Long term drift of voltage reference is unknown, can we guess?

itial tolerance spec

ions/en.DM00104043.pdf

inning of Life)

after 10 years

rom the numbers supplied?

e got insights?

"You can expect buried Zeners to perform better than 20ppm/?kh, and bandgaps between 20ppm and 50ppm/?kh"

initial tolerance spec

ations/en.DM00104043.pdf

eginning of Life)

ft after 10 years

from the numbers supplied?

ere got insights?

Wow, that is good info. A lot lower than I would had expected. Thanks

% initial tolerance spec

(Beginning of Life)

rift after 10 years

ft from the numbers supplied?

here got insights?

That's the kind of stuff we want more of in sci.electronics.design

s (Beginning of Life)

drift after 10 years

rift from the numbers supplied?

ne here got insights?

Maybe if they delve more into the physics of drift. But if all someone want s is a published number from a technical note, it's ... disappointingly vap id.

e:

te:

a 4% initial tolerance spec

ors (Beginning of Life)

he drift after 10 years

drift from the numbers supplied?

yone here got insights?

nts is a published number from a technical note, it's ... disappointingly v apid.

But as good as it is going to get. Before you can exploit your sophisticati on about the physics of drift, you have to know a lot about the constructio n of the junction that might be drifting, which happens to be a trade secre t.

Somebody at Linear Technology (now part of Analog Devices) can be told abou t the details of the construction, and apply their grasp of the physics of drift to make the kind of prediction we can use, without violating any trad e secrets.

I want to get my hands on a VBIC model for a transistor - pretty much any t ransistor would do - to verify the idea that "squegging" in the Baxandall C lass -D oscillator

depends on inverted transistor action. Spice simulations using the Gummel-P oon transistor model don't show squegging, and the VBIC model - which handl es inverted operation rather better - might. Sadly VBIC models are all comm erical in confidence.

Reference spec page:

A super-regenerative receiver is basically a squegging Colpitts oscillator. I had no trouble to obtain the squegging behaviour in the Spice simulation of one, using a BFR92A model published by Philips (now NXP).

Jeroen Belleman

And it can be quite misleading.

--sp

The problem with what Baxandall calls "squegging" in his class-D oscillator is that it seems to have nothing to do with the high frequency effect that he thought that he was seeing. It seems to have been a false analogy.

Boost the inductor as much as you like in a Class-D oscillator, and the Spi ce simulation won't squeg, but the switching transistors will start operati ng in inverted mode at about the point where Baxandall (and I) saw squeggin g happening in real life.

All you have to do is contact the appropriate foundry, and create a designer's account... if you qualify... probably all you incite at the foundries is copious laughter >:-}

(I currently have 120 foundry accounts ;-) ...Jim Thompson

y transistor would do - to verify the idea that "squegging" in the Baxandal l Class -D oscillator

l-Poon transistor model don't show squegging, and the VBIC model - which ha ndles inverted operation rather better - might. Sadly VBIC models are all c ommerical in confidence.

If I could find one that used a Cambridge Instruments electron beam tester. Samsung did have a Lintech electron beam tester that was sold to them by C ambridge Instrument, after I and a couple of other engineers had taken out a few of the bugs in it, but I don't know if they run a silicon foundry the se days.

Sure, and people paid you money to design integrated circuits that got made at those foundries. I designed gear that got used for other jobs, but that 's a while ago now. Since you are looking for work, you may be in the same state, despite the boastful claim.

I don't know much about it but a buried zener is not a "bandgap" AFAIK?

The buried zeners I have seen tend to be ~7.2V things like the LTZ1000, LM399 and LM723(?). They are indeed exceptionally stable.

I would not *automatically* expect the same performance from a bandgap, especially from one implemented on a process mainly designed for flash memory.

(Beginning of Life)

drift after 10 years

ift from the numbers supplied?

e here got insights?

