CLC filter in supply?

[FlexQ]

Hi Guys

I am having a discussion with a friend about my supply for my 2 x pa150 lm3876 amplifier.

He says that a coil in the supply will set the responcetime for the supply down so drasticly that it will sound terrible.

is that in any way correct?

Here is my supply diagram.

[IMG]http://img17.imageshack.us/img17/1882/tilh4a.jpg[IMG]

The coils is EI iron core, with the E and the I seperated by a thin nonelectric-material. It has a DC resistance at .2Ohm.

EDITED the image, it is quite big ...

[unclejed613]

first of all, you don't really need the coil, as the PSRR for this device is between 85 and 120db. the inductor's impedance is 300 ohms at 50hz, so it does smooth out a lot of ripple, but with such a high PSRR it's really not necessary.

[kaos]

Your bridge connections are wrong on your schematic.

[Magura]

35mH is way too much, 2mH would do just fine. Air core is to be preferred, as you have no issues of saturation and the like.

For a chip amp, an inductor may be a bit overkill, but not completely pointless.


Magura

[Magura]

Originally posted by FlexQ
Hi Guys

I am having a discussion with a friend about my supply for my 2 x pa150 lm3876 amplifier.

He says that a coil in the supply will set the response time for the supply down so drasticly that it will sound terrible.

is that in any way correct?

No, that is not correct.


Magura

[AndrewT]

and C1 & C2 will probably fail due to overheating.

[wakibaki]

Originally posted by FlexQ
He says that a coil in the supply will set the responcetime for the supply down so drasticly that it will sound terrible.

BS.

w

Your PSU is mad overkill. Any 4sshole with money can do that.

[unclejed613]

Originally posted by kaos
Your bridge connections are wrong on your schematic.

you're right, while everybody else was looking at the coils and bizarre arrangement of caps, the biggest, most obvious BOO-BOO of all went unnoticed, not to mention what would happen if he hooked the bridges up correctly.........
bridge rectifiers in series is always a bad idea, unless you happen to like fire and smoke....

it would work much better like this if he's using two transformers for the supply

[FlexQ]

Originally posted by unclejed613
first of all, you don't really need the coil, as the PSRR for this device is between 85 and 120db. the inductor's impedance is 300 ohms at 50hz, so it does smooth out a lot of ripple, but with such a high PSRR it's really not necessary.

Okay, havent thought of that (: When the amp is finally wired up, i will try with or without the coils, to see if i can measure any differences

Originally posted by kaos
Your bridge connections are wrong on your schematic.

Ah, havent seen that, it was just drawn fast so you could see it, it is wired up correctly IRL. So no worries

Originally posted by Magura
35mH is way too much, 2mH would do just fine. Air core is to be preferred, as you have no issues of saturation and the like.

For a chip amp, an inductor may be a bit overkill, but not completely pointless.


Magura

I just happened to have 4 laying around, therefore i used them

Originally posted by Magura



No, that is not correct.


Magura

okay (:

Originally posted by AndrewT
and C1 & C2 will probably fail due to overheating.

not if they are rated 160V

Originally posted by wakibaki


BS.

w

Your PSU is mad overkill. Any 4sshole with money can do that.

Why dident you just stay in bed when having a bad day?

Originally posted by unclejed613




you're right, while everybody else was looking at the coils and bizarre arrangement of caps, the biggest, most obvious BOO-BOO of all went unnoticed, not to mention what would happen if he hooked the bridges up correctly.........
bridge rectifiers in series is always a bad idea, unless you happen to like fire and smoke....

it would work much better like this if he's using two transformers for the supply

As said before, it was just drawn fast, so you could see my "bizarre" arrangement of caps and coils . Not for analyzing my bridge rectifiers. :P

But i got my answer. Thx guys

[AndrewT]

C1 & C2 are 1mF, what is the ripple rating?

The first caps in a CRC or CLC PSU are subjected to a very high ripple and this high ripple causes losses in the esr creating heat.

A small electrolytic cap located here cannot have sufficient ripple rating to survive very long. It will overheat and eventually fail.

If you connected 10off 1mF in parallel you will have increased the ripple rating by a factor of ten. That is more likely to survive than selecting a 10mF cap.