Why the APR Control Dehumidifies: Extended Compressor Runtime in Constant-Volume Systems

ABSTRACT

Constant-volume direct-expansion (DX) cooling equipment spends most of its operating hours at part load, where the compressor cycles on and off while the supply fan runs continuously. Published laboratory and field research has shown that this operating pattern severely degrades moisture removal: condensate retained on the coil evaporates back into the airstream during each off cycle, and net latent capacity at part load can fall below 50% of the steady-state rating [2, 3, 4]. The APR Control, a mechanical self-regulating external unloader, reduces effective compressor capacity at part load while holding suction pressure at its setpoint. This paper combines steady-state bench measurements of an APR-410-3 on a two-compressor 10-ton rooftop unit with the published part-load latent degradation framework to explain why the APR Control improves dehumidification in constant-volume systems. Bench data show (i) no measurable latent-capacity penalty from the APR Control when full capacity is demanded under a wet-coil latent load, (ii) continuous moisture removal throughout the APR Control's modulation range with the coil held above freezing, and (iii) measured compressor capacity reductions of 18.7% (both stages) and 37.5% (controlled stage alone), which for a fixed space load correspond to runtime extensions of approximately 1.23x and 1.60x. Longer on-cycles raise the runtime fraction, which published models and measurements identify as the dominant driver of delivered latent capacity in cycling constant-volume equipment. Rawal Devices' field experience with the APR Control spans more than thirty years, with space dehumidification the primary reason customers purchase the product; that record is reported here as field experience, not as a finding of this test program.

1. INTRODUCTION

The majority of packaged DX cooling equipment in light commercial service is constant-volume: the supply fan delivers a fixed airflow whenever the unit operates, and capacity is matched to load by cycling one or more fixed-speed compressors. Sensible load governs the thermostat, so at part-sensible-load conditions the compressor runs a diminishing fraction of each hour even when the latent (moisture) load remains high. Humid-climate shoulder seasons, high-occupancy spaces, and oversized equipment all produce exactly this combination: low runtime fraction, high moisture load. And surplus capacity is the normal condition, not the exception: equipment is selected against design conditions that occur only a small fraction of annual hours, and common sizing practice - rule-of-thumb cooling factors in place of full ACCA load calculations (Manual J/N), plus deliberate margin against the risk of undersizing - biases installed capacity further upward. The result, widely observed in the field, is equipment whose on-cycles at part load shrink to the order of minutes, too short for the coil to deliver meaningful moisture removal even as the thermostat is satisfied.

Henderson and Rengarajan [2] developed and Henderson [3] validated a latent capacity degradation model for precisely this equipment class, and a subsequent laboratory and field program by Shirey, Henderson, and Raustad [4] confirmed the underlying mechanism: moisture that condenses on the evaporator during the on cycle is not drained instantly, and when the compressor stops while the fan keeps running, the retained condensate evaporates back into the supply air. Net moisture removal at part load can fall to less than half of the equipment's steady-state latent rating. The problem is not the coil's steady-state performance; it is the fraction of time the coil spends cold and condensing.

The APR Control is a mechanical, self-regulating modulating unloader. When falling load pulls suction pressure below the setpoint of its Compression Ratio Reduction Valve (approximately 118 psig / 40 F saturated suction for R-410A, adjustable), the valve opens and diverts a portion of compressor discharge gas into a desuperheating chamber, returning it to suction as cool gas. The effect is a proportional reduction of effective evaporator capacity while the compressor continues to run [1]. In a constant-volume system, reducing compressor capacity at part load has a direct arithmetic consequence: the unit must run longer to meet the same space load. This paper assembles the measured bench evidence and the published latent-degradation framework into a single argument for why that longer runtime translates into better dehumidification.

Rawal Devices has applied the APR Control in the field for more than thirty years, and space humidity control is the primary reason customers purchase it. That field record motivates this paper but is reported as company field experience; every quantitative claim below is either a measurement from the referenced bench program, an explicitly labeled calculation from those measurements, or a citation to published research.

2. THE PART-LOAD DEHUMIDIFICATION PROBLEM IN CONSTANT-VOLUME EQUIPMENT

Three published findings frame the problem. First, latent capacity degradation at part load is real and large: with continuous fan operation, off-cycle re-evaporation of retained condensate can reduce delivered latent capacity to below 50% of the steady-state value [3, 4]. Second, the degradation is governed primarily by runtime fraction: the shorter the on cycle relative to the off cycle, the larger the share of condensed moisture that returns to the space instead of the drain [2, 4]. Third, the degradation is worst in exactly the applications where dehumidification matters most: humid climates at part-sensible load, where the equipment cycles frequently while the fan supplies continuous ventilation [4].

