2
\$\begingroup\$

I have a digital scale with built in auto off timeout function pictured, is there a way to hack modify this PCB to bypass the auto off function after timeout? So when batteries are inserted it will stay on until battery power is depleted. The E- S- S+ E+ go to load cell, ribbon cable goes to LCD, battery +- go to +3V- on lower photo lower left, and LED on lower photo upper right goes to backlight. I know it's controlled by the potted MCU but how could I perhaps mod the PCB to bypass?

EDIT: The pad right next to C3 which looks unmarked is normally 3V and when you press the tare button it goes to 0V, and PT21 is normally 3V and goes to 0V when you press the mode button. I tried PT20 and VPP and pressed the power button and they're always on 3 volts even when the power button is pressed and all of the pads are normally 3V even when the scale is powered off.

EDIT: I should mention that I need to keep the scale on for continuous weighing to monitor changes over a period of hours with a weight on the scale after it is zeroed out and there will be periods where the weight doesn't change enough to keep it from shutting off. I'll be recording the reading on camera so I won't be there to turn on if it goes off.

enter image description here

enter image description here

\$\endgroup\$
13
  • 2
    \$\begingroup\$ If you hold down the ON/OFF button does that disable the auto-off function? If so, perhaps could short-out ON/OFF button as modification. If holding down the ON/OFF button doesn't achieve the desired effect, guess a modification isn't possible unless you can reprogram the MCU. Do you know which MCU is used, or have access to the source code? \$\endgroup\$ Commented Oct 12 at 12:12
  • 3
    \$\begingroup\$ C'mon guys isn't this just a disallowed question due to it asking how a product might be modified to make it work without supplying a circuit: Questions on the use of electronic devices are off-topic as this site is intended specifically for questions on electronics design. <-- the rule I mention. The question raised is this: how could I perhaps mod the PCB to bypass so, unless someone has a schematic any answer is also an opinion and, they are soundly disbarred from being valid. \$\endgroup\$ Commented Oct 12 at 13:35
  • 2
    \$\begingroup\$ Re-programming likely not possible. An add-on circuit should be able to sense the back-light LED turning off, which would (as an output) re-trigger the "on-off" push button. How those push-buttons are sensed by the on-board MCU would have to be investigated. \$\endgroup\$ Commented Oct 12 at 14:12
  • 1
    \$\begingroup\$ @Andyaka <list_member_hat> --> I don't think so. We and the OP do not have a circuit diagram BUT I've used electronic digital scales since befre they existed :-) - and many here have used them for decades. My initial answer had some misadventures in it but was solid (enough :-) ) engineering design , and the OP says it provide him with a solution that worked. He plans to use a microcontroller to implement the solution and this Q&A have a good chance of satisfying the prime directive: "Provide a good persistent Q&A set that drives search engine traffic to the site". <hat_reset>. \$\endgroup\$ Commented Oct 13 at 7:10
  • 1
    \$\begingroup\$ @Andy <hat_as_above> --> I now see that the system says that your comment was made after my answer (noting that the system may lie). Serious amicable question: Did you not think that my answer demonstrated that the question was reasonable and/or effectively moved it into "a reasonable space". [[I'm (always) trying to learn. Dinna always happen :-) ]]. \$\endgroup\$ Commented Oct 13 at 7:16

3 Answers 3

Reset to default
2
\$\begingroup\$

Only maybe:

IF Chester's holding down the on/off button does not work, then:

  1. It is unlikely but possible that a signal on one of the other MCU leads will work. You have via's available on some leads. You can access then with a sharp probe - possibly having to remove conformal coating, for which a drill bit hand wielded should work. Applying battery or ground via 10k may work - but quite likely will not.

  2. IF pressing and releasing the on button or tare button occasionally keeps the scales on then a simple circuit that applies an occasional pulse of appropriate polarity, or possibly a floating short, would work. [Added: tare button is a bad idea]

  3. If there is a clock or similar signal on any MCU pins that may be able to be used to drive a MOSFET repeatedly to achieve the same aim.

  4. Added. Scale-subject-brain was obviously not working well enough: Pressing power once turns it off, of course. But pressing it twice turns it off and on. As long as it does not auto tare this works BUT it probably does auto-tare,
    so ... :-(.
    So operating mode button (which I did not mention directly) N times sounds most useful. If you are REALLY keen you may be able to monitor an LCD segment to ensure that it has returned to the correct setting. As long as the remote button control is open circuit when idle then using the panel control if needed to alter the mode will work.
    Note: Added after answer accepted BUT does not significantly alter the question:answer relationship.

Of these 2 is the most likely, and the hardest to implement.

