can use something like this:
>
> 0 VDD
> |
> _
> A
> |
> *---Ri------*--------Rf-----------*
> | | |
> GGG _ |
> A |
> | |
> | |\ |
> 0-------------RRRRR---*---RRRRR---|+\ |
> | \ |
> | >---*-*---0
> *---|- / |
> | | / |
> | |/ |
> | |
> *---Ri------*---Rf------*
> |
> GGG
>
> (Noninverting shown because inverting is reasonably obvious, and
> negative polarity left as an exercise for the reader.) You have to
> scale the resistances to fit your problem. The idea is that the first
> diode has nearly no voltage across it, and hence no leakage or
> capacitive current. This isn't a complete solution, because it costs
> you SNR eventually, but it's a lot better than nothing.
>
> The diode needs almost no breakdown voltage at all--certainly not 100V.
> Something like a BFT25 B-C junction (about 0.3 pF) will work great.
> (WDNNS PAD-1!)
>
***
That's an interesting idea, you actually bootstrapped the first diode.
Are you sure that any diode can do this? I mean I think I have done
some measurements with 1N4007 and even in ~0 volt, its leakage was high.
I don't remember the exact numbers, though I must find my papers.
> (*) Some chips, e.g. old CMOS->TTL converters use Zeners instead,
> because their inputs are intended to be overdriven by volts. Also, some
> rail-to-rail input op amps use charge pumps to allow them to run their
> input stages on more than VDD-VSS.
Thanks Phil Hobbs
Regards,
Ebrahim
Ebrahim
Hello John,
Thanks a lot :).
Your posts are very useful and give me a lot of inspirations.
Best wishes for you,
Regards,
Ebrahim
Phil Hobbs
Not a crisis. I used this trick ten years or so ago in the Footprints
system, which reduced the cost of a low-resolution thermal imager by two
orders of magnitude while maintaining competitive performance. It took
advantage of the photosensitivity of the LEDs as well, to supply a few
picoamps of bias current under processor control. Ironically, if it had
been only 0.5-1 order of magnitude cheaper instead of two, it would have
sold much better. Some people won't let you save them money. :(
(I've mentioned it here a few times before--for anyone who isn't tired
of hearing about it, the gory details are at
http://electrooptical.net/www/footprints/fpspie11.pdf.)
Winfield Hill
<jjlar...@highNOTlandTHIStechnologyPART.com> wrote:
> On Sun, 14 Jun 2009 03:46:42 -0700 (PDT), Winfield Hill
>
Very nice, John. Excellent!
Phil Hobbs
> Phil Hobbs wrote:
>> The diode needs almost no breakdown voltage at all--certainly not
>> 100V. Something like a BFT25 B-C junction (about 0.3 pF) will work
>> great. (WDNNS PAD-1!)
>>
> ***
> That's an interesting idea, you actually bootstrapped the first diode.
> Are you sure that any diode can do this? I mean I think I have done
> some measurements with 1N4007 and even in ~0 volt, its leakage was high.
> I don't remember the exact numbers, though I must find my papers.
Rectifiers intended for 60 Hz applications are very slow, both slow to
turn off and (interestingly) slow to turn on. They also have lots of
capacitance.
If you're doing something slow, and working at very low impedance
levels, then a 1N4007 could be the right part in this circuit.
Personally if it was as slow as that I'd probably be wanting to use a
Polyfuse and a Transzorb.
A diode has a nonzero conductance at zero bias--for an ideal diode, the
zero-bias conductance is Is*kT/e, as you can easily verify by
differentiating the diode equation. Bootstrapping will reduce the
current, but won't reduce the noise. Pick something sensible--you've
had several suggestions--and build something. Then tell us how it worked.
Cheers
Phil Hobbs
John Larkin
<pcdhSpamM...@electrooptical.net> wrote:
>Ebrahim wrote:
>> Phil Hobbs wrote:
><snip>
>>> The diode needs almost no breakdown voltage at all--certainly not
>>> 100V. Something like a BFT25 B-C junction (about 0.3 pF) will work
>>> great. (WDNNS PAD-1!)
>>>
>> ***
>> That's an interesting idea, you actually bootstrapped the first diode.
>> Are you sure that any diode can do this? I mean I think I have done
>> some measurements with 1N4007 and even in ~0 volt, its leakage was high.
