Electrical Gurus Needed :) ---- Measuring The Millivolt With High Accuracy

12346»

Comments

  • Take a gander at this one
  • Definitely remember the contact points of the probes as DigitalBob mentions. I found a couple more ATC fuses to test, and one of them from old stock or a different manufacturer seemed to have a layer of corrosion on the metal. It took a sharp probe to break through and make good contact. I was thinking a pogo test pin would work well to help assure a consistent force, but it would have to have a stiff spring and a really sharp point.
    432 x 432 - 72K
  • DigitalBob, Wow, a 100MΩ shunt resistor! Burr-Brown came out with all those solid precision state-of-the-art analog parts in the INA and OPA series. That is from 1995 or earlier, witness the +/- 15V isolated power supply.
  • Erna - Thanks for the info... That was definitely of interest to me, being somewhat of an automotive enthusiast.

    DigitalBob - It is nice to know that some accurate readings can be obtained from minimal parts.

    Tracy - While it is true, that I currently have most of my parts, I have been waiting on another order, before actually starting on this project. Those additional parts are scheduled to be delivered tomorrow. Then once those parts arrive, I will be able to put some of the project together, but then I will still have to order some more additional parts. I truly think that you would be impressed with my overall concept, but of course, as Peter mentioned, I am secretive :) It sure would be nice, if you were my next door neighbor, then we could collaborate on the design over the fence :)
    I found a couple more ATC fuses to test, and one of them from old stock or a different manufacturer seemed to have a layer of corrosion on the metal. It took a sharp probe to break through and make good contact.

    Corrosion is definitely an issue to be considered. Over the years, while working on quite a few automobiles, I have encounter fuses with corrosion.
    To check, I hooked it up to a constant current power supply (Agilent 6612) set at 0.219A.

    It sure would be nice to have a similar piece of equipment. I still have not figure out the lest expensive method for simulating various loads.


    Novel Solutions - http://www.novelsolutionsonline.com/ - Machinery Design • - • Product Development
    "Necessity is the mother of invention." - Author unknown.

  • I do like good power supplies! That's not to say that you can't DIY a nice current sink for your testing and development--Fat transistors and big heat sinks.

    The HP/Agilent 6612 power supply is a workhorse, 0-20 volts up to 2 amps, constant voltage or constant current, and it can regulate for both sourcing and sinking current. Another great feature is its two ranges for monitoring current, 0-20mA with 1µA resolution and 0-2A with 100µA resolution. So it can test both operating current and also standby current drain.

    All off ebay or Craig's list, I have also a 6611 (8V, 5A) and 6613 (50V,1A), as well as an old HP6033 (20V, 30A) and a Keithley 228, 4-quadrant supply (+/-10V, +/-10A, +/-100V,+/-1A). Those really help for multi-supply tasks, e.g., to characterize charge/discharge of solar power systems over a range of conditions.
  • That INA117 circuit that DigitalBob pointed out above uses back-to-back parallel diodes for protection of the input, like we have been talking about. What they pulled out of the bag of tricks is that the diodes are the collector-base junction of two 2N3904 transistors. Those junctions will have far lower leakage current than a signal diode like the 1N4148. Signal diodes are gold-doped for speed and have leakage in the nA range instead of pA. That's significant in that INA117 circuit with the 100MΩ shunt resistor. LED's too exhibit low-leakage in both reverse and forward low voltages , due to the very high bandgap voltage of the materials they are made of.
  • Tracy
    That INA117 circuit that DigitalBob pointed out above uses back-to-back parallel diodes for protection of the input, like we have been talking about. What they pulled out of the bag of tricks is that the diodes are the collector-base junction of two 2N3904 transistors. Those junctions will have far lower leakage current than a signal diode like the 1N4148.

    As mentioned, I will be putting together the circuit of diagram #2 that you have provided. So are you suggesting that I replace the back-to-back parallel diodes in diagram #2 with (2) 2N3904 transistors or is there something else that you are also hinting at?

    I might have some spare 2N3904s laying about, but I know that I have used a few over the years, so they may be gone by now...


    Novel Solutions - http://www.novelsolutionsonline.com/ - Machinery Design • - • Product Development
    "Necessity is the mother of invention." - Author unknown.

  • Bruce, signal diodes like the 1N4148 will be fine for your purpose. Is that what you have on hand? I wasn't hinting at anything. I was merely observing something about the circuit that DigitalBob posted from the Burr-Brown (now TI) data book. While that circuit is similar to yours, it is not directly relevant. It's resistances are several orders of magnitude higher and the currents correspondingly more sensitive to residual leakage in the diodes.
  • Tracy

    Thanks for the clarification.
    signal diodes like the 1N4148 will be fine for your purpose. Is that what you have on hand?

    I ordered (2) 1N4148 diodes and currently have them at my disposal.


    Novel Solutions - http://www.novelsolutionsonline.com/ - Machinery Design • - • Product Development
    "Necessity is the mother of invention." - Author unknown.

  • Tracy

    While drinking my coffee this morning, I started thinking that maybe this circuitry should have some type of over current protection.

    Of course I am not sure, so I am just asking.

    Let's assume that we are testing across a fuse which is unknowingly blown, with one side of the fuse supplying 12V and the other side of the fuse supplying a device which has short circuited directly to ground, with little to no resistance.

    Would the current circuitry of diagram #2 be able to handle this type of condition or should this circuitry also include a fuse to protect the test leads, electronics, etc...?


    Novel Solutions - http://www.novelsolutionsonline.com/ - Machinery Design • - • Product Development
    "Necessity is the mother of invention." - Author unknown.

  • The diodes serve that purpose, to protect the test circuit in case the full 12V battery voltage appears across the inputs. In the case of the second circuit, using the op-amp with a pair of 1kΩ resistors at the input, the fault current would be about 6mA and the diodes limit the voltage at the op-amp inputs to ~0.6V.
  • Tracy

    Thanks for easing my worry. I realize that was the plan of your circuit in the first place, I just did not know if you were thinking of a dead short.


    Novel Solutions - http://www.novelsolutionsonline.com/ - Machinery Design • - • Product Development
    "Necessity is the mother of invention." - Author unknown.

  • jmgjmg Posts: 12,641
    I'll bump this, as I see Microchip have a new new PAC1932 series

    This is low cost (82c at arrow) and has good precision and error numbers. It also has low bias currents, so is easier to protect with series R's.
    Data shows back to back Diodes+1K and 40V Zeners, as clamp elements.
    Quoted LSBs are 1.5uV on Current (100mV), and 488uV on Voltage (32V)

    With the offsets and limits here, this PAC1932 would also make a nice System sense element for a P2 Eval Board.
    It could read the currents over 10mV droppers to not disturb the supply, and resolve Vdd,Vio to < 0.5mV
  • jmg - Thanks for the bump and heads up on the new part...

    As it stands now, I have a selection of parts on the table, ready for assembly and experimentation, but as always, life gets in the way, and always bigger fish to fry.

    At this point, I must admit that I am a bit overwhelmed with to much to do, in so little time. I need four of me.


    Novel Solutions - http://www.novelsolutionsonline.com/ - Machinery Design • - • Product Development
    "Necessity is the mother of invention." - Author unknown.

Sign In or Register to comment.