Hacking A Fluke Multimeter To Serve Readings Over WiFi

Your multimeter is probably your most useful instrument if you work regularly with electronics. It goes with you everywhere, and is your first port of call in most cases when you are presented with a piece of equipment. And when you think about it, it’s a pretty amazing instrument. Multimeter technology has advanced to the point at which even an inexpensive modern device has functions that would have required a hefty budget a few decades ago.

There is still one thing affordable multimeters remain unable to do: they can’t log their readings for analysis on a computer. They’re an instantaneous instrument, just as they always have been.

Lord of Hackaday [Sprite_TM] decided to hack his multimeter to serve its readings over Wi-Fi. Rather than start with a throwaway meter from the bargain bin, he did it with a Fluke. The meter he chose was a Fluke 15B+, the company’s budget offering for the Indian and Chinese markets, since he had one spare.

Opening up the 15B+, he was presented with its processor concealed under a blob of epoxy and thus unidentifiable. Armed with the knowledge that other similar Flukes contain Fortune Semiconductor parts, he investigated as many data sheets as he could find from the same company and finally identified it as an FS98O24 one-time-programmable microprocessor. Sadly this chip has no serial port, but he did find an I2C EEPROM which he correctly guessed held calibration settings. Removing this chip gave him a meter with slightly off calibration, but also gave him a serial port of sorts.

Further detective work allowed him to identify the baud rate, and supplying random commands delivered him some that returned data packets. Eventually he identified a packet containing the states of the LCD’s segments, from which he could derive its displayed value. Connecting an ESP8266 module with appropriate software left him with a Wi-Fi connected multimeter. There was a little more refinement to his hack, he created a power management board to activate the ESP when needed, and a neat hack to display its IP address on the screen.

Multimeter hacks have featured several times here at Hackaday. We’ve had another serial port hack, or how about a remote display for another Fluke on a Gameboy Advance?

47 thoughts on “Hacking A Fluke Multimeter To Serve Readings Over WiFi

  1. I took a digital scale and hooked the power supply wires and connected them directly inline with the battery on my sell phone you can measure the flux in the OS and if you know what your doing you can map out … well let’s just say theirs allot you can do measuring the resistance of time and using a formula that involves the speed of lite

    1. I’m not quite sure what you’re trying to say.

      Measuring the flux in the OS?

      Also homophones and correct grammar, before someone else says it. (sell = cell, theirs = there’s, allot = a lot,lite = light)

        1. Lol yea I see what you mean but don’t take it out of context all I’m say is for instance even the most basic computers have to have something in the environment to start a boot cycle like on older computers their used to be a little speaker that would beep … this is what allows the Dll to maintain a logic loop keeping the internal clock from loosing it’s place in the registry in relations to the end user when it powers off and on.but now a days thing are a little more complex…and if you had the slightest idea of what I’m talking about you might understand what getting at e.g.(the fluctuation)it’s like having a decoder for the registry you can keep track of the servers and tell were something is coming from with out a IP adress , no pissing contests her though. Oh And you can also turn a multi metter into a geiger meter … lol

      1. Sorry it was late lol…I should have said when the scale is hook up in a capacitated environment when using it inline with the cellphones power supply you can see any shift in the phones parameters because of the frequency of the Crystal in the scale. You can see not only a negative but a positive shift as the phone is either pushing or pulling information from the cell towers. Depending on the time and the power consumption you can map out all sorts of thing the cell phone providers don’t want you to know, as well as thing I’m not going to mention in a public chat room….

      1. hay Rob don´´´´t be afraid to add any input on my comments maybe an argument some sort of contradiction or whatever but talking s**t is just tells me your a dumb **S.

  2. From the article, it looks like the multimeter HAS a serial port, normally used for calibration purposes.
    Sprite_TM desoldered the eeprom in order not to mess with the calibration that it stores… but it’s not the removal of the eeprom that gave him a serial port ;-)

    1. I would assume that the serial has a double duty for reading the calibration data on power up. The serial port functionality is always there as the firmware was written to respond to it. I2C devices would ignore stuff on the I2C until they see the I2C Start. By being sneaky with the design, you can share the signals.

      1. You can emulate an I2C EEPROM to fool the Fluke. By bit bang the GPIO lines on the ESP, you can make a read only I2C slave EEPROM device assuming that it can boot up fast enough to do so. (The contents of the desoldered EEPROM can be with an external programmer.)

        Write protect (active high) on I2C parts are right next to the Vcc pin. You can lift pin, put in a piece of kapton tape on the pad and solder blob in less 5 minutes if you choose so. Been there done that.

        1. That’s what I thought at first, but he contradicted my post implying that he *thinks* that it couldn’t co-exist and left it out. On the other hand, the picture does show the EEPROM is soldered in place along with the hack. So I guess he misunderstand my post?

