R4. Fans and airflow for a commercial freezer "semi-blast chiller"

Written 2026-08-30. All prices are listing prices observed on that date unless noted. Sources are given inline. Where I could not confirm a number I say so.

Scope: the air-movement half of the Tech Ingredients retrofit (video 1, https://www.youtube.com/watch?v=G1Syjh319VU, transcript at research/transcript2.txt), applied to a commercial reach-in upright or a commercial chest freezer in a commissary. Thermal mass (eutectic bottles) is covered elsewhere.

Summary

• The video's fan is the GESEXI slim radiator fan, 16 inch, 2-pack, Amazon B0DHKQS3DC, $59.99. Rated 12 V, 120 W (10 A), "CFM >= 1000", 2000 rpm, 10 blades, 3.27 in thick. The same family sells a 12 inch (80 W, "CFM >= 800", $28.99) and 14 inch (90 W, "CFM >= 900", $29.55). The listing describes exactly the reversal the video shows (flip the blade, swap the leads). No IP rating is stated anywhere on the listing.
• Running at 8 to 9 V cuts fan heat to roughly 40 percent of nameplate by the cube law: about 50 W for the 16 inch, 34 W for the 12 inch, at roughly 75 to 80 percent of full airflow. The video measured "about 80 percent of maximum airflow at less than half the power" at 8 to 9 V, which matches.
• A commercial reach-in already has forced air: a single ECM or shaded-pole evaporator fan in a back-wall plenum, on the order of 400 to 560 CFM (ebm-papst iQ 3612 with a 254 mm impeller, the motor True sells as part 989719). That gives a bulk cabinet velocity of 0.5 to 0.7 m/s if the flow were uniform, but measurements on display cabinets and fan-cooled freezers show shelf-level velocities mostly under 0.2 m/s away from the discharge. A blast chiller (Alto-Shaam QC3-40) moves 1,750 CFM through a 39.5 x 22.6 x 25.3 in cavity: 1.3 to 2.2 m/s average depending on the flow plane. Blast freezer design guides call for 1.5 to 3 m/s over packaged product and 3 to 5 m/s over bare product.
• Net: in a reach-in a booster fan roughly doubles to triples the surface heat transfer coefficient at the product (from a 10 to 15 W/m2K forced-but-slow baseline to 30 to 40 W/m2K in a 2 to 3 m/s jet). In a static chest freezer the baseline is natural convection at 5 to 8 W/m2K, so the same fan is worth 4 to 6 times. The video's 2x chill and 3x freeze speedup was measured in a fan-cooled household upright, so a commercial reach-in should land in the same range and a chest freezer should do better.
• Sealed fans that are actually rated for the environment: SPAL 30100467 12 inch (909 CFM, 6.9 to 7.4 A, IP68 motor, $93 to $112) for a 12 V build; ebm-papst iQ 3612 (115 V AC, IP42 or IP54, rated -40 to +50 C, $117 to $125 plus an impeller) if you want OEM refrigeration hardware. The cheap GESEXI fan is unrated for cold or moisture.
• Two configurations with parts and prices are in section 6. Both keep the PSU outside on a GFCI outlet and bring 12 V in through a flat cable under the door gasket or through the chest freezer drain plug.

1. The exact parts in the video

1.1 The fan: Amazon B0DHKQS3DC

Listing title: "GESEXI 16" Slim Fan Push Pull Electric Radiator Cooling Fan, Black, 2 Packs | Universal High Performance 12V 120W Motor with Fan Mounting Kit". https://www.amazon.com/dp/B0DHKQS3DC (seen 2026-08-30).

Price: $59.99 for two. That matches the video's "$50 for a pair".

Bullet text from the listing (quoted):

• "12V 120W 16" radiator engine cooling fan comes with mounting brackets ... CFM>=1000 CFM, Fan RPM=2000+-10% RPM. Fan Diameter (in): 16". Thickness (in): 3.2700". Number of Blades: 10."
• "The electric radiator cooling fan is reversible which can be used as a puller. Switch the push-pull working mode: only need to replace the power cord and turn the fan blade over to install. Please apply anti-loosening glue to the fixed nut to prevent the nut from loosening when turning the fan blades."

So the reversal the video shows (single nut, flip the blade, swap + and -) is the documented procedure. Use a drop of blue threadlocker on the nut as the seller says; a blade coming loose inside a freezer full of hotel pans is not a fun failure.

