Using Diode Mode Values from Boardview Files

Some boardview files include a number for every net: the diode-mode reading measured on a known-good board. On a dead board, comparing your readings with those numbers can point straight at a damaged rail or a leaky chip without powering anything. It is a powerful technique, as long as you understand where the numbers come from and how loosely to treat them.

What a diode-mode reading on a net means

In diode mode, a multimeter pushes a small current through whatever is between its probes and shows the voltage that develops, usually in millivolts or volts. On a laptop board, measuring from ground to a net (or from a net to ground, depending on convention) gives a value that depends on everything connected to that net: protection diodes inside chips, MOSFET body diodes, resistors, capacitors charging during the measurement.

The result is a kind of fingerprint for the net. A healthy rail gives a characteristic value; a rail with a shorted capacitor reads close to zero; a rail with a leaky chip reads lower than normal; an open connection reads as "OL" or very high. The underlying idea is covered in Diode Mode Measurements Explained.

Where the reference numbers come from

Reference readings in boardview files were recorded by someone — a service centre, a repair shop or a contributor — on a working board of the same design, usually with one particular meter and one particular probe polarity. That means:

• the meter matters — different meters push different test currents and show different values on the same net;
• the polarity matters — red probe on ground and black on the net gives a different number from the opposite arrangement;
• the board state matters — battery connected or not, CMOS battery fitted or not, daughterboards attached or not;
• the revision matters — a reading for one revision may not apply exactly to another.

Treat the stored numbers as a guide that is usually close, not as a specification.

Setting up a fair comparison

1. Use the same polarity as the reference. Most repair diode readings in boardview files are taken with the red probe on ground and the black probe on the net. If unsure, measure a few known-good nets and see which polarity matches the stored values.
2. Disconnect the battery and the CMOS battery unless the reference says otherwise. Both feed parts of the board and change readings.
3. Wait for the reading to settle. Large rails with many capacitors start low and climb as capacitors charge. Note the value after it stabilises.
4. Calibrate your expectations. Measure three or four rails that you believe are healthy and compare with the stored values. If your meter consistently reads 10% higher, apply that mentally to the rest.

Reading the comparison

The thresholds are rough guidance. A difference of 30 mV on a rail that normally reads 400 mV may mean nothing; a drop from 500 mV to 200 mV almost always means something.

A practical routine for a dead board

1. Start with the rails that matter first: adapter input, system rail, always-on 3.3 V and 5 V, then memory and core rails. The order follows the power sequence.
2. Measure each at a large capacitor or test point, found through the boardview.
3. Compare with the stored value and note anything far off.
4. For a low reading, highlight the net in the boardview to see every part on it, then split the rail at jumpers or inductors to narrow it down.
5. Move on to control signals only once the rails look normal; small signal nets vary more and are harder to judge.

Building your own reference values

Stored readings exist for only some boards. You can build your own:

• whenever a known-good board of a common model passes through your bench, measure the main rails and note the values with the meter you always use;
• record the conditions: polarity, battery disconnected, meter model;
• keep the notes per board code.

Readings taken by you, with your meter, are more useful than any stored value. Over time they become one of the most valuable resources in a repair shop.

When diode readings mislead

• Chips with internal regulators can make a net read unusually low on a healthy board. If two boards of the same design read the same "low" value, it is normal.
• Nets with only a pull-up resistor read the resistor value to another rail plus whatever that rail does, which can vary a lot.
• Large rails take time to settle; reading too early gives falsely low numbers.
• Corrosion can give readings that drift as you measure.
• Different board revisions may add or remove parts on a net, shifting the value.

Use diode values to decide where to look, then confirm with other methods: resistance to ground, voltage injection, thermal imaging, or simply powering the board with a current-limited supply and measuring rails.

Recording readings consistently

Whether you rely on stored values or your own, consistency is what makes them comparable. A simple record for each board code does the job:

• the meter model and the diode-mode test voltage if your meter shows it;
• the probe arrangement (which probe on ground);
• the board state: battery and coin cell disconnected, daughterboards attached or not;
• the point where each net was measured — a test point or a capacitor designator from the boardview;
• the value after it settles, rounded to the nearest 10 mV.

Measured this way, two readings taken a year apart on two boards of the same model are directly comparable, and small differences start to mean something.

Which formats carry diode readings

Readings are not part of any one boardview format; they are added by whoever prepared the file. Some boardviews distributed for repair work include them for most nets; many others have none. Your viewer shows them when present, usually next to the net name or in the pin information. If you are new to boardviews, read What Is a Boardview File and Why You Need One first.

FAQ

Why does my meter show different values from the boardview on every net?

Meters differ in test current and display. If the difference is consistent across healthy rails, it is your meter, not the board. Scale your expectations accordingly.

Should I use red on ground or black on ground?

Match the convention used for the stored values. Many repair technicians use red on ground and black on the net, but not everyone does. Measuring a few healthy nets tells you which one the file used.

Is a reading of 0.000 always a dead short?

A reading of a few millivolts almost always means a short or a very low resistance to ground. Confirm with resistance mode before you start removing parts.

Can diode readings find a bad chip that is not shorted?

Sometimes. A chip that leaks — partially damaged by liquid or a surge — often lowers the reading on the nets it connects to. It is a hint, not a proof.

Bottom line

Stored diode readings are a shortcut to suspicious nets on a dead board. Match the polarity and conditions of the reference, compare consistently, and treat large deviations as places to investigate rather than final answers. Build your own reference notes for the boards you see most, and use a diode mode comparison as one tool among several.