Module grant conditions — what `X8WWM02C` actually permits¶

Status: working document, not normative. The normative home is launch-plan.md §5. This file is the reading of the WM02C grant and the Washington Labs reply that produced it. Created: 2026-08-11, from emails/WL Labs Exchange.md and its attachment files/fanstel_doc.pdf — the FCC Form 731 Grant of Equipment Authorization for X8WWM02C. Corrects: FCC Certification.md §5 (the three routes), §7 (what to ask Fanstel) · FCC Lab Quote Responses.md §0, §2, §5 · launch-plan.md §5 obligation 2 and 3. Does not affect: FCC Respin.md. Every board change in it stands unaltered — none of this is drawable.

0. The headline¶

FCC Certification.md §5 asked the right question and stopped one step short. It treated the WM02C's certification as settled and framed the remaining work as arithmetic: can the exclusion ratio be brought under 3.0?

A grant is not a power number. It is a permission with conditions attached. The ratio was never the binding constraint. Reading the grant itself:

"The antenna's as listed in this application must be installed to provide a separation distance of at least 20 cm from all persons and must not be co-located or operating in conjunction with any other antenna or transmitter except in accordance with FCC multi-transmitter procedures."

⚠️ Corrected 2026-08-12 — this document dropped the exception clause it had itself quoted¶

The user pushed back on the co-location finding, on the grounds that the WM02C is sold as a WiFi companion module for the nRF54 series and a module that may not sit beside a BLE host would be an incoherent product. That objection is correct, and the evidence for it was already in the quote above: "except in accordance with FCC multi-transmitter procedures."

The sentence carries two conditions, and only one of them is unqualified. The table below has been corrected accordingly. The same qualifier appears verbatim in the u-blox clause quoted at Module Alternatives.md §4 and was walked past there too, so the workspace had the right text twice and the wrong reading in both places.

Grant condition	Qualified?	Watch	Anchor
≥ 20 cm from all persons	No — unqualified	Violated — wrist-worn, floored to 5 mm	OK — wall-mounted, stationary
No co-location with another transmitter	Yes — "except in accordance with FCC multi-transmitter procedures"	Permitted, conditional on following those procedures	Permitted, same

So the WM02C's modular approval does not cover the watch — but for one reason rather than two, and it is the exposure-category reason. Co-location beside the BM20C is a thing the grant anticipates and provides a route for. Being 5 mm from a wrist is not.

What "FCC multi-transmitter procedures" means in practice is the simultaneous-transmission evaluation in KDB 447498 D01 §4.4 and the co-location guidance in KDB 616217 — sum the exposure ratios of antennas that can transmit at the same time and compare against the same threshold, or demonstrate that they cannot transmit simultaneously. This workspace has already done that arithmetic: FCC Certification.md §5 sums BLE and 2.4 GHz WiFi and raises the coex-arbitration argument. That section was doing the multi-transmitter procedure while this one was calling co-location forbidden. They contradicted each other for a day; this is the resolution.

Fixing the part that is genuinely broken still requires touching the grant, and only the grantee can touch a grant.

And the same trap is very likely being set for the BM20C right now. §6 is the urgent section; everything before it is why.

✅ Resolved by events, 2026-09-22 — the trap sprang, and the reading above survives it¶

The BM20C granted on 2026-09-11, mobile, 20 cm (X8WBM20C — §6's dated block has the detail). So the sentence above was right, and §6's window is shut rather than urgent.

The part that matters here is that the table above now describes both modules. The BM20C's grant note was read verbatim on 2026-09-22 and it carries "except in accordance with FCC multi-transmitter product procedures" — the same exception, one word different. Co-location of the BM20C beside the WM02C on one board is permitted under both grants; the 20 cm exposure category is the unqualified half on both. The 2026-08-12 correction the user forced was not a reprieve specific to the WM02C; it is how these grants are written.

This document's title still says what X8WWM02C permits, and that is still what §1–§5 read. Everything in §0 now reads across to X8WBM20C unchanged — with one asymmetry that §6 states and is worth repeating: WiFi can be gated off-body, so wornness-gating is at least an argument for the WM02C. BLE cannot be, because transmitting while worn is the enforcement channel. The BM20C's permissive change has no design-side escape.

