SMPS design for power amps (Class D in particular)

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
Jump to Latest
G

Golgoth

Member
Joined 2004
Jaka Racman said:

...In this case you get zero voltage zero current turn on for free, so only turn off snubber is required. Because of higher peak current and high voltage (switch with 800V capability would be required) I thought that IGBT would be a better choice.
...
Click to expand...

Or maybe even better, an IGBT and a MOSFET paralleled, with the MOSFET switching off slightly later than the IGBT, as R Watson did for his 600W Flyback PFC, cf 3.6.2.3 of the ch3.pdf paper you linked to in post 26.

It was silly of me to think we could do without any turnoff snubber, especially in the isolated form of the Sepic where we will necessarily get a large positive voltage spike due to transformer leakage inductance refusing to stop conducting instantly.
 
J

Jaka Racman

Member
Joined 2002
Hi,

Genomerics, maybe you have not read it, but hard switching was never proposed. If lossless snubbers worked well in the days of the bipolars, I see no reason why this little forgotten technology would not work with IGBTs.

Saturation voltage of proposed IGBT is still a little high, but I would be glad to see a similar performing mosfet in TO220 housing, costing about the same.

Best regards,

Jaka Racman
 
G

Golgoth

Member
Joined 2004
Jaka Racman said:
If lossless snubbers worked well in the days of the bipolars, I see no reason why this little forgotten technology would not work with IGBTs.
Click to expand...

For the turn-off snubber, wouldn't a simple capacitor in parallel with the switch do the trick? Since the transistor is switched on at zero voltage (boundary conduction mode) it wouldn't have to empty the snubber capacitor at turn-on would it?
 
G

Golgoth

Member
Joined 2004
Jaka Racman said:
...
Fig 9a from slup100.pdf is what I had in mind when proposing nondissipative snubber. It is rather simple.
...
Click to expand...

Yup, makes perfect sense! Not quite sure but there might even be a way not to spit out the snubber's energy into the input (which would be bad for EMI): unless there is good reason not to do it, we could swap the coupling capacitor C and transformer primary L2 so C would have its foot at the transistor foot, and let the snubber spit out into the top of the capacitor (which would be around Vin).
 
G

Golgoth

Member
Joined 2004
Oops. After a closer a examination of its workings (below), if I am not mistaken this snubber will not be very effective in our case, as it will only start limiting rate of rise of collector voltage when the latter reaches max(0, Vin-Vout). So it ensures zero voltage switching only when Vin<Vout, which will not be the case at high line voltage i.e. where ZVS is most needed!

How this snubber works:
Let's suppose snubber capacitor Csn is empty initially, and Vin>Vout

1/ turn-on: Csn remains empty
2/ turn-off: Csn charges to Vout through top diode of snubber
3/ turn-on: Csn rings with Lsn for a half-cycle so it reverses polarity to -Vout (bottom diode of snubber blocks at this point)
4/ turn-off: top diode of snubber starts conducting when collector reaches Vin-Vout (which may be 250V), and then Csn charges to Vout again
5/ Goto 3
 
G

GregD

Member
Joined 2004
A and T Labs SMPS?

Searching around I found this company has an amplifier kit with a switching power supply. I was thinking that their switching power supply might work quite nicely with either the Zappulse or UCD amp modules. What do you think? Anyone have any experience with A and T Labs?
http://www.a-and-t-labs.com/K6_Sw_Amp/index.htm

Maybe I should post this in one of the Zappulse or UCD threads, but this thread seemed to be the most pertinent.
 
G

Golgoth

Member
Joined 2004
GregD said:
http://www.cadaudio.dk/pwmpricelist.htm

It looks like they sell the SMPS separately. On the above page they show a 1600VA SMPS for 256 euros. My concern was that the lowest output voltage they show is 60V. Would this be safe for the UCD400 modules since that is their max voltage?
Click to expand...

If the UcD400's OV detection trips (which it probably won't), you can always deactivate it.

But I wouldn't be so confident cadaudio will sell the SMPS separately, you better ask them.
 
tiefbassuebertr

tiefbassuebertr

Member
Joined 2005
tiefbassuebertragung.de
Most ultimate SMPS for Class-A, Class-AB and Class-D Audio Amplifier - where to find?

Until now the best SMPS, which I have heard in the last years, is the internal switch mode power supply of Halcro's DM38 - go to
Halcro dm38 power amplifier | Stereophile.com
But this approach is probably updated in the meantime due the age of several years. Some people claims, Linn in Scotland uses the best possible (own development) approaches for their SMPS.
Maybe one of the members have simplified schematics or such of detail.
Thank you for your information.
 
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
Top Bottom