ADR1000 Datasheet

[vmiguel]

Voltnuts Rejoice!

If you got you got, if not is too late

 

[The Soulman]

 

[TimFox]

Pray, what is this ADR1000 of which you speak?

[chuckb]

Feel free to share the datasheet here

[Vgkid]

I swear he had posted it, then deleted it.
Here you go:

[chuckb]

Does anyone have samples yet?

[chuckb]

if the ADA4084 opamp negative power connection is grounded, it is not the best device to use with this voltage reference,  When the common mode input voltage is between 0 and 0.5V the Bias current can suddenly change 200na. The offset voltage has a step change also. These effects are reduce by the gain of the internal transistors but there is no need for that instability in a precision circuit.

The new voltage ref looks great!

[branadic]

Does anyone have samples yet?

We have already seen first result on it

https://www.eevblog.com/forum/metrology/lowest-drift-lowest-noise-voltage-reference/

-branadic-

[Kleinstein]

The suggested ADA4084 is a nice OP, but likely not the best choice for the suggsted circuit. I have not found data for the low frequency current noise, but it could be a problem. The OPs see a rather high source impedance (some 60 K) and current noise thus matters.
The OP is not that critical, so that the difference may not be that obvious.

Chances are the classic LT1013 like in the standard LTZ1000 circuit would be the better choice.

6.6 V ouput voltage mean scaling to 10 V would be a factor very close to 1.5, which could be nice.

[tszaboo]

All right, a few notes:
Nice to see tempco susceptibility in the datasheet, that was something I almost had to guess from the LTZ1000's. They even put thermocouple effects in it, once again, this is great.
The circuit is very similar to the LTZ1000. The end result is lower noise on paper, but the peak to peak noise is almost unchanged. The 0.2ppm tempco seems worse, but I dont think it really is in practical applications. They probably built their circuit without those hermetically sealed VPG foil resistors, as most users will.
ADA4084: Probably a sales decision to put a twice as expensive opampm there, without a TI alternative. I dont blame the engineer.

6.6 V ouput voltage mean scaling to 10 V would be a factor very close to 1.5, which could be nice.

I was just thinking of that. Scaling 7.2V is odd, we either do PWM, or live with the issues of resistor scaling. Scaling 6.6V to 10, we can use the LT5400B and a trimmer, which is a lot cheaper than those bulk metal foil resistors with similar performance.

[ramon]

What a sad day! 

Needs 10mA more than LTZ1000A for the same setting.
And we still have those Kovar leads.

But great day for those who stocked old chips. I regret not having bought if 100 units when they were < $46.06.

[ramon]

Although it could be useful for a 0.48V reference (Dual LTZ1000A - ADR1000).

[Kleinstein]

The curve for the supply current is with a lower temperaure set point for the LTZ1000. Chances are this is just one example and the heater resitor on the LTZ1000 had quite loose specs. So the actual current needed for the LTZ1000A could scatter a lot, though many more can be more in the middle. At least the new heater specs are relatively tight - so the needed current is more predictable.

The noise is slightly lower - though this maybe more the higher frequency noise - not clear how the very low frequency noise behaves. There is quite some scattering anyway ! For some reason they show the noise curves with the heater inactive - so this may include thermal fluctuations.

The big plus is a lower TC before temperature control and thus less demand on the resistors. So the drift and TC contribution from the resistors is reduced by about a factor of 2.

[Echo88]

Thanks Vgkid 

There were rumors about a ADR1000-Version with integrated resistors and OPs, is that bollocks?

[chuckb]

Looks like Keysight can update the 3458A A9 Vref module now and offer a 1ppm / year factory option!
Two sealed Vishay BMF resistors could raise the voltage 10% to match the LTZ1000 voltage with no impact on drift.

With this lower Vref noise the old LT1001 (10nV / rt Hz) amplifiers on the ADC A3 PCB could need updating. I know some good choppers they could use.
The noise at the fundamental ADC frequency of 6Hz is now 1/2 of the old value. So that should help it's operation.

[Kleinstein]

For a direct replacement of an old LTZ1000 reference the 7-8% lower voltage may be an issue.  Using resistors for gain would still add to the drift, more than the resistors at the reference directly.

The drift specs are not that much different from the LTZ1000. The total circuit drift also include some contributions from the resistors - though less with the new ADR ref. Another point is that the typical value in the data-sheet are not yet that well tested. The DMM spec. numbers for the drift should also include hysteresis effects and an extra K factor of some 2 to 3 to get from typical values to expected upper limits. The logical 1st setp to better drift specs would be s slightly lower set temperature at the cost of lower maximum environmental temperature.

Adaption to the lower reference voltage at the 3458 could be done at the ADC:  replace the 7 to 12 V part included in U180 with 2 external resistors. The 2 resistors here don't have to be matched to the other resistors inside U180.
The LT1001 OPs in the 3458 ADC are only a small part of the ADC noise and drift. For 10 PLC mode the relevant frequency would not be 6 Hz, but 3 Hz (or 2.5 Hz with 50 Hz mains) - there are a few possible modern alternatives, even without using choppers.

For some reason the ADR1000 is not found easily on the analog website. So there may be a problem or delay and maybe still some change.

The external OP is the least problem - an LT1013 is not that expensive and large. The weak point is the need for the external resistors (e.g. 70K, 70K, 100 Ohms, 1 K, 13 K). The caps would be very hard to integrate on a chip.

[Echo88]

It wouldnt all fit in a TO99-package, but maybe in a TO3-package like the Apex PA88 https://www.richis-lab.de/images/Opamp/03x02.jpg 

[Kleinstein]

For many used with a relatively large temperature range the reference would run reasonable warm (like 60 C or even more). OP performance is better at low temperature.

With more parts that need power the oven would also need more power.  As a rule of thumb I consider oven power to be about 2 times the power of the circuit inside at least. So ideally only the parts that need the oven and are low power should be kept at temperature - other parts are better outside.
A smaller circuit also makes the oven faster and this helps to keep the capacitor reasonable small without going too high in impedance.

[The Soulman]

For some reason the ADR1000 is not found easily on the analog website. So there may be a problem or delay and maybe still some change.

Couldn't find it at all on the website, did something change or am I not looking hard enough?

[ScoobyDoo]

Couldn't find it at all on the website, did something change or am I not looking hard enough?

Hello folks,

The part is released but ADI will not take it to the web - something ADI rarely does.

Best regards
ScoobyDoo

[The Soulman]

Couldn't find it at all on the website, did something change or am I not looking hard enough?

Hello folks,

The part is released but ADI will not take it to the web - something ADI rarely does.

Best regards
ScoobyDoo

Interesting, will it only be available for "selected" customers?

[ScoobyDoo]

Interesting, will it only be available for "selected" customers?

The ADR1000 is intended for selected customers
The ADR1001 will probably be put online and made available through distributor channels.
The ADR1001 contains all necessary resistors and HW implementation is rather easy

Best regards
ScoobyDoo

[daqq]

The ADR1000 is intended for selected customers

That's confusing. Any idea why?

[ScoobyDoo]

Hello folks - the people from ADI did not divulge a rationale and so any word on this would only be pure speculation ...

However the good news is that most likely the ADR1000 IC team lead will give an online presentation during MM2021 ...

So you all have a good reason to attend MM2021 (Stuttgart) - ( if CoVid rules allow us ... )

Best regards
ScoobyDoo

[branadic]

Damn, now you've leaked the secret surprise talk that I could organize thanks to the help of my german ADI contact! 

-branadic-