1. Where this came from¶

Steve Koster, Washington Laboratories (stevek@wll.com), 2026-08-11 16:12 — the reply to lab quote #1, sent 2026-08-09. Full text in emails/WL Labs Exchange.md. Four claims:

"The module that you are using is certified as a mobile device… the radiating element of the radio has to be 20 mm from the body." He means 20 cm; the grant text says 20 cm and "mobile" is the FCC exposure category that carries exactly that distance. Read it as a typo, not a different number.
"The power output from this radio and a Bluetooth radio together will require SAR testing." SAR is confirmed, and note together — that is him assuming the two radios sum, which is one of the three questions FCC Lab Quote Responses.md §5 item 3 asked. Answered in passing and negatively. Get it said explicitly.
"The module will require what they call a class 2 permissive change. Only the Grantee can apply for a permissive change." The grantee is Fanstel Corporation, Taipei. Not you.
"It is more common for the manufacturer to allow a 'change of ID'… then you could apply for the permissive change to certify what you want to certify." This is the actual advice, and §5 works it through.

He offered a quote for the SAR testing ("a day or two") and a phone call.

One thing this reply discharges: FCC Lab Quote Responses.md §0 flagged that the WLL form was reached from their Part 87 (aviation) page and warned to confirm the scope. Steve's reply is scoped to Part 15 modular integration and RF exposure. That worry is closed — it was a landing-page accident, as suspected.

2. The grant, read closely¶

files/fanstel_doc.pdf — FCC Form 731, printed from a browser 2026-08-11 09:38 EDT.

Field	Value
FCC Identifier	`X8WWM02C`
Grantee	Fanstel Corporation, Taipei
Issued by	Kiwa Nederland B.V. (TCB), Apeldoorn NL
Date of grant	2024-09-13
Equipment class	Digital Transmission System
Notes	WIFI 6 Module
Modular type	Single Modular
Rule part	`15C`
Frequency range	2412.0 – 2462.0 MHz
Output	0.17 W (conducted)
Grant note	`CC` — "This device is certified pursuant to two different Part 15 rules sections."

Plus the standard modular conditions: sold to OEM integrators, installed by the OEM, end-users not given installation instructions, OEM must supply operating conditions for RF exposure compliance — and the 20 cm separation sentence quoted in §0, whose co-location half carries the multi-transmitter exception.

"Single Modular" matters. It means the module was certified standalone with its own shielding and antenna, and the OEM inherits the grant provided the conditions are met. It is the good kind of modular approval. It is also the reason the conditions are the whole story: there is no limited-modular escape hatch where the host is re-evaluated anyway.

2.1 Two things the attachment does not contain¶

The 5 GHz authorisation. The only rule-part row on the page is 15C at 2412–2462 MHz — US 2.4 GHz channels 1–11. But grant note CC says the device is certified under two Part 15 rule sections, and the second (15E, UNII, equipment class NII) is not on the page. The PDF is a one-page browser print and is very likely truncated.

This is not academic. FCC Lab Quote Responses.md §2 declares WiFi 5 GHz: 802.11a / n / ac / ax, UNII-1 and UNII-3 as supported modes, and FCC Certification.md §5 runs a 5 GHz exclusion calculation against them (ratio 7.3). If 5 GHz is not inside this grant, those modes have no authorisation behind them and the scope handed to three labs is wrong in a way no quote would catch.

→ Get the complete grant. Ask Steve, or pull it from the FCC OET equipment authorization search directly. fccid.io refuses automated fetches; the FCC's own search does not.

The grant conditions in detail. Certified antenna type and peak gain are not on the form — they are in the application exhibits. Needed for the exposure filing whichever route §5 takes.

3. The arithmetic, re-run at the grant's declared power¶

FCC Certification.md §5 used the datasheet's 18.24 dBm average (802.11b, 66.7 mW, → 84 mW with +1 dB tune-up) and got 26.2.