The oversizing context sharpens the problem. Cooling equipment is selected to meet a design load computed at outdoor design conditions that, by definition, occur only about 1% of annual hours; for every other operating hour the equipment carries surplus capacity relative to the actual load. Sizing practice widens that surplus. In place of a full ACCA load calculation (Manual J for residential, Manual N for commercial buildings), capacity is commonly selected from rule-of-thumb cooling factors - square feet of floor area per ton - and then padded with margin, because the professional consequences of undersizing are immediate and visible while the consequences of oversizing are chronic and hidden. The hidden consequence is humidity. An oversized unit satisfies the thermostat quickly and shuts down: within the thermostat-cycling framework used in the cited research, a unit at 50% runtime fraction already sees on-cycles on the order of ten minutes, and each increment of surplus capacity pushes the runtime fraction lower and the on-cycles shorter [2, 4]. Cycles of only minutes leave the coil little time cold and condensing, and the off-cycle re-evaporation mechanism returns much of what little was condensed. Space temperature reads satisfied; space humidity climbs.

Note the reach of this problem: it does not require a sizing error. Even a unit sized by a rigorous load calculation is oversized for nearly every hour it operates, because the design condition it was sized against is rare. Oversizing by rule of thumb merely deepens a part-load condition that correct sizing already guarantees.

It follows that, for a constant-volume system, any measure that lengthens compressor on-cycles at part load, without sacrificing latent capacity while running, moves the equipment up the runtime-fraction curve and recovers latent capacity that cycling would otherwise forfeit. The remainder of this paper shows, from bench measurements, that the APR Control does both: it does not sacrifice latent capacity while running, and it substantially lengthens the runtime required to meet a given load. In effect, its load matching converts surplus capacity - whether from rule-of-thumb sizing or simply from off-design hours - from a short-cycling liability into extended, continuously dehumidifying runtime.

The practitioner-facing form of this argument - oversized equipment, short on-cycles, and the humidity complaints that follow - was set out for service technicians in the trade press in 2020 [5]. The present paper supplies the measured basis.

3. TEST UNIT, DATA SOURCE, AND METHOD

All measurements are from a laboratory bench program, conducted in a psychrometric chamber, on a nominal 10-ton, two-compressor R-410A packaged rooftop unit. An APR-410-3, rated for 3.5 tons of capacity modulation, was installed on the first-stage 5-ton compressor only; the second compressor and its circuit carry no APR Control and serve as an internal control. The valve setpoint was approximately 120 psig (about 40 F saturated suction temperature). Runs were conducted with the valve either enabled (open eras) or isolated (closed eras), across matrix conditions of entering dry bulb/wet bulb, outdoor dry bulb, and airflow. Run-era classification follows the classification established for this dataset in the companion papers.

Method note on latent capacity: the bench logs total unit capacity (Qtci) from the airflow measurement system, together with entering and leaving dry-bulb and wet-bulb temperatures and standard CFM. Latent capacity values in this paper are derived from measured channels, not read from a directly logged latent channel: sensible capacity is computed as 1.08 x CFM x (entering DB - leaving DB), and latent capacity is Qtci minus sensible. An independent psychrometric cross-check, computing humidity ratios from the measured dry-bulb and wet-bulb pairs at an assumed sea-level barometric pressure, reproduces the same trends with a systematic offset of roughly 10%; all comparisons below therefore rely on differences and ratios between runs rather than absolute calibration. Repeatability of the derived latent value, taken from the two documented repeat pairs in the dataset (runs 130/131 and 234/235), is 2.6% to 4.4%, or 925 to 1,104 Btu/hr in absolute terms.

Exclusions applied in this paper, following the standing rules for this dataset: shakedown runs 1-4 and empty run 37 are excluded; run 254's suction superheat channel is excluded (not used here); the approximately 3,900 CFM full-airflow comparison is excluded from the matched-pair table because the closed-era run (3,953 CFM) and the open-era runs (3,753-3,780 CFM) differ in airflow and are therefore not a controlled pair; and runs 253 and 254 are excluded from the moisture-removal table because their entering dew point (approximately 38 F) is at or below coil temperature, making the coil legitimately dry - run 253's small negative derived latent value (-1,806 Btu/hr) is measurement noise on a dry coil, not negative dehumidification.

