Fridge not cooling - a case study with no happy ending

During this 2026 Spring Festival/Chinese New Year,
the fridge stopped cooling - a test icecube melted.

In a bizarre event, the wall socket energizing the fridge
ceased to work. I cannot replace it yet since the wires
were daisy chained at the socket.
I needed a Wago type device which I recently purchased.

BUT after a week or so, water turned into ice and it
was normal again . . . until it wasn't.

THEN my test icecube/chunk - in those coleslaw containers
you get from fast food outlets - melted again.

Bummer.

So I studied



How to Test & Replace a Defrost Timer in a Refrigerator | Repair & Replace

The rest of the videos in this series cover other failures.

In my case



HVACR: How A Refrigerator/Freezer Defrost Timer Works/Defrost Timer Troubleshooting/Sequence/Testing

Wow. My fridge, a National/Panasonic NR-B29T does not
use an electromechanical timer so it is good to know that
the 6 to 8 hour compressor/cooling period is interrupted
by 20 mins of defrost/heating. Were you aware of this?
I certainly didn't. The videos I watched never mentioned
this fact. So if my defrost cycles are too close together,
something is wrong.

These are the only two circuits exclusively energized.
And within each period, the defrost thermostat turns off the
heater when the ice is gone while the compressor shuts off once
the desired temperature is reached. So two circuits and 4 distinct
phases. Thus an idle period can follow a defrost if the thermostat
cuts out earlier because there is no more ice to melt.

Here are the readings I have taken with comments on what I did
or observed. I don't have the service manual so the behaviour
of the electronics has to be treated like a black box.

If I witnessed a recovery, chances are the fridge sensor
is still working although that does not rule out contact
corrosion or a marginal supply to the electronics interpreting
the sensor wrongly.

It is even more difficult when I don't have a baseline to
compare with BUT I challenge you, does anyone on this planet
have a baseline for their fridge?

Power meter shows normal current readings for
cooling, idle and defrost.
EXCEPT that the compressor now draws 512 mA
instead of 489 mA. The reading also depends on
the mains voltage. We are currently experiencing
a heat wave.

The power meter has been connected for ages and
I can tell the state the fridge is in. When
defrosting, the phase angle is zero.

When idle, it is 11 mA.

Water from freezer compartment does not flow
into the drain tube. It wets the floor.
I can't figure this out yet.
Now I know. The guide was not fully inserted
by the previous repairman. Now the defrost water
collects in the tray as expected.

13:49 19/04/2026 sick, cooling -1.5 12,0
15:51 19/04/2026 sick, cooling  1.3 11.4
21:25 19/04/2026 sick, cooling  3.8 12.5

14:26 20/04/2026       defrost  2.7 11.9
18:41 20/04/2026 sick, cooling  3.1 12.0

01:46 21/04/2026 sick, cooling  2.9 10.8
06:29 21/04/2026 sick, cooling  5.0 10.9
17:33 21/04/2026       defrost  5.6 13.9
20:48 21/04/2026 sick, cooling  4.4 13.0

08:01 22/04/2026 sick, cooling 11.1 13.6
10:08 22/04/2026 sick, cooling 11.1 14.0
15:19 22/04/2026 sick, cooling  3.7 12.1
16:34 22/04/2026 sick, cooling  2.9 12.0
22:28 22/04/2026 sick, cooling  6.1 12.3

06:59 23/04/2026 sick, cooling -0.7 11.0
08:58 23/04/2026 sick, cooling -0.6 12.0
19:34 23/04/2026 sick, cooling  0.6 11.0
21:04 23/04/2026 sick, cooling -1.4 10.9
21:29 23/04/2026       defrost  7.5 11.0

03:01 24/04/2026 sick, cooling -2.0 10.6
07:27 24/04/2026 sick, cooling -0.9 10.5
09:43 24/04/2026 sick, cooling -1.1 10.6
11:00 24/04/2026 sick, cooling -2.1 10.4
11:52 24/04/2026 sick, cooling  0.9 10.4
14:42 24/04/2026 sick, cooling -2.5 10.4
16:29 24/04/2026 sick, cooling -3.8  9.9
16:48 24/04/2026       defrost 11.3 10.4
21:43 24/04/2026 sick, cooling -2.6 11.0
23:20 24/04/2026       defrost 12.9 11.2

