Shop OBEX P1 Docs P2 Docs Learn Events
my gnd has 2v's on it — Parallax Forums

my gnd has 2v's on it

whiz100whiz100 Posts: 13
edited 2010-04-07 03:10 in Robotics
i was trying to workout why my circuit wasn't working correctly so I put a voltmeter across some of the components vcc and gnd's and noticed its not reading the correct voltages. after a bit more measuring I've found that the ground has 2v's on it. this i think is affecting the components functioning working.
i have several different components like a pic microcontroller on 3v, a ftdi module dual usb running off of 6v (the module does it's own power regulation), a webcam connected to the ftdi using the ftdi's ldo of 3v, a servo using 6v and a laser also on the 6v. each of the components run to a common ground.
I used to have the 6v regulator and another 3v ldo on the same board but because I built it wrong ( I didn't know I needed to put caps on the regulators) I've removed the regulators from the circuit board and built a separator board with the correct regulator caps. this second board now has the 6v, 3v and a ground wire connecting to the original board. if I measure the voltage outputs from the power board everything looks good. if I measure the ground from the power board to the ground of the original circuit there's the 2v. so I'm thinking the unused voltage from one (or perhaps all the components) is sitting on the ground and not disapearing through the ground quick enough (don't know if this sounds studip or not), therefore it's traveling backup the other components ground wires and causing the fault.
questions are:
1) is that even possible?
2) what should I do about it?
3) is that why I've seen some circuits designed that use resistors,capacitors or diodes between the component and ground?

thanks
Peter

ps. I've obviously new at this.

Comments

  • FranklinFranklin Posts: 4,747
    edited 2010-03-30 02:17
    1) Yes
    2) FIX it.
    3) Yes
    You are saying that between two ends of the same wire you measure 2v? Either the meter is bad or the wire is broken or too small for the job. What kind of current are you talking about?

    ▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
    - Stephen
  • ercoerco Posts: 20,256
    edited 2010-03-30 03:10
    Peter: You have what is technically known as a clusterfudge. Too much stuff all hooked up together incorrectly at once. Obviously something is wrong and items aren't working. Since you are new at this, there's no way for us to know where the wiring error is. Your best bet is to take everything apart now (hopefully before anything gets damaged) and hook up one thing at a time, carefully checking wiring, testing voltages and function at each step.

    ▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
    ·"If you build it, they will come."
  • whiz100whiz100 Posts: 13
    edited 2010-03-30 11:56
    Franklin

    the total current for the circuit is less than 150mA. I don't know the actual number because I don't have that information available for all of the compoent.
  • whiz100whiz100 Posts: 13
    edited 2010-03-30 12:11
    I'm going to redesign the board layout anyway because of the new power board. I must have done something to cause the fault when I removed the regulators and subsequently retro fitted the new board, before I did that it was working except when I had the laser and servo turned on at the same time. Hence the reason why I removed the regulators and added the support caps.
    I'm thinking I should build multiple boards to separate the logical functions that would make it more easy for me to test and fix. does that sound like a good idea?
    If I post the eagle files on this website would someone take a look and review them before I build them?

    thanks
    Peter
  • IRobot2IRobot2 Posts: 164
    edited 2010-03-30 12:58
    First off, any thing extra you can give us to help you figure out what your problem is would really improve your odds. Do you have a schematic at all? A good schematic is worth the time and effort for any project (let alone for what you have going on) and would help us - help you. I do not use eagle cad or I would take a look at that.

    This is just a complete shot in the dark but I bet your problem is somewhere around "...I removed the regulators and added the support caps." . Since we really don't know exactly what you are doing, I cant say... but that just seems like a good place to start looking for problems. Post some info on what you have (schematic, board layout, etc) and I am sure some one can help you out.

    ▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
    Alex Burke
    "It is not how smart you are rather, it is how you are smart." -Jon Campbell
  • whiz100whiz100 Posts: 13
    edited 2010-03-30 13:32
    attached I hope are the schematic and board.
    these are the original before I started to modify the board.
    so the current board has the 2 regulators on a separate board.
    I also remove the cap. next to the pic because it was causing problems.
    1613 x 1094 - 45K
    522 x 432 - 35K
  • John R.John R. Posts: 1,376
    edited 2010-03-30 15:39
    Your schematic is not readable, but based on the board image, you have "ground" coming in, going to one 4 pin header, the voltage regulator, and a one other device (possibly a transistor or voltage regulator?) That's it.

    I belive you want "ground" to be VSS for all components.

    With nothing soldered to the board, you should be able to put one probe of an Ohmmeter or continuity tester on your ground connection, and probe each and every "VSS" or "Ground" connection and see virtually zero Ohms (or hear a beep).

