[volt-nuts] Low-cost voltage reference questions

[Andreas Jahn]

Hello,

the below mentioned 3ppm/year are not easy to get (even with expensive 
references).

Your selected parts are both band gap references. Which are poor 
regarding long term stability + noise.
The best references (stability + noise) are buried zener references.
Plastic packages suffer from humidity effects. (even good references can 
have 10-15 ppm shift over 30% rH change).
So with high demands you would need a hermetically sealed reference (at 
best in metal can case)
Further effects are temperature hysteresis and mechanical stress from 
the PCB (or soldering).
Practically you have to pre-age the references (6-12 months) and to sort 
out the
less performing parts if you really want to go below 10 ppm stability.

Book recommendation:
Current Sources and Voltage References: A Design Reference for 
Electronics Engineers (Linden T. Harrison)

With best regards

Andreas

Am 23.11.2015 um 23:41 schrieb Russ Ramirez:
> Excellent points, especially the consideration of shipment effects
> year-round and across major temperature variations. Thank-you Charles.
>
> Russ
>
> On Mon, Nov 23, 2015 at 4:26 PM, Charles Steinmetz <csteinmetz at yandex.com>
> wrote:
>
>> Russ wrote:
>>
>> What is considered the break-over point of precision with low uncertainty
>>> versus cost to a group like this? Is there a rule-of-thumb for the cost of
>>> each additional digit of precision after N digits?
>>>
>> One person's opinion:
>>
>> To a group like this, I'd be inclined to say that interest begins at a
>> room-temperature (say, 20C +/- 3C) accuracy of 3ppm (i.e., guaranteed to
>> remain within 3ppm from 18-22C for at least one year after purchase).  3
>> ppm is 0.0003%.  There is at least one 10v reference with specifications in
>> this ballpark available at an asking price under $130 (I'm told the seller
>> has accepted offers significantly lower than this).
>>
>> If I sell someone a reference
>>> that I've ascertained is 2.50163v @70.3 F with a calculated uncertainty,
>>> is
>>> it valuable as a 0.1% reference even though the error may be much less,
>>> like +/- 0.08%?
>>>
>> I, for one, do not consider 0.08% to be "much less" than 0.1%.  One sneeze
>> and it's out of spec.  Indeed, I would consider a claim of 0.1% accuracy to
>> be bordering on fraudulent based on a calibrated measurement at 0.08%,
>> unless the spec was qualified as "within 0.1% at [temperature within 0.1C]
>> as is, where is -- no claim as to accuracy after it has been shipped to the
>> buyer."
>>
>> Speaking as someone with substantial commercial design experience, I would
>> never offer a voltage reference for sale as a claimed "0.1% standard" that
>> I did not have excellent justification for believing would stay below 0.05%
>> for a year over a several-degree range of temperature and multiple trips
>> across the country via commercial carriers.  I wouldn't expect to be able
>> to charge more than $10-15 for the product just described, and then only if
>> the nominal output voltage were 10v (I think you will find that there is a
>> very strong preference for 10v references over 5v, 2.5v, or other voltages).
>>
>> Best regards,
>>
>> Charles
>>
>>
>> _______________________________________________
>> volt-nuts mailing list -- volt-nuts at febo.com
>> To unsubscribe, go to
>> https://www.febo.com/cgi-bin/mailman/listinfo/volt-nuts
>> and follow the instructions there.
>>
> _______________________________________________
> volt-nuts mailing list -- volt-nuts at febo.com
> To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/volt-nuts
> and follow the instructions there.