Voltage sense with no current draw.

Discussion in 'The Projects Forum' started by tbillion, May 3, 2009.

  1. tbillion

    Thread Starter New Member

    May 3, 2009
    7
    0
    I was wondering if there was a way to measure voltage comming from a circuit without putting any ANY draw on the tested circuit. the reason i am asking:

    I am working on building a car. Called the Micarta Burner. (my-carta) i am basing the major components off of a car that was givent to me and well i scraped it in a weekend. I am left with the motor, the PCM, the transaxle, and the power distrabution system. I am trying to design in some very sensitive diagnostics to this car. so i would like to have a panel that contains some sort of display that will tell me if i have a blown fuse.

    Blown fuse = No voltage.

    But if i use a load to find out if there is a voltage i WILL disrupt the overal tuned ability of the vehicle.

    ANY ideas. I would like to use this not only for blown fuse but to check injection timing and a few other diagnostic purposes.
     
  2. studiot

    AAC Fanatic!

    Nov 9, 2007
    5,005
    513
    The only idea I have is that you haven't told us the whole story.

    Fill in some (lots of) details, like how can you measure ignition timing with a blown fuse indicator and how can a blown fuse indicator upset the vehicle main psu?

    A blown fuse monitor is easy, just put a light in series with the fuse. It will stay on while the fuse is good and go out when the fuse is blown.

    Then we may be able to help more.
     
  3. SgtWookie

    Expert

    Jul 17, 2007
    22,182
    1,728
    Regrettably, this will result in a dimly lit bulb and an underpowered circuit, as the bulb will have a voltage drop across it which depends upon the load and the wattage rating of the bulb.

    A P-ch enhanced MOSFET might work for this. Connect the source terminal to +V, the gate to the load side of the fuse. If the fuse is good, no current will flow through the MOSFET; if it's blown, the MOSFET's gate will be pulled low by the load, enabling current flow out the drain.
     
  4. PRS

    Well-Known Member

    Aug 24, 2008
    989
    35
    Here's an idea I just picked up from a guy who wanted to test for the presence of voltage without actually connecting a test wire. Wrap a number of turns of magnetic wire around the wire you want to check. The current in the wire you're testing will induce a current in the magnetic wire. Use this current with some sort of test device to indicate the presence of voltage or current as per your setup.
     
  5. studiot

    AAC Fanatic!

    Nov 9, 2007
    5,005
    513
    Hello Wookie,

    That rather depends upon the circuit as to which light to recommend, but it can be done.
    That is why I asked for more details.
     
  6. CDRIVE

    Senior Member

    Jul 1, 2008
    2,223
    99
    No matter how you slice it, doing this is quite unorthodox and definitely not SOP or recommended!:eek:
     
  7. studiot

    AAC Fanatic!

    Nov 9, 2007
    5,005
    513
    recommended by whom?

    Instead of criticising, how about posting your own solution so the OP can choose?

    The SOP for an illuminated tell- tale would be to connect it across the fuse so it is shorted by the fuse in normal operation and only illuminates if the fuse blows.

    However this method would violate the OP's apparent basic requirement

    as the lamp would present a load.

    If a simple series lamp would present an unaccptable voltage drop it is a simple matter to have it driven by relay which which only holds in whilst the main current is flowing.

    One problem with any tell-tale light is that the system in not failsafe in that it does not account for a lamp failure.

    I also have to assume, from what little the OP has said, that failure of this fuse does not actually prevent the engine running, merely disables some auxiliary circuit, albeit perhaps an important one.

    Which again brings me back to my request for more details if a more sophisticated approach is needed.
     
    Last edited: May 4, 2009
  8. CDRIVE

    Senior Member

    Jul 1, 2008
    2,223
    99
    OK, how's this? By the way, no offense was intended. There is simply a right and wrong way of doing anything.
     
  9. tbillion

    Thread Starter New Member

    May 3, 2009
    7
    0
    Ok i dont know how better to put this.

    I have a fusebox,(actually i have a UHJB and an IPJB <Underhood Junctionbox> and <Internal Periferial Junctiuon Block> )I have disected them and i am getting ready to integrate them into my newest beast (Picture included) The problem is that i canot disrupt the subcircuits that are powered by these Junction blocks. However i will be installing them in a place that is hard to reach/get to.

    Based on what i know (ohms law,watts law,the theorm of msgnetism) If i splice into the circuit i will create a drain of somesort and change the current requirements of the circuit, either causing some sort of Brownout type scenario or overloading delecate circuits on the machine. It is all part of an advances DIagnostics Module i am building for this car. As for the timing and all of that it would be effected but a change in circuit resistance, trust me i built most of it and it can be and behave very tempermentally. Further more i ould thing that GM totally didnt think of me when the Designed The PCM (powertrain Control module.)

    My ideas so far are of a dotmatrix display (seeing as i have 100 or so fuses to check) some sort of MUX and a processor that will periodically cycle through them all and if one cuircut is dead light up the corresponding matrix.

    I know how to do the display and the processore i guess my problem here is trying to find the right sensor. ya know I->P->O (input Process Output)

    SO to conclude all other subquestions ill give a scenario:

    You assume there is 12V 250ma running through a wire. The circuit that the wire is in has a strict resistance of 20 ohms. How do you findout if there is voltage running through the wire if you cannot modify the draw of the circuit or the resistance (i suppose those are hand in hand)

    As for the mosfet thing that sounds pretty good but wouldnt a Mosfet take some sort of current or something.

    and as for insuceing a voltage, i think it would be hard to sense with only 250ma in the wire not to mention messy for 100 circuits
     
  10. tbillion

    Thread Starter New Member

    May 3, 2009
    7
    0
    P.S. Whaty is SOP???
     
