R-600a No-Frost Refrigerator: Restriction or Undercharge?
A no-frost refrigerator in a care-home kitchen uses R-600a with a nameplate charge of 65 g; the compressor is rated 0.9 A. After 24 hours the freezer sits at −6 °C and the food compartment at 9 °C. Observations: the compressor runs continuously and is hot to the touch; the condenser is warm at the top and near ambient at the bottom; the suction line at the compressor is not sweating; the evaporator behind the panel shows a thin, even film of frost near the inlet and none at the outlet; amp draw is 0.42 A. About 40 mm from the strainer, the capillary shows a small damp, colder spot. Ambient is 30 °C, the unit is 6 years old, and the door gasket was replaced three months ago. The previous technician's note reads: "low charge - top up 20 g R-600a". The customer wants it fixed today because the kitchen serves 40 residents.
1. Is the "low charge" diagnosis supported? Name the tests that separate a partial restriction from an undercharge.
Read the worked answer
Not yet - both faults produce low suction, a partial frost pattern and a hot compressor, so the symptom list alone cannot decide it.
- Weigh the recovered charge. Recover into a labelled cylinder and weigh it. If roughly 65 g comes out, the system was never undercharged and the fault is a restriction; 30–40 g means a leak. This single test settles the argument and is the reason you recover instead of topping up.
- Temperature drop across the strainer/drier and the first 100 mm of capillary. A partial restriction gives a measurable drop of a few K - and the damp cold spot already visible at 40 mm is consistent with it. Undercharge starves the whole evaporator uniformly and creates no localised cold spot at the capillary inlet.
- Leak search with an electronic detector or soap at brazed joints, the flare, the schrader and the evaporator return bends, plus post-shutdown equalisation: a restricted system creeps back to equal pressure slowly, a starved one equalises normally.
The even frost stopping early is consistent with both faults, so it is not a discriminator on its own.
2. The previous technician's plan is to add 20 g of R-600a and leave. Explain the risk and set out the correct repair path.
Read the worked answer
R-600a is A3 - flammable - and 65 g is roughly two tablespoons of liquid, in a kitchen with gas appliances. Adding 20 g to a restricted system raises head pressure and floodback risk without restoring flow; the restriction stays and the compressor runs hotter. If the real fault is a leak, the added gas leaves the same way.
Correct path: recover the remaining charge with equipment rated for flammable refrigerant into a labelled cylinder, in a ventilated area with no ignition source within 3 m (no grinder, no relay chatter, no domestic vacuum cleaner motor). Cut out the filter-drier, clear or replace the capillary, fit a new drier of the correct rating, nitrogen-purge while brazing, leak-test, triple-evacuate to 500 microns, then charge by weight from the nameplate on a scale with 1 g resolution - pressure is a poor charge indicator at this charge size. Verify with a 24-hour pull-down, a full frost pattern, and a second leak check. Document the recovered and charged weights.
3. Why is the 0.42 A draw not evidence of an electrical fault, and what does it tell you about the running condition?
Read the worked answer
Current tracks load, not health. The compressor is pumping a small mass flow because the capillary is feeding almost nothing; less mass flow means less work, so current falls to 0.42 A against a 0.9 A rating. A winding fault normally shows the opposite - high current and an overload trip, a dead short, or an open winding reading infinite resistance - or nothing at all.
Still measure winding resistance and insulation resistance as a baseline before condemning the sealed system, because a compressor damaged by a previous overheating episode can read normally when cold.
The important operational point: a starved compressor is cooled by returning suction vapour, and there is very little of it. Running it for hours while "deciding" cooks the oil and can seize it. Power the unit down, recover, and diagnose from the recovered weight rather than from a longer running test.