On this page
- Treat Sheet Resistance as a Fired-System Property
- Choose the Resistance Decade for Geometry and TCR
- Manage Conductor Termination Effects
- Use Trimming Within a Stable Geometry Window
- Evaluate Noise and Stability for the Application
- A Worked Engineering Review
- Design Review Checklist
- Questions Engineers Commonly Ask
- Related Engineering Resources
- Closing Note
Prepared by Chipsimple Engineering Team, Engineering and technical content review
Published online August 10, 2026 · Reviewed August 10, 2026
ENGINEERING ARTICLE 10 / VOLUME 2
Ruthenium-Oxide Thick Film Resistors: Conduction, TCR and Stability
A thick-film resistor is not a homogeneous bar of metal. Conductive oxide particles form a distributed path within an insulating or semiconducting matrix. The contact between particles and the glass structure changes with composition and thermal processing. This is why two resistance decades can show different TCR or stability and why blending or refiring needs qualification rather than arithmetic alone.
Treat Sheet Resistance as a Fired-System Property
Paste composition, fired thickness and thermal history determine the conductive network. Consequently, label value is a target under specified processing, not a guarantee on every layout and furnace. The engineering value comes from understanding the direction and sensitivity of the effect, not from memorising a nominal number.
A robust drawing or process plan will use coupons that share conductor interfaces, substrate and firing sequence with the product. This makes the design intent visible and gives the supplier room to use a qualified material set without changing the function.
The result can be checked through sheet resistance distribution by lot and substrate position. Where possible, keep readings before and after the next thermal or mechanical operation; the delta often reveals more than the final value.
The failure pattern to investigate is systematic resistor offset after changing furnace loading or print thickness. Before changing materials, confirm orientation, lot history, measurement method and acceptance limits.
Choose the Resistance Decade for Geometry and TCR
The physical starting point is straightforward: Adjacent paste decades may meet the same nominal value with different aspect ratio and temperature behaviour. From there, a better geometry can be outweighed by an unsuitable TCR or process sensitivity. For ruthenium oxide thick film resistor, the important issue is the amount of process margin left after normal variation is included.
On the drawing and in the review record, screen candidate pastes against value range, geometry, TCR and trim requirement. Keep the requirement functional wherever possible, while making any safety- or interface-critical boundary unambiguous.
A production trial should capture resistance-temperature curves and as-fired capability on representative elements. Record the condition of the specimen and the next operation so that a later change can be traced to a specific stage.
If the team sees meeting room-temperature value while missing the system temperature error budget, pause before adding inspection or rework. First check whether the layout and the qualified process window are asking for contradictory outcomes.
Manage Conductor Termination Effects
In a stable manufacturing route, resistor-conductor overlap creates a composite current-transfer region. This means short elements and low values can be strongly influenced by contact geometry and metallurgical interaction. The observation is especially useful because it connects a visible feature to an electrical, thermal or mechanical consequence.
The control plan should use qualified termination overlap and consistent print-fire order. It should also name the drawing characteristic or coupon that represents the requirement, avoiding an instruction that depends on personal interpretation.
Evidence comes from length-series coupons that expose termination contribution. A trend across position or lot is usually more valuable than a perfect reading from one hand-picked sample.
Investigate nonlinear resistance-versus-length behaviour mistaken for random printing error. The pattern may identify a narrow process interaction long before the final circuit becomes an open, short or out-of-tolerance value.
Use Trimming Within a Stable Geometry Window
A useful way to frame the decision is this: Laser cuts concentrate current near the trim tip and remaining neck. As a result, large corrections can raise local power density and long-term drift. That cause-and-effect chain should remain visible when ruthenium oxide thick film resistor is reviewed with purchasing and quality teams.
The manufacturing boundary is protected when the team can target as-fired values close enough to final value and define maximum trim depth. This approach separates the customer's functional need from the supplier's machine-specific compensation.
Confirm the decision with post-trim stabilisation, powered testing and magnified cut inspection. Include both typical and boundary-condition specimens where the failure consequence justifies them.
The symptom resistors that pass immediately after trim but drift after ageing deserves a structured investigation. Check material lot, artwork position, thermal history and measurement setup before assigning a single cause.
Evaluate Noise and Stability for the Application
Geometry and material meet at this point: Not every resistor network is limited by nominal tolerance. Therefore, sensor, high-impedance and ratio circuits can be more sensitive to excess noise, humidity or tracking. A nominal specification that omits the interface is incomplete even if every individual value looks reasonable.
During release, identify absolute versus ratio accuracy and the relevant environment. Make sure the acceptance method measures the same physical feature that the design calculation assumed.
Useful confirmation includes noise, TCR, load-life or humidity testing appropriate to circuit function. Keep photographs or sections tied to part, revision, lot and orientation so they remain evidence rather than decoration.
One failure signature is using a general resistor specification that omits the actual error mechanism. It often becomes clear only when results are sorted by position, process stage or exposure instead of being combined into one average.
A Worked Engineering Review
A sensor network contains 200 Ω, 2 kΩ and 20 kΩ elements. Using one paste decade would force extreme geometries at both ends. A better review groups values into compatible paste ranges while keeping matched elements in the same family where tracking matters. Length-series coupons identify termination influence, and each group receives an as-fired target based on real distributions. Laser trimming is limited to a stable cut region. TCR is measured as a network requirement, not inferred from a single nominal paste property.
Design Review Checklist
- Use product-representative sheet-resistance coupons.
- Match paste decade to geometry and TCR.
- Qualify termination overlap and firing order.
- Limit trim depth and stabilise after trimming.
- Specify ratio, noise and drift where functionally important.
Questions Engineers Commonly Ask
Can resistor pastes be blended to obtain an intermediate value?
Some compatible systems permit blending, but TCR, print behaviour and stability may not interpolate simply. Treat the blend as a newly qualified composition.
Why do short resistors deviate from R = Rs × L/W?
Termination and current-spreading effects become a larger fraction of total resistance. Use correction data or length-series coupons.
Does laser trimming improve TCR?
Trimming sets resistance; it does not automatically improve material TCR and can create local current concentration. TCR must be verified separately.
Related Engineering Resources
Closing Note
Ruthenium-oxide resistor design is controlled by microstructure, geometry and process history together. Build the value and stability budget from measured product-representative data.