\$\endgroup\$
11
  • 1
    \$\begingroup\$ Pressing the tare button does keep the scale on until the next timeout cycle starting after the button is pressed. How could I design such a circuit? \$\endgroup\$ Commented Oct 12 at 13:17
  • 1
    \$\begingroup\$ But also I wouldn't want to use the tare function while it's in use because I need to keep the weight on the scale monitored for changes over a period of hours and there will be times when the weight doesn't change enough to cancel the auto off timer. \$\endgroup\$ Commented Oct 12 at 13:32
  • 2
    \$\begingroup\$ @wdbwdb1 Do you know how the buttons are connected? E.g. when pressed do they connect a pin on the MCU to 0V? From the picture of the PCB in the question the buttons appear to be surface mount, and maybe the PT20, PT21 pads and the pad adjacent to C3 are connected to one side of the buttons. If you have a multimeter can probe the voltage on these pads as the buttons are pressed, and edit the information in to the question? \$\endgroup\$ Commented Oct 12 at 13:32
  • 1
    \$\begingroup\$ Yeah the pad right next to C3 which looks unmarked is normally 3V and when you press the tare button it goes to 0V, and PT21 is normally 3V and goes to 0V when you press the mode button. I tried the other two pads and pressed the power button and they're always on 3 volts even when the power button is pressed and all of the pads are normally 3V even when the scale is powered off. \$\endgroup\$ Commented Oct 12 at 14:08
  • 2
    \$\begingroup\$ @wdbwdb1 1. Does pressing the power button occasionally keep it awake? 2. Does pressing the power button affect the displayed value IF it has not been zero'd with TARE? 3. Mode button could be operated with an "auto press system" to press it N times rapidly and then pause for most of a power-off cycle. \$\endgroup\$ Commented Oct 13 at 2:24
4
\$\begingroup\$

Unlikely to be possible unless they designed for such an option such as the power switch suggestion or there is some jumper option (nothing seems obvious).

Even if you cycled the power the scale would auto-zero.

You could possibly replace the board with an Arduino, an HX711 load cell ADC board and a display. Note that load cells "creep" if the load is constantly applied so don't expect accurate measurements after a long time.

\$\endgroup\$
5
  • 2
    \$\begingroup\$ Pressing on/off MAY just keep it onOR effectively do nothing (or not). Pressing Tare may keep it awake. Mod may swap beten imperial and metric, so pressing twce may work. \$\endgroup\$ Commented Oct 12 at 12:40
  • 2
    \$\begingroup\$ @RussellMcMahon Good points all. Some devices reset their power-off timer if any key is touched. Perhaps a short push of the power button even. \$\endgroup\$ Commented Oct 12 at 13:22
  • 2
    \$\begingroup\$ Mode key looks like a possibility. Possibly simulate a burst of presses to get back to original mode. If it resets the timeout. \$\endgroup\$ Commented Oct 12 at 14:17
  • \$\begingroup\$ As you answered, I'll ask - did you upvote the question? Someone who answered didn't. Obviously absolutely no compulsion. \$\endgroup\$ Commented Oct 13 at 7:40
  • \$\begingroup\$ The phantom downvoter strikes again. \$\endgroup\$ Commented Oct 13 at 7:44
3
\$\begingroup\$

There is an easy way to wake up the scale immediately after it shuts down. It's not exactly what you want, but perhaps would be an adequate solution.

When the scale is off, there's no excitation supply to the sensor, and E+ is zero. This could trigger a NE555C single-shot pulse that would trigger the ON/OFF button. The button could be "pressed" using a mosfet to short its terminals, or by driving the input directly - depending on how it's wired up in the scale.

An alternative would be to figure out what common "scale all in one" chip die is used, get the development tools from the chip manufacturer, and write new firmware. In all likelihood, the chip manufacturer provides sample firmware to make it easy for the customers to modify it to suit their product - so it would be even simpler then. Disabling the sleep function would be probably a change to a couple of lines at most.

The chip may be one-time programmable. In that case you'd need a new board with a packaged chip on it, instead of the bare die under a blob. At that point you can do it in whatever way suits you instead of using an obscure (to us) chip. A most basic Arduino with a suitable A/D converter attached would do an excellent job here.

\$\endgroup\$
3
  • 2
    \$\begingroup\$ Waking up typically won't work on this kind of scale which auto-zeros when powered up. Unless the weight is removed and replaced it will always read zero after powering up. And manufacturers typically set the lock bit after programming so competitors can't copy as easily so investigating the firmware is likely out. \$\endgroup\$ Commented Oct 12 at 15:55
  • \$\begingroup\$ As you answered, I'll ask - did you upvote the question? Someone who answered didn't. Obviously absolutely no compulsion. \$\endgroup\$ Commented Oct 13 at 7:40
  • \$\begingroup\$ @RussellMcMahon Of course. I compulsively upvote decent questions, and all questions I answer. A few days a year I run out of votes :) \$\endgroup\$ Commented Oct 13 at 12:09

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Start asking to get answers

Find the answer to your question by asking.

Ask question

Explore related questions

See similar questions with these tags.