The bench runs are open-loop: entering conditions and airflow were set by the test matrix and held, with no space thermostat in the control loop and no cycling. Every value reported here is a steady-state measurement under that arrangement.

4. MEASURED RESULTS

4.1 No latent penalty when latent load is present

The first requirement of a dehumidification argument is that the APR Control must not reduce moisture removal when the unit is running under a genuine latent load. Table 1 shows every matched open/closed pair at the wet-coil rating condition (80 F dry bulb / 67 F wet bulb entering, 95 F outdoor dry bulb; entering dew point approximately 60 F), by stage and airflow. At these conditions suction pressure sits above the valve setpoint, so the enabled valve is at or near its dormant state - which is exactly the field condition whenever full capacity is demanded.

Table 1. Derived latent capacity, matched open/closed pairs at 80/67 entering, 95 F outdoor. Full-stage differences are shown in percent; single-stage differences are shown in absolute Btu/hr because the small latent magnitudes make percentages misleading.

All eight pair differences are within, or marginally beyond, the 925-1,104 Btu/hr (2.6-4.4%) repeatability band established from the dataset's repeat pairs. It is noted plainly that all eight differences lean in the same direction (open slightly below closed); at face value the largest full-stage deficit is 3.3%. Even read uncharitably, a residual effect of that order is an order of magnitude smaller than the greater-than-50% latent losses that cycling operation produces in constant-volume equipment [3, 4]. The measured conclusion stands: with latent load present and capacity demanded, the APR Control leaves latent capacity essentially unchanged.

4.2 Moisture removal continues throughout modulation

The second requirement is that the coil keeps dehumidifying while the valve modulates. Table 2 lists the open-era runs in which the valve is at or beyond its setpoint (suction at or below approximately 120 psig) with entering dew point above coil temperature, spanning airflows from 1,499 to 2,999 CFM and saturated suction temperatures from 41.8 F down to 32.3 F - the last two runs (30 and 57) being the deepest modulation observed in the program, with the valve driven past its rated range.

Table 2. Wet-coil open-era runs with the valve actively modulating. Latent values are derived from measured channels as described in Section 3.

Every run in the table removes moisture continuously: leaving dew point sits 4 to 17 F below entering dew point, and derived latent capacity remains between roughly 20,000 and 42,000 Btu/hr. Run 55 is the most application-relevant case: at 1,501 CFM with a full 60.5 F entering dew point, the valve holds suction at its 120 psig setpoint and the unit removes 41,539 Btu/hr of latent - within repeatability of its closed-era twin - while pulling the supply dew point down to 43.4 F. At the modulation extreme (runs 30 and 57), the coil still removes approximately 20,000-21,000 Btu/hr of moisture. At no point in the open-era record does the coil reach freezing: the minimum observed saturated suction with the valve enabled is 32.3 F.

4.3 The steady-state contrast, stated honestly

This paper does not claim that the APR Control removes more moisture per running hour than an unmodulated compressor at the same instantaneous condition. The dataset contains the direct contrast: at the same deep part-load condition as runs 30 and 57, closed-era run 25 drives saturated suction to 24.2 F and removes 26,660 Btu/hr of derived latent - more per steady hour than the modulated 20,223-21,395 Btu/hr - precisely because the coil is colder. But that operating state is below freezing. On the bench it is a controlled half-hour snapshot; in the field, a coil held below 32 F accumulates frost, which progressively blocks airflow and ends moisture removal, or the unit trips its low-pressure safety and cycles off. The dataset's baseline census makes the exposure plain: of 154 documented closed-era runs, 10 ran below 32 F saturated suction (minimum 24.2 F), while none of the 26 documented valve-enabled runs did (minimum 32.3 F). The unmodulated compressor's apparent steady-state latent advantage at deep part load is an advantage the equipment cannot actually keep; the APR Control trades it for an operating state the equipment can hold indefinitely.

5. RUNTIME EXTENSION IN A CONSTANT-VOLUME SYSTEM

This section is labeled explicitly: the capacity reductions below are measured; the runtime multipliers computed from them are modeled from those measurements. No thermostat-cycling test was conducted in this bench program, and no cycling measurement is claimed.

