ASTM D7843 MPC varnish test: what it tells you and what not

ASTM D7843-25e1 measures insoluble color bodies on a membrane as a ΔE value, showing varnish potential, not deposits already on surfaces. The reading holds for undyed in-service turbine oil sampled hot from a running unit, trended against its own history and read with RULER and RPVOT trends. ASTM lists D7843-25e1 as the active edition as of 2026-10-13.

• ASTM D7843-25e1 reports the membrane patch color as a ΔE value in the CIELAB color space and is not appropriate for turbine oils with dyes (ASTM International, 2025).

• Published descriptions of sample conditioning give 60 to 65 °C for 24 hours, then 68 to 74 hours in the dark before filtration (Hong and Jang, 2023; Pravda et al., 2024).

• Membrane patch color became darker with increasing incubation period in laboratory tests on turbine oil (Kon et al., 2018).

• Varnish may start to build up while remaining antioxidants are as high as 60% of the new oil level (Lantos et al., 2022).

This reading applies to undyed in-service turbine oils, the scope of ASTM D7843. For hydraulic fluids the method has no standardized basis, so results there are trends against the same fluid only. An MPC value never replaces inspecting servo valves, filters and the reservoir for deposits.

What does ASTM D7843 measure, and how?

ASTM D7843, membrane patch colorimetry (MPC), measures the color of insoluble oil degradation products that a fine membrane retains from an in-service turbine oil sample, and reports it as ΔE in the CIELAB color space (ASTM International, 2025). The scope states: "This test method extracts insoluble contaminants from a sample of in-service turbine oil onto a patch and the color of the membrane patch is analyzed by a spectrophotometer." It also states that the method "is not appropriate for turbine oils with dyes." As of 2026-10-13, ASTM lists D7843-25e1 as the active edition, after D7843-12, -16, -18, -21 and -25 (ASTM International, 2025).

The conditioning step decides what the patch shows. In published descriptions, the sample is heated to 60 °C for 24 hours and held 68 to 74 hours in the absence of light, then filtered through a 0.45 μm membrane (Hong and Jang, 2023). A hydraulic oil study following ASTM D7843-21 gives 65 °C for 24 hours and 68 to 72 hours of incubation (Pravda et al., 2024). The heating matters because oxidation products behave differently with temperature. Kon et al. (2018) found that oxidation products became low-molecular and soluble in turbine oil when the oil was hot, and high-molecular and hard to dissolve when the oil was cooled and stored. A darker patch therefore means more material that drops out of this oil at room temperature, which is the varnish potential, not a count of deposits in the machine.

How do you read MPC results?

MPC results are read as a trend in ΔE against the same oil's history, next to RULER (Remaining Useful Life Evaluation Routine, ASTM D6971) antioxidant levels and RPVOT (rotating pressure vessel oxidation test, ASTM D2272) retention. ASTM D7843 itself frames the use this way: "The results from this test are intended to be used as a condition monitoring trending tool as part of a comprehensive program, as outlined in standards such as Practice D4378." A single ΔE tells you how much insoluble material the sample dropped under test conditions. The sampling temperature history, the antioxidant reserve and the oxidation reserve tell you why, and what to do next. Both companion results are ratios to the same new oil. RPVOT retention is the used-oil result divided by the new-oil result, expressed as a percentage (Hong and Jang, 2023). RULER compares the antioxidant peak areas of the in-service oil with those of its original formulation (Lantos et al., 2022).

The table below decides whether to resample, treat the oil or inspect for deposits by reading ΔE with the RULER and RPVOT trends. Band values are the ones the cited sources report; your laboratory's bands may differ.

The row that most often changes a decision is the third: when ΔE climbs while RULER still reads high, the oil is already making varnish material, so waiting for the antioxidant result to fall delays treatment (Lantos et al., 2022).

Why can MPC read low when varnish deposits already exist?

MPC reads low while varnish deposits exist because the test sees only the degradation products still carried in the oil sample, not those already laid down on servo valves, coolers, pipe walls and the reservoir floor. Turbomachinery International (2013) describes two field patterns. After an oil change without a flush, the article reports: "The new oil has a low MPC value, yet varnish is still throughout the lubricant system." After an outage in which the oil stays cool for weeks, degradation products precipitate into reservoir deposits, and a sample drawn then can show a low MPC value even though deposits remain; the author calls this improper sampling.

The chemistry explains the pattern. Kon et al. (2018) found that cooled, stored turbine oil turned its oxidation products high-molecular through hydrogen bonding and made them hard to dissolve, and that hydrogen-bond cleavage of oxidation products ended after the oil reached at least 60 to 70 °C. Duchowski, quoted in STLE (2016), notes that in dead zones or cooler spots varnish "possibly will not simply be re-dissolved into the fluid."

Top-ups can also pull ΔE down. In a simulated 40% fresh-oil refresh, MPC fell to ΔE 16 against 21 expected from a linear dilution model, and matched the dilution model when the sample stood 168 hours instead of 72 hours (Lantos et al., 2022). The authors attribute this to slower re-precipitation in diluted oil. Record outage, oil change and top-up dates on every sample.

