I need help to make the TOP227 run!

Discussion in 'The Projects Forum' started by Lindembruck, Feb 28, 2016.

  1. Lindembruck

    Thread Starter Member

    Feb 28, 2016
    32
    1
    I need a help to make the TOP227 get run!
    I bought five units and I have the Transformer ready including circuit.

    Problem:
    TOP227 only flash to start but do not get run with power.
    All five what I bought, don't get the 100 KHz, only around 7 KHz as show

    Power input, 310 V

    Pictures Oscilloscope Print Screen:
    Short pulses, transformer pins
    FFT, Transformer pins
    Long wave, control pin to ground
    PCB protoboard down side
    PCB protoboard up side
    Schematic from datasheet annexed page 7

    Keys: TOP221 TOP222 TOP223 TOP224 TOP225 TOP226 TOP227
     
  2. DickCappels

    Moderator

    Aug 21, 2008
    2,656
    632
    Are you using an isolation transformer so that you are not at risk accidentally providing a current path from the circuit to ground?

    There are few parts so these things usually work right away if all the parts are good and properly assembled. It is usually a good idea to bring the AC voltage up slowly so that you can troubleshoot the circuit with a minimum of voltage -that could give you time to react to a problem fast enough to prevent damage.

    First, did you buy the parts from a reliable distributor?

    Then check a few other things:

    Is the voltage on the control supply (lower secondary circuit) enough for the feedback to work? How much current is going through the LED. What is the output voltage?

    Check your transformer to see whether it was designed for 100 kHz operation at the currents and voltages you expect to see in the circuit.

    It might be useful for you to post a schematic with voltages measured around the circuit.
     
    Lindembruck likes this.
  3. Lindembruck

    Thread Starter Member

    Feb 28, 2016
    32
    1
    Hi!
    Thanks to reply me!
    I will check all your list and give you each one feedback.
    For now, my insulated transformer picture made by me with four insulated levels more protections.
     
  4. Alec_t

    AAC Fanatic!

    Sep 17, 2013
    5,801
    1,105
    Don't see a transformer??
     
  5. Lindembruck

    Thread Starter Member

    Feb 28, 2016
    32
    1
    HI!
    The black box from a old Musteck UPS. The insulated system is inside it.
    Thanks
     
  6. RichardO

    Well-Known Member

    May 4, 2013
    1,235
    384
    The transformer on your prototype board looks like one designed for power line frequencies such as 50 or 60 Hz. This switcher runs at more like 100 KHz. You need a high frequency transformer.

    I saw these "3-pin wonders" many years ago. I was not directly involved but, as I remember, they were tricky to use. I think the transformer was a pot core with a specific magnetic gap. Also, I dont see any ground plane on your boa

    Read and re-read this information from page 8 of the data sheet:

