razor e300 speed control help please

Discussion in 'The Projects Forum' started by daveyjones97, Nov 14, 2011.

  1. daveyjones97

    Thread Starter Active Member

    Oct 21, 2011
    46
    1
    hi, im new electronics and stuck with this: a while ago i got a razor e300 scooter off ebay. (yes i know better...) long story short it would cut out under load. i "fixed" it by buying another speed controller but really want to understand how motor control circuits work. my multimeter has a diode test voltage of 2.6v which aac tells me isnt enough to switch the fets which as far as i can tell arent shorted open. i measured the resistance across gate source and drain. there was some light smoke residue around one of the fets but i wiped it off but its been in my rainy day bin for 6months now. i cant see any damage through fire or otherwise to any part of the circuit now. it looks like a dual layer board which complicates things.

    what i want to know is can you test a fet reliably when its still in a circuit and what would be a best guess as to why it would cut out at approx half load? its a black heatsink esc and there are only 2 fets, it controls a brushed motor. again i cant see anything on either fet to suggest theyre dud.
    thanks
    ps i have one of those kids spring terminal project electric kits, could i rig this up to test fets somehow?
     
  2. daveyjones97

    Thread Starter Active Member

    Oct 21, 2011
    46
    1
    just been looking at the razor esc with a magnifying glass and noticed a white blob of what looks like pva glue over 5 pins of a 16 pin ic. im struggling to make out the writing on it apart from 494c.
    suggestions please.
     
  3. SgtWookie

    Expert

    Jul 17, 2007
    22,182
    1,728
    A TL494 is a 16-pin PWM controller IC.

    Here is a datasheet for it:
    http://www.ti.com/lit/ds/symlink/tl494.pdf

    You said that you measured the resistance across the gate, source and drain. You should have measured open circuit between those three. If you used the diode function, you may have read a good diode from the source to the drain.

    Without CLEAR images or part numbers, it will be difficult to do anything to help you.
     
    daveyjones97 likes this.
  4. daveyjones97

    Thread Starter Active Member

    Oct 21, 2011
    46
    1
    http://i1096.photobucket.com/albums/g325/daveyjones97/DSCN5619.jpg

    http://i1096.photobucket.com/albums/g325/daveyjones97/DSCN5618.jpg

    http://i1096.photobucket.com/albums/g325/daveyjones97/DSCN5620.jpg


    people have suggested taking the fets out of the circuit to test them, as i only have a little experience of soldering do i need to be careful about iron power and how far up the fets legs i get solder? i dont want to ruin the fets with hamfistedness only to find another component was the problem. i have soldered before but only 12v automotive wiring not circuit boards i do have a low power small iron. also this looks like a dual layer board would i be better off cutting the fets off at the surface rather than trying to un/resolder them through 2 layers. thanks for your patience guys.
     
  5. SgtWookie

    Expert

    Jul 17, 2007
    22,182
    1,728
    daveyjones97 likes this.
  6. daveyjones97

    Thread Starter Active Member

    Oct 21, 2011
    46
    1
    ok , just wondering which components would be favourites to test first and if it would be neccessary to remove them from the board to do so. im not against making up a circuit to test the esc but is that likely to be needed? if so are there any schematics available? i really appreciate your help as i am out of my depth here, even if im finding it fascinating. please be honest if novice fault finding this esc is not gonna happen. thanks
     
  7. wayneh

    Expert

    Sep 9, 2010
    12,154
    3,061
    Do you see how that sounds? Smoke residue is not something to ignore - it's a valuable clue. I'd remove whichever component that was (the rectifier I'll bet?) from the board and test it. It's not hard to remove a component - I use a trigger-type large soldering gun with a tip that has a straight edge, and I lay that hot edge against all 3 pins at once while pulling gently on the part. I use my fingers. With a little practice, you can pop them right out before you get burned. :eek: If you're there long enough to feel the heat, you're going too slow.

    When it comes time to re-insert the part, resist the urge to push. You can rip the trace off the backside if you force the pins through. Better to desolder the holes and let the pins slide in easily.
     
  8. daveyjones97

    Thread Starter Active Member

    Oct 21, 2011
    46
    1
    yes, initially i got as far as wiping away the helpful evidence before deciding repair was futile and tossing it in my rainy day bin. not clever! having now decided i cant kill a dead esc so i might as well try and learn something im only being held back by knowledge of (de)construction circuit board methods, hence rummaging round the web begging shamelessly for advice. gonna read as much as i can and as soon as i get a chance il unsolder the fet and schottky diode (next thing i have to research testing of) and go from there. as ever a sincere thanks to everyone helping me.
     
  9. SgtWookie

    Expert

    Jul 17, 2007
    22,182
    1,728
    When you go to remove the MOSFET, it would be a good idea to use a hemostat with long jaws in order to:
    1) Keep the gate, source and drain electrically connected to help prevent damage from static electricity
    2) Help keep the MOSFET itself from getting too hot by removing the heat during soldering.

