A Very Quick and Simple Measurement that May Solve Your Intermittent GE Refrigerator Problem
[ This article was also posted at http://appliancejunk.com/blog/ where you will find other articles of interest ]
At www.fixyourboard.com we repair a lot of GE fridge control boards with various component failures but there is one failure mode that we see often. This turns out to be responsible for a variety of seemingly random intermittent fridge problems that tend to be very hard to diagnose. Well, we came up with a very simple tool and procedure to isolate this problem on-site. It only requires one measurement with a DVM, nothing needs to be disconnected from the controller, only the control board cover panel need be removed.
Here I will describe the problem, how to construct this simple tool for about $3 and 10 minutes of your time, and how to make the measurement. This applies to the wide range of control board part numbers that look similar to this:
First: The Problem and The Test Strategy
The motherboard contains a 13.5 Vdc power supply that powers the relays, fans, dampers, etc as well as power to the temperature control board and dispenser board. There is a frequently occuring component failure that causes this power supply to still work fine under light loads, but lose its regulation ability under heavier loads. So suddenly, depending on the state of the fridge and what devices the motherboard is trying to energize, things go from working fine to … not so good … or major failure.
The test strategy is to make a voltage measurement on the connector (J4, pins 2-3) that supplies 13.5 Vdc to the other boards, AND TO DO SO UNDER FULL LOAD.
Next: The Test Load Device
Measure the voltage with a 25 ohm/10W shunt load to ferret out the problem (keep reading to see how to make the shunt load). A good board will see less than a few tenths Volt droop on this supply under load. A defective board will see a large drop in voltage (several Volts) when applying the load.
To be clear: If you see a supply voltage less than about 13.3 Volts under this condition then the motherboard has a problem. If the supply holds up under load, then the supply is good, but there could still be other problems with the motherboard. In a future blog I will deal with this. This is a quick NO-GO test.
If you think the supply may be bad but it’s not obvious, then also do a no-load measurement to check the difference. There should less than a few hundred millivolts difference between no-load and full load.
How to Contruct the Test Load
Here we build a 25 ohm, 10 Watt shunt load that plugs into the standard 3/4 inch spaced banana jacks on most DVMs. (OK, not all DVMs use this standard jack spacing, in which case you can add some short banana patch cords, or get a better meter). Following is a sequence of images showing how to construct the load, along with part numbers from www.mouser.com.
Last: Where to Measure
Backprobe connector J4 as shown below (the 2nd and 3rd pins from the edge). This is the 13.5 Vdc power supply to the other boards.
- Backprobe the 13.5 Vdc supply on the connector J4, between pins 2 and 3 without the test load
- Record the actual voltage (usually around 13.5 to 13.6 V)
- Plug in the test shunt load between your meter and probes
- Backprobe the same points and record the voltage under load
- If you see more than a few tenths of a volt drop under load, then the motherboard has a problem
- We can repair the control board for you at http://www.fixyourboard.com