The grant declares 0.17 W = 170 mW = +22.3 dBm conducted, and the grant's number is the one that governs. That figure sits with the datasheet's peak column (20.51–23.11 dBm), not its average column, which is consistent — a DTS grant declares the maximum the module is authorised to emit, and the WM02C-Power-Setting_radiotest_prj.txt backoff is what brings the shipped average below it.

Re-running KDB 447498 D01 §4.3.1 at the wrist device's floored 5 mm separation:

(170 mW / 5 mm) × √2.437 GHz  =  34 × 1.5611  =  53.1      against NT = 3.0

53, not 26. Seventeen times the 1-g threshold; seven times the 10-g extremity threshold.

Sanity check on the grant's own internal consistency, at the 20 cm it was certified to:

(170 mW / 200 mm) × 1.5611  =  1.33   ≤ 3.0    →  EXCLUDED

Which is exactly why Fanstel certified it as a mobile device and why the 20 cm condition is on the form. The grant is coherent; the product is outside it.

The power ceiling that would exclude at 5 mm is unchanged — 9.6 mW = +9.8 dBm, from FCC Certification.md §5. What changed is the size of the gap: 12.5 dB of backoff below the grant's declared maximum, not 8.4 dB below the datasheet's average.

4. What this does to the three routes¶

FCC Certification.md §5 offers three ways out and calls the first one free. All three now pass through a permissive change, except possibly the third — and the third is the whole game.

Route (§5)	Was	Is
1. Cut WiFi TX power in firmware to ≤ +9.8 dBm	"Free — costs ~8 dB of range, no hardware, no schedule"	Not free. You are not the grantee and cannot redeclare the module's maximum power. And cutting power does not lift the 20 cm condition — that is a separate clause, not a consequence of the power figure. Reduces what you test; does not remove the permissive change.
2. Test SAR	"Book it, budget it, 1–2 weeks"	Still that, plus change of ID plus a Class II PC. The lab test is now the smaller half.
3. Ship BLE-only (WM02C depopulated)	"SAR-excluded by calculation alone"	Still the clean escape, and now worth far more. One transmitter: no multi-transmitter procedure, no WM02C grant to satisfy, no PC. Conditional on §6.

Corrected 2026-08-12 alongside §0: route 2's permissive change is needed for the 20 cm exposure category, not for co-location — co-location is permitted under the grant's own exception, via the multi-transmitter procedures. That narrows the scope of the PC and it removes one of the two things Fanstel would have to be asked to cover, but it does not remove the PC. Route 1 remains unavailable for the reason given: an integrator cannot redeclare a module's certified maximum.

4.1 The claim in §5 that is now unsafe¶

"A BLE-only watch is SAR-excluded by calculation alone. No SAR test, no phantom, no robot, no 1–2 weeks, no five-figure line item."

That rests on arithmetic — ratio 2.5 against a threshold of 3.0 — and ignores the grant condition, which is the exact mistake the WM02C just exposed. It is true if and only if the BM20C's grant permits portable, body-worn use. Nobody has asked. See §6.

4.2 The cost frame has moved¶

launch-plan.md §12.5 prices the Class-II-PC-with-grantee-cooperation scenario at $15k → $35k and presents it as the failure case — what happens if you deviate from a grant and get caught.

For a WiFi watch a permissive change is now the normal path — but not at that price. See §5.2: the $15k → $35k figure prices a full intentional-radiator retest, which is route 3. A C2PC that changes only the exposure category, evidenced by an exclusion calculation rather than a SAR measurement, is a paperwork filing of a few thousand. The machinery is the same and the dependency on a vendor in Taipei is the same; the bill is not. (Corrected 2026-08-12 — this section was the source of the "almost infeasible" reading.)

4.3 This collides with a decision made yesterday¶

2026-08-11: v1 ships with WiFi, decided on the grounds that WiFi carries the WATCH_REMOVED notification (firmware_spec_v3_nrf.md §1.5) and that dropping it reopens the Sunrise Lock bypass. The reasoning was sound and is not disturbed by anything here.

But it was decided against a §5 that priced the WiFi problem as "8 dB of range, or a SAR test." The actual price is a change of ID, a Class II permissive change, a SAR test, and Fanstel's cooperation on the first of those. That is a materially different number and a materially different schedule risk.

Re-open, not reverse. The WATCH_REMOVED argument still stands and no new information touches it. What changed is the other side of the ledger, and the decision was made without it.

5. Getting a body-worn watch authorised — the four routes and the actual procedure¶

Reprioritised 2026-08-12. Everything below was written about the WM02C. It matters more for the BM20C, for a reason §6 states and this section did not: wornness-gating can rescue the WiFi radio, and nothing can rescue BLE. The watch transmits BLE while worn — that is the enforcement channel and the whole product. There is no duty-cycle argument, no separation claim, no gating trick. If the BM20C is granted mobile-only, a permissive change on the BLE radio is unconditional. The WM02C is the negotiable one; the BM20C is not.

5.0 The principle, stated once because it keeps being rediscovered¶

Being exclusion-eligible does not authorise body-worn use. A BLE radio at +8 dBm clears the 5 mm SAR-exclusion threshold with a ratio of 2.82 — and if its grant says 20 cm, installing it in a watch is still outside the authorisation. The ratio and the grant condition are independent gates and you must pass both. This is the whole WM02C lesson, and it is why the filing scope, not the power number, is the thing to chase.

The four routes¶

	Route	Needs Fanstel?	Cost shape
1	Fanstel files a Class II PC on their own grant	Yes — entirely	Cheapest if they agree; they have no reason to
2	Change of ID, then your own C2PC ← Steve's recommendation	One letter	Grantee code + CoID + C2PC filings
3	Certify the finished watch as its own intentional radiator	No	Full Part 15C/15E testing — most expensive, fully in your control
4	Ship BLE-only on a portable-granted module	No	~$6k, and `BLK-15` dies — but costs `WATCH_REMOVED`

Route 1 — Fanstel files it themselves. Only the grantee may file, so you have no standing. Fanstel has no commercial reason to prioritise a one-board customer, the filing lives on their FCC ID forever, and every future change — a new host, a different co-located radio — goes back through them. Worth asking; do not plan around it.

Route 2 — change of ID, then your own permissive change. Fanstel writes one letter; you become grantee of record for your own FCC ID covering the same module hardware, and file your own permissive changes from then on. One letter buys permanent autonomy, which is why it is the standard OEM route and why Steve steered you to it. The procedure is §5.1.

Route 3 — certify the composite device. Nothing obliges you to rely on modular approval at all. You can certify the assembled watch as your own intentional radiator: full Part 15C (and 15E if 5 GHz is in scope) on the finished product, your own grant, no Fanstel letter, no CoID, no dependency on a vendor in Taipei. It costs more and takes longer. Its value is not as a plan but as a position — it is the difference between asking Fanstel for a favour and asking them for a discount on your alternative, and it means a refusal from them is expensive rather than fatal. Nobody had written this down before 2026-08-12.

Route 4 — BLE-only. Covered in Module Alternatives.md §5.1, and stronger than it looks now that X8WBC15 is confirmed portable. It costs the WATCH_REMOVED transport, which is what firmware_spec_v3_nrf.md §1.5 decided against paying.

5.1 The change-of-ID procedure, in order — and the three steps people miss¶

Get your own FCC Grantee Code — start this now. An FCC application in its own right (an FRN, a COREs account, a fee on the order of tens of dollars). It is separate from anything a TCB or a lab does, it has lead time, and nothing downstream can be filed without it. It is also the only step that costs almost nothing and can be done today, before any question is answered.

2026-09-23 — there is now a company to hold it. Impulse Gear LLC was organised in Virginia effective 2026-09-23 (SCC ID 12081863; receipt in TaxReturns/registration/). A grantee code is issued to whoever registers the FRN, and every FCC ID filed under it carries that grantee's name for good — so register the FRN to the LLC, not to you personally. A business FRN is registered against the entity's tax ID, which makes an EIN the first step now, not the FRN. Moving a grantee code between holders later is possible and is paperwork nobody needs. 2. Fanstel's letter. A specific document addressed to the TCB, authorising a Change in Identification for identical hardware. This is the only step that depends on someone else's goodwill. 3. A TCB files the Change in ID. ⚠️ Not the test lab. Certifications are granted by a Telecommunication Certification Body; a test lab performs testing. Some organisations are both — Element/PCTEST is — and whether Washington Laboratories is a TCB has not been confirmed. Ask Steve directly; it decides whether he can carry this end to end or whether you need a second organisation. 4. Then the Class II Permissive Change, on your grant.

⚠️ The step most people miss, and it is step 3's whole character: a Change in ID copies the original grant's conditions verbatim. You would end up holding your own mobile-only, 20 cm grant. The CoID fixes nothing by itself — it moves the steering wheel to your side of the car. Step 4 is the step that changes the authorisation, and the CoID exists only so you can file it without going back to Taipei each time.

5.2 What the permissive change actually costs — probably far less than §12.5 says¶

launch-plan.md §12.5 prices this at $15k → $35k, and that number has been read as the cost of this route. It is the cost of a full intentional-radiator retest, which is route 3, not route 2.

A C2PC that changes only the exposure category, supported by an exclusion calculation rather than a SAR measurement, is a paperwork filing — TCB fees on the order of a few thousand, not chamber time. For a BLE radio that clears exclusion at 5 mm, no SAR test needs to sit inside it at all. The realistic route-2 bill is: grantee code, the CoID filing, the C2PC filing, and the Part 15B SDoC you owe on the finished product regardless.

That materially changes the 2026-08-11 reaction that this was "almost making the entire project infeasible." The alarming number was the wrong number for the likely path.

5.3 ⚠️ The catch worth a phone call¶

FCC KDB 178919 (Permissive Change Policy) is reported to say that a permissive change may not be used to amend an application where the device was represented as being for mobile use only and SAR was not included. If a TCB reads it that way, mobile → portable is not a Class II at all — it is a new certification, and the cost moves by roughly an order of magnitude, landing you on route 3 whether you chose it or not.

Secondary sources split on this. Some classify an RF-exposure change as Class III or as requiring new certification; Steve, who is a lab engineer looking at the actual grant, says Class II. He is more likely to be right than a summary is, but this is the single question with the largest dollar figure attached to it, and it is free to ask.

The primary text was not read directly — apps.fcc.gov and fccid.io both refuse automated fetches. Read KDB 178919 D01 v06 §"Class II" before relying on either reading, or have Steve confirm on the call.

6. ⛔ The question nobody asked — and it expired on 2026-09-11¶

What happened, recorded 2026-09-22¶

The filing closed and the grant issued on 2026-09-11: X8WBM20C, original equipment, DTS, 2402–2480 MHz, ≈8.4 dBm conducted / 15.4 dBm EIRP, and mobile — "at least 20 cm from all persons." Nobody saw it for nine days; the question below was never put to Fanstel and drafts/fanstel-exposure-category.md has been written and unsent since 2026-08-12.

Score the three bullets under this block honestly, because two of them landed and one did not:

✅ "The BLE-only fallback is not SAR-excluded by calculation alone" — correct, of the BM20C. The arithmetic still clears (2.5 vs 3.0); the grant does not care. Note the shorthand has to name a module from here on, because X8WBC15 is evaluated portable and the phrase stays true of it.

✅ "Every route in §4 dies, including route 3" — correct. The watch cannot use either radio without a permissive change on somebody's grant.

✅ The struck co-location line stays struck. The grant note was read verbatim on 2026-09-22 and carries the multi-transmitter exception (§0). Striking it in advance, on 2026-08-12, on the strength of how the WM02C's was written, turned out to be right.

What replaces this section is SEP-13 — one change-of-ID conversation with Fanstel now covering both modules, on SHIP-05's by-May deadline. The three questions at the end of this section are still the right three; question 1 needs rewriting before it goes, from "will you scope the filing" to "will you file the permissive change, or authorise a change of ID."

The one input that improved. The grant declares ≈8.4 dBm conducted — far below the WM02C's declared 0.17 W — and a Class II permissive change argues from the declared maximum. This is the most favourable number this problem has, and it did not exist before 2026-09-11.

How will Fanstel certify the BM20C? (The section below is left as written on 2026-08-11/12. It was the correct argument at the time and it is the record of a cost that was real.)

They told you on 2026-08-04 that FCC testing completes "this month." That filing is open right now.

If they file it the way they filed the WM02C — mobile, 20 cm — then:

The BLE-only fallback is not SAR-excluded by calculation alone. FCC Certification.md §5, launch-plan.md §5 and BLK-15 all assert that it is. The arithmetic clears (2.82 ≤ 3.0, corrected 2026-08-12); the grant would not.
Every route in §4 dies, including route 3. The watch could not use either radio without a permissive change on somebody's grant.
~~The co-location clause would then apply from both directions on the same PCB.~~ Struck 2026-08-12 — co-location is permitted under the grant's own multi-transmitter exception (§0), so this was never the consequence. The exposure category is the whole of the risk here.

The user's prediction, 2026-08-12, and what the evidence does and doesn't support¶

"The BM15C, if I'm not mistaken, is certified for mobile-only, despite the fact that at 5 mm its radio output isn't enough to require SAR testing. Fanstel did this for legal liability reasons — they don't want to be responsible for every host PCB. This makes me think it's highly likely the BM20C will be the same."

The mechanism is right and it is the standard reason vendors file mobile-only. Portable authorisation obliges the grantee to specify and police host installation conditions; "20 cm, integrator's problem" is the cheap defensible default. u-blox's identical boilerplate on the NORA-B2 is the same instinct from a much larger company.

Two things the evidence says, and one it doesn't.

⚠️ BM15C and BC15 are different modules with different grants, and only the BC15's has been read. X8WBM15C's conditions are unknown — the prediction may well be right and nothing in this workspace confirms it. That naming confusion has been live since 2026-08-09 and this is the second decision it has touched.

X8WBC15 is not mobile-only, which cuts against the generalisation: its RF exposure exhibit states "According to §2.1093 this is a Portable device" and runs the 5 mm exclusion. Fanstel filed both categories on that part. So mobile-only is a per-part decision, not house policy — which is precisely why the BM20C question is worth asking rather than assuming.

If the prediction is right anyway, it is worse than the WM02C problem, per the note at the top of §5: BLE must transmit while worn, so no design change rescues it.

Corrected 2026-08-11 by Module Alternatives.md §3 — this is a real risk but not the one it first looked like. An earlier draft read the WM02C's 20 cm condition as Fanstel's house style and inferred they would repeat it. That inference is wrong twice over:

The WM02C had no choice. At 170 mW, 2.4 GHz WiFi cannot be exclusion-qualified below about 57 mm. Its 20 cm is physics, not habit. The BM20C at +8 dBm can be qualified at 5 mm (ratio 2.5), so Fanstel has a genuine choice there that they did not have on the WM02C.
The 20 cm clause is boilerplate across the whole industry. u-blox carries it word for word on the NORA-B2 — multi-transmitter exception included. It says nothing about a vendor.

What survives, and it is enough: a module vendor decides at filing time whether to seek approval for portable host platforms, and it is a choice, not a given. u-blox took that option on the NORA-B2 — "approved for installation into mobile and/or portable host platforms" — which proves it is available on a +8 dBm BLE part. Fanstel may or may not have scoped their filing that way, and nobody has asked. If they file mobile-only, the fallback loses the property three documents rest on; if they file portable, it keeps it. One sentence in an open filing decides it.

Right now, before their filing closes, is the only cheap moment to influence it. Asking a grantee to scope an in-flight filing for portable body-worn use costs them a paragraph. Asking after the grant issues costs a permissive change — the same one this whole document is about.

drafts/fanstel-followup.md does not ask this. It has eight items, none of which is the exposure category. It has been drafted since 2026-08-10 and unsent. Three items to add before it goes:

Will the BM20C be certified for portable / body-worn use (< 20 cm), with the RF exposure evaluation included in your filing? Say plainly that it is going into a wrist-worn device. Ask while the filing is open.
Will you grant a Change in Identification for the WM02C, and for the BM20C once granted?
The WM02C grant declares 0.17 W; your datasheet table says 18.24 dBm average. Which governs the OEM's exposure filing, and does the published WM02C-Power-Setting_radiotest_prj.txt backoff reduce the certified maximum or only the typical output?

Item 1 is the one on the critical path. It has overtaken the grant-number chase that BLK-15 was built around, because a grant number for a mobile-only grant is a number that does not help.

7. What each document has to change¶

Document	Change
`FCC Certification.md` §5	Add the grant-condition gate above the arithmetic — the ratio was never binding. Re-run 2.4 GHz at 170 mW (53.1, not 26.2). Re-score the three routes per §4. Mark the BLE-only claim conditional on §6.
`FCC Certification.md` §7	Add §6's three Fanstel items. Item 1 outranks the existing item 1.
`FCC Lab Quote Responses.md` §0	The Part 87 caveat is discharged. Add: the WM02C is certified mobile-only, 20 cm — say it on every form, because the lab will read the grant. (Corrected 2026-08-12: do not say "no co-location." The grant permits it in accordance with FCC multi-transmitter procedures, and claiming otherwise on a quote form scopes work you don't need.)
`FCC Lab Quote Responses.md` §2	Hold the 5 GHz modes until §2.1 is resolved.
`FCC Lab Quote Responses.md` §5	The scope text now omits the permissive change, which is most of the job. Add it before labs 2 and 3 go out. Item 3's BLE-only delta is no longer a contingency — it is the decision.
`FCC Lab Quote Responses.md` — lab table	Lab 1 replied 2026-08-11.
`launch-plan.md` §5	Obligation 2 and 3 have merged: staying inside the grant conditions is the RF exposure problem, not a separate risk beside it. §12.5's $35k scenario is the baseline path for a WiFi watch.
`FCC Respin.md`	Nothing. Every item stands. §4 already says the respin does not fix SAR; it is now also true that it does not fix the grant.

Labs 2 and 3 should not go out on the current scope text. FCC Lab Quote Responses.md §0 is right that identical scope text is what makes three quotes comparable — but that text asserts a 5 GHz mode that may not be granted and omits the permissive change that is now most of the work. Three comparable quotes for the wrong job are worth nothing.

8. Next steps¶

Today

Reply to Steve. Accept the SAR quote, and pin the scope while he is engaged: - Send the complete grant — is 5 GHz (15E/UNII) inside X8WWM02C? (§2.1) - Confirm change-of-ID over a Fanstel-filed PC, and whether WLL handles the filing. (§5) - The KDB 178919 question — is mobile → portable genuinely a Class II here, or will a TCB push it to a new certification? (§5.1) - The two still-unanswered §5 questions: wrist as extremity (N_T = 7.5 rather than 3.0), and coex arbitration vs summing — he glanced off the second in his reply. - Quote the BLE-only variant as a real alternative, not a contingency line.
Send the Fanstel email with §6's three additions. Time-sensitive in a way nothing else here is.

This week

Hold labs 2 and 3 until the scope text is fixed. (§7)
Re-run FCC Certification.md §5 against 170 mW and the grant-condition gate.
Take Steve's call. Most of what is open above is TCB judgement rather than rule text, and a call resolves §5.1 and the extremity question together for free.

Sources¶

emails/WL Labs Exchange.md — Steve Koster, Washington Laboratories, 2026-08-11 16:12
files/fanstel_doc.pdf — FCC Form 731 Grant of Equipment Authorization, X8WWM02C, granted 2024-09-13 by Kiwa Nederland B.V.
FCC KDB 447498 D01 General RF Exposure Guidance v06 §4.3.1 — the exclusion formula
FCC KDB 178919 D01 Permissive Change Policy v06 — not read directly, see §5.1
FCC KDB 996369 D02/D03 — modular approval and OEM integration
datasheets/WM02C+Product+Specifications.pdf — Ver 1.00, Sep 2024, for the power table the grant supersedes

Module grant conditions — what X8WWM02C actually permits¶