L
Admin
[Technical article / FAQ, rfforum Editorial Team]
A "temporary" SMA-to-2.92 or N-to-3.5 saver left on a VNA port for months often becomes the largest uncontrolled uncertainty in the chain — while everyone argues about IF bandwidth and averaging.
1. Why savers drift
- Mate cycles and side-load bend the centre conductor and oval the outer
- Mixed-connector families (SMA mated to 3.5 / 2.92 under torque) transfer wear into the more expensive sexed port you thought you were protecting
- Contaminants and skin oils raise contact resistance; you see it first as unstable S11 residual, not as a dramatic open/short
2. Bench habits that catch it early
- One saver per port, labelled; never borrow across stations without a residual re-check
- After any drop, cable yank, or "just one wrong mate," measure a known thru / airline before the next DUT
- Retire on residual vs baseline, not only on a calendar or mate counter — counters lie when operators share adapters
- Prefer a sacrificial sexed saver on the instrument; keep the DUT interface on a short, inspected adapter that gets cleaned every shift
3. Report line
Write "saver stack + serial/lot" into the measurement header the same way you write cal kit ID. Two labs disagreeing by a few tenths of a dB are often disagreeing about two different invisible adapters.
Discussion: do you retire savers on mate count, on thru residual vs baseline, or only after a visible pin injury — and have you ever traced a week of "DUT drift" to a borrowed adapter?
— rfforum 编辑部 / Editorial Team
A "temporary" SMA-to-2.92 or N-to-3.5 saver left on a VNA port for months often becomes the largest uncontrolled uncertainty in the chain — while everyone argues about IF bandwidth and averaging.
1. Why savers drift
- Mate cycles and side-load bend the centre conductor and oval the outer
- Mixed-connector families (SMA mated to 3.5 / 2.92 under torque) transfer wear into the more expensive sexed port you thought you were protecting
- Contaminants and skin oils raise contact resistance; you see it first as unstable S11 residual, not as a dramatic open/short
2. Bench habits that catch it early
- One saver per port, labelled; never borrow across stations without a residual re-check
- After any drop, cable yank, or "just one wrong mate," measure a known thru / airline before the next DUT
- Retire on residual vs baseline, not only on a calendar or mate counter — counters lie when operators share adapters
- Prefer a sacrificial sexed saver on the instrument; keep the DUT interface on a short, inspected adapter that gets cleaned every shift
3. Report line
Write "saver stack + serial/lot" into the measurement header the same way you write cal kit ID. Two labs disagreeing by a few tenths of a dB are often disagreeing about two different invisible adapters.
Discussion: do you retire savers on mate count, on thru residual vs baseline, or only after a visible pin injury — and have you ever traced a week of "DUT drift" to a borrowed adapter?
— rfforum 编辑部 / Editorial Team
