Generating a 10v rail

[cosmicray]

I have a need to generate a +10v rail. It is unlikely that this needs to be a precision 10v rail, mostly because it will be driving the Vdd of a HEF4001B (Quad 2-input NOR gate, Vdd range up to 18v), and thence on to the gate of a AO3400 mosfet (Vgs ± 12v). My first inclination was to use a LM317L and the circuit from the National datasheet (below) for the High Stability 10V Regulator. Then I saw the market price for the LM329, and decided to look at other options. LM329 stocks are low, even at the elevated prices.

Option 1 seems to be set up the LM317L as a normal voltage regulator, make sure to include the caps for stability, and a small trimpot so adjust the voltage closer to 10v. Input rail to the LM317L will be 17v-20v battery, current draw will be well within the limits of the LM317L. Input-Output Voltage Differential will not exceed 10v, which is also within the LM317L charts.

Option 2 would be to adapt the National datasheet circuit to use a conventional Zener instead of the LM329.

Stability is an objective, but a 10v precision not so much.

Looking for thoughts on how to proceed.

[pcprogrammer]

Why not just use a trusted UA7810 voltage regulator to do the job. https://www.datasheet4u.com/datasheet-pdf/SEMTECH/7810/pdf.php?id=1097636

[Ian.M]

Also look at various manufacturerers 78L10 regulators (nom. 100mA).   You should be able to get about 50mA continuous out of one with 10V drop without overheating it, which shoud be plenty for gate drive of a logic IC and a SOT-23 MOSFET.

https://www.digikey.com/short/9nz4tf8z

[TimFox]

If the calibration accuracy and precision of the reference are not critical, but you want stability, you do not need to consider the datasheet line for output voltage (which includes production spread), just the other lines for change in output voltage with load current, input voltage, and temperature.
Normal Zener diodes can have substantial incremental resistance, so their voltage will vary considerably with current.
If you want to use a Zener-like shunt regulator, the TL431 with fixed 1% resistors might be a better choice  https://www.ti.com/lit/ds/symlink/tl431.pdf  depending on the current level needed.

[PCB.Wiz]

I have a need to generate a +10v rail. It is unlikely that this needs to be a precision 10v rail, mostly because it will be driving the Vdd of a HEF4001B (Quad 2-input NOR gate, Vdd range up to 18v), and thence on to the gate of a AO3400 mosfet (Vgs ± 12v).

If you already say it is for gate drive, and precision is not important why do you then do this ?

My first inclination was to use a LM317L and the circuit from the National datasheet (below) for the High Stability 10V Regulator.

Any half-decent regulator you can buy these days will likely be fine for gate drive.
The 78L10 (and 78L12, a bit more common ) is a good choice, fine for HEF4001 + gates.

If you are running a ADC off this and need better than 4% , lcsc have Holtek HT75A0-2 and ME6203A100PG at 10V/1% specs.

[Alex Nikitin]

Connect R1 to the output of the regulator not the input, and reduce the value to 1K. And don’t forget capacitors in parallel to input/output (according to the chip data sheet), and preferably in parallel to the zener too!

Cheers

Alex

[Ian.M]

Why aren't you simply using AZ1117C-ADJ and two 1% resistors?  You've got an unnecessary 5% Zener in there and are throwing away stability vs Vin by not biasing it from Vout.

[magic]

Or 78L10 and zero resistors

6.8V zener diodes have some 2mV/°C thermal drift, so at your luckiest you will have a relatively stable thermometer.

[PCB.Wiz]

Parts are somewhere in transit. BOM cost is about 5-cents. If I wanted to get closer to 10.0v, all I need to do is bin a bunch of zener diodes until I find one doing breakdown slightly below 6.8v.

The Iq of AZ1117 is 4mA typ 6mA max, so it can provide the zener current and you do not need the resistor.
You can buy 5% zeners (C6v8) and 2% zeners (B6V8) and 1% zeners (A6V8)

Or, you could use the -ADJ version of the AZ1117 and get your 10V that way.

[SiliconWizard]

There are tons of LDOs out there that can give you a 10V output with good stability, many adjustable, fewer fixed (but they do definitely exist). Some with pretty low Iq, way lower than a few mA.
I've used the SPX3819, which has very low noise and Iq < 100µA. It happens to be cheap as well.

If you were looking for a negative voltage, options are more limited.