Fairchild IC Unlock: Which Families We Can Read

Here’s the short answer most people need first: the large majority of Fairchild part numbers are analog, discrete, or logic devices with no user-programmable memory inside, so there is nothing to unlock. A genuine Fairchild IC unlock job is almost always one of three things — a serial EEPROM or flash sitting next to a Fairchild power stage, a legacy Fairchild-branded microcontroller or programmable logic part, or a part that became an ON Semiconductor device after the 2016 acquisition and is still marked Fairchild on the board.

So the useful question isn’t “can you unlock Fairchild chips.” It’s “which chip on my board actually holds the code, and is that part readable.” That’s what this page sorts out.

Why “Fairchild IC unlock” is usually a mislabeled request

Fairchild Semiconductor built its name on discretes and analog: MOSFETs, IGBTs, gate drivers, optocouplers, PWM controllers, LDOs, the 74AC/74VHC logic families, the FAN-series power management parts. None of those store firmware. A FAN7601 doesn’t have a program image inside it, and neither does an FDP-series MOSFET or an FOD optocoupler.

What usually happens is this. An engineer opens a dead controller board, sees a big Fairchild marking on the largest IC, and assumes that’s the brain. Nine times out of ten the actual code lives in a small 8-pin SOIC serial EEPROM, a SPI flash, or a separate microcontroller from a different vendor entirely — an STM8, a PIC, an 8051 derivative — sitting a few centimeters away.

That’s good news for you, because the parts that do hold the code are usually the well-characterized ones. It just means the first job is identification, not extraction.

The families that genuinely can hold code

• Serial EEPROM and flash on Fairchild-heavy boards. 24Cxx, 93Cxx, 25-series SPI flash. These carry calibration tables, run-hour counters, motor parameters, and sometimes a whole bootloader image. Readable in most cases, and the most common real outcome of a Fairchild unlock inquiry.
• Legacy Fairchild-branded microcontrollers. Older 8-bit parts from the pre-2000 catalog, including Fairchild-labeled 8051-architecture devices. These use classic lock bits or security fuses rather than modern multi-level read protection, which generally works in your favor.
• Fairchild programmable logic. Older PAL/PLA-class devices and one-time-programmable arrays. These need a different approach entirely — you’re recovering a fuse map, not a hex file.
• ON Semiconductor parts still marked Fairchild. Post-acquisition, several product lines carried over. If the die is an ON Semi design, feasibility follows the ON Semi rules, not the old Fairchild ones.
• Configuration memory paired with Fairchild power stages. In LED drivers, SMPS, and motor drive modules, the Fairchild part switches the power and a tiny memory holds the setpoints. That memory is your target.

What we do with the board, in order

The process is deliberately front-loaded on identification, because the wrong target chip wastes everyone’s time.

1. Read every marking. Full top marking, date code, package, and any secondary line. Fairchild logos appear on relabeled and remarked parts often enough that we treat the marking as a hypothesis, not a fact.
2. Map the board’s data paths. The engineer traces SDA/SCL, SPI CS/CLK/DI/DO, and any programming header back from the candidate memory. If a chip has no data bus going anywhere, it isn’t holding your firmware.
3. Confirm the die where the marking is doubtful. For questionable or heavily worn parts, a decap and optical inspection tells us what’s actually inside before we commit to an approach.
4. Non-invasive read attempt first. Standard programming interfaces, factory test modes where documented, in-system reads where the board permits. This is where most EEPROM and legacy 8-bit work finishes.
5. Escalate only if needed. Semi-invasive and invasive work is quoted separately, because it consumes the part and it changes the cost picture significantly.
6. Verify the dump. A read is not a result until it’s checked. We look at reset vector plausibility, interrupt table structure, string and lookup-table regions, blank-fill patterns at the end of memory, and — where a second identical unit exists — a byte-level comparison between two dumps. A dump that programs into a blank part and boots the board is the strongest verification available, and we prefer to get there.

Honest limits on Fairchild IC unlock work

This is the part competitors skip. Here’s where we say no or “maybe.”

OTP and mask-ROM parts. Some legacy Fairchild devices are one-time-programmable or mask-programmed. Mask ROM contents can sometimes be recovered optically after delayering, but that is a laboratory-scale job with real cost, and it is not always successful on small geometries or with damaged die.

Physically damaged parts. A chip that took a surge, shows a cracked package, or has a blown bond wire may be unreadable regardless of protection. Heat-stressed flash cells lose charge. If the board died from an over-voltage event, tell us — it changes our estimate.

Encrypted memory contents. Reading a 25-series flash is often straightforward. Reading it doesn’t help if the payload is encrypted and the key lives in a secure element or in a fused key store somewhere else. We’ll tell you if a dump looks high-entropy end to end, because that’s the signature.

Unclear markings with no reference unit. Sanded-off tops, laser-blacked packages, and house-numbered parts add an identification stage. Sending a second board, or even a known-good unit, shortens this a lot.

What changes the answer: die revision, whether the security fuse was actually set or the customer just assumed it was, whether the programming pins are still accessible on the board, package type, and whether you can supply more than one sample.

Which target to expect, by symptom

What to send us, and what you get back

A useful quote takes about five minutes of your time to set up. Send:

• Clear photos of both sides of the board on a dark, non-reflective background, whole board in frame, in focus. Then one close-up of the suspect chip with the marking legible — angle the light so the laser etch reads.
• Board dimensions and rough layer count if you know it.
• The exact chip marking, transcribed line by line, including the date code.
• One line on what you need: a raw binary dump, a hex file you can reprogram into replacement parts, or firmware plus a full board reconstruction so you can build again.
• Any history — did it fail under load, is the part obsolete, do you have a second working unit.

What comes back from us: a feasibility read (yes / probably / unlikely, with the reason), the approach we’d use, an estimated lead time, and a price. If the answer is “unlikely,” we say so rather than taking the job and stalling. Pricing on this kind of work is driven by target chip family, protection mechanism, whether the package has to be opened, how many samples you can supply, and whether you also need the board itself reconstructed — which is why we quote per case instead of publishing a rate card.

Why people order this work

The realistic reasons, all legitimate: the original hex file and source disappeared with a departed contractor or a dead build machine; a production line runs on a controller nobody makes anymore and spares are gone; a Fairchild power part went end-of-life and the design has to be migrated to a current equivalent, which means understanding what the firmware expects; or a failure analysis needs the actual memory contents, not a guess about them.

We work on the basis that you hold the rights to the design or have the owner’s authorization, and every project runs under NDA. That’s the whole of it — we’re not going to lecture you, and we’re not going to help anyone clone a competitor’s product.

Because a Fairchild inquiry so often turns out to be a different vendor’s microcontroller, it’s worth knowing our coverage elsewhere: the STMicroelectronics families we can read come up constantly on power boards using Fairchild switches, PIC and Microchip unlock work covers most of the legacy 8-bit controllers we see, and Nuvoton and 8051-derivative parts account for a lot of appliance and driver boards. If your target turns out to be programmable logic instead, CPLD and FPG

Related reading

• Oct 2026dsPIC33C Crack | Protection Levels & Recovery Guide
• Oct 2026Microchip IC Unlock | PIC & dsPIC Families We Read
• Oct 2026dsPIC33C Unlock | Feasibility, Cost & Lead Time