07:20 25/04/2026 sick, cooling -2.5 10.5
09:04 25/04/2026 sick, cooling -0.4 12.8
13:59 25/04/2026 sick, cooling  9.0 11.0
18:30 25/04/2026 sick, cooling -0.5 10.9
20:22 25/04/2026         idle  12.6 11.1

04:54 26/04/2026 sick, cooling -2.0 10.3
08:34 26/04/2026 sick, cooling  0.7 10.8
12:12 26/04/2026       defrost  0.9 10.2
13:39 26/04/2026 sick, cooling  2.3 11.0
15:09 26/04/2026 sick, cooling -0.6 10.9
18:17 26/04/2026 sick, cooling -1.4 11.8
22:25 26/04/2026 sick, cooling -0.8 11.8

04:30 27/04/2026       defrost  6.5 10.1
07:58 27/04/2026 sick, cooling -2.1 10.8
10:03 27/04/2026 sick, cooling -2.1 10.5
12:35 27/04/2026 sick, cooling -3.4 10.2
13:59 27/04/2026 sick, cooling -3.9 10.5
17:42 27/04/2026 sick, cooling  0.2 12.5

02:27 28/04/2026 sick, cooling -0.1 10.6
07:28 28/04/2026 sick, cooling -4.3  9.5
08:54 28/04/2026       defrost -1.1  9.3
10:42 28/04/2026 sick, cooling 11.3 10.4 no ice at freezer evaporator
13:43 28/04/2026 remove fridge control flap
14:24 28/04/2026 sick, cooling  2.2  8.2
14:35 28/04/2026 sick, cooling  1.9  7.5
15:33 28/04/2026 sick, cooling  1.4  7.0
17:20 28/04/2026 sick, cooling  0.0  5.1
17:38 28/04/2026 sick, cooling -0.3  4.9
19:23 28/04/2026 sick, cooling -1.9  3.2 ***
20:22 28/04/2026 sick, cooling -0.3  4.3
22:39 28/04/2026       defrost  2.9  5.1

02:44 29/04/2026 sick, cooling -0.9  3.4 ***
08:03 29/04/2026 sick, cooling  9.4  7.3
08:20 29/04/2026 sick, cooling -0.4  4.0 ***
09:32 29/04/2026 sick, cooling -0.8  3.7 ***
10:52 29/04/2026 sick, cooling -1.8  3.1 ***
defrost ?
12:47 29/04/2026 sick, cooling  1.8  5.6
13:29 29/04/2026 sick, cooling -2.8  2.3 ***
14:45 29/04/2026 sick, cooling -2.8  2.3 ***
16:10 29/04/2026 sick, cooling -3.2  2.0 ***
16:46 29/04/2026          idle 11.6  6.5
17:45 29/04/2026 sick, cooling -0.9  2.7 ***
defrost 20min
18:04 29/04/2026 sick, cooling  8.4  7.9
18:14 29/04/2026 sick, cooling  0.4  5.2
18:45 29/04/2026 sick, cooling -2.7  2.6 ***
20:35 29/04/2026 sick, cooling 11.1  7.3

02:18 30/04/2026 sick, cooling -3.3  2.0 ***
07:xx 30/04/2026       defrost  0.5  3.3
07:36 30/04/2026 sick, cooling  8.1  7.0
07:45 30/04/2026 sick, cooling -0.1  4.4
09:34 30/04/2026       defrost  0.0  3.7 ***
10:00 30/04/2026 sick, cooling  0.1  4.6
10:16 30/04/2026 sick, cooling -1.4  3.4 ***
11:24 30/04/2026 sick, cooling -1.3  3.4 ***
13:00 30/04/2026 sick, cooling -2.5  2.5 ***
16:21 30/04/2026 sick, cooling -0.4  4.6
18:15 30/04/2026 sick, cooling -3.5  1.6 ***
19:24 30/04/2026 sick, cooling -3.6  1.5 ***
19:39 30/04/2026 in defrost
19:48 30/04/2026 sick, cooling 11.9  7.6
20:55 30/04/2026       defrost  3.3  4.9
21:08 30/04/2026 sick, cooling  8.5  9.5
22:22 30/04/2026       defrost  7.3  6.1
23:52 30/04/2026 sick, cooling  1.7  5.3