How well does MPC predict servo valve sticking?

MPC predicts servo valve sticking only as a risk indicator: a rising ΔE shows more varnish-forming material in the oil, but whether a valve sticks depends on the unit's design, duty cycle and cold zones. Turbomachinery International (2013) states that "MPC results of 30 dE could pose a high reliability risk for one power station while the same results could translate into little operational concern for another station." The article lists factors that raise sensitivity: newer large frame gas turbines with precise servo controls, peaking or cycling operation, a shared bearing and control oil reservoir, and hydrogen seal oil systems.

The same turbine oil that flows around 200 micron bearing clearances also has to pass three micron servo valves (STLE, 2016). In cycling units, the reservoir may stay hot while oil in the hydraulic circuit cools to ambient, letting degradation products precipitate in the control zone (Turbomachinery International, 2013).

Laboratory work supports MPC as an early oxidation signal, not a valve-failure forecast. Kon et al. (2020), measuring patch color with a colorimetric patch analyzer on laboratory-aged oil, report that the method is able to "monitor the decrease of the RPVOT residual rate from the early stages of oxidative deterioration." A review notes that offline methods such as MPC "are not effective to diagnose varnish contamination immediately" (Hong and Jang, 2023). None of the sources reviewed here gives a ΔE value that guarantees or rules out sticking, so set the action band per unit from its own stiction history.

Does MPC work on hydraulic oils?

MPC can be run on hydraulic oils, but ASTM D7843 is written for in-service turbine oils, so a hydraulic ΔE result has no standardized precision or limits and should be trended only against the same fluid. Pravda et al. (2024) state that the D7843 standard "is only applicable for in-service turbine oils," and that the relevant hydraulic standard is ASTM D4898, a gravimetric method. ASTM D4898-23 weighs insoluble contamination retained on a membrane, including "both particulate and gel-like matter, organic and inorganic" (ASTM International, 2023). Weight and color are not substitutes.

The same study tested HM 46 hydraulic oil from five rubber vulcanizing presses at incubation periods from 12 to 120 hours (Pravda et al., 2024). Four of the five groups showed ΔE rising with longer incubation, group 1 only marginally and group 3 ranging from 20.28 to 38.38, while group 2 showed no clear link. Highly degraded samples clogged patches and filtered unevenly, and the authors concluded that incubation can be shortened below 72 hours for visibly degraded hydraulic oils without losing accuracy (Pravda et al., 2024).

Two limits from turbine practice carry over. ASTM D7843 extracts the insoluble contaminants in the sample onto the patch (ASTM International, 2025), so dirt and wear particles may add to the patch color (inference); pair ΔE with a particle count. The RULER method does not cover antiwear hydraulic oils either: ASTM D6971 applies to non-zinc turbine oils with rust and oxidation inhibitors "but not antiwear additives" (ASTM International, 2022). For a hydraulic system, trend ΔE against the same fluid and follow the fluid maker's guidance.

Common mistakes with MPC results

The most common mistakes with MPC results are treating one ΔE band as a universal limit, comparing results from different conditioning, and ignoring particles in the sample. Chokelarb et al. (2022) report that ASTM D4378-20 suggested a warning limit at ΔE 30. Lantos et al. (2022) describe a consensus of ΔE above 30 as condemning, above 20 as alarming and below 15 as safe. Turbomachinery International (2013) says most labs treat results above 40 as critical and that guidelines "will vary between different test labs." A cleanup target of ΔE under 15 is quoted in STLE (2016). Use one laboratory's bands; act on the trend.

MPC rose when the same oil stood 168 hours or four months instead of 72 hours before filtering (Lantos et al., 2022). Patch color also became brighter when the oil was mixed with solvent or filtered hot (Kon et al., 2018). Results across laboratories or procedures do not compare directly. Ask for the incubation time with every result. Light matters too: the standard incubation runs in the absence of light (Hong and Jang, 2023), and one study used shielded containers because UV light exposure can heighten deposit formation in oil samples (Pravda et al., 2024), so use light-blocking bottles.

Because the method collects all insoluble contaminants on the patch, not only varnish (ASTM International, 2025), read ΔE with the ISO 4406 particle code. Conventional colorimetric analyzers read only reflected light from the patch, while the colorimetric patch analyzer (CPA) method uses both reflected and transmitted light (Hong and Jang, 2023), so ΔE values from the two instrument types should not be compared directly (inference). A dyed oil is outside the method's scope (ASTM International, 2025).

What tests should run alongside MPC?

RULER antioxidant analysis, RPVOT, a particle count and, where the laboratory offers it, a soluble-varnish measure should run alongside MPC, because each answers a question that ΔE cannot. ASTM D4378-24 is the practice for in-service monitoring of mineral turbine oils, and its values "are purely indicative," to be read with equipment type, workload and top-up level (ASTM International, 2024).

