Page 2 of 3 FirstFirst 123 LastLast
Results 11 to 20 of 22

Thread: TL494 halbridge like ATX problem

  1. #11
    it seems to me your gate drive is lacking. you should have active pull up and pull down with higher power transistors like BD139/140 pair in each TL494 OUTPUTS.

  2. #12
    I set some voltage on voltage potentiometer, and then i increase the current. I've got 1ohm loading resistor. Up to 2Amps there is this crap, but over 2 amps it works stable and perfectly
    https://www.hobbyelektro.eu/spinany-zdroj-0-30v-0-6a/ - there is the actual, working schematic. I tried to use bigger feedback capacitor (on Your schematic 68nF ).
    There is the crap on feedback(pin 3 of 494)(green) - ac coupled and pink is pulses on the 494 output. (Can see how it turn switching on/off and thats actually audible - the freuency changes (5-12kHz?), but over 2amps output, it dissapears)
    1592492626.jpg

  3. #13
    .... Silvio's Avatar
    Join Date
    Apr 2015
    Location
    Malta
    Posts
    864
    Blog Entries
    9
    @FIFI-22 Have you tried to decouple the feedback rail going to pin 3? It might shunt down any oscillations. A series filter formed by a capacitor and resistor in series between the feedback line and ground. (33nF and 820R) should get a frequency close to 5.8Khz. This should dump out this oscillation hopefully.
    Have you built the pcb or you have a lot of wires flying around? If so these are all antennas and pick up stray inductance and capacitance. At least for the feedback loop use a twisted pair of wires to help out. also use a star ground.

  4. #14
    Thanks for super fast reply
    The pcb is made from the link given before, it has some star grounding. I made it, becaouse author claimed, that it worked. I will try adding filter and post what changed, but the problem might be somewhere else, i have never seen such a filter in ATX (but they dont have current control). I will also try feedback, from schematic, which You gave me before, but i like to have current setting down to like 10mA (to power small leds etc.) Is there a possibility to achieve that?

  5. #15
    Sorry for posting one after another,
    Made some tests, adding filters doesnt work, hovewer I got rid of last feedback and changed it to this from Your post. It works better, no more oscillations. The problem is that when I want low output power , the 494 has such small pulse width, that transistors cant get fully open, so they get a bit hot. Also it sometimes happens, that switching action stops and starts again (with lower power), is there a way to not allow the 494 create such small pulsewidths and not waste that power? I am thinking right?

  6. #16
    .... Silvio's Avatar
    Join Date
    Apr 2015
    Location
    Malta
    Posts
    864
    Blog Entries
    9
    Quote Originally Posted by fifi_22 View Post
    Sorry for posting one after another,
    Made some tests, adding filters doesnt work, hovewer I got rid of last feedback and changed it to this from Your post. It works better, no more oscillations. The problem is that when I want low output power , the 494 has such small pulse width, that transistors cant get fully open, so they get a bit hot. Also it sometimes happens, that switching action stops and starts again (with lower power), is there a way to not allow the 494 create such small pulsewidths and not waste that power? I am thinking right?
    Well getting to such small pulse width eventually the system goes in discontinuous mode that means it start switching on and off with such a small current. It is not so easy to get to such low currents hence having also the availability of the smps to go up the 5 amps also. It will all end up in the output choke that has to be quite large in inductance like being 300uH or so to get to small minimum currents. This will eventually create a large voltage drop when it comes to high current and a lot of headroom has to be taken into account otherwise it will not maintain the voltage at the higher current. The chip itself I believe is not designed to go in discontinuous mode. In that case a flyback topology will perform better having a greater duty cycle like 10 to 90% rather than HB topology which the chip can only go to around 48% or so.

    If you want your psu to be that versatile then you can adapt a linear stage using an LM317 which this will have the pwm voltage following the output voltage by 4v more than the output voltage. The LM317 current can be enhanced using a pnp transistor like MJE2955 to boost up the current. This system will further filter your output noise as well and heat dissipation will be kept to a minimum thus the voltage at the input of the linear stage will always be 4v or so as the pwm output voltage will follow the linear stage with 4v more according to the output voltage setting.
    Regarding current regulation this now can be put on the linear stage using an opamp to to sense the output current through say a 0.2 ohm resistor.

