Attempt at half bridle with 555's

Discussion in 'The Projects Forum' started by magnet18, Mar 30, 2011.

  1. magnet18

    Thread Starter Senior Member

    Dec 22, 2010
    1,232
    124
    Alright, I've noticed that everyone wants to make <high voltage device>'s, particularly with the 555, but all the 555 design's are no good because they only allow one MOSFET to be driven. So, I'm trying to build a decent circuit that uses half bridle/bridge (not sure what it's technically supposed to be called) using only 555/556's and MOSFET's, preferably not the really expensive kind.
    Heres my first design, I realize that the duty cycle for U1 wont be 50% with those values, so I'm going to adjust them tomorrow.

    So, since it is inconceivable that I actually draw this up right in one night, what do you guys see that needs fixed? I've attached the .sch file as well.
    [​IMG]

    I'm trying to be sensitive to the HV issue going on right now, so I've adjusted labels to be generic :rolleyes:

    on an entirely unrelated note (in case you can't put 2&2 together), http://www.youtube.com/watch?v=vCI5RozQdTY
     
  2. CDRIVE

    Senior Member

    Jul 1, 2008
    2,223
    99
    Me amigo, not quite sure what you're trying to accomplish as I only superficially scanned the schematic. It looks like you're trying to design some sort of push - pull output to the transformer. My guess is that this project is an extension of your flyback circuit with the arc dancing to your audio input. As is, the outputs to the FETs are not push pull if that's what you're trying to do.

    BTW, why can't the output of one 555 drive more than one FET? Are you referring to parallel connected FETs with both drains connected to the flyback xfmr?
     
  3. Wendy

    Moderator

    Mar 24, 2008
    20,764
    2,535
    I don't think what you are doing is what you think you are trying to do.

    If you want a push pull you need either a transformer with a centertape (this will let you use two MOSFETs), or a H bridge for your MOSFETs.

    The latter will feed your transformer twice the P-P voltage as your power supply. I don't think the former would have much in the way of benificial effects, even if you did have a center tap on the transformer.

    You can drive multiple MOSFETs with one 555, if you are trying to increase the drive. The increase will be minimal though, MOSFETs are pretty good about giving it their all, and their on impedance is very small. Is your MOSFET getting hot or something?

    What you have really doesn't make much sense to my eyes. You are using a CMOS 555, which has notoriously low drive compared to a conventional 555. If you were using a single conventional 555 at 12VDC it would handle any number of MOSFETs.

    Another major problem with your design, MOSFETs need low ohmage resistors at the gate to damp ringing. Ringing at this point will cause huge problems, of the type you may be blaming the 555 for.

    You are using the MOSFETs source to drive the other 555 circuits in a manner similar to a voltage follower. This does not work well, even a little. It is playing to MOSFETs weaknesses amoung other things. I've mentioned it before, but you are going to have to take the time to study transistor theory if you want to design with the suckers. It is a pain, but your knowledge gaps are showing. When I went to college I went through a similar experience, I had a fair background, but what I didn't know was killing me.


    A side note, you are using this schematic, but the way you have drawn it it is not obvious...

    [​IMG]

    This is the standard way it is drawn, it slows people down if you draw it a different way. Your choice of course, but if you look in a 555 datasheet you will see the above schematic. I noticed this in your other thread with the working plasma speaker.
     
    Last edited: Mar 30, 2011
  4. jpanhalt

    AAC Fanatic!

    Jan 18, 2008
    5,675
    899
    Hi Robert, it is nice to finally meet you.

    Your schematic is a little hard to read because it is too wide. Adding to some of the comments above, on the right, the two mosfets are essentially in parallel. Paralleling mosfets for greater power is common practice and can be done with a single driver.

    A 555 might have the current capability to drive multiple, low-capacitance mosfets, but driving several parallel mosfets is usually done with a dedicated, high-current driver.

    As for your push-pull/half H-bridge, the top side is usually a P-channel mosfet or an N-channel with a special, boosting driver.

    As for use of depletion mode mosfets, that is somewhat unusual and they are a bit harder to find. Here's a nice overview: http://www.eetimes.com/design/analog-design/4009878/An-introduction-to-depletion-mode-MOSFETs

    John
     
  5. CDRIVE

    Senior Member

    Jul 1, 2008
    2,223
    99
    The comments made regarding unconventional schematic structure are very important here because I, like Bill and John, did not find your schematic easy to follow.

    What Spice/CAD are you using that made such a large image?
     
  6. Wendy

    Moderator

    Mar 24, 2008
    20,764
    2,535
    This I can answer. It is Express PCB. First thing I did was create a 555 in it's library when I started using it.

    [​IMG]

    If I knew how to export the component I would attach it to this thread.

    When I draw a schematic I do not use Express. I use M/S Paint and my templates, because clarity is vital. Come to think of it, I also made the ground symbol (they didn't have it). :D
     
  7. CDRIVE

    Senior Member

    Jul 1, 2008
    2,223
    99
    Uh, you mean the "Earth Ground" symbol, or is that a "Chassis Ground" symbol? :D:D

    For those that feel compelled to reply to this statement, see this thread..
    http://forum.allaboutcircuits.com/showthread.php?t=52178
     
  8. SgtWookie

    Expert

    Jul 17, 2007
    22,182
    1,728
  9. magnet18

    Thread Starter Senior Member

    Dec 22, 2010
    1,232
    124
    OK, maybe I should clarify, I'm trying to reduce the load on the MOSFET(s), because I currently need to keep it on ice. (literally, without an icecube directly on the MOSFET, it blows. poof. magic blue smoke.
    I'll make the 555 up as recommended, I just used that one because it was the only 555 in ExpressSCH.
    After doing more research, I have reached the conclusion that I need to do more research to find out if what I'm doing will help much or not...
    (I also noticed that I wired the reset's wrong :rolleyes:)

    Yes, this is a continuation of my other project, and I'm trying to make a decent one using only 555's and minimal MOSFET's.