Current draw at 12 V: 120 W / 12 V = 10 A nameplate. The listing does not state amps separately. Treat "CFM >= 1000" as a marketing number; reputable 12 inch automotive fans at 7 A do about 900 CFM free air (SPAL, below), so a 10 A 16 inch fan doing 1000 to 1200 CFM free air is plausible, but there is no test data.

Motor sealing: not stated. No IP rating, no temperature range. The video's claim that these motors are "sealed" is generic to automotive radiator fans; it is not on this listing.

Size variants in the same listing family (all seen 2026-08-30, all "GESEXI" or unbranded, same design):

Nominal amps: 16 in 10 A, 14 in 7.5 A, 12 in 6.7 A.

Real outside diameter including the shroud is about 0.4 in more than the nominal size (a similar 12 inch fan lists 12.44 in overall, https://www.k2-industries.com/products/universal-12-pull-push-electric-radiator-engine-cooling-fan-w-mount). Plan on a 16.5 in square for the 16 inch.

1.2 The PSU: Amazon B09YSJQWRG

Jesverty SPS-3010, "0-30V 0-10A Adjustable Switching DC Regulated Bench Power Supply with High Precision 4-Digit LED Display, 5V/2A USB Port, Coarse and Fine Adjustment". https://www.amazon.com/dp/B09YSJQWRG (seen 2026-08-30).

Price: $47.49 shown as the buy price; a $49.99 figure also appears on the page, probably the list price.

Specs from the listing: 0 to 30 V, 0 to 10 A, display shows V, A and W (resolution 0.01 V, 0.001 A, 0.1 W), automatic CV/CC crossover, 7.1 x 3.35 x 6.1 in, about 2.5 lb, short-circuit and overload protection, temperature-controlled internal fan. The listing warns that the included test leads are "merely a give-away" and to use leads with real current capacity for long-term high-current use; at 10 A that means 14 AWG or heavier.

The 10 A ceiling is exactly the 16 inch fan's nameplate at 12 V, so at 12 V the PSU would be at its limit. At 9 V the fan draws roughly 7.5 A (see 1.4), which is comfortable. Historic deal prices for the SPS-3010 family ran $27 to $42 (https://plasticfantasticdeals.com/jesverty-sps-3010x-variable-switching-dc-power-supply-0-30v-0-10a/), so it is worth watching.

Why the video liked it: the wattage readout tells you the heat you are dumping into the box, and the knob lets you find the efficiency knee. For a production install you do not need either.

1.3 Cheaper fixed-voltage alternatives

All Amazon, seen 2026-08-30:

• SHNITPWR 12 V 10 A 120 W desktop brick, 5.5 x 2.5 mm barrel, B07PWZQ33N, $20.99. https://www.amazon.com/dp/B07PWZQ33N. Enough for one 16 inch fan at 12 V or two 12 inch fans at 9 V via a controller. The listing itself warns the supply "will be damaged soon" if the load exceeds 10 A, so do not run two 16 inch fans on it.
• PHc 9 V 5 A regulated wall adapter, 2.1 x 5.5 mm, B006QZ7YRU, $8.29. https://www.amazon.com/dp/B006QZ7YRU. 45 W. Enough for the 12 inch fan at 9 V (about 34 W, 3.8 A) with margin. Not enough for the 16 inch at 9 V (about 50 W, 5.6 A).
• RioRand 12 to 40 V, 10 A PWM brushed-motor speed controller, B007TH4EN6, $9.99. https://www.amazon.com/dp/B007TH4EN6. Rated 8 A continuous, 10 A max, 13 kHz, knob, fuse. Put it between the 12 V brick and the fan and set the knob for about 75 percent. Same effect as the bench PSU at 9 V for $31 total. PWM at 13 kHz on a brushed fan may whine audibly; if it does, turn the knob up or down a little.

Do not use 115 V AC fans inside the box unless they are refrigeration-rated (section 3). Everything in the 12 V path is SELV, which is the main reason to keep the fan side at 12 V.

