Shop OBEX P1 Docs P2 Docs Learn Events
Over voltage damage to pin through voltmeter? — Parallax Forums

Over voltage damage to pin through voltmeter?

sccoupesccoupe Posts: 118
edited 2018-07-19 17:42 in Propeller 1
If I connect a voltmeter between a 12v power supply and a propeller pin set to logic low, will it damage the pin? The propeller and power supply share the same ground.

Thanks

Comments

  • I'm trying to figure out why you would do that?
  • I personally wouldn't. But i'd like to know of any consequences without needing to fry a micro to find out.
  • If you have a digital voltmeter it should have a resistance of more than a mega-ohm, and there shouldn't be a problem with measuring the voltage between a 12 volt power supply and a propeller pin. However, if you have an analog voltmeter you might have a low resistance if you select a low voltage range. On my analog voltmeter the resistance is about 6 K-ohm for the 0.25 volt range. The 10-volt range is 250 K-ohm, and the 50-volt range is greater than 1 M-ohm. Check your voltmeter manual to see what the impedance should be.

    Just to be safe you might want to put a 10 K-ohm resister in series with your voltmeter test leads to ensure that you don't run too high of a voltage/current to the Prop pin. Or better yet, just measure the pin voltage from ground. I would also wait to hear from some of the hardware gurus on the forum just to be sure.
  • If the pin is an output, it will clamp the voltage at Vdd or Vss, and if it is an input, the Prop substrate diode will normally clamp it at around Vdd+0.7V. Not a problem in either case, although there can be special circumstances. I agree with what Dave said about a classic voltmeter where the pointer moves due to the current produced by the device under test, it is a low resistance path that could be dangerous for the test suggested. Better get yourself a modern meter with an input resistance of at least 10MΩ! With that, the current from the 12V supply thru the voltmeter into an input pin on a Prop powered at 3.3V would be limited to 4V / 10M = 0.4µA, which is quite safe for the Prop substrate diode.

    You say voltmeter, but if it is a multimeter, be sure by no mistake is it left on a current measuring range when you make that hookup to 12V! Sure smoke!

    While on this subject, there is a potential issue with ohms, continuity and diode-check measurements. Multimeters vary quite a bit in the voltage that they put behind those measurements. For example, I never use my fancy Agilent 6.5 digit 34401A multimeter for troubleshooting Prop boards for continuity, because on all its ohms and continuity ranges it puts out more than 7 volts behind the measurement, at up to 100 microamps current. While the substrate diodes on the Prop (and most ICs) can absorb that kind of current, it can be a disaster when a Prop circuit is unpowered or is running on RCslow. Even 100 µA can overflow the power supply and subject the entire circuit to damaging voltage. For that kind of troubleshooting, I reach for an old Beckman Tech-300 that uses less than 0.4V for its ohms measurements, which is safe, and it has the added advantage that it does not forward bias diodes (except on its diode test range), which is an advantage for troubleshooting. Even a cheap Chinese import multimeter I have (labeled "Excel"). keeps the potential on its ohms ranges at under 3 volts. It is definitely something to be aware of. It's no joy to have a circuit ruined by the test instrument.

  • The issue I see is than any capacitance or internal protection component in the meter could cause
    significant currents to flow briefly.
    One would hope though that the meter is basically has > 1M resistor is series with anything internal,
    so only the stray inductance and capacitance of the leads could cause issues.
  • Datasheet, brand, etc. on voltmeter would be helpful. Is it cheap china or fluke etc.
  • I see the original post as only telling half the story. Did you actually damage a pin and how or is this only hypothetical? What range was the meter on etc? If you're going to ask a question then supply any necessary information up front. "I'd like a burger" is a question that requires more information which is quick and easy over a counter but a forum needs it spelt out up front otherwise we are just playing a guessing game.
  • Lots good info here. A remote user tested a circuit that measures a 0-3.3v signal and two outputs are either logic high or low based on the signal range. The supply voltage to the propeller circuit was 12v with the proper regulating to 3.3v. A second power supply with a common ground was used to simulate the 0-3.3v signal. The user then used a multimeter (not a voltmeter) expecting to measure 12v, or not between the propeller output and the 12v supply based on whether or not there was logic ground. It was designed to measure between output and VCC for 3.3v or 0v. While everything worked as expected when it left me, it doesn't work this way with the remote user now. So, based on what i read here, he may have been ok by measuring this, but perhaps he could have powered everything up, hooked up the meter and then spun the dial to the voltmeter setting and possibly caused damage. The unit is on the way back in any case. Will find out for sure. I'm just not good enough at designing something to protect against, "Why would anyone do that?". :)
  • Protecting against "Why would anyone do that?" is actually pretty difficult - I'm currently working on a design to be used in factory systems with up to 24v supply, and I'm designing it to tolerate 60V AC to any pair of pins without damage.
  • It is always pretty hard getting the ab'user to tell you exactly what they did and what they may even have as a sneaking suspicion that caused the problem. Nonetheless, once you get the unit back you can test it but more than likely they spun the dial around past (or from) the low current range and immediately blew the pin.
  • It depends on the meter. Most meters have a different set of jacks to read current to prevent mishaps.
  • I have about 8 meters floating about my workshop and quite a few of the more common ones have the same jacks for voltage and low current, with a separate jack for 10A.
  • Tracy AllenTracy Allen Posts: 6,656
    edited 2018-07-22 02:00
    Then, after an amps/milliamps measurement, one forgets and makes a voltage measurement. The Jacks for the probes can be separate or they can be the same, it doesn't matter, human error. At best, pop goes the meter fuse (Oh! fubar, no replacement fuse on hand!). Been there done that.


  • Both output pins were fried. Processor swap and all working again and sent back out. Thanks all.
  • I remember "serviceman" type articles in electronics magazines where the author would tell this long story about a situation or a customer, so as to paint a picture which gave us readers the insight that he had in this regard. So when he made a statement like "one day this customer walked in" then it would be like we were in his place and had all the background so everything that followed made sense. Not many of us are writers but usually we can tell a story which is so much better for us readers rather than sparing the detail which makes it so much harder to know how and what to answer :)
  • There are still such articles. Take EDN's popular series, Tales from the Cube. The devil is in the details, or, sometimes, the forest for the trees.

  • Most multimeters have a 7V crawbar using 2 BJT , could this pass through higher voltages?
    https://www.eevblog.com/forum/beginners/multimeter-input-protection-what-are-these-bjts-doing/
  • jmgjmg Posts: 15,140
    tonyp12 wrote: »
    Most multimeters have a 7V crawbar using 2 BJT , could this pass through higher voltages?
    https://www.eevblog.com/forum/beginners/multimeter-input-protection-what-are-these-bjts-doing/

    Those should be 'on the other side' of the main series divider resistor, so they may change the apparent resistance, but should not jump from uA to mA.

    Most multimeters have 1~10M input dividers, but some better ones remove the down side resistor on the lowest range, to go > 100M, and some have an optional lower load

    The most likely oops, is 'range-right, but probes in wrong place' - which everyone denies ever doing... ;)
Sign In or Register to comment.