    @John, thats my name, lets not wear it out :)
     
  10. magnet18

    Thread Starter Senior Member

    Dec 22, 2010
    1,232
    124
    Ah, if N channel use negative voltage to shut the flow off, that means P channel must use positive to shut it off, right?

    [EDIT]
    nevermind, kept reading, I guess so.
    If I was going for the same inverting logic effect of a PNP transistor, I would use P-channel Depletion, correct?
     
  11. jpanhalt

    AAC Fanatic!

    Jan 18, 2008
    5,675
    899
    I'm not really sure what you are asking. With N-channel, the gate must be positive relative to the source to turn on. With a P-channel, the gate is negative relative to the source to turn on. Note, with an N-channel device, the source is usually connected to ground or is at least at lower potential than the drain. In the P-channel, the source is at higher potential than the drain. The intrinsic diode will give a good clue as to which way to connect them.

    I do not think you need to use depletion mode mosfets. If you look at the spec sheet for IRFZ vs. similar IRF mosfets you will see lower capacitance. You might get by with driving two in parallel with a 555, but a dedicated driver would be better.

    Your circuit does not have a gate resistor. I don't recall seeing any oscillation ("ringing") in the gate tracings, so one might not be needed. If you put the mosfets in parallel, it is common practice to put a small resistor on each gate. It will help balance gate turn-on time. That resistor can be anywhere from a few ohms to 10 or 20 ohms. The International Rectifier app. notes and others discuss some of the nuances of gate drives. From memory, I believe Fairchild recommended smaller gate resistors than IR. For high-power drives that I have built, I just used something around 10 Ω.

    John
     
  12. CDRIVE

    Senior Member

    Jul 1, 2008
    2,223
    99
    Have you tried reducing the duty cycle?
     
  13. Wendy

    Moderator

    Mar 24, 2008
    20,764
    2,535
    I kinda suspected as much. Remember my comment about needing gate resistors to damp ringing? The reason ringing is a problem is it keeps the MOSFET in its analog region. If the MOSFET is not switched cleanly the end result is heat, lots of it. Keep It Simple Stupid is a real design philosophy. You need to look closely how you are doing things as well as what you are doing.

    Add a 2.2 or 4.7Ω resistor to those gate, right where the gate is on the MOSFET package. If you need resistors PM me.
     
  14. hgmjr

    Moderator

    Jan 28, 2005
    9,030
    214
    Just as noise indicates inefficiency in mechanical systems, heat indicates inefficiency in electrical systems.

    As cdrive has stated you probably are driving the coil for a period of time that is too long. You probably do not know the inductance of the coil you are driving but if you did and the dc resistance of the coil was nearly zero ohms, then you could determine the time you could safely drive it.

    The formula that governs the safe on time of the inductor is:

    \frac{\Delta i}{\Delta t}\ =\ \frac{V}{L}

    The lefthand expression is the change in current with respect to time and the righthand expression is the applied constant voltage V across inductor with inductance L.

    In effect, if you can rapidly switch the applied voltage across the inductor say from zero volts to say 1V then you can take that voltage and divide it by the inductance L and you will get the rate at which current will build up in the inductor. Keep in mind that your driver has only a finite amount of current it can switch without coming out its turned-on state. Once that finite max current is reached then the switching device will not be able to sustain the on state and a voltage will develop across the device. The small voltage across the device times the high current will cause heat to buildup rapidly in the switching device. Then poof.

    hgmjr
     
  15. magnet18

    Thread Starter Senior Member

    Dec 22, 2010
    1,232
    124
    wellz, ill give those ideas a shot and see how it fares. Would a scope dampen the ringing? Because it might have been me but it seemed to work better with the scope attached...

    My goal is a little further than just make it work for me this time though. I want to end up with a design that's really simple, reliable, and that has a satisfying volume, so that pretty much anyone with something with a CRT in it to scrap, some 555's (and maybe a couple MOSFETs), and a free afternoon can build a plasma speaker that is reliable, decent sounding, and long lasting. (health issues created by plasma aside).
    Theres a lot of circuits out there that look simple, but don't work very well at all, or work great, but are complicated enough that troubleshooting can take as long as building the darn thing.
    and almost anything that works well needs chips the average person doesn't have laying around unless they are into PWM.

    Essentially I want to make the plasma speaker a less daunting task, completable in a weekend with radioshack parts and a little goodwill shopping. (and without requiring a scope to troubleshoot the darn thing)

    @hgmjr, thanks, i'll look into that.
     
  16. Wendy

    Moderator

    Mar 24, 2008
    20,764
    2,535
    The act of observation changes what you are observing.

    The MOSFET is a capacitor on the gate. Not huge, but there. The length of wire is an inductor. The ringing is there, and the extreme input resistance of a MOSFET is all you need to pick up on it.
     
  17. magnet18

    Thread Starter Senior Member

    Dec 22, 2010
    1,232
    124
    Heisenberg uncertainty principle, right.

    I believe you, I'll try it out soon (along with some other stuff I shouldn't discuss ;))
     
Loading...