    It looks rather complicated in a way but it is not. The voltage of the PWM will follow the output with a 3.2v zener diode + series diode (3.9v)and a small transistor which senses the output on the linear stage and gives feedback to the TL494 with 4v more. This will give adequate headroom for good regulation. As the load increases on the output so does the pulse width and the headroom voltage will be maintained all the time up to a maximum of 5 or 6A depending what the smps could give.

    Keep in mind that sometimes you cannot have it all with simple circuits and more elaborate circuits are needed to get such results. With the circuit I gave you your minimum current will be 100mA. Getting below this current as I explained will need to reconsider the circuit all over again. You have a way out but, You can use an LM317 in constant current mode in series with your output of the smps. You can look up the datasheet of the 317 and you will see how easy this is to achieve.

    Regards Silvio

  7. #17
    Again thanks for comprehensive reply, I learned so much!

    So, We can say, that converter works, I will characterise it, and post schematic, photos etc. on the forum in thanks for help.
    The idea with adding linear postregulator is excellent, I want to try that, even if this will take months, I learned so much until now, and I want to continue! This psu was ment to be as a bench power supply, and this additional regulation would be good, efficiency doesnt matter for me (at least at small powers).
    I will try making this regulator automatically switching in the output, only in small currents (<1A), but in larger ones, putting the regulation on the 494, is that a good idea? Also i think LM317 is a good starting point, but 4V dropout voltage is a bit large, I will serach for regulator with smaller dropout voltage. Current regulation on 317 could also be tricky, maybe I should build such regulator from descrete coponents?

    Again, thanks for help!

    Regards,
    Fifi

  8. #18
    .... Silvio's Avatar
    Join Date
    Apr 2015
    Location
    Malta
    Posts
    864
    Blog Entries
    9
    I can give you an idea how to combine the LM317 to regulate the feedback for the TL494 . The switching regulator (tl494) will follow the input to the LM317 with a little more voltage (4V) This will help with low dissipation in the pass transistor. Less than this voltage the voltage will not be so stable at high current say 6A. I have built this circuit and it works for me.

    Here is my dual smps followed with linear regulator.

    Link https://www.youtube.com/watch?v=uXzrmLrwPcg&t=1s

    Regards Silvio

  9. #19
    Great build! What topology is switcher operating with? I will defenetly try to play with regulator. I love that idea! But, regarding to actual stage of my psu, I'm a little bit confused with tl494 Feedback circuit, I will post, what exactly i'm not sure tomorrow.

    PS. I changed current feedback to work with inverting opamp, so the voltage on current shunt will not disturb output voltage.

  10. #20
    .... Silvio's Avatar
    Join Date
    Apr 2015
    Location
    Malta
    Posts
    864
    Blog Entries
    9
    Quote Originally Posted by fifi_22 View Post
    Great build! What topology is switcher operating with? I will defenetly try to play with regulator. I love that idea! But, regarding to actual stage of my psu, I'm a little bit confused with tl494 Feedback circuit, I will post, what exactly i'm not sure tomorrow.

    PS. I changed current feedback to work with inverting opamp, so the voltage on current shunt will not disturb output voltage.
    The topology used is Half bridge as this is a modified ATX supply. You will find some schematics that I made during the build but I never came to make a proper one. They are all drawn in pencil. You will surely understand what I finally came up with the description I prepared for you.

    There is something wrong in the website and I cannot upload pdf files. Send me your mail address and I will e-mail the file to you.

    Regards Silvio.

Similar Threads

  1. TL494 Lab PSU
    By bogdan2011 in forum diysmps
    Replies: 27
    Last Post: 07-07-2019, 02:29 PM
  2. Tl494 smps
    By noorworkshop in forum Switching Power Supplies Informations
    Replies: 26
    Last Post: 06-13-2018, 06:59 PM
  3. TL494 Class D Amp
    By noorworkshop in forum Class D
    Replies: 1
    Last Post: 05-15-2018, 02:49 PM
  4. Replies: 7
    Last Post: 07-24-2014, 06:20 PM
  5. 1kW halbridge smps with PWM control
    By mych3 in forum diysmps
    Replies: 4
    Last Post: 07-22-2011, 09:37 PM

Tags for this Thread

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •