L
Admin
[Technical article / FAQ, rfforum Editorial Team]
A microstrip hairpin or edge-coupled filter that comes back 50-100 MHz low (or high) is often not a ceramic εr surprise. Etch undercut, plating build-up, and a stack-up card that listed "finished line width" as the simulated width move the coupled sections first. The schematic was never the patient.
1. What the etch owns
- Coupled-line filters are more sensitive to gap and edge profile than to a small length error. Undercut narrows the strip and widens the gap; both lower coupling and shift the passband.
- A fab drawing that quotes "0.20 mm line / 0.15 mm gap, as drawn" is not the same as the etched and plated geometry. If the EM used the drawing, the board will not.
- Soldermask over tight gaps changes effective εr and coupling. A first article without mask can "pass" and then move after coating.
2. Freeze list before you retune the layout in the CAD
- Finished (etched + plated) width and gap from a microsection or a coupon, not only the Gerber number
- Stack-up: foil type, plating, and whether the sim used as-drawn or as-etched edges
- Mask on/off A/B on the same coupon if the production part is masked
- A one-line coupon of the coupled section, or at least a TDR / unpopulated filter S21, before a full ECO
3. Habit
Never freeze "εr was wrong" from a single shifted filter with no etch measurement. Archive finished geometry and mask state next to every first-article passband claim. If the coupon with measured etch lands on the measured S21, stop blaming the laminate lot.
4. Practical gate
- Measure finished line/gap on the failing panel
- Re-sim with that geometry (and mask if present)
- Only then decide between a compensation etch factor, a gap tweak, or a real material change
Discussion: on your last "filter landed off frequency" ticket, did the fix land in etch factor, soldermask, or a real εr - and do you require a finished-geometry coupon before a layout respin?
— rfforum 编辑部 / Editorial Team
A microstrip hairpin or edge-coupled filter that comes back 50-100 MHz low (or high) is often not a ceramic εr surprise. Etch undercut, plating build-up, and a stack-up card that listed "finished line width" as the simulated width move the coupled sections first. The schematic was never the patient.
1. What the etch owns
- Coupled-line filters are more sensitive to gap and edge profile than to a small length error. Undercut narrows the strip and widens the gap; both lower coupling and shift the passband.
- A fab drawing that quotes "0.20 mm line / 0.15 mm gap, as drawn" is not the same as the etched and plated geometry. If the EM used the drawing, the board will not.
- Soldermask over tight gaps changes effective εr and coupling. A first article without mask can "pass" and then move after coating.
2. Freeze list before you retune the layout in the CAD
- Finished (etched + plated) width and gap from a microsection or a coupon, not only the Gerber number
- Stack-up: foil type, plating, and whether the sim used as-drawn or as-etched edges
- Mask on/off A/B on the same coupon if the production part is masked
- A one-line coupon of the coupled section, or at least a TDR / unpopulated filter S21, before a full ECO
3. Habit
Never freeze "εr was wrong" from a single shifted filter with no etch measurement. Archive finished geometry and mask state next to every first-article passband claim. If the coupon with measured etch lands on the measured S21, stop blaming the laminate lot.
4. Practical gate
- Measure finished line/gap on the failing panel
- Re-sim with that geometry (and mask if present)
- Only then decide between a compensation etch factor, a gap tweak, or a real material change
Discussion: on your last "filter landed off frequency" ticket, did the fix land in etch factor, soldermask, or a real εr - and do you require a finished-geometry coupon before a layout respin?
— rfforum 编辑部 / Editorial Team
