DC-DC converter

[Ryssen]

I have built this DC-DC convereter,and it´s working fine,but what if I want say 40volts/1.5 A,can I use a larger inductor for that?
And the switching frek is 16Khz I can hear it when I use it for an amp,can I use a LC or a RC filter?What values should that be?

[Pabo]

If you want to simulate your converter, try [URL=http://henry.fbe.fh-darmstadt.de/smps_e/smps_e.asp[/URL]

[Pabo]

Sorry, didn't work

http://henry.fbe.fh-darmstadt.de/smps_e/smps_e.asp

[Ryssen]

Nice site,my schema uses a bobin with no core,on the site it´s a
core,what´s the difference.

And I might be blind but I see nothing about the filter.

[rayban68]

Why can't you just drop Ct to something like 350 picofarads, and then the switching frequency is up over 30kHz and you won't be able to hear it.. as per the appnotes page 4. I think its on the datasheet too

http://focus.ti.com/lit/an/slva143/slva143.pdf


As far as using a core, I think that just makes the inductor physically smaller.

[li_gangyi]

using a core will reduce inductor size...but be careful of saturation...as for higher voltages...just change the configuration of the feedback resistors...the formula can be found inside the datasheet I guess...it will also depend on the voltage reference.

[rayban68]

So you are currently using this supply on an amplifier? how well does it perform? I ordered some samples and want to use this chip for a +/- 20 to 25 volt powersupply, at about 3 amps from the +12v in a car. I am wondering how well this sort of a setup would work, compared to a standard SMPS. I would like to avoid winding my own transformer and such, and this chip looks like the easy solution. Do you (ryssen) or any one else know how this would perform with a speaker as the load?

I know that voltage doublers and the like don't perform well with the variable impedance load that a speaker presents to the power supply, so I would like to be sure that this works before I get all crazy and make some pcbs.

Thanks

[li_gangyi]

looks promising...but I dunno about the current though...seems a bit on the low side.

[Ryssen]

and it´s working fine

I am using a IRF830 instead of a BUZ 10,and and at low volume on the amp,it is 30 volts,but when i turn the volume up the voltage drops to say 12v,like the input volt.I will order a Buz 10 and try if it gets better. The amp is loaded with 3 ohms (2x40watts ic bridge amp)maybe the load is to low,and I wil try to
lower the osc cap for higher frekvency.will the current output get lower with higher frekvency?

[rayban68]

I don't think that the capacitor will affect the current output. That appnote I posted has the formula for peak current as Ipeak=2*Iload*(Vout/Vin). In your case that would be Ipeak=2*3*(30/12)=15Amps. So you need to be sure you have a pretty fat transisitor in there. Looks like it is supposed to a high power bjt, maybe something like ON Semi 2N3772. Not sure exactly what will work there.

Also, your current limiting resistor appears to be the two .1 ohm resistors in parallel, giving .05 ohms. The formula for current limiting from that appnote is 0.5/(Ipeak) . (At least that is what it looks like to me, on page 14) that means you are limiting current at 10amps. Either add another .1ohm in parallel to give you 15 amps, or just get rid of them and connect the two pins together, giving no current limiting. (Altho you run the risk of letting out some smoke i suppose)

I think (hope?) that the reason your output is dropping to 12 volts is primarily because of the current limiting resistors. Before you replace your exsiting transistor, play with these values a bit, and if that doesn't work, then get a new transistor.