1
\$\begingroup\$

I have built a ETD44 flyback with 11 turns on the primary and 144 turns on the secondary. I am using 0.75 mm paper as spacers and the core material is PC40. The diode on the output side is a FEP30 from Vishay. I am operating this with 10 V input voltage and 50 kHz switching frequency.

I am doing 2 measurements with my oscilloscope: Primary voltage at the FET Drain and output voltage accross a 120 Ohm resistor. I can see that when the FET Drain voltage goes to 0V (the FET turns on) the output voltage also goes to 0V. When the FET turns off the output voltage goes to 10 V. The Voltage accross the FET rises to 20V while its off and it stays constant there until the next turn-on, despite the input voltage being only 10V.

I have sketched what my analog scope shows:

enter image description here

The setup looks like this:

enter image description here

When I disconnect the resistor and only connect the scope to the output and when I ramp up the voltage to 12V input I see this (10 V/div, 10 uS/div, 0V for the bottom signal, FET Vds, is the solid line just below the dotted line, for the top signal, Flyback output, 0V is the center line):

enter image description here

I would have expected a much higher output voltage with a decay afterwards. Same thing for the voltage accross the FET. What went wrong here and why is it showing this behaviour, and most importantly what can I do to fix this so the Flyback works properly and gives the expected output voltage? I have tested without a spacer and in that case when I raise the voltage higher (up to 15V) the flyback starts to emit a high frequency ringing noise.

\$\endgroup\$
12
  • 2
    \$\begingroup\$ Please show the schematic. \$\endgroup\$ Commented Nov 25 at 18:34
  • \$\begingroup\$ Have you observed appropriate dot notation in your flyback circuit and what attenuation do your scope probes have? Show the circuit \$\endgroup\$ Commented Nov 25 at 18:37
  • \$\begingroup\$ @Ferrybig Schematic added \$\endgroup\$ Commented Nov 25 at 18:51
  • \$\begingroup\$ @Andyaka No idea about the attenuation, its a Hameg HM 204. Probe is a P6100. I also tried to use a cable that connects directly to the BNC input of the scope, that doesnt change anything though. \$\endgroup\$ Commented Nov 25 at 18:55
  • \$\begingroup\$ Your secondary winding has to be connected the other way around. \$\endgroup\$ Commented Nov 25 at 19:02

1 Answer 1

Reset to default
1
\$\begingroup\$

The basic problem you have is that the output load is too current-hungry for this to work without extensive redesign. R1 (the load) is 120 Ω and by my calculation you expect to see over 120 volts on the output. If we stick with 120 volts then that's a load current of 1 amp and, that naturally requires a current on the primary side that is over 20 amps at some point.

Of course this is dependent on duty cycle and inductances but, if we assume 20 amps peak are needed on the primary then the IR530 MOSFET looks to be inadequately rated.

Then there is the core itself. It has to store sufficient energy in the magnetic field to service the load's energy requirements and the ETD44 may not be adequate however, you say it is gapped (initially) so maybe it will just about work.

Nevertheless you are still not providing sufficient details to nail this down to core problems, winding losses and MOSFET inadequacy so, here's an image from my basic website that attempts to follow what you did and, I assumed a primary inductance of 100 μH for this: -

enter image description here

It's running close to 50% duty cycle in continuous conduction mode (CCM) and so, as a design, this seems right for the turns ratio of 11:144 (the reciprocal is 0.076 in the image above). But, you can see that the peak primary current is 25 amps and this might easily be the problem. MOSFET power dissipation: -

enter image description here

25 amps drops around 2.4 volts and dissipates 60 watts but, it only does this for 50% of the time hence MOSFET is dissipating as heat 30 watts. That is far too much.

\$\endgroup\$
6
  • \$\begingroup\$ The 120 ohm resistor was just for testing as without a load it works even worse. I am using a film capacitor there now and the output voltage doesn't go beyond 40V for 12V input voltage. The final design should push 200-250W through. I was trying to keep the scope of my question limited to this issue, if you want to understand the full picture then this is what I am trying to achieve: ietresearch.onlinelibrary.wiley.com/doi/10.1049/… (Just with a higher input voltage, so my turn ratio differs). \$\endgroup\$ Commented Nov 25 at 22:00
  • \$\begingroup\$ I agree that the 120 Ohm resistor is not an adequate load for 120V output voltage, the issue here is that the flyback itself is not working as intended though, I can do adjustments on the components but right now I don't have to worry about other components (and their overheating) if the flyback itself is not working properly. I am currently not working with near enough power to cause any kind of relevant heat dissipation, I wish that something would get hot as that would be a sign of something hitting the limit, but right now I am just seeing a lower than expected output voltage. \$\endgroup\$ Commented Nov 25 at 22:04
  • \$\begingroup\$ You really are not recognizng the wisdom of this site to solve your problems by hiding things that are totally relevant to your problem. Your circuit is in tatters and your supply voltage has changed etc.. You have also changed the MOSFET so, just what are your plans for the next round of changes? To be honest I have lost faith in you stating the problem and am backing away from helping you. \$\endgroup\$ Commented Nov 26 at 0:22
  • \$\begingroup\$ What am I hiding? Of course I am not sitting back and waiting for the solution to magically appear in my inbox, I keep trying things and while doing that I might come across things that also seem wrong. I turn the voltage of the power supply up and down all the time to see how the output is affected. The problem is that I don't know what the problem is: Could be wrong gap size, something wrong with the way I wound the transfomer, saturation, it could even be the supplier of the cores shipping something that's not PC40 at all! I can only state what I am seeing, what I am doing and what happens. \$\endgroup\$ Commented Nov 26 at 0:35
  • \$\begingroup\$ Half my answer was based on the MOSFET being what you originally said and now you've changed it or revealed it to be a different type. That is very rude and counterproductive. This is why I am not offering further help on this question. \$\endgroup\$ Commented Nov 26 at 9:27

Start asking to get answers

Find the answer to your question by asking.

Ask question

Explore related questions

See similar questions with these tags.