GMCC KSK89D35UEZ3 Rotary Compressor: R32/R410A Technical Data, DC Inverter Performance and Selection Guide

The GMCC KSK89D35UEZ3 is a hermetic DC inverter single-cylinder rotary compressor rated at 8.9 cm3/rev displacement, 2,790 W cooling capacity, 715 W input and a coefficient of performance of 3.90 under the manufacturer's SEER60 test condition. It accepts R32 and R410A.

Author: AUTHOR_NAME, AUTHOR_JOB_TITLE, refrigeration and air conditioning equipment sourcing specialist at Qishanr Technologies. Reviewed against the GMCC Rotary Compressor Product Manual 2024. Published YYYY-MM-DD. Updated 2026-09-24.

What is the GMCC KSK89D35UEZ3?

The KSK89D35UEZ3 is the second of two 8.9 cm3/rev single-cylinder DC inverter models that GMCC prints in the R32 and R410A section of its current rotary compressor manual, and it sits in this site's GMCC rotary compressor range. It shares its swept volume, its 240 mm shell height and its 8.1 mm suction and 9.8 mm discharge connections with KSK89D32UEZA31, and the two rows differ only in their rated operating point. Where the first row is rated at 2,795 W for 728 W of input, this one is rated at 2,790 W for 715 W, which is a 1.56 percent efficiency advantage bought with a 0.18 percent capacity give-up.

That trade is the most useful fact about this part number, and it is the reason it exists. A buyer comparing the two model strings sees two numbers that are nearly the same length and one digit apart, and could reasonably assume the later row is simply a corrected version of the earlier one. It is not. The manufacturer has printed both rows as current products in the same table, under the same group heading, at the same displacement, and the difference between them is a deliberate choice about where on the capacity and efficiency curve the compressor should sit. For a system designer, that choice is the whole selection decision, because a compressor is not chosen for displacement alone but for how much cooling it delivers at the electrical input the design can afford.

The table below collects the published figures for this part number. Each row names where the figure comes from, because a compressor data sheet is only as useful as its traceability: a capacity figure without its test condition is not a specification, it is a number.

Two entries in this table deserve a note before they are used. The absence of a run capacitor is meaningful rather than missing data: the manufacturer prints a numerical capacitor value for every fixed-speed row in the same manual and a dash for every inverter row, so the dash is consistent with the inverter drive the group heading names. The test condition code SEER60 is printed as a label and is the seasonal efficiency condition the manufacturer applies to this group; it is not a measurement of any particular installed system.

The capacity per unit displacement figure, 313.5 W per cm3/rev, is derived here rather than published. It is included because comparing compressors of different displacement is otherwise impossible, and because it makes the trade with the sibling row explicit. It must not be quoted as a manufacturer rating.

Technical specifications and what each number is based on

A specification table in a selection guide has one job beyond listing values: it has to make clear which values are measured, which are printed and which are computed. The table below separates those three cases for every figure this guide uses.

The two arithmetic checks matter more than they might appear. The manufacturer prints a metric capacity, an imperial capacity, an input and a coefficient of performance for every row, and those four numbers are not independent: the imperial figure is the metric figure converted at 3.412 Btu per watt-hour, and the coefficient is the capacity divided by the input. Both relations hold exactly on this row, to the rounding of the printed figures. That does two things. It confirms that the row was transcribed correctly from the source, and it confirms that the row belongs to the group heading printed above it rather than to a neighbouring group, because rows read from the wrong group would not reproduce their own internal arithmetic.

A second kind of check applies to the whole group, and it is where the manual has something to disclose. Of the rows printed on the same page under the same family of headings, one row in the twin-cylinder column, KTF400D64UMTA at 39.8 cm3/rev, prints an imperial capacity that exceeds its own metric capacity converted at 3.412 by about 26 Btu/h. That row is not the row for this model, it is in a different cylinder class and a different part of the table, and it is not corrected or used as evidence here. It is disclosed because a reader comparing several rows in the same document is entitled to know that not every printed figure in it survives the arithmetic check, and because the alternative, silently showing only the rows that reconcile, would misrepresent the source as cleaner than it is.