    Hemostat:

    [​IMG]

    They come in a wide variety of sizes. They are very handy when fiddling with electronic bits. If you overheat a semiconductor device, it will be destroyed.

    MOSFETs are very sensitive to static electricity. If the voltage between the gate and source terminal exceeds ±20v, it will be destroyed.

    In contrast, you can't feel a discharge from your fingers until several hundred volts of static builds up first.
     
    Last edited: Nov 16, 2011
  10. wayneh

    Expert

    Sep 9, 2010
    12,154
    3,061
    I highly agree; I think the hemostat is the most valuable tool lacking from most toolboxes. Should be standard, along with one of those telescoping magnetic pickup tools.
     
  11. daveyjones97

    Thread Starter Active Member

    Oct 21, 2011
    46
    1
    thanks guys, i assumed over heating any small component would kill it, id only seen those hemostats in the medical field before but im going to buy a cheap one from maplin asap. do i need to worry about the middle layer of this board connecting properly when i reassemble it? its a dual layer board. also no luck yet finding how to test the schottky diode, would it be the same test as for the mosfet? i have had a look for specific tests or information today but drawn a blank except for http://www.allaboutcircuits.com/vol_3/chpt_3/4.html and wikipedia which i havent had time to read properly yet. thanks
     
  12. daveyjones97

    Thread Starter Active Member

    Oct 21, 2011
    46
    1
    ok, so i seem to have a good schottky diode removed from the board. i realise this can only end badly but im doing it as a learning excercise as much as anything now.
    as yet nobody wants to make an educated guess at what might have caused this controlled to cut out at approx half load (amps, not speed). ive tested the diode and it appears to be good, http://forum.allaboutcircuits.com/showthread.php?t=62230 from what ive read it could be a dodgy fet, or a wild stab in the dark, could the mysterious blob of white stuff be creating resistance on the pins of the main controller ic and messing up the input signal? i have no idea what it is but its always been there, but should it be? in the picture the blob is on the right of the circuit board on the left of the ic that has a yellow tag to the right of it, just under the 3rd wire from the right at the top of the board: http://i1096.photobucket.com/albums/g325/daveyjones97/DSCN5620.jpg any ideas welcome.
     
  13. SgtWookie

    Expert

    Jul 17, 2007
    22,182
    1,728
    The white blob is just some adhesive.

    Closest to the foreground, there are four black rectangles with two leads on their left sides, and one on the right side; those are transistors. Was there any smoke around them?
    The cylindrical things with the orange body and black ring on one end are diodes.
    The one cylinder that is red with two other color stripes (green and blue, perhaps?) is a Zener diode.
    all of those tiny black rectangles with silvery ends are resistors.
    The rectangles that are taller and tan or brown in the middle are ceramic capacitors.


    Where was the smoke residue on this board?
     
  14. daveyjones97

    Thread Starter Active Member

    Oct 21, 2011
    46
    1
    no obvious reside, ive finally found my loupe (magnifying thing) and will examine it thorougly tonight. the obvious original sooty smoke reside i wiped off 6 months ago as above but im sure it was localised to either the schottky or the mosfet and the tab on the inside of the heatsink from the alu case that the esc was housed in. the heatsink got weighed in around 6 months ago with some other scrap i had collected, yes i need to help myself here! what i forgot to mention was the previous owner had a bbq and all his mates had a fun time racing about on it in the height of summer consuming a whole charge in one go. a month or so after that his kids got bored of it and it was left without charge for 6 months, hence when they went to use it wouldnt work well, and me thinking spares repairs meant sulphated batteries. unfortunately i bought it new batteries too which made no difference. (crying into wallet)
     
  15. daveyjones97

    Thread Starter Active Member

    Oct 21, 2011
    46
    1
    alsp this wouldnt be the first thing ive owned that had a bizarre fault which makes me wonder if the blob of glue might be causing resistance on the controller ic pins fooling that into cutting power, but i must point out im new to the world of circuitboards and components so that might be wildly wrong. thanks as always
     
  16. daveyjones97

    Thread Starter Active Member

    Oct 21, 2011
    46
    1
    still at work, but just flicking through the pics i linked to and noticed the top surface of the power supply(?) big black cube thing looks unusual. could that have lightly smoked internally, dusting the fet and schottky in soot and giving the top of the case a melted look? the writing on it is:TIANBO coil: 24vdc NO:40A 14vdc TRKP-L-S-H1. the case wont come off, i tried out of curiosity the other day but depending on opinions here i think i might have a look. sealed for life? il be the judge of that!
     
  17. daveyjones97

    Thread Starter Active Member

    Oct 21, 2011
    46
    1
    for the benefit of anyone finding this thread, as far as i can tell the fet, diode and the relay (black box) are all fine with no evidence of smoke! i opened the relay box. any further fault finding would be galaxies away from my abilities, thanks to sgt wookie for help and patience, but i dont think we're (me?) gonna get to the bottom of this one.
     
Loading...