[volt-nuts] Some questions to zeners (1N823-1N829)

WarrenS warrensjmail-one at yahoo.com
Tue Jan 29 11:30:58 EST 2013


Andreas

Nice work.
May I suggest one of the next steps you work on is improving the short term 
noise and resolution over say a couple hrs, when at constant temperature, by 
a factor of ten or so, as ultimately that will be what limits the 
performance.

For the long term ageing problem, which will be your major issue after you 
get all the 'easy' things worked out.
Make that a separate issue by using a separate unpowered part(s), i.e don't 
use the working ADC ref.
Power the long term ageing part up only as needed to auto cal the ADC.
With the long term reference mostly off, this can significantly help (fix) 
your  power and ageing issue.
That will make the near impossible task of finding a low power, low long 
term drift voltage ref much easier.
And yes, it will have to be a part with good turn on repeatable and 
hysteresis.
But that will be a heck of a lot easer than estimating the long term ageing 
rate of a constant on part that is in an uncontrolled environment to 1PPM / 
year.

For the long term reference part, one thing to consider is to use the 
average of say 10 aged and selected zeners.
The power on duty cycle can be made so low (<<< 10%)  that you can afford 
the 75 ma or so they take when on.


ws

********************
*******************
Hello Warren  and Volt Nuts:

for the first: from the experiences of Warren my project seems to be 
feasible.
I'm working now since 2008 on my "project".
Eliminating step by step the drawbacks of stability.
And learning a lot of real and not ideal parts.

... snip

With best regards

Andreas





More information about the volt-nuts mailing list