1.4 Heat dumped into the box at 8 to 9 V

Every watt the fan draws ends up as heat inside the freezer (motor loss plus air friction). For a brushed DC fan, speed is roughly proportional to voltage and shaft power to speed cubed, so:

The video's own measurement (anemometer at fixed distance, voltage sweep) found "at about 8 to 9 volts this thing will move about 80 percent of its maximum air flow and use less than half of the power", which is the same cube-law knee. The transcript does not give the fan's actual watts at 9 V, so the numbers above are estimates.

For scale: a Beverage-Air FB23HC-1S 23 cu ft reach-in freezer has a 3/4 HP system rated 2,097 BTU/h, which is 615 W of cooling (spec sheet dated 12/25, https://store-2tiw84ejdt.mybigcommerce.com/content/spec-sheets/Vista-Series/FB23HC-1S.pdf). A 35 to 50 W fan is 6 to 8 percent of that, small but not nothing; the household freezer in the video had about 90 W compressor input at 50 percent duty, so there the fan was a much larger fraction, and it still paid off.

2. Does a commercial reach-in already move air, and how fast

2.1 What the manufacturers say

Reach-in uprights are all forced-air. Chest freezers are static (cold-wall coils in the liner, no fan).

• True T-23F-HC (27 in wide, 23 cu ft, 1/2 HP, 3.7 A, -10 F): spec sheet says "State of the art, electronically commutated evaporator and condenser fan motors. ECM motors operate at higher peak efficiencies and move a more consistent volume of air" and "Rear airflow guards prevent product from blocking optimal airflow." Wire shelves are 22 7/8 in x 23 1/4 in, which is a good proxy for the interior cross-section (about 3.7 sq ft). https://d3ld6frh4bdurh.cloudfront.net/pdf/598-T23F.pdf. Price $5,019.24 at https://www.plantbasedpros.com/product/true-t-23f-hc-one-section-reach-in-freezer/ (seen 2026-08-30).
• True's current evaporator motor is the ebm-papst iQ 3612 (True part 989719 = ebm-papst 55330.01530), $125.00 on Amazon B06XQLC9N5; the generic ebm-papst 55330.01210 (115 V, 15 W, 1550 rpm) is $117.41 on B07Y8TY2GQ (both seen 2026-08-30). Older True cabinets use 9 W shaded-pole motors (True 800401/800448/922036); an aftermarket 9 W motor with an 8 in blade is $68.55 (Amazon B07TDS85L3).
• ebm-papst iQ datasheet (https://ebmpapst.se/sv/dat/media_manager/pid/2970/product-datasheet/datasheet.pdf): iQ 3612, 115 V, 1550 rpm, 10 W shaft, ambient -40 to +50 C, IP42 with horizontal shaft, IP54 optional. With a 254 mm impeller at 22 degree pitch it moves 700 m3/h (412 CFM) for 15 W input; at 28 degrees, 950 m3/h (559 CFM) for 23 W. Those figures are at 230 V/50 Hz (1300 rpm); the 60 Hz version runs 1550 rpm, so add about 19 percent. Reach-ins use one of these, or one 9 W motor with a 6 to 8 in blade at maybe 100 to 200 CFM on older units.
• Arctic Air AWF25 (25 cu ft, 3/4 HP, 7 A, interior 26 1/4 W x 21 D x 50 H in): "a larger fan motor and blades to recover temperatures quicker after the unit has been opened." $2,199.00 at https://www.webstaurantstore.com/arctic-air-awf25-white-single-door-reach-in-freezer-25-cu-ft/123AWF25.html (seen 2026-08-30).
• Turbo Air M3F19-1-N (18.7 cu ft, 3/8 HP, 4.5 A, -10 to 0 F): the M3 manual says "Evaporator fan motor(s) will run after all doors are completely closed" and "Good air flow in the freezer unit is critical. Be careful to load product so that it neither presses against the back wall, nor reaches within four inches from the evaporator compartment." https://turboairinc.com/wp-content/uploads/2020/01/M3-Reach-Ins_HC.pdf. Two things matter for this project: the stock fan stops whenever the door is open, and the back 4 in is a no-load zone.
• Beverage-Air FB23HC-1S: 3/4 HP, 2,097 BTU/h, 6.34 A, 23.32 cu ft, three shelves, epoxy-coated evaporator coil. No airflow figure on the sheet.

None of the reach-in spec sheets publish CFM or shelf velocity. The iQ 3612 datasheet is the closest thing to a number.

