Connect with us

Motorbike KERS problem

Discussion in 'Sensors and Actuators' started by Tom Niebel, Nov 13, 2015.

Scroll to continue with content
  1. Tom Niebel

    Tom Niebel

    7
    0
    Nov 13, 2015
    Hi all

    First post here, i'm studying for my Motorsport engineering degree which is effectively just an applied mechanical course so my electronics knowledge is there but not very deep :eek:

    We have a Project to implement the regenerative aspect of KERS on a r6 motorbike. We have a small 2.2kw outrunner motor as a test motor. The aim to rectify and regualate this before charging capacitors.

    When you short out 2 or 3 of the leads from the motor, the motor becomes a brake and you can feel this, however when i built a bridge rectifier with 6 schottkey diodes the motor no long brakes when the two 'dc' terminals from the rectifier are shorted. I understand that the torque in the motor is proportional to current which matches up with the motor locking with a dead short. I was expecting the same behaviour by shorting the wires after the regulator.

    I have 60A 600v diodes which I suspect may be the problem as they will have some leakage and 1-5% of 600V is alot for a DC system.

    Hooking the output of the regulator up we get 5V which is expected from the 200kv motor hooked up to a cordless drill at 1500rpm - ish however the measly 1A is causing no noticeable braking on the motor

    Any help would be greatly appreciated

    Thanks

    Tom Niebel
     
  2. Tom Niebel

    Tom Niebel

    7
    0
    Nov 13, 2015
    Here is a photo of my set up and rectifier.
    The rectifier is connected to a power meter then after that a small 12V dremel as a load
    [​IMG]
     
  3. duke37

    duke37

    5,343
    753
    Jan 9, 2011
    "Hooking the output of the regulator up we get 5V which is expected from the 200kv motor"
    200kV - that is some motor:)

    Any leakage in the diodes will add, not detract from the braking effect. The diode voltage drop will be significant at 5V input.
    Have you got the diode bridge connected correctly?
     
  4. Tom Niebel

    Tom Niebel

    7
    0
    Nov 13, 2015
    The kV in this instance isnt Kilo Volts but represents the voltage increase in relation to the rotational velocity. the rpm per volt, so at 1000 rpm we are achieving 1000rpm / (200rpm/volt) = 5V
    Sorry if you knew this but the units of kV confused me at first so best to clear it up.

    hmm ok, I plan on making a system with smaller diodes and will double check the circuit with someone else. I have checked the circuit with peers however as it is soldered i cannot do a diode check to see if any are faulty.

    Am i right in thinking that we should get full braking torque when the two rectifier wires are shorted ?

    Thanks duke37!
     
  5. CDRIVE

    CDRIVE Hauling 10' pipe on a Trek Shift3

    4,960
    648
    May 8, 2012
    I sure wish I knew that over the past 50 + years. If I did I wouldn't have wasted my time wringing out all those suspect bridges. :)

    Chris
     
  6. duke37

    duke37

    5,343
    753
    Jan 9, 2011
    What you want is very low diode voltage drop, smaller diodes will have a higher voltage drop so will not help.
    You could get the data for the diodes you have and work out the speed needed to overcome the voltage drop. Below this speed there will be negligible braking. Remember that the current has to go through two diodes in series with perhaps 2V drop.

    Check the circuit again. There should be three anodes connected to one output and three cathodes connected to the other output. Each phase should connect to one anode and one cathode. A neater layout would make it easier to see.:)
     
  7. Tom Niebel

    Tom Niebel

    7
    0
    Nov 13, 2015
    ok, so I should be ok with the diodes then. Maybe i need to desolder and rewire the rectifier then, i planned on a different heat sink design which is why they now look messy. I have ran this through a regulator/rectifier unit from a motorcycle aswell and achieved simular results however slightly worse. We have a dyno in which will be able to spin the motor to 5K so that is the next step for us to try to see if the low rpm is the issue.
     