RULER (ASTM D6971) measures the hindered phenol and aromatic amine antioxidants left in the oil. The standard adds that it does not "measure the overall stability of an oil" (ASTM International, 2022). Turbine oils are commonly used until remaining active antioxidants reach 25% of the new oil (Lantos et al., 2022), but varnish can appear earlier, which is why MPC and RULER are read together.

RPVOT (ASTM D2272) assesses the remaining oxidation test life of in-service oil and is not intended to compare the service lives of new oils of different compositions (ASTM International, 2022). Its reproducibility of plus or minus 22% suits large drops over time, not small changes (Hong and Jang, 2023).

A soluble-varnish measure covers what MPC leaves out. Chokelarb et al. (2022) monitored soluble and insoluble color bodies at three power plants and warn that relying only on insoluble MPC can mislead maintenance decisions. STLE (2016) also lists the ultracentrifuge test with MPC and RULER as a combined approach. Lubinpla applies this kind of combined reading to a customer's own documents and site history and returns one cited document for review.

Key takeaways for reading MPC results

• ASTM D7843 MPC measures varnish potential in the oil sample, not deposits already on surfaces.

• A low ΔE after an outage, an oil change without a flush or a large top-up can hide existing deposits.

• ΔE bands differ by source and laboratory; trend against the same oil and the same laboratory.

• Read MPC with RULER, RPVOT and a particle count; a rising ΔE with high RULER is an early warning.

• For hydraulic oils, MPC has no standardized basis; trend it only against the same fluid.

Lubinpla applies this kind of reasoning to your own documents and site history and drafts the document for your review before it goes out.

Sources checked 2026-10-13.

Sources

1. ASTM International. ASTM D7843-25e1 Standard Test Method for Measurement of Lubricant Generated Insoluble Color Bodies in In-Service Turbine Oils using Membrane Patch Colorimetry. 2025. https://www.astm.org/d7843-25e01.html

2. ASTM International. ASTM D4378-24 Standard Practice for In-Service Monitoring of Mineral Turbine Oils for Steam, Gas, and Combined Cycle Turbines. 2024. https://www.astm.org/d4378-24.html

3. ASTM International. ASTM D6971-22 Standard Test Method for Measurement of Hindered Phenolic and Aromatic Amine Antioxidant Content in Non-zinc Turbine Oils by Linear Sweep Voltammetry. 2022. https://www.astm.org/d6971-22.html

4. ASTM International. ASTM D2272-22 Standard Test Method for Oxidation Stability of Steam Turbine Oils by Rotating Pressure Vessel. 2022. https://www.astm.org/d2272-22.html

5. ASTM International. ASTM D4898-23 Standard Test Method for Insoluble Contamination of Hydraulic Fluids by Gravimetric Analysis. 2023. https://www.astm.org/d4898-23.html

6. Kon T., Honda T., Sasaki A. Influence of Solvents, Oil Temperature, and Incubation Period on Membrane Patch Color. Tribology Online 13(5). 2018. https://www.jstage.jst.go.jp/article/trol/13/5/13_225/_article

7. Kon T., Honda T., Sasaki A. Estimation of the Oxidative Deterioration of Turbine Oil Using Membrane Patch Color. Advances in Tribology. 2020. https://doi.org/10.1155/2020/1708408

8. Pravda S., Sedenka D., Blata J., Hrabovsky L. Influence of incubation period on membrane patch color for insoluble contaminant measurement in degraded hydraulic oils. Frontiers in Mechanical Engineering 10. 2024. https://www.frontiersin.org/journals/mechanical-engineering/articles/10.3389/fmech.2024.1366543/full

9. Hong S.H., Jang E.K. Varnish Formation and Removal in Lubrication Systems: A Review. Materials 16(10). 2023. https://pmc.ncbi.nlm.nih.gov/articles/PMC10222146/

10. Chokelarb W., Assawasaengrat P., Sitton A., Sirisithichote T., Sriprom P. Soluble and Insoluble Varnish Test Methods for Trending Varnish Buildup in Mineral Turbine Oil. Journal of the Japan Institute of Energy 101. 2022. https://www.jstage.jst.go.jp/article/jie/101/12/101_242/_article

11. Lantos A.B., Wooton D., Lauría S., et al. (WearCheck Argentina). Assessing Oxidation Condition and Lubricant Refreshment in Turbine Oils. WearCheck Technical Bulletin, Issue 81. 2022. https://wearcheck.com/resources/techdoc/WZA081.pdf?rev=1753322448

12. Turbomachinery International. Interpreting results for the new ASTM standard for varnish potential (D7843). 2013. https://www.turbomachinerymag.com/view/interpreting-results-for-the-new-astm-standard-for-varnish-potential-d7843

13. Van Rensselar J. The unvarnished truth about varnish. STLE Tribology and Lubrication Technology. 2016. https://www.stle.org/files/TLTArchives/2016/11_November/Feature.aspx

Written by the Lubinpla team from the sources listed and checked against them before publication.