How to read the KSK89D35UEZ3 model number

The manufacturer publishes a naming legend for its rotary compressor range, and it is worth decoding this part number against it, because two segments of the string carry selection information that the table does not repeat.

Two positions in this string cannot be fully decoded, and that is a statement about the manufacturer's documentation rather than about the compressor. The manufacturer publishes two naming legends in the edition reviewed, one for the general model construction and one for special specifications, and neither of them defines the third position of this string or the trailing markers after the power code. The displacement block and the power code can be decoded because the printed table independently confirms both, which is the standard this guide applies: a decoded segment is only reported as decoded when a second source agrees with it.

That leaves the two digits 35 in the trailing markers. They are not a displacement, because displacement is carried by the preceding block and the table agrees with it. They are not a refrigerant or a structure code, because those positions are already taken. Nothing in the manufacturer's published legends assigns them a meaning, so nothing is assigned here.

The power code is worth one further check, because it can be tested against output the manufacturer did not intend as a naming check. GMCC part numbers carry the supply class in a fixed position, and this site's own model pages print both the part number and the drive type the page states for it. On the pages available for comparison the position holds: the models whose string carries D are the pages that print a DC inverter drive, and the models whose string carries V are the pages that print a 50 Hz single-phase supply. Two examples from the same KSK89 displacement family are the KSK89D59UEZC page, which prints a DC inverter drive for a string carrying D, and the KSM89V1VDZ page, which prints a 50 Hz single-phase supply for a string carrying V.

One caveat belongs with this table. Several of these pages state the drive type in a row that is labelled for voltage rather than for drive, so the value a reader sees under the heading Voltage on those pages is a drive description such as DC inverter rather than a voltage figure. Two of the pages checked print a cylinder count in that same row that contradicts the cylinder count printed in their own bullet list, and a third prints a cylinder count in a model string that carries the single-cylinder structure code. Those are defects in the site's own product templates, they are not data about this compressor, and this guide does not reproduce any of them as a specification. The table above uses those pages only for the drive class they state, which is the one thing all of them agree on and which the manufacturer's table independently confirms through the group heading.

Where the KSK89D35UEZ3 sits in GMCC's 8.2 to 10.3 cm3/rev group

The manufacturer prints this model inside a group of single-cylinder DC inverter rows rated under one heading for R32 and R410A at the SEER60 condition, and this site indexes the family inside the wider rotary compressor category. Reading the whole group rather than the single row is what makes the positioning visible, because the group shows how the manufacturer spends displacement, shell height and efficiency across a family that was clearly designed as a range rather than as a set of unrelated parts.

Every row in this table is a manufacturer's printed row from the same page and the same group heading, and every row has been checked against its own internal arithmetic. This model's row is in bold. The rows are shown so that a buyer can see the shape of the range rather than a single point in it.

Four things are visible in the table. The first is that the same displacement does not imply the same rating, which the two 8.9 cm3/rev rows demonstrate directly. The second is that the manufacturer reuses one mechanical envelope across a wide band: rows from 7.5 to 14.0 cm3/rev share a 240 to 260 mm height, an 8.1 mm suction connection and either a 9.8 or a 12.9 mm discharge. The third is that connection diameter does not track displacement in a simple way, because the 8.2 cm3/rev row carries a 6.15 mm suction while rows above and below it carry 8.1 mm. The fourth is the pattern that matters most for selection, and it is examined in its own section below: at a constant displacement, this range offers several different efficiency levels rather than one.

A note on scope. The group table above is the single-cylinder R32 and R410A set printed under a DC inverter heading at the SEER60 condition. Two further model strings are printed in the same group but are absent from the table, KSK53D15UEZ3 and KSN108D34UEZ3, because their numeric cells do not survive text extraction from the manual; they are named here so that the table is not read as the complete group. It is not the whole dealer catalogue and it is not every GMCC compressor at these displacements. Rows from other groups in the same manual, including the fixed-frequency and the higher-pressure sections, are not comparable with these rows and are deliberately absent, because a capacity figure from another group is measured at a different condition and combining the two would produce a comparison that looks precise and is not.

KSK89D35UEZ3 against KSK89D32UEZA31 at the same displacement

This is the comparison a buyer of either part number actually needs, because the two rows are the only 8.9 cm3/rev single-cylinder inverter entries under this group heading, and a designer who reaches this displacement has to pick one.

