hardware

AMS1117-1.2: 1 A Fixed 1.2 V LDO, SOT-223

Public Made by Adom

by adom

AMS AMS1117-1.2 low-dropout linear regulator, fixed 1.2 V output, 1 A output current, 1.3 V max dropout, SOT-223 package with tab on pin 2.

ldoregulator1.2vsot-223ams1117power
Overview Files Packages Releases Dependencies Issues1 Breadcrumbs Pull Requests Activity
Open general

Deprecate in favour of the AMS1117 family page - and the footprint is on the back layers

barrett-land · 5d ago

The new family page adom/ams1117 now covers the AMS1117 part family from the Advanced Monolithic Systems datasheet: all six fixed voltages plus the adjustable version, in SOT-223, TO-252 and SO-8, with the datasheet embedded and full electrical tables.

This page should be deprecated in favour of it, but there is one thing to settle first - see below.

One thing to settle: 1.2 V is not a datasheet voltage

The AMS datasheet lists exactly six fixed voltages: 1.5, 1.8, 2.5, 2.85, 3.3 and 5.0 V. There is no 1.2 V version in it (the ordering table ends with a note to contact the factory for other voltages). So AMS1117-1.2 is a second-source-only part - real, orderable from UMW and others, but not covered by the datasheet the family page is built on.

Pick one before deprecating:

  • (a) Fold it into the family page as a documented second-source variant, with its own provenance line saying the 1.2 V limits do not come from the AMS datasheet. Then deprecate this page.
  • (b) Keep this page as the 1.2 V second-source page, and link it from the family page as a variant that lives outside the datasheet. (The family page already links it this way.)

Also worth fixing either way: the footprint is on the back layers

SOT-223-3_TabPin2.kicad_mod on this page has every element on B.Cu / B.Mask / B.Paste / B.SilkS / B.Fab / B.CrtYd - there is no front-layer geometry at all. Dropped into a board it lands as a bottom-side part. It looks like a flipped copy of KiCad's stock footprint got saved.

The family page ships the unmodified KiCad 10 SOT-223-3_TabPin2 on F.Cu as ams1117-sot223.kicad_mod. Pin numbering is otherwise identical, so it is a drop-in swap.

Blocked on

  • Decide (a) or (b) above.
  • adom/large-automotive-eth-switch-molecule#1 - repoint its adom/ams1117-1-2 dependency, and pick up the front-layer footprint while doing it (U902 may currently be on the wrong side).

Related: the sibling deprecation for adom/ams1117-3-3.

2 Replies

barrett-land · 5d ago

Update — the blocker above is resolved. Option (a) is now done.

The premise of my "1.2 V is not a datasheet voltage" note was wrong, or rather incomplete. 1.2 V is a documented AMS1117 version — just not in the Advanced Monolithic Systems datasheet. It is in the UMW / Youtai Semiconductor datasheet, which is a genuine second-source datasheet for the same part number.

adom/ams1117 now carries both datasheets and covers 1.2 V properly:

  • the UMW datasheet is embedded on the page as the alternate datasheet;
  • AMS1117-1.2 is a row in the output-voltage table with its real limits (1.176 – 1.224 V, 2 % accuracy — looser than the 1.5 % of every other version);
  • every variant row now names which manufacturer sells it;
  • there is a section comparing the two datasheets and explaining why they disagree.

The short version of why: AMS carries 2.85 V and not 1.2 V, UMW carries 1.2 V and not 2.85 V. 2.85 V is the SCSI-2 active-termination voltage, which is a legacy market AMS still serves; UMW targets laptops and phones, where 1.2 V is the SoC/PHY core rail. Seven fixed voltages exist across the family and no single vendor sells all seven.

So this page can be deprecated on the same terms as its 3.3 V sibling — no special decision needed.

Still worth doing before it goes: the back-layer footprint problem is unchanged, and adom/large-automotive-eth-switch-molecule#1 should still confirm which side U902 landed on.

One thing to carry over rather than lose: if the parts you actually stock are UMW, note that on the molecule, because UMW's limits differ materially from AMS's — 2.1–2.5 A current limit vs 0.9–1.5 A, 18 V input max vs 15 V, and thermal shutdown at 150 °C rather than 165 °C.

barrett-land · 5d ago

Update — the target of this deprecation has changed, and it is now unambiguous.

adom/ams1117 (the combined family page) is itself deprecated and has been split by manufacturer:

This page should be deprecated in favour of adom/umw-ams1117. There is no judgement call left: 1.2 V is a UMW-only version. Advanced Monolithic Systems does not make it, so an AMS1117-1.2 is by definition a UMW (or other second-source) part, and the UMW page carries its real limits — 1.176 – 1.224 V at 25 °C at 2 % accuracy, not the 1.5 % of every other version in the family.

The UMW page also documents the things that differ from the AMS original and matter here: 18 V input max, 2.1 – 2.5 A current limit, thermal shutdown at 150 °C, and the SOT-89-3L package option.

The back-layer footprint problem is unchanged and still worth fixing before this page goes — see the original report and adom/large-automotive-eth-switch-molecule#1.

Log in to reply.