2.2 Velocity at the shelf

Take a True T-23F class box: one evaporator fan at roughly 400 to 560 CFM in a back-wall plenum discharging across the top and returning at the bottom, into a cabinet cross-section of about 3.7 to 4.0 sq ft. If that flow were spread uniformly over the cross-section the bulk velocity would be 100 to 150 fpm, 0.5 to 0.7 m/s. It is not uniform. Measurements on comparable cabinets:

• Commercial refrigerated display cabinet with a rear perforated plate: horizontal velocity from the plate "was low (<0.2 m/s) for all shelves"; only the air curtain at the discharge grille reached 0.6 m/s. https://www.sciencedirect.com/science/article/abs/pii/S0140700714000334
• Fan-cooled (frost-free) domestic refrigerator CFD and PIV: up to 4 m/s right at the duct outlets, "0 to 0.1 m/s" in the lower parts. https://www.sciencedirect.com/science/article/pii/S2590123025000350
• Domestic refrigerator compartment: about 0.25 m/s maximum along the cold wall. https://www.sciencedirect.com/science/article/abs/pii/S0140700708000777

So: expect 0.1 to 0.5 m/s over most of the product in a loaded reach-in, with a faster band near the plenum discharge, and a hotel pan of stock sitting in a pocket of nearly still air if it is pushed toward the door.

2.3 Blast chiller velocities for comparison

• Alto-Shaam QC3-40 reach-in blast chiller spec sheet: "Air circulation: 1,750 cfm", interior 39-1/2 x 22-9/16 x 25-1/4 in, 216 lb capacity. https://cdn.beedash.com/AltoShaam/3042ef891b127d41b813f388d0e8c998af3afca9.pdf. Across the front plane (6.9 sq ft) that is 253 fpm, 1.3 m/s average; through the side plane that the pan rack presents to the evaporator (4.0 sq ft) it is 442 fpm, 2.2 m/s. Local velocity between pans will be higher.
• Blast freezer design guide (https://ingener.by/refrigeration-systems/food-processing-refrigeration/frozen-food-processing/blast-freezers/air-velocity-requirements/): packaged goods 1.5 to 3.0 m/s, unpackaged bulk 3.0 to 5.0 m/s, IQF 4 to 6 m/s. Heat transfer coefficient about 25 to 32 W/m2K at 2 m/s, 33 to 42 at 3 m/s, 46 to 57 at 5 m/s. "Doubling air velocity increases h by approximately 1.75x, reducing freezing time by 43%", while fan power rises with the cube of velocity.
• Research on industrial air-blast freezing puts typical tunnel velocities at 3 to 10 m/s (https://www.sciencedirect.com/science/article/abs/pii/S0260877426001305). One study found no further gain in h above 2 m/s for its geometry (https://www.researchgate.net/publication/307702119_AIR_VELOCITY_PROFILES_IN_AIR_BLAST_FREEZERS_FILLED_WITH_BOXES_OF_FRUIT_PULP_MODELS). For a covered hotel pan or a stock pot, the air-side coefficient stops being the bottleneck somewhere around 2 to 3 m/s; after that the product's own conductivity limits the rate.

2.4 What the booster fan actually adds

Rough surface coefficients (order-of-magnitude, from the ranges above and standard natural-convection values):

So a booster fan in a reach-in is worth about 2 to 3x on the air-side coefficient at the pan; in a chest freezer it is worth 4 to 6x. The video measured 2x for chilling and 3x for the freezing plateau in a fan-cooled household upright, which already had some forced air, so those numbers are a fair expectation for a reach-in. A chest freezer should do at least as well as the video, probably better, because it starts from a worse baseline.

One thing the video is right about that spec sheets confirm: the reach-in's own fan is sized to hold temperature and recover after a door opening (Arctic Air's wording), not to strip heat from a 70 C pan. It also switches off when the door opens (Turbo Air).

