I'm in the process of drawing pcb layout for a couple of two channel OPA549 based amps. They will be put inside some small speakers along with active crossovers and bsd correction. The circuits will be powered from 18-0-18 120VA transformers. How much output power can I expect (2 OPA549 per toroid)?
However, I have several questions:
- The OPA549 has poor PSSR. How should I deal with this? There's no possible way to include active regulation of the supply due to limited space. The OPA549 is chosen cause I have some spare ones laying around.
- The active crossover will be powered from the same supply as the OPA549s. Three OPA134s will be used - or maybe I'll make a 24db/okt with one OPA134 and two 2134. I don't know yet. These opamps have fairly good PSSR, but the rail ripple will be rather significant because of the OPA549s, I suppose. If I'm wrong, please tell me. And the rail voltage is to high for OPA134. Should I actively regulate the supply down to about +-15v (would require heatsinking of the regulators, which the limeted space doesn't allow)? Or could I just use some active rail splitters to split the positive or the negative supply with ground as 0 to +-12v?
I hope you understand what I mean...
Maybe I should draw a schematic...
EDIT: forgot to mention.. each OPA549 will have a seperate pair of 4700uF electrolytics.
However, I have several questions:
- The OPA549 has poor PSSR. How should I deal with this? There's no possible way to include active regulation of the supply due to limited space. The OPA549 is chosen cause I have some spare ones laying around.
- The active crossover will be powered from the same supply as the OPA549s. Three OPA134s will be used - or maybe I'll make a 24db/okt with one OPA134 and two 2134. I don't know yet. These opamps have fairly good PSSR, but the rail ripple will be rather significant because of the OPA549s, I suppose. If I'm wrong, please tell me. And the rail voltage is to high for OPA134. Should I actively regulate the supply down to about +-15v (would require heatsinking of the regulators, which the limeted space doesn't allow)? Or could I just use some active rail splitters to split the positive or the negative supply with ground as 0 to +-12v?
I hope you understand what I mean...
Maybe I should draw a schematic...EDIT: forgot to mention.. each OPA549 will have a seperate pair of 4700uF electrolytics.
OPA549 2* at 120VA 2*15V transformer (my computer speaker amp)
measures with load resistors on each channel:
1kHz
10 Ohm
0,02%THD ~ 10W
0,33%THD ~ 13W
5 Ohm
0,05% THD ~ 22W
3,33Ohm
0,2% THD ~ 31,5W
1,8 Ohm
0,22% THD ~ 43,5W
caps are for each rail (+ and -)
2700uF
2*1000uF
3,3uf Film
0,1uF Film
in parallel.
measures with load resistors on each channel:
1kHz
10 Ohm
0,02%THD ~ 10W
0,33%THD ~ 13W
5 Ohm
0,05% THD ~ 22W
3,33Ohm
0,2% THD ~ 31,5W
1,8 Ohm
0,22% THD ~ 43,5W
caps are for each rail (+ and -)
2700uF
2*1000uF
3,3uf Film
0,1uF Film
in parallel.
Thank you for the numbers, Till.
While I'm at it, I may just as well add another question to the list. As noted above, each OPA549 will have a seperate pair of electrolytics and bypass capacitors, but will share rectifier bridge. What would I benefit from using seperate rectifiers apart from lowering the stress on the rectifier if only one is used, considered the application (bi-amping of a two way)?
While I'm at it, I may just as well add another question to the list. As noted above, each OPA549 will have a seperate pair of electrolytics and bypass capacitors, but will share rectifier bridge. What would I benefit from using seperate rectifiers apart from lowering the stress on the rectifier if only one is used, considered the application (bi-amping of a two way)?
this is a philosophical question, better diskuss with those people who spend their time listening to differences of different brands of resistors. I use on transformer and one bride rectifier for the + rail and one for the - rail made from some discrete diodes for both channels. I would build depending on the parts you allready have on hand.
Hmm, thanks. I only have to 25A rectifiers at the moment, and 2 35A. The latter are overkill and they are physically too large.
What is the benefit from using a rectifier for each rail? I've seen people claim, that seperate rectifiers for each amp will provide some of the benefits from using completely seperated supplies (including seperate transformer). I suppose this is voodoo, or?
EDIT: In my Harman/Kardon HK680 seperate bridges for the two channel are used along with seperate electrolytics. But not two bridges for each supply.
What is the benefit from using a rectifier for each rail? I've seen people claim, that seperate rectifiers for each amp will provide some of the benefits from using completely seperated supplies (including seperate transformer). I suppose this is voodoo, or?
EDIT: In my Harman/Kardon HK680 seperate bridges for the two channel are used along with seperate electrolytics. But not two bridges for each supply.
You will find enough diskussion and good reasons in this board by using the search function. I will not give comments whats sounding better etc. One bridge for the positive and one for the negative side is needed to have some real ground inbetween (0V). In case you use only one bridge for both sides and make a virtual ground by two resistors and some caps, this ground will go up and down with against the load, not really stable ground. So better use one bridge for + and one for - . Separate bridges for each channel are not needed. I use 30A diodes bwt.
sobazz said:
The 549 is a fine chip, and of the several I have made, I have never noticed any audible, ( or even noticable on my cheap, old scope), PSRR problems.
I used dual bridges, with a 1000uf cap bypased with a .1uf at the bridge, and an extra 1000uf at the chip. Somewhere on the board there are some photos, but I'm not quite sure where!
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.