    Key Application Considerations
    General Guidelines
    • Keep the SOURCE pin length very short. Use a Kelvin
    connection to the SOURCE pin for the CONTROL pin
    bypass capacitor. Use single point grounding techniques at
    the SOURCE pin as shown in Figure 9.
    • Minimize peak voltage and ringing on the DRAIN voltage
    at turn-off. Use a Zener or TVS Zener diode to clamp the
    drain voltage below the breakdown voltage rating of
    TOPSwitch under all conditions, including start-up and
    overload. The maximum recommended clamp Zener
    voltage for the TOP2XX series is 200 V and the
    corresponding maximum reflected output voltage on the
    primary is 135 V. Please see Step 4: AN-16 in the 1996-97
    Data Book and Design Guide or on our Web site.
    • The transformer should be designed such that the rate of
    change of drain current due to transformer saturation is
    within the absolute maximum specification (ΔID in 100 ns
    before turn off as shown in Figure 13). As a guideline, for
    most common transformer cores, this can be achieved by
    maintaining the Peak Flux Density (at maximum ILIMIT
    current) below 4200 Gauss (420 mT). The transformer
    spreadsheets Rev. 2.1 (or later) for continuous and Rev.1.0
    (or later) for discontinuous conduction mode provide the
    necessary information.
    • Do not plug TOPSwitch into a “hot” IC socket during test.
    External CONTROL pin capacitance may be charged to
    excessive voltage and cause TOPSwitch damage.
    • While performing TOPSwitch device tests, do not exceed
    maximum CONTROL pin voltage of 9 V or maximum
    CONTROL pin current of 100 mA.
    • Under some conditions, externally provided bias or supply
    current driven into the CONTROL pin can hold the
    TOPSwitch in one of the 8 auto-restart cycles indefinitely
    and prevent starting. To avoid this problem when doing
    bench evaluations, it is recommended that the VC power
    supply be turned on before the DRAIN voltage is applied.
    TOPSwitch can also be reset by shorting the CONTROL
    pin to the SOURCE pin momentarily.
    • CONTROL pin currents during auto-restart operation are
    much lower at low input voltages (< 36 V) which increases
    the auto-restart cycle time (see the IC vs. DRAIN Voltage
    Characteristic curve).
    • Short interruptions of AC power may cause TOPSwitch to
    enter the 8-count auto-restart cycle before starting again.
    This is because the input energy storage capacitors are not
    completely discharged and the CONTROL pin capacitance
    has not discharged below the internal power-up reset
    voltage.
    • In some cases, minimum loading may be necessary to keep
    a lightly loaded or unloaded output voltage within the
    desired range due to the minimum ON-time.
    ReplacingTOPSwitch with TOPSwitch-II
    There is no external latching shutdown function in
    TOPSwitch-II. Otherwise, the functionality of the
    TOPSwitch-II devices is same as that of the TOPSwitch family.
    However, before considering TOPSwitch-II as a 'drop in'
    replacement in an existing TOPSwitch design, the design
    should be verified as described below.
    The new TOPSwitch-II family offers more power capability
    than the original TOPSwitch family for the same MOSFET
    RDS(ON). Therefore, the original TOPSwitch design must be
    reviewed to make sure that the selected TOPSwitch-II
    replacement device and other primary components are not over
    stressed under abnormal conditions.
    The following verification steps are recommended:
    • Check the transformer design to make sure that it meets the
    ΔID specification as outlined in the General Guidelines
    section above.
    • Thermal: Higher power capability of the TOPSwitch-II
    would in many instances allow use of a smaller MOSFET
    device (higher RDS(ON)) for reduced cost. This may affect
    TOPSwitch power dissipation and power supply efficiency.
    Therefore thermal performance of the power supply must
    be verified with the selected TOPSwitch-II device.
    • Clamp Voltage: Reflected and Clamp voltages should be
    verified not to exceed recommended maximums for the
    TOP2XX Series: 135 V Reflected/200 V Clamp. Please
    see Step 4: AN-16 in the Data Book and Design Guide and
    readme.txt file attached to the transformer design
    spreadsheets.
    • Agency Approval: Migrating to TOPSwitch-II may require
    agency re-approval.
     
  7. Lindembruck

    Thread Starter Member

    Feb 28, 2016
    32
    1
    I make the transformer with a ferrite core EE28 for 110 KHz including wire diameter for the same frequency.
     
    RichardO likes this.
  8. Lindembruck

    Thread Starter Member

    Feb 28, 2016
    32
    1
     
  9. RichardO

    Well-Known Member

    May 4, 2013
    1,235
    384
    Ah. That view makes all the difference. :)

    Now, about your grounding... It seems that you have several small wires running about that might not be right.

    Key Application Considerations
    General Guidelines
    • Keep the SOURCE pin length very short. Use a Kelvin
    connection to the SOURCE pin for the CONTROL pin
    bypass capacitor. Use single point grounding techniques at
    the SOURCE pin as shown in Figure 9.
     
  10. Lindembruck

    Thread Starter Member

    Feb 28, 2016
    32
    1
    Thee wires are for the oscilloscope and they are connected in:

    • ground input
    • capacitor 47 uF control +
    • transformer pin drive control
    • ground output
    • board to board yellow, signal feedback to the insulated optocoupler led side
    It have a 1.5KE200CA - 1500 Watt Mosorb Zener Transient Voltage Suppressors 200V
     
  11. Lindembruck

    Thread Starter Member

    Feb 28, 2016
    32
    1
    HI!

    In the datasheet, (you can download it in the first post here), page 15, we have a test circuit.
    I make it and to control it, a LM1117 Adj with a trimpot in the place of the variable battery. Picture annexed.
    The chip got run at 100 KHz in a especial condition but after understand it, is the same in the datasheet information:
    • it start when the control get supply around 5.36 volts in a pulse by a current through a resistor, not in a low speed up.
    • If the control pin down to under around 5.36 volts, the oscillator get down to around 4.7 KHz, but not a true 4.7 KHz but 100 KHz in pulses with interval sequences when the control pin get a triangular wave around the 4.5 to 4.8 V.
    • if the level of the control pin get up then around 5.36 V in short time, the output drive control get full 100 KHz drive.
    • if the current in the control arise, the drive wave change to negative pulses instead the around 30% large negative pulses.
    Now, I will get a good look in the main circuit but before, I will get a night rest.

    Thanks for today.
     