>> I don't remember the exact numbers, though I must find my papers.
>
>Rectifiers intended for 60 Hz applications are very slow, both slow to
>turn off and (interestingly) slow to turn on. They also have lots of
>capacitance.
We make one gadget whose operation involves forward biasing a big P-N
junction diode by 48 volts.
John
Phil Hobbs
Yeah, but you're insane.
Cheers
Phil Hobbs
John Larkin
As I've mentioned, the damned RatShack banana terminal things turned
out to be conductive to the tune of several nA, so I had to add the
lexan insulator after the thing was all built. That was messy... chips
and cutting fluid everywhere in my circuit.
And I recently received my ebay Keithley electrometer. It has a nice
mirror-scale meter with lowest range 1e-14 amps full-scale.
I'm doing a preamp for a gadget that's a 10 pF source, an antenna of
sorts, audio to 20 MHz, nanovolt signals, no DC allowed to accumulate
on the sensor (screws up spins or something), horrible transmit
overloads, runs in UHV. So I'm groveling around with fA's and nV's and
surface-mount Tohm resistors.
John
John Larkin
<pcdhSpamM...@electrooptical.net> wrote:
Thanks, but it doesn't come natural; a guy's gotta work at it.
John
mi...@sushi.com
I'm not so sure about zener input protection. Generally when you need
to go above the rail, you just use a N mosfet with gate tied to source
and the drain going to the pad. You depend on the breakdown of the fet
to provide input protection. [If you are trying this in layout,
antenna rules may apply.]
Every time I investigated chip damage after ESD testing, it was always
due to over-voltage versus over current. I'm not a fan of high fields
(as found in reverse bias) in semiconductors. They seem to find the
weak spot in the junction and zap it. Forward biased junction might
have hot spots, i.e. current hogging, but in general they are pretty
rugged.
mi...@sushi.com
A manufacturer sells a device that passes a minimum breakdown. It is a
one sided specification. That is, a minimum voltage is specified, but
not a maximum.
Maybe the breakdown voltage will be much higher than where you want it
to be.
I see all sorts of ideas in this forum on how to cheat and make a part
do something that it is not intended to do. Fine, but you are on your
own. I've seen specifications of parts change wildly over the years.
Case in point was a SC filter which had a pretty good THD. The THD was
not a guaranteed specification, but there was an internal spec to make
sure things didn't go wild. Well somebody changed epi vendors, and the
THD went about the internal limit. Since the parts didn't have a THD
spec, they went out into the public, though the old epi vendor was
used again.
If you want a bullet proof design, take every line item without limits
and blot them out with a sharpie, then go design.
George Herold
<jjlar...@highNOTlandTHIStechnologyPART.com> wrote:
> On Sat, 13 Jun 2009 21:48:27 +0430, Ebrahim <Engineer....@gmail.com>
> wrote:
>
>
>
>
>
> >Phil Hobbs wrote:
> >> Ebrahim wrote:
>
> >> device might be quite interesting in applications, but I don't know of one.
> > > Even femtoamp leakage devices (e.g. the venerable LMC660) have
> > > protection diodes. You can use them for resetting integrators, by
> > > briefly inverting the power rails (with suitable current limits). Win
> > > Hill posted a circuit for that in this very group a few years back.
>
> >***
> >Hello Phil
>
> >Look at data sheet of OPA627, It has shown equivalent schematic. I do
> >not see any protection diodes in inputs.
>
> current.
>
> If not, use the c-b junctions of a BFT25.
>
> John- Hide quoted text -
>
> - Show quoted text -
Cool, Thanks John I'm going to have to probe the inputs of some
opamps. I've always assumed they had protection diodes on the front
end. I should be able to find the series resistance if I'm careful.
George H.
George Herold
>
> - Show quoted text -
Wow, Thanks Phil, I've scribbled the circuit down in my notebook and
I'll analyze it if I ever need low leakage protection.
George H.
Phil Hobbs
Transzorbs are pretty bulletproof--its job would be just to limit the
overvoltage across the series protection resistor long enough for the
polyfuse to switch. The resistor and the input diodes would protect the
input devices, at least if it were done right.
Cheers
Phil Hobbs
Michael A. Terrell
Phil Hobbs wrote:
>
> Yeah, but you're insane.