Do pay attention. Wanderer posted a link to a Linear Technology application note

and the excerpt he posted with the link clearly distinguished between band- gaps and buried zeners. It would be nice to have something on the Fet-based references that Analog Devices invented which fit in between.

e:

a 4% initial tolerance spec

rs (Beginning of Life)

e drift after 10 years

drift from the numbers supplied?

one here got insights?

on note

d-gaps and buried zeners. It would be nice to have something on the Fet-bas ed references that Analog Devices invented which fit in between.

More Vapid Engineering

"The XFET architecture offers performance improvements over bandgap and bur ied zener references, particularly for systems where operating current is c ritical, yet drift and noise performance must still be excellent. XFET nois e levels are lower than bandgap based bipolar references operating at an eq

C (allowing easier compensation when required), and the series has lower hy steresis than bandgaps. Thermal hysteresis is a low 50 ppm over a ?

inally, the long-term stability is excellent, typically only 50 ppm/1000 ho urs."

I think I'll start a company called Vapid Engineering and search engine cal led Matthew 7-7:8:)

ote:

, a 4% initial tolerance spec

rors (Beginning of Life)

the drift after 10 years

m drift from the numbers supplied?

nyone here got insights?

ks

K?

p,

tion note

and-gaps and buried zeners. It would be nice to have something on the Fet-b ased references that Analog Devices invented which fit in between.

uried zener references, particularly for systems where operating current is critical, yet drift and noise performance must still be excellent. XFET no ise levels are lower than bandgap based bipolar references operating at an equivalent current, the temperature drift is low and linear at 3-8 ppm/? ?C (allowing easier compensation when required), and the series has lower hysteresis than bandgaps. Thermal hysteresis is a low 50 ppm over a ?

. Finally, the long-term stability is excellent, typically only 50 ppm/1000 hours."

alled Matthew 7-7:8:)

This one is fun.

"Lies About Long-Term Drift" "Some manufacturers are now touting phenomenal long-term drift specifications, based on accelerated high temperature testing. THIS IS A DELIBERATE LIE! Long-term drift cannot be extrapolated from accelerated high temperature testing. The only way long-term drift can be determined is to measure it over the time interval of interest. The erroneous technique produces numbers that are wildly optimistic and uses the Arrhenius Equation to derive an acceleration factor from elevated temperature readings"

ce, a 4% initial tolerance spec

errors (Beginning of Life)

w the drift after 10 years

erm drift from the numbers supplied?

anyone here got insights?

anks

AIK?

gap,

ash

cation note

band-gaps and buried zeners. It would be nice to have something on the Fet

-based references that Analog Devices invented which fit in between.

buried zener references, particularly for systems where operating current is critical, yet drift and noise performance must still be excellent. XFET noise levels are lower than bandgap based bipolar references operating at a n equivalent current, the temperature drift is low and linear at 3-8 ppm/

wer hysteresis than bandgaps. Thermal hysteresis is a low 50 ppm over a ?

ice. Finally, the long-term stability is excellent, typically only 50 ppm/1

called Matthew 7-7:8:)

It's not exactly a lie, it's just a quick way of getting something indicati ve.

The problem is that to use the Arrhenius equation, you have to have to know the activation energy of the process that creates the drift, and it doesn' t have to be the same process at high temperatures as at low temperatures.

It's better than nothing, but that's about it.

What, you want me to read all the way to the end of the sentence?

I think the 723 is more of a surface zener, but I'm not sure.

The geophysics community generally likes the LT1021-7 buried zener.

Cheers

Phil Hobbs

It's an emitter-base breakdown

...Jim Thompson

-- | James E.Thompson | mens | | Analog Innovations | et | | Analog/Mixed-Signal ASIC's and Discrete Systems | manus | | STV, Queen Creek, AZ 85142 Skype: skypeanalog | | | Voice:(480)460-2350 Fax: Available upon request | Brass Rat | | E-mail Icon at

I'm looking for work... see my website.

Thinking outside the box...producing elegant & economic solutions.

My uncertainty is whether the device is far enough from the oxide interface that it's quiet like a buried zener. I suspect not, but I haven't measured it. I should, but then my 250 LM329s arrived yesterday, so I'm not that motivated. ;)

Cheers

Phil Hobbs