  11. leftyretro

    Active Member

    Nov 25, 2008
    394
    2
    Standard Operating Procedure. In context here, the orthodox or more normal or standard way of doing something.

    Lefty
     
  12. tkng211

    Well-Known Member

    Jan 4, 2008
    65
    2
    The MOSFET just takes a short charging time (in msec range) with the maximum current of approx. 0.012mA (0.005% of 250mA) if the LED of the circuit is energized from another source or through some branch circuit so the current flows through the LED won't affect the data/ reading. I believe it can be negligible.
    If you wanna get some real isolation between the indicator circuit and the current (250mA) carrying wire, you may insert a small coil wound magnetic core(EI or UI type) in the wire circuit, with a hall element IC put inside the gap between the cores as a sensor. The coil can possibly be less than 100 turns and the its DC resistance could be less than 0.3 ohm.
     
    Last edited: May 5, 2009
  13. CDRIVE

    Senior Member

    Jul 1, 2008
    2,223
    99
    One hundred fuses!!! It would have been nice to know that from the get go! :eek:
     
  14. tbillion

    Thread Starter New Member

    May 3, 2009
    7
    0
    lol... yeah ann some cicuits use 2 or 3 at the same time. its really weird. but if it was just like 4 or 5 fuses i totally wouldnt waste my time.

    ok so imma try this whole mosfet thing. to have this question evelve a bit is there a multiple mostfet package? like more than one in a DIP20 or something... do you suppose i could use a mosfet driver?
     
  15. studiot

    AAC Fanatic!

    Nov 9, 2007
    5,005
    513
    Anyone installing a fusebox in a place that is hard to get to is asking for trouble.

    I am having a hard time trying to understand what you might need 100 off 250 milliamp fused connections for, even in a full sized F1 car.

    I also wonder if fusing is your best protection strategy or perhaps do you need a rethink?

    I also doubt that Mosfets would survive an F1 environment.

    As I said, manufacturers have been coiling connecting wire around solenoid plungers for more than a century, and using the magnetism generated to activate a switch. Auto manufacturers have been particularly adept at this, as I'm sure you know if you are some sort of auto technician.
     
  16. tbillion

    Thread Starter New Member

    May 3, 2009
    7
    0

    No offense bro but you seem to come off a bit condescending with the whole some sort of auto technician. No big though.

    Also the 250ma was an example. Some loads are in excess of 30A my point is well i have a few points. I cant move the fuse box due to aerodynamic constraints, and well GM originally designed the wire harness to be only 2 feet long. It is designed like and F1 its end production will be a street car and weekend racer... its technical class is an F2000. as for a fuse panel in an all to hard to get to place well when you consider that a fuse never goes bad and as long as a circuit is sound you never have to replace it then i pose the question as to whether or not there is such a thing as "a bad place for a fuse box"

    either case that is all off topic. Im gonna run with the mosfet thing or other high impedance solution i will post my results here. i think this idea has marketability if you take into consideration that finding a blown fuse is always a pain in the a**
     
  17. tbillion

    Thread Starter New Member

    May 3, 2009
    7
    0
    ok so i tried the schematic from CDRIVE. maybe i dont get it but it wouldn't light the LED in my simulator and always blew the fuse. The object of the game is to have no Load so i decided to figure out what makes for no load. In class tofay we covered High impedance logic probes. I based this off of that.

    the input impedance is 200MΩ so far i have come up with a rather parasitic load of
    .02 microamperes = 2.0 × 10-8 amperes


    pretty cool i say. seeing as i have no clue on truly what i am doing in the big terms. here is the all analog deal. i will post the final here. Maybe i dont get how mosfets work but i tried that one and i didnt get any luck. Maybe you could explain it a bit CDRIVE?
     
  18. CDRIVE

    Senior Member

    Jul 1, 2008
    2,223
    99
    I modeled the circuit with 2N6755's which are N Channel Enhancement FETs. These FETs are power FETs but you only need low power switching FETs. I used these because my dopey spice library listed nothing but power fets.

    If your fuse is blowing then something is very wrong at your end. If you look at my circuit you will see two parts. The one on the left (Circuit Under Test) is your car battery, pwr switch, fuse and load. There is also a switch across the load (SW-SPST-1) which I used to blow the fuse when in Interactive Mode in my spice. None of the components in that box are part of the monitor circuit.

    Circuit Operation:
    Power switch (SW-SPST-2) is closed and fuse is closed.
    T1 gate sees 6V through resistive divider R3 & R2. This makes the T1 conduct and pulls the Drain near ground (0V) potential. This, in turn keeps T2 OFF (not conducting) and LED OFF because it sees no gate voltage.

    When the fuse blows the gate of T1 sees no gate voltage which brings its drain up to 6V through the voltage divider of R4 & R5. This 6V is applied to the gate of T2, which turns on T2 (Cathod of LED pulled to GND) and lights the LED through current limiting resistor R6.

    Your posted circuits:
    The circuits you posted all use BJT's which would not provide the quiescent micro current draw that you desire.
     
  19. CDRIVE

    Senior Member

    Jul 1, 2008
    2,223
    99
    I find this statement humorous. Do some research on a little known invention known as the Proximity Fuse. That's what I call a hostile environment.
     
  20. studiot

    AAC Fanatic!

    Nov 9, 2007
    5,005
    513
    The field is wide open for you to elaborate so we can all share the joke.

    I will watch from the sidelines while you solve the F2000 problem. I'm always happy to learn something.
     
Loading...