You say that like its a bad thing.
--
You can't have a sense of humor, if you have no sense!
mi...@sushi.com
<pcdhSpamMeSensel...@electrooptical.net> wrote:
I am referring to on-chip protection and failure analysis I've done.
I've used transorbs, but only on power rails. In any event, putting
zeners on chip isn't done to my knowledge, at least in MOS/BiCMOS.
See the slide "ground gate NMOS"
http://www.icims.csl.uiuc.edu/icap00/sjoshi_icap00.pdf
Phil Hobbs
> Phil Hobbs wrote:
>> Yeah, but you're insane.
>
>
> You say that like its a bad thing.
>
>
delusions actually work.
Phil Hobbs
>>>> Something like a BFT25 B-C junction (about 0.3 pF) will work great..
I'm sure it isn't done anymore, but check out the CD4049UB for an
example. http://www.fairchildsemi.com/ds/CD%2FCD4049UBC.pdf, page 2.
John Larkin
<pcdhSpamM...@electrooptical.net> wrote:
>Michael A. Terrell wrote:
>> Phil Hobbs wrote:
>>> Yeah, but you're insane.
>>
>>
>> You say that like its a bad thing.
>>
>>
>Not at all. John is the most entertaining wild man I know--his
>delusions actually work.
>
By design. Some of the most creative people have also been
schizophrenics. In schizophrenia, the barriers between brain
compartments become leaky, and ideas diffuse into different contexts.
Very often, they make no sense when they get there. Moderated
schizophrenia is great for systems and electronic design, churning up
all sorts of ideas. People can even do it in groups.
The next stage is to winnow out the practical ideas from the insane
ones. Last step is brutal, disciplined, cost-effective, no-risk
implementation of the one that will sell best.
Not many people are good at spanning this range. But it can be taught
and practiced.
"Crazy like a fox" is the popular version.
John
Jim Thompson
Schematic "representations" on data sheets are often just that,
"representations". The "equivalent" IS done using a capacitively
coupled NMOS.
...Jim Thompson
--
| James E.Thompson, P.E. | mens |
| Analog Innovations, Inc. | et |
| Analog/Mixed-Signal ASIC's and Discrete Systems | manus |
| Phoenix, Arizona 85048 Skype: Contacts Only | |
| Voice:(480)460-2350 Fax: Available upon request | Brass Rat |
| E-mail Icon at http://www.analog-innovations.com | 1962 |
I love to cook with wine Sometimes I even put it in the food
Phil Hobbs
I'm not surprised that it isn't a normal zener--but those parts don't
have normal protection diodes, which is where I was going with it. The
input of a CD4049 isn't the most robust thing in the universe, largely
because of having the weird protection network.
Jim Thompson
<pcdhSpamM...@electrooptical.net> wrote:
Most "modern" CMOS parts don't have classic diodes, particularly on
output pins.
Phil Hobbs
What do they use exactly? Unless they're SOI, there's still the
substrate diode.
Cheers
Phil Hobbs
Jim Thompson
<pcdhSpamM...@electrooptical.net> wrote:
On the outputs, yes there are the body diodes... _unless_ they are
tri-state-when-unpowered, then things get a bit weird.
You'd be stunned at the complexity of today's CMOS protection
circuits. I'll dig thru my drawings and see if there's something I
can legally post.
David DiGiacomo
Jim Thompson <To-Email-Use-Th...@My-Web-Site.com> wrote:
>>> I am referring to on-chip protection and failure analysis I've done.
>>> I've used transorbs, but only on power rails. In any event, putting
>>> zeners on chip isn't done to my knowledge, at least in MOS/BiCMOS.
>>>
>>> See the slide "ground gate NMOS"
>>> http://www.icims.csl.uiuc.edu/icap00/sjoshi_icap00.pdf
>>>
>>
>>I'm sure it isn't done anymore, but check out the CD4049UB for an
>>example. http://www.fairchildsemi.com/ds/CD%2FCD4049UBC.pdf, page 2.
>
>"representations". The "equivalent" IS done using a capacitively
>coupled NMOS.
We are talking about parts that were designed 40 years ago. The RCA
COSMOS process was not dense enough to use MOSFETs for protection devices
(couldn't fit 6 more on the chip), so they used zeners. No doubt they
felt comfortable with that because of experience with their consumer
bipolar ICs that had on chip zener regulators.