06:40 01/05/2026 sick, cooling -2.3  2.6 ***
07:01 01/05/2026 sick, cooling -3.7  1.5 ***
09:45 01/05/2026 sick, cooling  0.3  4.7
11:14 01/05/2026 sick, cooling -2.3  2.6 ***
12:12 01/05/2026 sick, cooling -1.3  3.3 ***
13:07 01/05/2026 sick, cooling  4.0  6.5
15:11 01/05/2026 sick, cooling -2.6  2.8 ***
16:05 01/05/2026 sick, cooling -0.3  4.4
16:39 01/05/2026 sick, cooling -2.4  2.9 ***
18:04 01/05/2026       defrost  9.4  7.1
18:29 01/05/2026 sick, cooling -0.5  4.1
19:50 01/05/2026       defrost  2.1  5.1
20:24 01/05/2026 sick, cooling  0.5  5.9
22:42 01/05/2026       defrost  1.8  4.9

defrost
05:37 02/05/2026 sick, cooling 10.4  7.4
06:12 02/05/2026 sick, cooling -1.5  3.4 ***
07:25 02/05/2026 sick, cooling  4.1  7.0
09:12 02/05/2026 sick, cooling  3.1  7.4
09:42 02/05/2026       defrost  4.1  6.9
12:04 02/05/2026       defrost  1.3  5,1 getting more defrosts
13:11 02/05/2026       defrost  6.8  7.5
13:53 02/05/2026 sick, cooling -0.7  4.1
15:30 02/05/2026 sick, cooling  1.5  6.0
16:14 02/05/2026 sick, cooling -1.5  3.5 ***
17:32 02/05/2026       defrost  2.8  5.5
18:10 02/05/2026 sick, cooling  0.5  6.3
23:15 02/05/2026 sick, cooling -2.7  2.5 ***

05:23 03/05/2026 sick, cooling -0.8  3.6 ***
05:53 03/05/2026 sick, cooling -2.3  2.5 ***
05:54 03/05/2026       defrost
06:49 03/05/2026       defrost  1.8  4.5
08:04 03/05/2026 sick, cooling  2.4  6.1
08:30 03/05/2026 sick, cooling -2.0  3.0 ***
12:17 03/05/2026 sick, cooling -1.6  3.4 ***
13:08 03/05/2026 sick, cooling -0.1  4.6
13:57 03/05/2026       defrost  3.0  5.0
15:45 03/05/2026 sick, cooling  2.5  6.6
18:12 03/05/2026 sick, cooling  5.1  8.8

03:16 04/05/2026 sick, cooling  7.5  9.1
first time noticing a climb from 489 to 512 mA
05:41 04/05/2026 sick, cooling  2.8  7.2

06:20 04/05/2026 sick, cooling  7.3  8.8
09:12 04/05/2026 idle 11.3 5.0 something wrong with these numbers
09:13 04/05/2026 sick, cooling 11.4  9.2
10:30 04/05/2026 sick, cooling  0.1  5.1
13:45 04/05/2026 sick, cooling  1.6  6.3
14:13 04/05/2026       defrost  1.4  5.1
14:49 04/05/2026 sick, cooling -1.0  4.0 ***
15:59 04/05/2026          idle 11.5  8.3
16:32 04/05/2026          idle 11.8  8.3
19:26 04/05/2026 sick, cooling 12.2  9.4
20:39 04/05/2026 sick, cooling  0.4  5.3