The arithmetic of this trade is worth stating plainly, because it is the whole basis of choosing between the two. Thirteen watts less input and five watts less capacity produces a coefficient of performance that is one and a half percent better. Expressed per unit of swept volume rather than per unit of capacity, the design gives up one point eight percent of its input intensity and holds almost all of its capacity intensity. That is the signature of an efficiency-optimised variant rather than of a smaller machine: nothing about the physical envelope changed, and the capacity that was given up is a fifth of the input that was saved.

For a system designer the practical consequences are specific. Because the capacity falls slightly, a design that was sized exactly at the 2,795 W point of the sibling row has to absorb five watts, which is far inside the tolerance any real system already carries. Because the input falls by thirteen watts, the same design draws less power at the same duty. Because the envelope, both connection diameters and the mounting are identical, neither the pipe work nor the chassis has to change, so the substitution is a data-sheet decision rather than a redesign. And because both are inverter rows, the capacity can be modulated in either case, which means the five watt difference is a difference in the shape of the modulation range rather than in the ability to modulate at all.

There is one further consideration that decides the choice for many projects, and it is not on the data sheet. Optimising a compressor's rated point for efficiency typically shifts where the efficiency peak sits on the operating map. Two rows at the same displacement with different rated points are not evidence that one dominates the other across the whole envelope; they are evidence that the manufacturer has characterised the two differently. A selection should therefore be made against the part-load conditions the system will actually see, which means asking the manufacturer for the performance map rather than choosing on the single catalogue point. The catalogue point is sufficient to shortlist the two rows. It is not sufficient to award one of them. This site carries a selection guide for the sibling row, KSK89D32UEZA31 guide, and for a smaller displacement member of the same family, KSK82D22UEZA31 guide.

Does this range offer several efficiency levels at one displacement?

It does, and this is the most commercially useful pattern in the group, because it means a displacement does not lock a project into a single efficiency. The clearest evidence is a family of six rows at 9.8 cm3/rev that share one identical rated capacity and one identical imperial capacity, and differ only in the input they consume to reach it.

Six rows, one displacement, one capacity, and an input that falls from 784 W to 758 W. The coefficient rises from 3.92 to 4.05, which is a 3.3 percent efficiency improvement across the family with no change to the duty the compressor is rated to deliver. The two rows at the top of that efficiency sequence both move the discharge connection from 9.8 mm to 12.9 mm, which is a system-level consequence rather than a data-sheet one: a buyer selecting the most efficient member inherits a larger discharge line.

Nothing in the six model strings explains the progression through the published legends, because the segments that change are the trailing variant markers rather than the decoded positions. The progression is therefore reported here as a pattern in the manufacturer's printed data, not as a rule derived from the naming convention. What can be said from the printed data is what the table shows: the manufacturer's range carries multiple efficiency levels at a fixed displacement and a fixed capacity, and the only way to identify which level a particular part number delivers is to read its row.

The same pattern appears in the two other displacement families in the group. At 12.0 cm3/rev the two rows share a capacity of 3,730 W and differ in input by 25 W, taking the coefficient from 3.85 to 3.95. At 14.0 cm3/rev the three rows share a capacity of 4,350 W and differ in input by 30 W, taking the coefficient from 3.85 to 3.95. In both cases the efficiency step is bought without touching the displacement or the rated capacity, which is what allows a product family to be repositioned between efficiency tiers on a shared platform.

For the KSK89D35UEZ3 specifically, the consequence is that it is the more efficient of the two 8.9 cm3/rev rows, and that a designer who needs more efficiency than it offers cannot get it by staying at 8.9 cm3/rev under this heading. The next efficiency step in the group is at a different displacement, which means accepting a different capacity class and a different envelope. That is a real constraint and it should be established before the compressor is committed to a platform, not after.

Which chassis does the KSK89D35UEZ3 share?

The KSK89D35UEZ3 sits inside a mechanical envelope that the manufacturer reuses from 7.5 cm3/rev to 10.3 cm3/rev, which is a displacement spread of about 37 percent across one 240 mm shell and one pair of connection diameters. That is the practical meaning of a chassis in this context, and it is the single most useful thing a product planner can take from the range: a family of air conditioning products can be moved up or down the capacity band without redesigning the pipe work, the mounting or the cabinet space allocated to the compressor.