    If this is not the case, than "something" is not good in whoville...

    ▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
    John R.
    Click here to see my Nomad Build Log

    Post Edited (John R.) : 3/30/2010 3:45:14 PM GMT
  • whiz100whiz100 Posts: 13
    edited 2010-03-30 15:50
    the ground goes to
    the big 6v regulator, to the right that looks like a transistor is a 3.3v ldo
    the icsp header
    the pic via a cap 0.1uF (cap removed from actual board) - on the other side of the pic is the gnd that goes to the headers for the laser, 10kpot and the parallax continuous rotation server
    the cn3 (which is an ftdi ft2232h mini module - follows on to the cn2
    the 4 pin header is connected to a webcam (from sparkfun - SEN-09334 JPEG Color Camera - UART Interface)
    there is a transistor (mps222a) which turns the laser on and off

    the pic sends the pwm to the servo, turns the laser on and off, reads the adc from the pot. the pic also connects rx1/tx1 to the ftdi
    the ftdi connects to the webcam on rx2/tx2
    the 1k resistor is for the pic icsp

    i really need to put a resistor on the laser somewhere probably a 30ohms to drop the 6v down to 3.2

    but yep that's it.
  • whiz100whiz100 Posts: 13
    edited 2010-03-30 15:52
    i've attached the eagle files, because the images didn't come over so well.
  • John R.John R. Posts: 1,376
    edited 2010-03-30 16:12
    I don't have access to Eagle, but the board image you posted does not support your verbal description.

    Also, ground (or any other connection) cannot go "through" a cpacitor. A capacator is connected in parallel between two points to "stabilize" voltage (not technically correct, but may help you understand in this case).

    Do you have a "bare board"? Have you tried probing it as described?

    ▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
    John R.
    Click here to see my Nomad Build Log
  • whiz100whiz100 Posts: 13
    edited 2010-03-30 16:19
    which part of the description doesn't match the image? "t the board image you posted does not support your verbal description."
    that could where i'm going wrong, i think i does something when it doesn't.
    i was using the 0.1uF cap to decouple. what actually happen was the cap sucked out all the current and didn't have enough power for the pic. so I removed it.
  • John R.John R. Posts: 1,376
    edited 2010-03-30 16:30
    I have to retract my comment, I misread one of the board traces as a "schematic style" no connection crossing instead of a "board trace". Your ground looks OK from what I can tell, but I don't know the pinouts of the devices, and am not in a position to delve in much deeper.

    If the capacitor drained all your avaialble current, I'd suggest "something" is not right.

    I might go back to probing a bare board, and make sure you have what you think you have...

    ▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
    John R.
    Click here to see my Nomad Build Log
  • whiz100whiz100 Posts: 13
    edited 2010-03-30 16:38
    ok, thanks,

    I'll to build out the separate boards as I said early. that will help me find the problem. think there's also the possibility I fried something which is now causing the problem.
    the uC being the obvious choice because I've removed everything else except the usb module.
  • Chris SavageChris Savage Parallax Engineering Posts: 14,406
    edited 2010-03-30 22:07
    erco said...(trimmed)
    Peter: You have what is technically known as a clusterfudge.
    Erco...nice one...I have to remember that for my own site.

    ▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
    Chris Savage

    Parallax Engineering
    ·
  • whiz100whiz100 Posts: 13
    edited 2010-04-07 01:49
    I've removed the power supply components and made that into a separate board, I've also removed the webcam and the user module into another separate board.
    I've check they are working (so pleased they still do). but before I make the next board I wanted to check with you brilliant people
    1) should I put a diode across the servo (parallax continuous servo)?
    2) does it matter what side I put the laser connected to the transistor (mps2222a general purpose) ie. on the collector side or on the emitter side?
    3) I don't have a resistor on the base of the transistor does it matter?

    on the image the following are:
    VDD = 3.3v
    V+ = 6v


    thanks
    P
    2170 x 1207 - 35K
  • James NewmanJames Newman Posts: 133
    edited 2010-04-07 03:10
    Diode across one of those types of servos shouldn't be necessary, since they have alot of internal circuitry for ease of use.

    Yes, since your transistor is a npn type, you need a positive voltage on your transistors base to the emitter. If your load is on the emitter side, then your voltage drop is below after the emitter. This means you would need a higher voltage than what is powering the laser to turn on your transistor. If the load is on the collector side, then your voltage at the emitter will be low (I assume vss) meaning anything higher than that on base should be easy.

    Haven't looked at your transistor, but (non mos type) transistors are actually current controlled devices. This means that the current from base to emitter scaled by the transistor gain is the current through the transistor. You need to make sure you don't pull too much current from your uC, and a resistor is a good way of ensuring this.
Sign In or Register to comment.