Electronics > Projects, Designs, and Technical Stuff
Design of SMPS - Not sure what these spikes are being caused by.
MagicSmoker:
--- Quote from: Zog on September 21, 2019, 11:34:28 am ---...
Should I also include that huge inrush when it first starts up and getting to voltage as part of the selection process ? I am guessing yes. ie: I should take the peak to peak of the whole simulation and not just when it has settled down.
--- End quote ---
Yep. Inrush causing the boost choke to saturate is a major problem at higher power levels, but at this power level simply using the controller IC's soft-start function will deal with it.
--- Quote from: T3sl4co1l on September 21, 2019, 10:30:45 am ---
--- Quote ---as it has typically fallen to 50% of the unbiased value
--- End quote ---
better look up what that means.
--- End quote ---
Hmmm... I didn't think that was worded ambiguously. If you look at the datasheet for any of these commercial chokes* they will both state the inductance at zero current - ie, unbiased by DC - as well as the inductance at rated current or saturation or the like. In this part number I suggested earlier the unbiased inductance is 18uH, the rated current is 10.1A, but the inductance at that current has fallen to 8.5uH, or 47% of the unbiased value. This means that the real current rating isn't anywhere close to 10.1A.
* - a note on terminology: a choke is an inductor designed to tolerate DC bias without saturation.
Zog:
Sorry man. It's just that I am reading so much stuff it's kinda making my head spin.
Thank you for spelling it out for me. edit: better have a very close look at those curves.
Well. I have had a crack at making a 1:10 using this
https://au.mouser.com/datasheet/2/597/da2032-463371.pdf They do in all fairness to them say it is made for capacitor charging with a different chip. But I though I would try to see how it went.
Err... ugly. It's fine at say 20mA's or so .. but boy does it not like to draw a lot of current above that.
I bought one using the above inductor off ebay. Fine up to 40mA's but errr.. yer. After I blew it up I removed all the chips off and had a good look at the layout though, so all was not wasted. Very nice.
Attached for your amusement.
MagicSmoker:
The DA2034 is a great little flyback transformer, but if that is what you are thinking of using then you might as well design a flyback rather than screw around with the tapped boost, dual/cascaded boost, etc.
Note that the dot ends of each winding in a flyback are inverted with respect to the other. E.g. - if the dot end of the primary is connected to the supply voltage then the non-dot end of the secondary goes to the output diode.
You can also use this transformer in the tapped boost by connecting the dot end of the primary to the 12V supply, the non-dot end of the primary to the dot end of the secondary, and the non-dot end of the secondary to the output rectifier. If you find the switch blowing up at a tiny fraction of expected power output then the most likely cause is that you've connected the dot end to the dot end (or the non-dot end to the non-dot end).
Also note that you need some kind of clamp across the primary to protect the switch in a flyback; the RCD type is preferred, but even a series fast diode + zener or TVS diode can work. If using the latter choose a zener/TVS voltage at least 50% higher than the reflected output voltage (which is the output voltage divided by the secondary:primary turns ratio; e.g. - 200V / (10:1) = 20V; set clamp voltage to 30V or higher).
This is probably a lot to digest, but I can't distill 20+ years of SMPS design into a few forum posts, even if you paid me.
Zog:
Thank you once again for your valuable tips.
I am designing a flyback at the moment with this.
https://www.coilcraft.com/pdfs/hexa-path.pdf
The flexibility is great and I am getting good results so far.
The snubber is noted. Though the application notes are a little confusing.
I am going for this one.
Not this one, if that even is a snubber.
Zog:
I think I might have something that works BUT. Now I am worried about wattages.
I should of course be using Average for this right ? Peak to peak is insane.
Another question too if I may. I am trying to get the model as close as I can and my datasheet mentions leakage and interwinding capacitance. Leakage I can do with the K factor but where do I put in the interwinding capacitance ?
In the primary or the secondary or both ?
https://au.mouser.com/datasheet/2/597/da2032-463371.pdf
P.S.
I am sure you guys know about this little magic bean. https://www.analog.com/media/en/technical-documentation/data-sheets/8304fa.pdf
200V right out of the box with no external mosfet.
Looks a little tricky to get right though and not enough amps for me. The application notes are an interesting read though all the same.
The transformer they recommend is right on the edge with the amps so I don't really know how they do it.
Navigation
[0] Message Index
[#] Next page
[*] Previous page
Go to full version