Two limits on that reading should be stated. Sharing a printed envelope is not a guarantee of interchangeability, because the mounting feet, the accumulator and the internal clearances are not printed in the table used here, and a substitution within the envelope still has to be confirmed against the manufacturer's drawing for the specific part. And the envelope does not extend without limit: the rows adjacent to this band move to a 250 or 270 mm height at the same or a smaller displacement, so the flat envelope is a property of this band rather than of the whole family.

Why is this compressor on R32 and R410A, and what does that mean for a system?

The group heading for this row names both R32 and R410A, and that is not a formality. It means the manufacturer rates one mechanical platform for two refrigerants that differ in pressure, in volumetric capacity and in the safety classification they carry, and a buyer has to decide which of the two the system will use before the compressor is ordered rather than after.

The table is deliberately short, because the manufacturer's compressor manual does not print a refrigerant comparison and this guide does not invent one. The entries that are drawn from published sources are the composition of R410A and the A2L classification of R32, both of which come from the standards and regulatory sources registered at the end of this guide. The design pressure ordering follows from the refrigerants' own properties rather than from a compressor figure.

What the compressor table does establish is where the responsibility sits. Because one platform is rated for both refrigerants, the compressor is not the component that decides the refrigerant; the system is. A buyer replacing a compressor in an existing R410A unit must replace it with an R410A unit, because the oil and the metering device in that system were selected for a blend at a given pressure. A buyer designing a new platform can choose either, and the choice then propagates into the charge limit, the leak detection and the siting of electrical components through the A2L requirements.

The European prohibition schedule is the other half of that decision, and the thresholds are what the substance does or does not clear rather than the dates alone. R32 has a 100 year global warming potential of 675, which sits below the 750 threshold applied to single split air conditioning with a charge under 3 kg from January 2025. It sits above every 150 threshold in the schedule, and those arrive for self-contained air conditioning and heat pumps and for split air-to-water systems in 2027, for split air-to-air equipment up to 12 kW in 2029, for larger split systems in 2033, and as a prohibition on any fluorinated greenhouse gas in split systems up to 12 kW in 2035. The values in this paragraph are published by the European Commission's assessment of climate-friendly alternatives to F-gases, which draws them from the annexes of Regulation (EU) 2024/573, and the German Environment Agency publishes a global warming potential table keyed to the same regulation for reference values across substances. The role of refrigeration and cooling demand in the wider market is set out in the International Energy Agency's report The Future of Cooling.

One boundary belongs in the same paragraph. This guide describes where R32 stands as a substance. It does not state that any particular appliance complies with anything, and no certificate held by this compressor is claimed here. A2L is a classification that places obligations on the finished product, and the completed unit is assessed against the product-level safety standard rather than against the compressor standard alone.

Which standards apply to a compressor like this?

Pointing at a standard is not a claim that this compressor has been assessed against it. The compressor standard covers the compressor as a component. The appliance standards cover the finished product, and it is the finished product that carries the refrigerant charge and therefore the refrigerant safety obligations. The performance rating standard is a measurement convention rather than a safety requirement, which is why the same compressor can carry two capacity figures under two conventions without either being wrong. This guide also cites three International Organization for Standardization entries by number rather than by link, because the standard catalogue entries could not be retrieved programmatically during the preparation of this guide; the standards are named so that a reader can locate them directly.

Which applications suit the KSK89D35UEZ3?

The rating basis is the first thing to settle, because it decides which applications the published data actually supports.