00:14 05/05/2026 sick, cooling -0.2  5.0
06:10 05/05/2026       defrost  3.9  6.5
06:22 05/05/2026          idle 11.4  8.1 no cooling phase?
06:24 05/05/2026 sick, cooling 10.4  8.3
06:53 05/05/2026 sick, cooling -0.1  4.5
07:01 05/05/2026       defrost  6.0  7.4 haywire !
07:24 05/05/2026 sick, cooling  0.7  5.3
12:23 05/05/2026 sick, cooling  2.8  7.3
12:41 05/05/2026 sick, cooling  0.5  5.3
14:37 05/05/2026 sick, cooling  0.4  5.8
14:43 05/05/2026       defrost  3.6  9.8
17:45 05/05/2026 sick, cooling  0.6  5.4
20:17 05/05/2026       defrost  5.8  8.4
22:26 05/05/2026 sick, cooling  1.8  6.7

01:37 06/05/2026       defrost  0.0  5.3
06:57 06/05/2026          idle 11.9  9.0
10:27 06/05/2026 sick, cooling  2.4  6.8
11:18 06/05/2026 sick, cooling  2.7  7.0
12:01 06/05/2026 12.9  9.9
DISCONNECTED. No Happy Ending.

Hisense open-close door
0.7  7.8
1.0 15.4
3.7  6.7
2.8  8.9
6.7 10.8
5.2  9.5
4.3 10.3
5.1 11.9
4.2 10.6
4.2  9.7

So, where to start?

Firstly the FIRST reading of the day. This gives the fridge
all night long to reach the lowest temperature.

Most readings taken here allow for at least 30 mins after the door
has been opened. The temperature probes are placed just
inside the door which make for wild readings once the
door is opened because of the lack of thermal mass and the
condensation from the outside air. But they are placed
to be out of the way of accidental strikes and lack of
cable restraints and the length of the cable.
It is what it is.

And then there are those blips which indicate a defrost
happened but I cannot be sure.

I mark "***" at those readings which meet the requirements
for both the freezer and fridge compartments IE 0C or below
for the freezer and less than 4C for the fridge.

The barrier happens at:

"13:43 28/04/2026 remove fridge control flap"

This opens the passage of freezer air directly into
the freezer. This is also when I reseated the drain
thingy connecting the Freezer Compartment to the
manifold in the fridge compartment leading outside to
the drain hose. No blockage found by sticking in a
cable tie.

It used to be that at the maximum settings, lettuce
would ice up in the fridge.

It is clear that before, there are no *** while
after removal of the flap, I now get ***.

The fridge can clearly do no better and is just
limping home.

When we moved into this house end-1972, we brought
along the Kelvinator and my wife recalls we switched to
this smaller fridge between 1995 and 2003.
So about 2 decades of continuous use.

The current draw increase and the poor reading at the
freezer vent (my car air-conditioner was pumping out
at least -10C) leads me to conclude that the compressor
needs to be replaced at the very minimum for food safety.

The fridge failed exactly this way a few years ago and
replacing the board fixed it. I don't think the sensors
or the board are at fault this time.

Perhaps it is a failing run capacitor. The second hand
unit we saw advertised has apparently been sold so
there is nothing to compare.

So we bought a Mitsubishi. The radiator runs along the
body shell just like the National. We waited 3 hours after
delivery before powering on. Is seemed sluggish but
overnight it cooled to expectations. Water froze in the
freezer. No hurry to place a stone in the plastic cup
with a chunk of frost from the Hisense icebox.

One day I will improve the Hisense by placing a PC fan
inside to circulate the air.  

Before we chose the replacement, we asked for a photo
of the circuit diagram. I bet no other consumer does that.

Edward, your log is better than most service calls I get, and I think your conclusion is close. In order, I would suspect:

1. Power and heat rejection first. That dead, daisy-chained socket, the heat wave, and a dusty condenser or tired condenser fan can all starve or overheat the compressor before anything exotic fails.
2. Defrost running too often. Defrosts crowding together in early May point to the board or a drifting sensor, even with a board replaced once before.
3. Airflow. Temperatures only reached your target marks after you removed the fridge control flap, so a stuck damper or weak evaporator fan was strangling cold-air flow.
4. Sealed system last. The current creep from 489 to 512 mA, with poor pull-down, fits a tired run capacitor, a weak compressor, or low refrigerant.