          It’ll need a I2C start as well as matching I2C address + Write before the EEPROM would do anything. Now if Fluke had it designed correctly, then the condition of ruining that is very low. They are in the Test Instruments business.

          1. Ah, I put the EEPROM back after figuring out the protocol: the multimeter won’t measure anything useful without it, and the specific commands I use in the final product don’t seem to mess with the calibration.

      1. The only thing I can think of that you would need to shield yourself from when taking measurements on a multimeter is making a nuclear bomb or power plant ,or some type of new optical display….

        1. Hey, they could make it True RMS and stick all the features of a 1000 dollar meter for virtually the same price, but then no one would buy the 1000 dollar meter, would they?

          1. Well in a meter like the one I got from harbor freight for 10$ if you open it they have built in extras. If you pull the knob off and align the points differently theirs different functions…theirs also spots on the circuit board to add more precise measuring instruments…in a tool like a multimeter it has set potentials a high resistance and a low resistance (semiconductors). These are made from different materials found in the earths environment from metals to gas and minerals. They all have a different reactant rate when a current is induced causing a reduced rate of flow in the current. Now if you break it down these measurements can be taken down the the molecular levels if need be…the question you need to know is who sets these standard levels.the thing about that is the multimeter is the main idea in the process in how a registry works on a computer it measures a momentarily current when electricity is applied. Now The ? at hand. To gather data that can be transferred to the SD card your going to need something that works statically to read write information like a hard drive which is consistently looping. A multimeter has set functions with set tangents that have coefficients based on the semiconductors ability to accept the electrons rate of flow through the circuitry. now what is set at a balance (0) on a multimeter and what is set at (0) in a computer are two different things the zero in the meter is sold state does not move and what is set in a computer is consistently shifting. I did a experiment and reprogrammed the multimeter using different resisters from my work at tarra systems I had about 50 different typs I then tried to map out my smartphones micro processor chip output levels of resistance,itwould cause the display to loop out and the output reading would shift causing it to display a different reading if I did it in the right sequence I noticed a pattern. Depending on the sequence it would expand the sequence or it would reduce the sequence like fragmentation expansion or defragmentation reduction. Simply if you want to measure the flow of electricity/data and store a record on a removable drive, Then Yull need a smart motherfucker like me. Theirs allot of legal bullshit though if it is put on the market I’m sure someone is going to try and sue someone for copywrites on patents but theirs a way around that it’s got allot to do with freedom of speech, I know it sounds like two different subjects , but their more closely related than you think and you just happen to find the guy that has the answers to the ? “But”, yes theirs a but,27trillion$ reasons why it hasn’t been done.universal porthole service on a global level can become very pricey…

  3. So.. after he was done did he re-connect the EEPROM then? Or did this hack mean he couldn’t have both the ESP and the EEPROM at the same time? “Rather than start with a throwaway meter from the bargain bin, he did it with a Fluke.” He didn’t turn it into one did he?

    Also, with the calibration done entirely at factory and just stored… I wonder how these things age…

    I also wonder… would it have been easier to just grab the signal going to the LCD and ‘read’ the numbers from there.

    1. Yes he reconnect it. He only removed the EEPROM to avoid messing up the content (and so the calibration of the multimeter) while sending radom data to the multimeter. Didn’t know that calibration is now done in software but it’s quite logical. The problem is that you can no longer calibrate your multimeter yourself, you would need the protocoll etc instead of a simple screwdriver. :-(

      1. yes you can program it yourself their should be a place on the board some were when they set the parameters the first time. your going to need a whole array of different resisters you can go to mouser.com to find it. i worked at a digital scale company programming
        and building LED displays for all sorts of applications using scales… you will have to measure each resister in a certain sequence. you will have to figure out the rest … its like loading the defaults back in the device your just installing set potentials back to factory setting sordaspeek.

    2. No you can’t do that ,it is a momentary measurement if you connect anything to it it will make the displayed reading jump all over the place,because the meter is not only reading the measured output of whatever it is your measuring it will also pick up anything inline with the digital or analog output reading on the display…

  4. on a side track: I haven’t found on the market a portable (battery powered) DMM to which I can plug in an (u)SD flashcard and have the measurements logged, say over a few days, out in the field.
    Serial connections + laptop for logging the DMM measurements don’t cut it: clumsy, unreliable, too short battery life of the laptop, etc.

    any friendly pointers/hints please?

  5. oh Jenny List , this summary suggest you didnt read second page of Sprite_TM writeup, or whats worse you read it but didnt understand? :(
    there is a serial port, eprom was disconnected to make sure its not overwritten while fuzzing.

Leave a Reply

Please be kind and respectful to help make the comments section excellent. (Comment Policy)

This site uses Akismet to reduce spam. Learn how your comment data is processed.