LM399 circuits from the datasheet, which one is better?

[dmmguy]

Hi everyone,

Both taken from the LM399 datasheet; which one is better?

Thanks

[KT88]

A salways: It depends...
The right one is optimized for portable (battery operated) use at the cost of higher noise (lower Iz) and drift (it takes some time for the heater to settle).

[branadic]

I would say: "None of them, use the bootstrapped solution instead."

-branadic-

[Gyro]

That would be the second one then wouldn't it? The 200k resistor is just there to make sure it starts up.

[dietert1]

Yes, the circuit shown in the lower right is the usual bootstrap circuit. The 200k resistor is for startup and the zener current comes in from the output via the 5k resistor. A standard value for that resistor would be rather 3 KOhm to pass about 1 mA from 10 V to 7 V.

Regards, Dieter

[Andreas]

That would be the second one then wouldn't it? The 200k resistor is just there to make sure it starts up.

Thats right. (but the 5K resistor should be more in the 3K range to get the specified 1mA current through the LM399)

For real precision cirquit I would improve the trimming scheme (like on other precision references like the AD587).
A pot with rather high T.C. should not be in series with usually better precision resistors.

And also the heater voltage should be stabilized for maximum stability if ppm level is needed.

With best regards

Andreas

[bdunham7]

And also the heater voltage should be stabilized for maximum stability if ppm level is needed.

I've always been curious--is something like an LM7815 considered good enough for all purposes or is it ever worthwhile to use the 10V output with a comparator setup to stabilizer the heater input even more?

[Andreas]

Some results are in the LM399 thread.
e.g. here
https://www.eevblog.com/forum/metrology/lm399-based-10-v-reference/msg443181/#msg443181
so you can decide by yourself for some ppm/V how much stabilisation is needed for your application

with best regards

Andreas