3. Fans rated for cold and wet duty

Reading the table:

• For a 12 V build the SPAL is the only fan here that combines real airflow with a published sealing rating. It costs 2.5x the GESEXI. Its 12 inch size fits any reach-in shelf and the 909 CFM figure is a manufacturer test number, not a marketing floor.
• The Noctua and Wathai fans are sealed and cheap but tiny. Four NF-A14s in a 2 x 2 array give about 430 CFM for $132 and 6 W of heat. That is a reasonable "low heat, low noise" option for a reach-in where you can afford to lose one shelf position to a plywood or polycarbonate fan board.
• ebm-papst and Elco/Fasco motors are what the cabinet already contains. They run on 115 V AC, need an impeller (254 mm blades run about $15 to $30 from refrigeration parts houses; not priced here), and need proper grounded wiring inside a wet box, which means a licensed install if you care about liability. Their advantage is a -40 C rating and a bearing that will not stiffen.
• Washdown circulators are walk-in tools. Nothing 20 in and 115 V belongs inside a reach-in.
• I did not find any fan sold as an "NSF food zone" fan in a size that fits a reach-in. The GESEXI and SPAL fans are painted steel and plastic with an unrated finish; treat them as equipment, not food-contact surfaces, and keep a pan or rack between fan and food.

4. Practical install issues in a commercial box

4.1 Getting 12 V through the wall

Options, best first:

1. Chest freezer drain. Commercial chest freezers have a defrost drain plug in the floor or low on the front (Atosa MWF9016GR: "convenient defrost drain for easy cleaning", https://therestaurantwarehouse.com/products/mwf9016gr). Pull the plug, pass a two-conductor cable, and seal the hole with a rubber grommet or a blob of food-grade silicone. You lose the drain until you unseal it, which only matters at defrost time.
2. Under the door gasket with a flat cable. Magnetic gaskets will close over a flat cable of about 2 mm. Amazon carries flat 12 to 24 V DC cables sold for Starlink Mini installs, $23 to $49 depending on length (search "Flat DC Power Cable for Starlink Mini", seen 2026-08-30); their conductor gauge is not published and Starlink draws under 3 A, so they are fine for a 12 inch fan at 9 V (about 4 A) and marginal for a 16 inch. A 3D-printed gasket gap filler exists specifically for this (https://www.thingiverse.com/thing:6713969, "Wire Seal Gasket"). Route at the hinge-side bottom corner where the gasket compression is highest and the cable barely moves when the door swings.
3. Existing light or thermostat port. Reach-ins have a light switch, a temperature sensor and sometimes a door switch wire entering the liner. On foamed-in-place cabinets those wires are foamed in and there is no spare conduit; drilling the liner voids the warranty and breaches the vapor barrier. Do not do this on a cabinet you care about.

Whatever the route, the failure mode is the same: any gap at the gasket lets in humid kitchen air, which freezes on the frame; "Gaskets can't seal to ice or frost" (https://www.gasketguy.com/blog/heater-wires). Check the cable spot for frost weekly for the first month.

4.2 PSU outside, GFCI

Keep the PSU or brick outside. The Jesverty has an internal fan and is not rated for -20 C or condensation, and a 120 V device inside a wet, cold box is a shock hazard when someone reaches in with wet hands.

2023 NEC 210.8(B) requires GFCI protection for receptacles in "areas with sinks and permanent provisions for food preparation, beverage preparation, or cooking", covering 125 to 250 V single-phase receptacles at 50 A or less and 150 V or less to ground (https://www.jadelearning.com/blog/2023-nec-section-210-8b-gfci-protection-other-than-dwelling-units/; https://www.eaton.com/us/en-us/markets/residential/nec-2023-updates/gfci-210-8b-other-than-dwelling-units.html). A commissary kitchen is that. If the outlet by the freezer is not already GFCI, use a plug-in GFCI adapter or have the outlet swapped. The 12 V side is SELV and needs nothing.

Fuse the 12 V line at 10 A near the PSU. The RioRand controller has a 10 A fuse built in; the bench PSU has current limiting.

4.3 Guarding the blade

Radiator fans ship with a finger-safe shroud on the motor side and an open blade on the flat side. If the flat side faces the product (the video's orientation), a cook reaching for a pan can put fingers into 2000 rpm. Put a wire fan guard on the open face (automotive 12 in and 16 in fan guards exist; not priced here) or mount the fan under a wire shelf so the shelf is the guard. In a chest freezer, mount it on a wall behind a wire basket.