Also, they did not bother with "representations". If you look at die
photos of old RCA parts like the 4000A series, they match the datasheet
schematics.
Of course it's likely that second sources changed the protection circuits
(don't they always mess with them?).
--
David DiGiacomo, San Francisco, CA da...@slack.com
Phil Hobbs
Well, we Norwegian Blues stun easily. ;)
mi...@sushi.com
Web-Site.com> wrote:
> On Mon, 15 Jun 2009 12:28:40 -0400, Phil Hobbs
>
>
>
>
>
>
>
> <pcdhSpamMeSensel...@electrooptical.net> wrote:
> >Jim Thompson wrote:
> >> On Mon, 15 Jun 2009 11:29:40 -0400, Phil Hobbs
>
> >>> Cheers
>
> >>> Phil Hobbs
>
> >> Schematic "representations" on data sheets are often just that,
> >> "representations". The "equivalent" IS done using a capacitively
> >> coupled NMOS.
>
> >> ...Jim Thompson
>
> >I'm not surprised that it isn't a normal zener--but those parts don't
> >have normal protection diodes, which is where I was going with it. The
> >input of a CD4049 isn't the most robust thing in the universe, largely
> >because of having the weird protection network.
>
> >Cheers
>
> >Phil Hobbs
>
> Most "modern" CMOS parts don't have classic diodes, particularly on
> output pins.
>
> ...Jim Thompson
> --
> | James E.Thompson, P.E. | mens |
> | Analog Innovations, Inc. | et |
> | Analog/Mixed-Signal ASIC's and Discrete Systems | manus |
> | Phoenix, Arizona 85048 Skype: Contacts Only | |
> | Voice:(480)460-2350 Fax: Available upon request | Brass Rat |
>
> I love to cook with wine Sometimes I even put it in the food
A lot of times you just plop in an "off" transistor when a diode would
do, simply to get past layout verification tools.
Occasionally you see this done with a P-FET do discharge a node using
a parasitic diode from the positive rail.
Jim Thompson
[snip]
>
>A lot of times you just plop in an "off" transistor when a diode would
>do, simply to get past layout verification tools.
Yep. An NMOS, gate to ground thru a resistor, source to ground, drain
to pin.... sometimes fancier gimmicks.
>
> Occasionally you see this done with a P-FET do discharge a node using
>a parasitic diode from the positive rail.
I have seen 10-15 device structures for protecting outputs when they
are configured as tri-state-when-unpowered ;-)
...Jim Thompson
--
| James E.Thompson, P.E. | mens |
| Analog Innovations, Inc. | et |
| Analog/Mixed-Signal ASIC's and Discrete Systems | manus |
| Phoenix, Arizona 85048 Skype: Contacts Only | |
| Voice:(480)460-2350 Fax: Available upon request | Brass Rat |
| E-mail Icon at http://www.analog-innovations.com | 1962 |
Jim Thompson
Forgot a point, seen only recently... a special schematic symbol that
describes layers to satisfy the LVS.
Fred Bartoli
Why the resistor?
>> Occasionally you see this done with a P-FET do discharge a node using
>> a parasitic diode from the positive rail.
>
> I have seen 10-15 device structures for protecting outputs when they
> are configured as tri-state-when-unpowered ;-)
>
> ...Jim Thompson
--
Thanks,
Fred.
Jim Thompson
>Jim Thompson a �crit :
>> On Mon, 15 Jun 2009 15:13:04 -0700 (PDT), "mi...@sushi.com"
>> <mi...@sushi.com> wrote:
>>
>> [snip]
>>> A lot of times you just plop in an "off" transistor when a diode would
>>> do, simply to get past layout verification tools.
>>
>> Yep. An NMOS, gate to ground thru a resistor, source to ground, drain
>> to pin.... sometimes fancier gimmicks.
>>
>
>Why the resistor?
Rise-time dependent clamping.
>
>>> Occasionally you see this done with a P-FET do discharge a node using
>>> a parasitic diode from the positive rail.