• Residential split air conditioning and small multi-split systems. At 8.9 cm3/rev with 2,790 W of rated capacity under a seasonal efficiency condition, this is a single split or small multi-split compressor, and the inverter drive allows the capacity to be modulated to the load rather than cycled.
• Applications where the listed capacity is close to the design point. The five watt gap to the sibling row makes this part useful where a design point sits just under the needed capacity and the efficiency matters more than the last few watts.
• Product families that need one envelope across several capacity classes. Because the 240 mm envelope and the 8.1 mm and 9.8 mm connections are shared from 7.5 to 10.3 cm3/rev, a family can span that band without changing the chassis, the pipe work or the cabinet allocation.
• Platforms being positioned between efficiency tiers. The six-row 9.8 cm3/rev sequence in the same group shows that this range is designed to be moved between efficiency levels at a fixed capacity, and the two 8.9 cm3/rev rows are the same idea applied at this displacement.
• New builds choosing between R32 and R410A. The platform is rated for both, so the refrigerant decision can be made on the system's compliance position rather than on compressor availability.
• Not for a duty that depends on a refrigeration rating condition. The published rating for this row carries a seasonal efficiency condition, which is an air conditioning basis. A buyer sourcing this part against a refrigeration or freezer duty point, or against a low-temperature evaporating condition, does not have a rating for that duty in this document, and should not treat the 2,790 W figure as one. The manufacturer prints separate low-pressure and refrigeration sections, and the correct part for such a duty has to be selected from those.

That exclusion is not a technicality. The manufacturer's section heading for this page addresses air conditioning compressors, and the condition code attached to the rating is a seasonal efficiency condition rather than a refrigeration pull-down condition. This site's own product pages for members of the same family list refrigeration and chiller among their applications, and that listing is a statement about what a compressor can be built into rather than a published rating at a refrigeration condition. The two claims are not in conflict, but only one of them is supported by a printed capacity figure, and it is the air conditioning one.

Selection checklist

1. Confirm the refrigerant the system will use. The platform is rated for R32 and R410A, and the choice changes pressures, the metering device and the compliance position. Settle it before ordering, not after.
2. Check the drive class. This part number carries the DC inverter power code. A fixed-speed drive board will not run it, and an inverter board selected for a fixed-speed compressor will not either.
3. Match the duty point to the published rating basis. The published figures carry a seasonal efficiency condition, which is an air conditioning basis. If the design point is a refrigeration or low-temperature evaporating condition, request the rating for that condition rather than extrapolating from this one.
4. Compare against the sibling row before committing. The other 8.9 cm3/rev row in the same group is rated 5 W higher in capacity for 13 W more input. That trade decides the selection more often than any other single figure.
5. Verify the envelope against the drawing. The 240 mm height and the 8.1 mm and 9.8 mm connections are printed, but mounting feet, accumulator position and internal clearance are not printed in the table used here. Confirm them against the manufacturer's drawing for the specific part.
6. Confirm the connection diameters against the system. The suction and discharge figures are 8.1 mm and 9.8 mm. Rows adjacent in the same range move to a 12.9 mm discharge, so a change of tier can change the pipe work.
7. Ask for the part-load performance map. Two rows at one displacement with different rated points are characterised differently. A single catalogue point shortlists a compressor; it does not award one.
8. Establish the compliance position for the refrigerant. The global warming potential of R32 is above every 150 threshold in the European schedule. The dates and thresholds are in the section above and should be checked against the market the product will be sold in.
9. Request the figures the table does not print. Mass, oil charge, running speed, locked rotor current and sound power level are absent for this row in the edition reviewed. A system design that needs any of them needs them from the manufacturer.
10. Request the current document before finalising. The figures here come from the edition registered at the end of this guide. Ratings are revised, and a compressor selected on a superseded table is a project risk rather than a sourcing saving.

Frequently asked questions

Is the KSK89D35UEZ3 a fixed-speed compressor or an inverter compressor?

It is an inverter compressor. The group heading printed above its row names a DC inverter single-cylinder structure, its power code position carries D, which the manufacturer's legend defines as DC inverter, and its run capacitor cell is printed as a dash, which is how the same manual prints every inverter row and how it never prints a fixed-speed row. A fixed-speed drive should not be used with it.

What is the difference between the KSK89D35UEZ3 and the KSK89D32UEZA31?

They are the same displacement and the same envelope with different rated points. KSK89D35UEZ3 is rated at 2,790 W for 715 W of input, a coefficient of performance of 3.90. KSK89D32UEZA31 is rated at 2,795 W for 728 W, a coefficient of performance of 3.84. The first gives up 5 W of capacity to save 13 W of input, which is a 1.56 percent efficiency advantage. Both rows are printed as current products under the same group heading, so the difference is a design choice rather than a revision.

Is there a more efficient option at the same displacement?

