Semi E136

SEMI E136 defines a test method for verifying an RF generator's output-power accuracy — how closely the power the generator actually delivers at its output connector matches the power it reports to the host, measured against a traceable reference load, independent of whether the system settles quickly (E135) or whether the whole delivery path including matching network and cables stays efficient (E113). A generator can have excellent transient response and sit in a perfectly matched system and still be systematically wrong about how much power it's actually putting out — its own internal power sensor can be miscalibrated, drift with temperature, or simply read differently from what a traceable reference measurement would show at the same connector — and none of E113's path-efficiency framing or E135's settling-time framing would ever catch that, because both assume the generator's own reported number is correct in the first place.

Generator reports a power value to the host
E113 assumes the path is efficient; E135 assumes it settles cleanly — neither checks if the number itself is right
SEMI E136 — RF-generator output-power test methodCompares reported power against a traceable reference measurement
Standardized test load and reference power meter
Same measurement conditions across generator vendors
Defined accuracy tolerance across the generator's rated power range
Not just one spot-check at a convenient setpoint
Host can trust the generator's reported power number itself
A precondition every downstream recipe, APC loop, and fault threshold depends on

The reason output-power accuracy needed its own standardized test method is that every other RF standard in this family — E113's whole-path delivery, E114's cable testing, E115's matching-network efficiency, E135's transient response — implicitly treats the generator's self-reported power reading as ground truth, when in reality that reading comes from an internal sensor subject to the same calibration drift any measurement instrument faces. If a process recipe is defined in terms of "run at 500W" and the generator's internal sensor reads 500W while actually delivering 470W, every downstream specification built on top of that number — matching-network efficiency calculations, APC feedback corrections, fault thresholds — inherits an error none of them were designed to detect, because they all take the reported value as their input rather than independently verifying it. E136 is the test that checks the one assumption everything else in the RF standards stack depends on.

The E136 Test Bench, Connector to Connector path efficiency and settling time both assume the generator's own number is correct RF generator internal sensor reads: 500W drift/miscal risk lives here output connector Traceable reference power meter independently measured actual: 470W Delta -6% outside tolerance at this setpoint Repeated across the generator's full rated power range, not one spot check 100W: ✓ 300W: ✓ 500W: ✗ 800W: ✓ 1000W: ✓ E113/E114/E115/E135 all trust the generator's own reported number E136 is the one test that checks whether that number itself is right

E136 is what lets a fab trust the one number every other RF standard in this family quietly assumes is correct — by testing a generator's reported output power against a traceable reference measurement across its full operating range, it catches sensor drift and miscalibration that no amount of path-efficiency analysis (E113), cable integrity (E114), matching-network tuning (E115), or transient-response testing (E135) would ever expose on its own.

Take SEMI E136 further

Ask the copilot about this term, or have our engineers assess it against your process.

Ask CFSGPT about this → Request a technical assessment → Open the CFS app