Pulse Generator

Discussion in 'General Electronics Chat' started by warp, Nov 12, 2014.

  1. warp

    Thread Starter New Member

    Jan 20, 2014
    18
    0
    Hi,
    I am trying to build a dc pulse generator with a 50% duty cycle but I would settle for a bit less.
    It needs to be about 3MHz or more, but Im not too fussy about the wave form right now.
    It doesn't have to be too square. I dont mind if it comes out triangular even.

    I am looking at a LMC555CN which the data sheet says capable of three MHz astable, a TC4426A, fet driver, and an IRF540N fet.

    The fet and its driver are little bit hard for me to get given my personal situation right now, so I would like to be pretty sure that Im ordering the right thing.

    My question is, can these three components be used together to achieve my desired outcome.

    Thanks
    Regards
    Wil.

    .

    I forgot to mention that I will be switching 12 volts.

    Thanls
     
    Last edited: Nov 12, 2014
  2. bertus

    Administrator

    Apr 5, 2008
    15,645
    2,344
    Hello,

    Why do you need such a high frequency for the pulse generator?
    The gate capacitance of the fet might give you troubles at that speed.

    Bertus
     
  3. warp

    Thread Starter New Member

    Jan 20, 2014
    18
    0
    Hi Bertus,

    That was one of my concerns.
    It is for an experiment. I want to pulse a transformer and induct a standing quarter or half wave into an open ended secondary without having to have too longer wire for the secondary.
     
  4. bertus

    Administrator

    Apr 5, 2008
    15,645
    2,344
    Hello,

    What pulse transformer are you using?
    Keep in mind that a transformer is an inductive load.

    Bertus
     
  5. warp

    Thread Starter New Member

    Jan 20, 2014
    18
    0
    I often wind my own transformers but I am also going to try a 2 KV xennon trigger trans as well.

    Hi
    Assuming the above mentioned fet and driver work ok together what kind of frequency could I expect from that arrangement.

    Thanks
    Wil
     
    Last edited by a moderator: Nov 12, 2014
  6. alfacliff

    Well-Known Member

    Dec 13, 2013
    2,449
    428
    whaty voltage are yo supplying to the fet? voltage and current specs are necessary. sounds like a solid state tesla coil. might google (or whatever) solid state tesla coil for ideas.
     
  7. bertus

    Administrator

    Apr 5, 2008
    15,645
    2,344
    Hello,

    The switching speed of the TC4426 is dependend on the gate capacity of the driven mosfet.

    [​IMG]

    As you can read from the other picture the gate capacity of the IRF540 can be upto 1700 pF:

    [​IMG]

    Bertus
     
  8. warp

    Thread Starter New Member

    Jan 20, 2014
    18
    0
    I was hoping to feed the gate with 12 volts also as well as switching 12 volts. I was hoping the gate driver would amplify the voltage to that, but this would be ny first using gate drivers so I will have to play with it. I have used NE555 before directly bias IRF540 from pin 3 with 3906 pnp to help switch off the fet. This worked ok to about 100 to 150K then things started to go pear shaped.
     
  9. GopherT

    AAC Fanatic!

    Nov 23, 2012
    5,987
    3,733
    Pear shaped? Do you mean that your inductively loaded square-wave signal didn't look like a square beyond 150kHz? And more like a pear? Or does pear shaped mean something else?
     
  10. warp

    Thread Starter New Member

    Jan 20, 2014
    18
    0
    Thanks for your time posting the info Bertus. it does look like those nanoseconds are starting to add up when I look at all 3 rise and fall times and delay ect.

    @alfacliff Yes something like a tiny little Tesla coil. At the moment Im driving it with a PN222A wirelessly with a clip lead clipped to the base. It is going at 270KHz.
    Kind of wierd that frequency counter on my muktimeter went off before I plugged it into the circuit. Unfortunately I have no frequency control so I dont know whether
    it is pulse transformer type resonance or seconday quater or half standing wave resonance.
     
  11. warp

    Thread Starter New Member

    Jan 20, 2014
    18
    0
    Sorry about the slang. Ill stop that, it makes for to much confusion. I dont have a scope so I cant describe the wave form but according to my multimeter at 100K it has a duty cycle of 52% and at 240K it is up to 75% and thats just off pin3 of the NE555 I was too scared to see what was coming out of my fet. Im not too fussy about the wave form but even I couldnt have that. I would like to keep the duty cycle about 50 but much rather less than more. I would be happy enough just to make some kind of impression at small MHz frequencies.
     
  12. ScottWang

    Moderator

    Aug 23, 2012
    4,853
    767
    Using a 24 mhz crystal oscillator module and divided by 4 and 2, then you can get a 3 mhz 50/50 cycle.
     
  13. alfacliff

    Well-Known Member

    Dec 13, 2013
    2,449
    428
    thnat would be the quarter wave resonance. the best frequency for your coil if it were used as a tesla coil. if you go higher, you will probably have an insulation breakthrough at the quarterwave point of the frequency you are trying to use.
     
  14. warp

    Thread Starter New Member

    Jan 20, 2014
    18
    0
    Hi
    ScottWang Thanks How variable is this arrrangement for tuning purposes.

    alfacliff Thats interesting. I thought the frequency might have needed to be higher. Using rough figures from memory I had a 1MHz wavelength at about 250 meters then divide by 4 then multiply by the velocity factor in wire of ???. I lost a bit of data when i reloaded windows. I havnt measured the secondary on my xennon trigger coil yet. Perhaps the electrical wave is compressed when coiled due to the magnetic flux in the coil.

    The 270 K does seem a little fast for pulse transformer res. Thats shaping up to be good news for me electronically as the frequencies needed might be much lower than I anticipated. Any thoughts welcome.
    Thanks
    Wil
     
  15. Alec_t

    AAC Fanatic!

    Sep 17, 2013
    5,773
    1,103
    Personally, I'd use a FET with lower gate capacitance and lower Rds(on).
     
Loading...