  8. CDRIVE

    CDRIVE Hauling 10' pipe on a Trek Shift3

    4,960
    648
    May 8, 2012
    Eureka! It just dawned on me what the problem is. The regenerative EMF from the motor is the same polarity as the source. Looking at it from the motor's point of view, the (+) motor terminal is looking into the Cathodes of the bridge. Therefore braking current will be blocked when shorting the AC input terminals of the bridge.

    In the event that I'm misinterpreting you please post a schematic.

    Chris

    Edit: A schematic is in order regardless of my interpretation. ;)
     
    Last edited: Nov 15, 2015
  9. Tom Niebel

    Tom Niebel

    7
    0
    Nov 13, 2015
    Hi Chris

    This is the diagram I have used (not mine)
    [​IMG]

    I understand the basics of EMF and inductance however thats where I start to leave my comfort zone.

    If I have the terminals wired up backwards on the grouped anodes and cathodes, would this not just reverse the polarity of the system? (is this what you mean?)

    Thanks
     
  10. duke37

    duke37

    5,343
    753
    Jan 9, 2011
    The bridge diagram as shown is correct but when the diodes face each other on the heat sink, the anodes will be on the right on one row and on the left on the other row. The amount of linkage wire should be minimal.
    Have you got this right?
    The positive output will be connected to three cathodes and the negative output will be connected to three anodes. There is no need for several crossing wires.
     
  11. Tom Niebel

    Tom Niebel

    7
    0
    Nov 13, 2015
    Sorry for the huggee photo, the two sides are symmetrical (apart from reversing the anode and cathode) and this is the clearest photo I have.
    [​IMG]
    The reason for the bad layout is they were intended to be placed on a block like this photo below however they were getting warm so decided to fold them out on some sheet to increase the surface are and add thermal paste on the back. The photo below shows a taped up version of the initial idea.
    [​IMG]
     
  12. CDRIVE

    CDRIVE Hauling 10' pipe on a Trek Shift3

    4,960
    648
    May 8, 2012
    Somehow I missed that this is a 3PH source and bridge. This explains the 6 diodes that had me scratching my head. None the less it doesn't change my comments in my last post.

    Please mark on the schematic the points you're shorting when attempting to brake the motor.

    Chris
     
  13. CDRIVE

    CDRIVE Hauling 10' pipe on a Trek Shift3

    4,960
    648
    May 8, 2012
    I went back and re-read this thread from post #1 hoping to glean why you're demonstrating Inertia Generated EMF with a hand drill.

    Why aren't you demonstrating this more similar to it's end use? In other words I would be demonstrating Generator/Braking effect by first fitting the motor with a substantial Flywheel. I would then power the motor from your Source/Bridge through a SPDT switch with the common connected to motor (+), pole1 to Bridge(+) and pole2 to Bridge(-). Pole1 position runs the motor while Pole2 position shorts and brakes the motor.

    Heck you could use a SPDT(CO) switch to further demonstrate no braking switch position.

    Chris
     
  14. Tom Niebel

    Tom Niebel

    7
    0
    Nov 13, 2015
    That was the initial plan to use a car fly wheel and this way lock-ups could be tested to effectively have a anti-lock regenerative braking system however as we are students the uni decided that a flywheel doing 6000rpm wasn't safe :(
    The plint dyno in the background of the photo is able to spin us up to 6k in a secure enclosure so we are going to use that to drive our motor on friday hopefully!

    The project is mainly based on the mechanical installation and design process involved which they will use to grade us, and they have said we only need to recover the energy not to deliver it and the delivery will be a separate topic for another group, else we could have used the mosfet quadrant driven option to do both aspects. We may still resort to that method however it involves a control system that wouldnt be needed for the method tried above.
     
Ask a Question
Want to reply to this thread or ask your own question?
You'll need to choose a username for the site, which only take a couple of moments (here). After that, you can post your question and our members will help you out.
Electronics Point Logo
Continue to site
Quote of the day

-