Not under this group heading. The two 8.9 cm3/rev single-cylinder inverter rows printed in this group are the pair compared above, and the more efficient of the two is the one described here. The next step up in efficiency inside the same group appears at a different displacement, which means a different capacity class and a different envelope.

Can I use the published 2,790 W figure for a refrigeration or chiller duty point?

No, not without a rating for that condition. The figure is published under a seasonal efficiency condition, which is an air conditioning basis, and the section this row is printed in addresses air conditioning compressors. The manufacturer prints separate sections for other duty ranges, and a refrigeration or low-temperature selection should be made from those. This site's own pages for related models list refrigeration among their applications, which is a statement about what a compressor can be built into rather than a published capacity at a refrigeration condition.

What refrigerant does the KSK89D35UEZ3 use?

The group heading names R32 and R410A, so the platform is rated for both. R32 is a single substance with a safety classification of A2L and a 100 year global warming potential of 675. R410A is a blend containing R32 and R125 and carries the A1 classification. Because both refrigerants are handled with polyol ester oil, the oil family is common to the two, but the metering device, the charge procedure and the product-level safety requirements differ, and the refrigerant must be fixed before the compressor is ordered.

Where can I obtain the official GMCC document for this model?

The manufacturer publishes its rotary compressor documentation through its own product centre and its downloads page, where the product manual is distributed after a form submission. The edition this guide was checked against was read from a distributor-hosted copy of the GMCC Rotary Compressor Product Catalog 2024, which is named here as a third-party host rather than presented as a manufacturer address. Further GMCC model guides are indexed in this site's GMCC model guides on this site.

Does this compressor carry a certificate for any particular market?

No certificate held by this compressor is claimed in this guide, and none was sighted during its preparation. The standards named above are the frameworks a compressor and the appliance built around it are assessed against; naming a framework is not a certification claim. Market entry documentation should be requested from the manufacturer or the distributor for the specific market and application.

Sources and verification notes

The figures in this guide come from the primary register below. Displacement, capacity in both units, input power, coefficient of performance, height and both connection diameters for KSK89D35UEZ3 are taken from one row of the manufacturer's table and the group heading printed above it. Sibling model figures are presented as sibling data and are never offered as data for this part number.

• Manufacturer product centre, the route to the current product categories and the model index: GMCC's official product centre
• Manufacturer downloads page, where the rotary compressor product manual is distributed behind a form: GMCC official downloads page
• The edition this guide was checked against, hosted by a distributor: GMCC Rotary Compressor Product Catalog 2024
• An earlier manufacturer edition on a document platform, used only as a version-enrolment reference: FAWAZ-GMCC Rotary Compressor 2018 catalogue
• Standards and certification bodies, for the four levels of the standards framework: IEC 60335-2-34:2024, IEC 60335-2-89:2019, ANSI/ASHRAE Standard 34, AHRI 540
• Regulation and research, for the refrigerant position and the prohibition dates: European Commission's assessment of climate-friendly alternatives to F-gases, Regulation (EU) 2024/573, global warming potential table, The Future of Cooling
• Site structure and related model pages referenced in the comparison tables: the GMCC rotary compressor range, the rotary compressor category and the individual model pages linked in the model-number section
• Selection guides for adjacent options on this site, for a displacement above this family or for a change of refrigerant: EKPM310D85UMTR and the non-fluorinated alternative DSG320S1UFT

Four points about evidence limits belong here. First, a second-row check on the page this model is printed on found one twin-cylinder row, KTF400D64UMTA, whose printed imperial capacity exceeds its own metric capacity converted at 3.412 by about 26 Btu/h. That row is not the row for this model, is in a different cylinder class, and is neither corrected nor used as evidence here. Second, the test condition code SEER60 is printed as a label in the group heading but is not defined in the edition reviewed, so this guide reports that the rating carries the condition without claiming what the condition consists of; the operating points behind it should be requested from the manufacturer. Third, the group table does not print a mass, an oil charge, a running speed, a locked rotor current or a sound power level for this part number, so none of those is quoted anywhere in this guide. Fourth, the two model string segments that the manufacturer's published legends do not define, the third position and the trailing variant markers after the power code, are reported here as undecoded rather than assigned a meaning.