Integrated Rectifier Bridge VS Rectifier Diodes

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

I started looking at amplifier chips a few days ago, and found this site. It has been a great source of help, just reading it.

Now to my question. It seems to me that most people use four diodes for rectification (is that a word in english?). Is there a reason to not use a integrated rectifier bridge (if that's what it's called in english, what I mean is a chip with four diodes connected as a rectifier bridge)?

Erik
 
You can see the bridge here. There has been a lot of discussion on the use of normal rectifier diodes vs Shottky rectifiers, or high speed diodes. But personally I like the bridge, basically anything thats nicely packaged and closely thermally coupled will have an edge over discretes is my belief. So...
What is not shown in the PCB is provision for some snubbers.. I plan to put them directly on the pins, on the solder side.

ajju
 
Use the integrated bridge rectifiers. Less work, usually higher current rating, cheaper, easier to replace, etc.

There's really no reason to use seperate diodes.

Some people claim that faster diodes work better, but this is 60Hz - it doesn't matter since even the slowest diode in the world will be many times faster than this. Snubbing caps across the pins is a good noise prevention measure if you feel like it, usually not required but cheap so why not.

Some also claim that it makes a "sonic" difference if you get the individual diodes. These are usually the same people that claim that a $500 oxygen-free copper "super duper" mains power cord to feed your equipment will improve the sound of your amp. :rolleyes:

But they are right to some degree -> it does make a sonic difference. No sound will be coming out of your speakers when you can't afford to pay the electric bill becuase you spent your money on expensive high speed diodes :D
 
gchackle said:
There's really no reason to use seperate diodes.
Some people claim that faster diodes work better, but this is 60Hz - it doesn't matter since even the slowest diode in the world will be many times faster than this. Snubbing caps across the pins is a good noise prevention measure if you feel like it, usually not required but cheap so why not.

Hehehe :D And why with the cheap snubber caps? Some people don't rely on numbers. I have been happy using MUR120/140/160/480/860 etc. May be it is not because the 50ns, may be it is just because Motorolla has great machines. But who cares with the machines?

Some says price doesn't lie. Well, may be not true but motorolla diodes are not expensive when we talk about high end systems. If I cannot hear the difference in the separate diodes, I will still have no reason to not use it, except the relative difficulty in assembling. As with snubber caps, I cannot hear the difference, but I have a reason to not use it. Because you should not add anything to the system unless you know what it can deliver.
 
Sci said:
What about paralell connection of two bridges? Any problems there?

Erik

None...however, one of them will undoubtedly carry all the current. It will be difficult to balance the currents without some series resistors. The one with the lower vf dice will conduct and heat up, lowering it's vf even more..

One of sufficient capacity will work fine.

Cheers, John
 
jneutron said:


None...however, one of them will undoubtedly carry all the current. It will be difficult to balance the currents without some series resistors. The one with the lower vf dice will conduct and heat up, lowering it's vf even more..

One of sufficient capacity will work fine.

Cheers, John

OnSemi has a double-diode "in one" the MBR20100 -- which balances the current. It's in a TO-220 package like the MUR860.
 
jackinnj said:


OnSemi has a double-diode "in one" the MBR20100 -- which balances the current. It's in a TO-220 package like the MUR860.

Good info.

Where I used to work, we would build mil bridges, and three diodes were used per leg in the three phase units, a total of 18 individual 3 amp devices in each unit.

We would match them to one millivolt, and they had to be from the same diffusion run. In addition, heat sinking for the three in parallel had to be virtually identical, there had to be good thermal communication among them, and each had to have identical current feed..so current paths were critical.

We also made this 100 amp extremely low theta jc unit, with about 100 diodes in parallel..unit was about 3 by 5 inches, 3/4 inch thick. all diodes, to within a millivolt..

To try to match devices that close is beyond most, but if you can get them from the manu already matched, they will work in parallel really well, as long as thermal is good and wiring doesn't introduce asymmetry..

Cheers, John
 
jneutron said:


Good info.

Where I used to work, we would build mil bridges, and three diodes were used per leg in the three phase units, a total of 18 individual 3 amp devices in each unit.

Long Island -- some very interesting electronics manufacturers there -- my 2kV Kepco power supply -- just a curiousity -- was from Kepco -- got my first scope from Grumman. Is AIL still out there?

well, I have the power supply test bed almost finalized -- it has the option of injecting noise or pulses into the primary or secondary. somehow we will be able to produce empirical results which clear the haze.
 
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.