28 VDC Aircraft Bus Testing: The Second Programme Inside MIL-STD-704F

Aviation & 400 Hz · 5 min read · 37 cited facts

Programmes budget for the 400 Hz source because it is the conspicuous purchase, then discover the DC work late. That is not carelessness — it is how MIL-STD-704F reads. The AC power types take the headlines; the 28 V and 270 V clauses sit in the same standard, apply to the same equipment, and need a rig the AC source cannot provide. If your equipment takes a DC feed, you are running a second test programme whether or not it was budgeted as one.

The scale of that second programme is set by one table and three envelope figures, plus two operating states that take the bus below anything a bench supply defaults to.

The steady-state window is not the test

Any laboratory supply reaches 29.0 V and plenty reach 280.0 V, and if the steady-state window were the requirement the DC programme would cost nothing — which is roughly what tends to get budgeted for it. The window is where the bus lives; the test is what the bus does on its way out of it. The ripple terms deserve a second look too: ≤1.5 V on the 28 V bus and ≤6.0 V at 270 V are conditions your source must be able to impose on the equipment deliberately, which means ripple injection capability, not merely a quiet output.

Reaching 50 V is not running Figure 13

This is where 'the supply reaches the voltage' and 'the source runs the envelope' come apart. A supply that can be set to 50 V is not a source that can hold 50 V for 12.5 ms and only 12.5 ms, drop to 18 V for 15 ms, and land back inside 22.0–29.0 V with the timing intact. The abnormal case is harder still: V = 31.38 + 0.931/t is a curve against time, and reproducing it means sequencing voltage with millisecond resolution rather than stepping between setpoints. The instrument requirement that falls out of 704F is a DC source to at least 50 V for the 28 V bus and 350 V for the 270 V bus, with ripple injection — a different specification from a bench supply at any power level.

Emergency and starting go below the window on purpose

Electric starting takes the bus to 12 V; emergency operation holds it as low as 16 V. Equipment that must stay functional through either state gets tested there, not at nominal, and brown-out recovery at 12 V is a genuinely different piece of equipment behaviour from riding through a momentary sag. These are the clauses that decide whether your DC source needs clean, programmable control at the bottom of its range rather than merely headroom at the top.

If DO-160 is also on the programme

The civil document adds its own DC matrix rather than reusing 704F's. Category B and Z equipment faces an engine-start profile — a drop to 10 V followed by a 0.3 V/s ramp for 35 s up to 20.5 V (§16.6.1) — and DC ripple runs <4 Vp-p on the 28 V bus and <16 Vp-p at 270 V across 10 Hz–30 kHz (Ch.18 Fig 18-3/18-4). Same physical bus, second set of clauses, and a 35-second ramp is a different source capability from a millisecond transient.

One scope note before the budget meeting. If your equipment takes only the AC feed, none of this applies, and a DC transient rig is money you should not spend. And the test procedures themselves are not in 704F — they live in MIL-HDBK-704-1 through -8, numbered by power type. Write the handbook method numbers into the statement of work: 'tested to 704F' on its own commits nobody to Figure 13.