>>
>> I have seen 10-15 device structures for protecting outputs when they
>> are configured as tri-state-when-unpowered ;-)
>>
>> ...Jim Thompson
...Jim Thompson
Robert Baer
> Robert Baer wrote:
>
>> * "Equivalent schematics" AFAIK have never shown protection diodes,
>> and in many cases will imply possible operation in regions that just
>> cannot be achieved with a real device.
>> Almost all analog and digital devices have a reverse polarity
>> protection diode "built in" under the bonding pad and many have the
>> same at the outputs; this appears to be standard practice and is
>> ASS-u-ME-ed known and therefore not alluded to by most people in the
>> industry.
>> In a number of cases, an additional diode may be added to the supply
>> rail and/or the doping of the pad-diode may be adjusted for a (say)
>> zener breakdown of 5-8V.
>> Cannot say about that as i have not tried to measure this
>> presumption - and again industry is fairly quiet..
> ***
> Robert:
> They do look at Op27
> datasheet(http://www.analog.com/en/other/militaryaerospace/op27/products/product.html)
> it also has shown equivalent schematics. I can show you many BJT opamp
> that have shown protection diodes and also can show you many ultra low
> bias current opamps without protection.
>
> If it is assumed to be known there should be some references to show
> that can prove this. I couldn't have find one. I'm not expecting you to
> show me a reference but I expect people that accuse me of "going beyond
> competence" just because I posted something that I assume is true to
> provide references to back their claims. I have provided my references:
> 2 datasheets op27 and opa627(I can post many more).
>
> Regards,
> Ebrahim
>
The protection diodes shown are between inputs.
Related to inputs, note those to the substrate (V-) and if
implemented, those to the supply are most definitely not shown.
For both digital and analog devices, note protection diodes to the
substrate are not shown.
And diodes to the supply (if added) seem to be almost state secrets..
Jim Thompson
There's a good OP27 model on my website... one I did in the early
'90's.
...Jim Thompson
--
| James E.Thompson, P.E. | mens |
| Analog Innovations, Inc. | et |
| Analog/Mixed-Signal ASIC's and Discrete Systems | manus |
| Phoenix, Arizona 85048 Skype: Contacts Only | |
| Voice:(480)460-2350 Fax: Available upon request | Brass Rat |
| E-mail Icon at http://www.analog-innovations.com | 1962 |
Obama... Recklessness Cloaked in Righteousness
James Arthur
> On Mon, 15 Jun 2009 11:22:32 -0400, Phil Hobbs
> <pcdhSpamM...@electrooptical.net> wrote:
>
>> Michael A. Terrell wrote:
>>> Phil Hobbs wrote:
>>>> Yeah, but you're insane.
>>>
>>> You say that like its a bad thing.
>>>
>>>
>> Not at all. John is the most entertaining wild man I know--his
>> delusions actually work.
>>
>
> By design. Some of the most creative people have also been
> schizophrenics. In schizophrenia, the barriers between brain
> compartments become leaky, and ideas diffuse into different contexts.
> Very often, they make no sense when they get there. Moderated
> schizophrenia is great for systems and electronic design, churning up
> all sorts of ideas. People can even do it in groups.
>
> The next stage is to winnow out the practical ideas from the insane
> ones. Last step is brutal, disciplined, cost-effective, no-risk
> implementation of the one that will sell best.
I'm not sure how other people's brains work, but mine is some sort
of massively-parallel possibility generator, followed by a
does-it-make-sense filter--cultivated over time--then some I/O.
At least that's its creative process. Possibly related to the
left-handed, quasi-ambidextrous thing--that changes brains.
Thoughts aren't in any spoken language, but in some sort of parallel
machine language where multiple ideas fire in 300mS, which, if
interesting, are sampled / latched in snapshots, separated, sorted,
parallel-to-serial converted, then translated into language for
expression.
Lots of times the thinking parts sit back and listen, amused,
while the talking part spools output.
Talking fast enough to voice the chaos is impossible by factors.
The filter's the difference between madness and sanity.
Weird, huh?
--
Cheers,
James Arthur
John Larkin
Yes. There seems to be some massively parallel idea generator that can
be carefully primed and then let loose. It definitely runs
subconscious, background, often during sleep. It's not quantitative
but can make some pretty good guesses.
Why work at designing when you can do it painlessly in background?
I suspect that DNA designs, say, antibodies by a similar algorithm,
some sort of massive, random set of cross-correlation experiments.
John