Simple checks first: unplug it, confirm full voltage at a known-good outlet, clean the coils, verify both fans spin freely, and test or swap the run capacitor, after discharging it safely.

Call a pro for anything sealed-system: refrigerant, brazing, or a compressor swap on a fridge this old is rarely economical, especially once food safety is at stake.

While trouble shooting refrigerators or suspect power grids I will place a penny or stone on top of ice cube in IceCube tray in the freezer. If the unit fails long enough to thaw the ice so the rock sinks I know the food is suspected of going bad.

johny ortan wrote:Edward, your log is better than most service calls I get, and I think your conclusion is close. In order, I would suspect:

1. Power and heat rejection first. That dead, daisy-chained socket, the heat wave, and a dusty condenser or tired condenser fan can all starve or overheat the compressor before anything exotic fails.
2. Defrost running too often. Defrosts crowding together in early May point to the board or a drifting sensor, even with a board replaced once before.
3. Airflow. Temperatures only reached your target marks after you removed the fridge control flap, so a stuck damper or weak evaporator fan was strangling cold-air flow.
4. Sealed system last. The current creep from 489 to 512 mA, with poor pull-down, fits a tired run capacitor, a weak compressor, or low refrigerant.

Simple checks first: unplug it, confirm full voltage at a known-good outlet, clean the coils, verify both fans spin freely, and test or swap the run capacitor, after discharging it safely.

Call a pro for anything sealed-system: refrigerant, brazing, or a compressor swap on a fridge this old is rarely economical, especially once food safety is at stake.

01 We used a different socket - same weak refrigeration.
     This is Japanese. The heat exchanger is the body of the refrigerator so no dusty condenser or fan.
02 It could be the sensor but the readings trend down. It doesn't go all over the place.
03 The airflow is not strangled.
    There is only one fan and I can feel it.
04 Ha ha ha. The pro is not interested. We couldn't wait any further and got a second-hand unit
     without those bells and whistles and internet connectivity. This unit is a bit smart - the fan stops when
     the door is open and the lights are LED so no need for my stash of replacement bulbs.

I used to repair supermarket type cash registers, Automated Teller Machines all the way up to machines
the customer needed a US Government Licence for. If you know you know. With the latest mainframes
you could power down that processor and swap it while the rest of the mainframe continues
running. My skill set is no longer needed.

J. Syme wrote:While trouble shooting refrigerators or suspect power grids I will place a penny or stone on top of ice cube in IceCube tray in the freezer. If the unit fails long enough to thaw the ice so the rock sinks I know the food is suspected of going bad.

My trick is to place some ice cubes in a transparent plastic cup. It evaporates so I
occasionally top up.

Fair points, Edward — the body-shell condenser is a good catch, I shouldn't have assumed dusty coils without asking the model first. Main thing is you're cooling again. And a man who kept cash registers, ATMs and mainframes alive has nothing to prove to a fridge — enjoy the second-hand unit.

johny ortan wrote:Fair points, Edward — the body-shell condenser is a good catch, I shouldn't have assumed dusty coils without asking
the model first. Main thing is you're cooling again. And a man who kept cash registers, ATMs and mainframes alive has nothing to prove to a fridge — enjoy the second-hand unit.

There were other threads or portions of threads in Permies dealing with poor air
conditioning and refrigeration. I thought I would deliver an all encompassing
account that would help others as well as myself when it happens again,.

Especially about the 8 hour cooling period before defrost. I wish I had
known about that sooner. I must have seen over 20 YouTube videos
before anyone mentioned that.

Not wanting to guess and buy part after part to test, I found the same
model - obviously second-hand - but that was sold off months ago.
So swapping parts to isolate the failure was not possible.
My last shot and that door was closed.

Any idea why the defrost cycles appear so frequently?
A failing timing capacitor?

The increase in current draw may be due to the reduction in supply
voltage when all the street lights come on or something. I have
noticed that. I only record the current - not both so the view
may be distorted.  I don't have any device that can sample continuously.
 
I have a Raspberry Pi 2B but there are are far more urgent matters
to attend to.

Time was running out so we went for a second-hand unit.

It is still running to this day.