Measured capacity reduction: at the deep part-load condition, total unit capacity with both compressors running falls from 87,531 Btu/hr (closed, run 25) to 71,201-74,479 Btu/hr (open, runs 57 and 30), a reduction of up to 18.7%. On the controlled stage alone, the matched pair 252/253 shows 37,646 Btu/hr closed against 23,531 Btu/hr open, a 37.5% reduction. Both figures reconfirm, in the current data file, the modulation-depth results established in the companion papers.

Modeled runtime consequence: in a constant-volume system, the runtime fraction required to meet a given space load is approximately the ratio of load to capacity. A capacity reduction of 18.7% therefore extends runtime for the same load by a factor of about 1/(1 - 0.187) = 1.23; a reduction of 37.5% on a single-stage system extends it by about 1/(1 - 0.375) = 1.60. These multipliers are arithmetic on the measured steady-state capacities, and they represent the upper end of the modulation range - the valve is proportional, so the runtime extension scales continuously with how far load has fallen.

The published framework supplies the final step. Henderson and Rengarajan's model and the subsequent laboratory and field validation identify runtime fraction as the dominant variable in delivered latent capacity for cycling constant-volume equipment with continuous fan: latent delivery falls steeply as runtime fraction drops, because a fixed quantity of retained condensate re-evaporates during every off cycle regardless of how short the preceding on cycle was [2, 3, 4]. Moving a system from short cycles toward long, continuous operation at reduced capacity therefore recovers latent performance disproportionately: the coil spends more of each hour cold and condensing (Section 4.2 shows it condenses throughout modulation), and fewer off cycles occur in which to give the moisture back.

6. SYNTHESIS: WHY THE APR CONTROL DEHUMIDIFIES

The argument assembles from four parts, each labeled by its evidentiary standing. Measured: the APR Control imposes no resolvable latent-capacity penalty when capacity is demanded under latent load (Section 4.1). Measured: the coil continues removing moisture across the entire modulation range, holding above freezing at all times (Section 4.2). Measured capacity reduction with modeled runtime arithmetic: at part load the APR Control reduces effective capacity by up to 18.7% (both stages) or 37.5% (controlled stage), extending runtime for a given load by up to roughly 1.23x and 1.60x respectively (Section 5). Published: delivered latent capacity in cycling constant-volume equipment is governed by runtime fraction, with off-cycle re-evaporation capable of erasing more than half of rated latent performance [2, 3, 4].

A constant-volume unit with the APR Control at part load therefore runs longer, cycles less, keeps its coil continuously below the space dew point, and never enters the frost regime that terminates moisture removal in the unmodulated baseline. The argument applies with particular force to oversized equipment - which, given prevailing sizing practice and the rarity of design conditions, describes most installed constant-volume equipment during most of its operating hours: the APR Control's load matching turns the surplus capacity that drives short cycling into the extended runtime that drives dehumidification. Each element of that sentence is either measured in this program or established in the published record. This mechanism is consistent with Rawal Devices' more than thirty years of field application, in which space dehumidification has been the primary reason customers purchase the APR Control; that field record is reported as company experience and stands outside the measured scope of this paper.

7. SCOPE AND LIMITATIONS

All findings apply to the APR Control, a mechanical self-regulating valve, as tested; no claim is generalized to the electronically controlled APR-E. All measurements are steady-state bench results from a single R-410A rooftop unit with the valve on one of two compressors; no cycling, annualized, or field-performance measurements are claimed. Latent capacity is derived from measured channels as described in Section 3 rather than logged directly, and comparisons rely on differences between runs rather than absolute calibration. The runtime multipliers in Section 5 are arithmetic consequences of measured capacity reductions, not measured cycle tests. The consistent slight negative lean of the matched-pair latent differences in Table 1 is reported rather than suppressed; it is within or marginally beyond the dataset's repeatability band and does not alter the conclusions.

8. CONCLUSIONS

In constant-volume DX equipment, part-load dehumidification is lost primarily to compressor cycling, not to any deficiency of the coil while running. Bench measurements show the APR Control preserves latent capacity when capacity is demanded, continues removing moisture throughout its modulation range while holding the coil above freezing, and reduces effective compressor capacity by measured margins that extend runtime for a given load by up to approximately 1.6x on the controlled stage. Published research establishes that this runtime extension is precisely the lever that governs delivered latent capacity in this equipment class. The combination explains, on measured and published grounds, why the APR Control dehumidifies.

REFERENCES