  12. Lindembruck

    Thread Starter Member

    Feb 28, 2016
    32
    1
    Hi!

    I did the tests as you can see in the test protoboard.
    The transformer was made by me to work with 110 KHz in especial to this chip included wire diameter for the same frequency.
    Tomorrow, I will get a open eye look in the circuit and will get the answers and write here as you ask.
    Please, take a look in posted pictures and storyline.

    Thanks to your approach.
     
  13. Lindembruck

    Thread Starter Member

    Feb 28, 2016
    32
    1
    Hi all!

    Now, the imput looks some desired result but no DC output.
    The feedback looks like the output not rectified because don't know for now but in way to solve.

    • Red channel, transformer pin driven by TOP227
    • Yellow channel, capacitor 47 uF + pin, before resistor to the TOP227 pin control
    • 1.5KE200CA - 1500 W Mosorb Zener Transient Voltage Suppressors 200V >>>> GETTING HOT
    • TOP227 get a little warme
     
  14. Lindembruck

    Thread Starter Member

    Feb 28, 2016
    32
    1
    This is:

    Channels with 5 seconds Persist active

    Channel red in AC mode connected to the Transformer imput by the TOP227
    Channel yellow DC mode connected to the end Output

    Pink lines:
    First one up in the mark of 312 V
    Second down in the mark of 5.20V

    When TOP227 pulse comes, the output get a shake and transformer get a speed oscillation with around 600KHz

    First image trigger over 100KHz TOP227 drive pulse
    Second image trigger over 600KHz center oscillation

    Next problem now, is to get no ripple result in the final output.
     
  15. DickCappels

    Moderator

    Aug 21, 2008
    2,656
    632
    It looks like the TOP227 is only on for a very short period of time. That means that there is a good chance that it is driving a short circuit.

    You might discover something interesting if you check all of the diodes and capacitors -one could be shorted (use and ohmmeter) or more likely be backwards. Similarly, you might have the phase on one or more of the transformer windings reversed.

    While talking about the transformer, did you verify the inductance of the primary? A shorted turn, the wrong ferrite, or too large of a gap could be the problem. Come to think of it, too little of a gap could (but is not likely to) be the problem.
     
  16. Lindembruck

    Thread Starter Member

    Feb 28, 2016
    32
    1

    --------------------------

    Hi!

    Looks like have a "ghost" in the circuit!

    If I connect the prob in the circuit ground, the noise is the same!
    Look both picture from the same measurement!

    About the bobines can have crossing connections, only if the control bobine but the polarity is right. It do the positive output to up over the input ground.
    The output bobine is a double bobine with full wave rectifier and the signal go to the optocoupler diode over its own ground.

    The oscilloscope is connected to the new one and ground specially made exclusively to this work bank!

    Thanks
     
  17. Lindembruck

    Thread Starter Member

    Feb 28, 2016
    32
    1
    ----------------------------------

    I discovered a thing.

    The noise in the output is up only if the red prob is connected to the transformer input without this red prob ground cable connected!
    With the red prob unconnected, the noise looks like this pictures with now with two scales.

    Thanks
     
  18. DickCappels

    Moderator

    Aug 21, 2008
    2,656
    632
    The noise on the scope is normal and does not tell us much.

    Do you have a short or backwards polarized component in the circuit?

    If you measure the primary inductance is it what you expected to see?
     
  19. Lindembruck

    Thread Starter Member

    Feb 28, 2016
    32
    1
    -----------------------
    Mister, thanks!

    I have the same schematic as in the picture in the first post here but with:
    • In the place of the P6KE200 I use a 1.5KE200CA now with the BYV26C where when I start here, I has removed because the caracteristics of the 1.5KE200CA.
    • In the place of the zener diode, I use a TL431
    • The optocoupler is a EL357NC
    All others in the circuit is the same like there.

    I do not have the L measurer device. It is in the way... The same as in the picture annexed here.
    I only calculate the transformer and take a look the wire frequency table.
    It is a EE29 with 29 x 14 x 11.5 and I cant find, for now, the paper with the data about windings etc. The five volts output have 6 or 7 turnes.
    All wires diameter for the minimal 110 KHz.

    Maybe, the transformer core is less then the TOP227 power drive and it got resonance with a verry good ferrite.

    Thanks for now
     
    Last edited: Mar 1, 2016
  20. Lindembruck

    Thread Starter Member

    Feb 28, 2016
    32
    1
    Next step:
    • Synchronise the transformer internal frequency with the TOP227 shoot to both be in same fase
    • Calibrate the TL431 to get the 12V projected output
    • Develop a filter to clear more possible the noise.
    Thanks all for now!
     
Loading...