4.4 Frost and ice on the fan

Warm food brings water vapor. In a stock reach-in it frosts the evaporator over a day; the cabinet's defrost cycle handles it (True: "shortest possible defrost cycle"; Turbo Air: "Smart defrost system will defrost as needed"). A booster fan does two things to that:

• It moves the vapor to the coil faster, so the coil frosts faster during a chill run. Expect more defrost events on chill days. This is not damage, just work.
• Fan blades and the motor housing also collect frost, especially the blade leading edges. A brushed automotive motor with a sealed housing tolerates it; the imbalance from uneven frost is the thing to watch. Wipe the blade at the weekly gasket check. Cover the pan or use a lid whenever the product allows; it cuts the vapor load and the frost.

Never run the booster fan with the door open. The stock evaporator fan stops on door-open for a reason (Turbo Air manual): a fan running with the door open pumps cold air out and humid air in. Either wire a magnetic door switch in series (reed switch, a few dollars) or put the on/off switch outside where the person opening the door can see it.

4.5 Cold starts at -25 C

A fan that sits idle at -25 C between chill runs has to start with stiff grease. Refrigeration motors are rated for it (ebm-papst -40 C, Elco -30 C, Fasco with -40 F oil). SPAL rates its brushless line at -40 C ambient and its brushed motors are automotive parts that start in Canadian winters, but I could not fetch the brushed datasheet to confirm a figure. The GESEXI listing says nothing. A brushed motor that will not start at 9 V may start at 12 V; the PWM controller or bench PSU gives you that option. If a fan hums and does not spin, cut power within seconds; a stalled brushed motor at 10 A heats fast.

The simplest fix is to not cold-start: leave the fan running at a low setting (5 to 6 V, a few watts) all the time, which also keeps the air mixed and the blade frost-free. The heat penalty is under 10 W.

4.6 Vibration, shelves, noise

The video zip-ties the fan flat on a wire shelf with the tabs tucked so the shelf above stays usable. That works on commercial epoxy-coated wire shelves too. A 16 inch fan at 2000 rpm with an unbalanced blade will walk a wire shelf; use four zip ties at the corners of the shroud, pulled tight, and a strip of closed-cell foam tape under the shroud.

Noise: none of these fans publish dB(A). A forum measurement put a 16 inch curved-blade radiator fan at "approximately 80 dB with the hood open" at full voltage (https://www.camaros.net/threads/electric-fan-noise-level.480655/); listing claims of "30 dB" for the same class are not credible. At 9 V it is much quieter (fan noise drops steeply with speed) and the freezer is an insulated closed box, so what you hear in the kitchen is a hum, not a roar. The video's "this thing really roars" was at 12 V with the door open.

Weight: a 16 inch fan is about 4 lb, no issue for shelves rated 110 lb (Turbo Air).

5. What commissaries use instead

Prices from https://www.webstaurantstore.com/3071/cooling-paddles.html and product pages, seen 2026-08-30:

Notes:

• "Ice wand" searches on Amazon return only these paddles (Rapi-Kool is marketed as "Ice Paddle and Cooling Wand"). "Cool Blue" and "Chill Blaster" did not turn up as products; if those are names someone mentioned, they may be house names for a paddle or a countertop blast chiller.
• Paddles are the right tool for liquids because they put the cold inside the pot; a fan only works on the outside. The combination (paddle for the first 30 minutes, then shallow pans in the fan freezer) is what most kitchens actually do.
• Shallow hotel pans and ice baths cost nothing beyond ice. Ice cost is the hidden expense: a 20 lb bag per stock pot adds up. That is the argument for the eutectic bottles in video 2.
• Real blast chillers, for scale: cheapest on WebstaurantStore is the Techfrost JOF23 countertop, 15 lb capacity, $5,400.00; Alto-Shaam QC3-3 countertop, 36 lb, $11,020.02; Techfrost E5 reach-in, 44 lb, $11,599.00 (https://www.webstaurantstore.com/search/blast-chiller.html, seen 2026-08-30). American Panel HurriChill roll-in units run $38,487 to $92,359 (https://www.culinarydepotinc.com/, per search results). American Panel and ThermalRite sell modular blast chillers "assembled in the field, the same way a typical walk-in is erected" (http://www.americanpanel.com/modularchillers.html); prices are quote-only.
• I did not find a retrofit "blast chill fan" product for walk-ins or reach-ins from any foodservice supplier. The nearest things are walk-in evaporator coils (a 2-fan, 12,000 BTU/h, 1,400 CFM 115 V unit cooler was listed on eBay per search results) and washdown circulators (section 3). The market gap is real; the video's approach is the retrofit.

