On this page
Removing a divider's source does not guarantee that its output approaches zero smoothly. Capacitance across the arms stores charge, and a fast falling input can drive the output below its reference before resistance restores the settled state. Opening a connector can instead leave a different part of the network floating. Protection review must identify what changes electrically at the instant of disconnection, not treat every event as an ideal zero-voltage command.
System boundary
This review covers stored-charge redistribution at a divider-to-receiver interface during specified connection changes. It does not define safe discharge times, permissible high-voltage access or complete protection compliance.
Integration interfaces
| Interface | Required input | Thick film role | Validation owner |
|---|---|---|---|
| Source connection to divider | Actual switching impedance, opening sequence and initial voltage history. | Identify passive resistance and capacitance boundaries. | Qualified source-system designer. |
| Divider output to receiver | Input model, power state, clamps, cable and reference. | Provide the drawing-defined output network. | Analog front-end owner. |
| Disconnection to service state | Independent isolation, discharge and measurement-validity requirements. | Support component interpretation without granting access authorization. | Electrical-safety authority. |
Integration risks
| Risk | Control or verification | Validation owner |
|---|---|---|
| A falling positive input creates negative output stress. | Include initial charge and both-polarity receiver limits. | Analog designer. |
| An opened return invalidates the assumed discharge path. | Analyze each actual contact state. | System electrical owner. |
| A zero or recovered reading is mistaken for safe access. | Use the approved independent safe-state determination. | Qualified service authority. |
System integration decisions
- Record the capacitor voltages immediately before the disconnect event.
- Distinguish a source forced low from an open source, open return or removed receiver.
- Evaluate both polarities of input stress and the actual clamp-current destination.
- Keep a modeled low-energy waveform separate from high-voltage safety verification.
Identify which connection changes first
Draw the upper divider terminal, lower return, output connection and receiver reference. Mark the switch or connector contacts that actually open and their possible sequence. Disconnecting the source from its supply is not necessarily equivalent to forcing the divider input to zero through a low impedance.
A retained cable, compensation capacitor or connected instrument can preserve an electrical path after another contact opens. Include those objects in the pre-event and post-event schematics. A single schematic labelled power off omits the topology change that determines where charge can move and which voltage the receiver experiences.
Retain the established charge before the event
A normal transfer-function calculation often starts with zero stored energy. A disconnection usually begins after the divider has been energized, so the upper and lower capacitances already carry voltages. Preserve those initial conditions when solving the changed circuit. Otherwise the calculation can miss an output excursion with the opposite polarity from the measured source.
Record whether the preceding input was settled or changing. The same opening event after a short pulse can begin from a different capacitor state than an opening after a long hold. A screenshot showing only the final zero reading cannot establish the maximum transient or the charge remaining on an isolated segment.
Calculate the immediate feedthrough for one explicit case
Consider a low-energy hypothetical divider with an initial 5 V input and a settled DC ratio of 0.1, giving 0.5 V output. Let upper capacitance C1 be 20 pF and lower capacitance C2 be 80 pF. If the input is then forced rapidly from 5 V to zero while the lower return remains fixed, charge conservation predicts an immediate output change of −1 V, leaving −0.5 V before resistive recovery.
The capacitive fraction is 0.2, larger than the DC fraction of 0.1. That mismatch explains the negative excursion. A real source has finite edge speed and a real receiver may clamp, so the ideal jump is a pre-conduction prediction, not a measured waveform. These values are illustrative and do not define a safe test or a high-voltage product rating.
ΔVout,fast = [C1/(C1+C2)] ΔVin; Vout,after = Vout,before + ΔVout,fast
- C1: effective capacitance between input and output
- C2: effective capacitance between output and the fixed lower return
- ΔVin: imposed fast input change
- Vout,before: output immediately before that change
Lumped linear capacitances, fixed lower return, negligible resistive charge transfer during the edge and no clamp conduction during the prediction. Opening a source is not the same boundary as forcing it low.
Change the model when protection begins to conduct
If the predicted output crosses a receiving device's permitted input boundary, include the applicable clamp or protection model. The node can no longer follow the unconstrained capacitive prediction. Current may flow into a rail, return or external protective component, and each destination has its own allowable state.
Input structures differ between devices and power states. Do not substitute a universal diode voltage or transient-current allowance. Protection guidance requires attention to the actual analog front end and its stress limits. Series impedance, clamp capacitance and recovery behavior can also alter the normal measurement response, so protection and accuracy must be reviewed together.
Use separate models for different opening events
The following table prevents an apparently similar service action from inheriting the wrong electrical boundary. None of these observations authorizes access to an energized or insufficiently discharged assembly.
| Event | Boundary that changes | Required analysis |
|---|---|---|
| Input forced low with return intact | Known fast voltage transition | Initial-state capacitive feedthrough and clamp response |
| Upper source contact opens | Input segment may float | Remaining discharge and capacitive reference paths |
| Lower return opens first | Output reference can move | Receiver-relative voltage and unintended return paths |
| Receiver cable is removed | Lower capacitance and loading change | Charge state of both disconnected segments |
| Receiver supply disappears but source remains | Input structures may change state | Power-state loading and aggregate rail injection |
| All commands indicate off | Physical energy state remains unverified | Approved independent isolation and discharge determination |
Keep transient recovery separate from valid measurement
After the event, resistance may restore a node toward its new settled voltage, but the time constant depends on the post-event topology. A resistor that was a discharge path before a contact opened may no longer be connected afterward. Do not reuse the original divider's time constant without checking the remaining circuit.
A clamped or recovering output is not a calibrated measurement of the source. The consuming electronics need an appropriate validity state rather than a plausible scaled number. Recovery of communication does not establish analog recovery, and a zero display cannot independently prove absence of hazardous voltage. The system owner defines both measurement readiness and safe service determination.
Verify the model through an approved low-energy boundary
Begin with circuit analysis and a suitable low-energy surrogate where it preserves the relevant capacitance ratios, timing and protection topology. Record input, output and relevant rail behavior on a common time base. Include probe capacitance in the model because it can materially reduce or reshape the very excursion being measured.
Any verification with the actual high-voltage source, connector opening or stored energy requires qualified personnel and an approved safeguarded procedure. This guide does not provide an energized disconnect sequence. The useful engineering output is the predicted event envelope and unresolved protection requirements for the responsible test team to evaluate safely.
Control the disconnect interface through revisions
Retain the pre-event state, contact sequence, capacitance inventory, predicted polarity and protection-current destination with the divider drawing. Identify which cables and receiver configurations the result covers. A connector or receiver change can alter the transient even if the printed resistance ratio is unchanged.
After compensation, protection or software changes, verify the relevant falling and opening events again under the approved plan. Keep normal transfer fidelity, transient electrical stress, recovered measurement validity and safe access as separate conclusions. ChipSimple's passive divider review does not replace the integrator's complete high-voltage safety acceptance.
Send the before-and-after disconnect schematics
Include initial charge and the contact sequence that changes the circuit.
- Divider arms, capacitances, cables and receiver input model.
- Initial voltage history and actual source/opening behavior.
- Return and output contact sequence with possible intermediate states.
- Protection-current destinations and component stress requirements.
- Approved verification boundary and independent safe-service method.
The drawing-upload form loads as you reach this section.