6. Two configurations

Fan CFM figures below are the sellers' claims derated to 9 V (x0.75). Real numbers will be lower for the GESEXI; the SPAL figure is a test figure.

6a. Chest freezer (Atosa MWF9016GR class, 60 x 26.5 x 34.5 in exterior, about 54 x 21 x 27 in interior)

Mount one 16 inch fan vertically on the end wall above the compressor step, blade flipped and leads reversed so air exits the flat face, blowing along the length of the box. Product goes on a wire rack or upturned basket in the stream, 12 to 24 in from the fan. The end-wall cross-section is about 21 x 27 in, 3.9 sq ft, so a 16 inch fan (1.4 sq ft) is 36 percent of it, close to the video's 40 percent rule. The return path is the remaining 64 percent around the fan and back under the rack.

Floor-mounted blowing up is the alternative (the video's orientation) but a chest freezer's horizontal cross-section is 7.9 sq ft, so a single 16 inch is only 18 percent of it and the plume spreads and stalls. Two 16 inch fans on the floor (the 2-pack) get to 36 percent; that is the choice if the food is in baskets and you cannot give up an end.

Parts:

Total: about $75 with the brick and PWM, $90 with the bench PSU, $135 with the SPAL.

Expected air speed: about 750 CFM (claimed) at 9 V from a 1.4 sq ft face is 540 fpm, 2.7 m/s at the fan face; 2 to 3 m/s over product within 12 to 18 in; bulk velocity along the box about 190 fpm, 1 m/s. Heat added about 50 W (16 in) or 38 W (SPAL 12 in). Baseline before the fan: static, natural convection only, so this is the bigger win of the two.

6b. Reach-in upright (True T-23F class, about 23.5 in wide inside, 22 7/8 x 23 1/4 in shelves)

Option 1, the video's layout: fan lying flat on the bottom shelf or the floor, flat face up, blowing up through the wire shelf above it. The pan sits on that shelf, directly over the fan, 4 to 8 in from the blade. The shelf is the finger guard. A 16 inch fan is 16.5 in square and fits a 22 7/8 in shelf with room for return flow at the sides and front; it is 35 percent of the cabinet cross-section (about 4 sq ft). Keep the rear 4 in clear per Turbo Air and do not block the True rear airflow guards.

Option 2, for a box with a shallow interior (Arctic Air AWF25, 21 in deep): 12 inch fan on the same layout, 12.4 in square, 26 percent of cross-section. Less air, half the heat, fits under any shelf.

Parts:

Total: about $75 to $95 with the 16 in fan; about $45 with the 12 in fan and the 9 V adapter.

Expected air speed: 16 in at 9 V, about 750 CFM claimed, 2.7 m/s at the face, 2 to 3 m/s at a pan 6 in above it, tapering to 1 m/s at the top of the cabinet as it spreads; bulk cabinet circulation about 190 fpm (1 m/s) versus the stock 0.1 to 0.5 m/s at the shelf. 12 in at 9 V, about 600 CFM claimed from 0.8 sq ft, 3.8 m/s at the face (optimistic; the SPAL's tested 909 CFM at 12 V gives 3 to 4 m/s at 9 V), bulk about 0.8 m/s. Heat added 34 to 50 W, 6 to 8 percent of a 3/4 HP system's capacity.

Either way the pan must be uncovered or in a shallow hotel pan to benefit; a lidded stock pot gets only its outside cooled and the fan's advantage shrinks toward the paddle-plus-shallow-pan baseline.

7. Open questions

• The GESEXI fan's actual CFM, amps and cold-start behavior are unmeasured. A $30 anemometer and the bench PSU's ammeter would settle it in an afternoon, which is what the video did.
• No reach-in manufacturer publishes shelf air velocity; the 0.1 to 0.5 m/s figure is inferred from cabinet studies and the iQ 3612 datasheet, not measured in a True or Turbo Air box.
• SPAL brushed-motor operating temperature range was not fetched (Summit and JEGS block scripted access); the -40 C figure is from SPAL's brushless datasheets.
• The Schaefer 24CFO-SWDS at $159.99 on Amazon looks too cheap for a stainless washdown fan; confirm before relying on it.
• Fan guards and 254 mm impellers for the ebm-papst motor were not priced.