Ultrafast Diode bridge substitution

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
I've come to a delema in my car amp project planning.
I definatly need fast diodes for the rectifing but I don't have access to any bridges that even mention fast.
the Biggest diode I can get are 1 amp 100v (or 1000v bu it wno't make 2 much difference).
Can these be used for 6amp purposes with heatsinking?
how many would be needed to pull this off?
 
fr0st,

No, you can't. But, if they were ultrafast too you can put six of them in paralell to achieve your current needs. It's not the best option, in my opinion and I've never tried it before with six diodes, but, considering your information, you have no choice.

Regards
 
See the Michael Percy Audio catalog for a listing of available high speed rectification diodes and bridges. He lists IXYS 20A, 600V high speed bridges but they are expensive: $17 US. I just bought some discreet IR HEXFRED 8A, 600V diodes from Digikey for $1.32 US each (qty = 10). Don't know about sources for these parts on your side of the globe.

Ren
 
bahh.... I should never post when tired
I found some TO 220 15a fast diodes in a electronics catalogue of a place near me which suits me perfectly.
I still have a question about the core sizing.
can a 3.5cm core withstand 400watts?
It will have cooling (fan forced) and suitable gauge winding.
Thanks for your help
 
fr0st said:
I still have a question about the core sizing.
can a 3.5cm core withstand 400watts?

I assume when you say 3.5cm, you mean the area of the core in square centimeters? If so, then it depends on the average path length of the core. (ie, it's basically the mass of the core, not just the area)

I have a core here, it's area is about 1.38cm squared, and the effective length of the core is about 13cm long. It's a torroid about 4.8 cm od, 3.2cm id, by 1.7 cm thick and I've tested it at 1300W. So if you're anywhere near that size at 50Khz, I'd say 400W should be no problem.

BTW, if anyone knows how to calculate the power a core is good for, I'd be interested in knowing how do to it myself. I've always used the WaAc method from the Mfr specs. I still think that there is a better way of doing it. (Maybe what I really need is a good book on magnetics for SMPS supplies???)

-Dan
 
dkemppai said:


I assume when you say 3.5cm, you mean the area of the core in square centimeters? If so, then it depends on the average path length of the core. (ie, it's basically the mass of the core, not just the area)

I have a core here, it's area is about 1.38cm squared, and the effective length of the core is about 13cm long. It's a torroid about 4.8 cm od, 3.2cm id, by 1.7 cm thick and I've tested it at 1300W. So if you're anywhere near that size at 50Khz, I'd say 400W should be no problem.

BTW, if anyone knows how to calculate the power a core is good for, I'd be interested in knowing how do to it myself. I've always used the WaAc method from the Mfr specs. I still think that there is a better way of doing it. (Maybe what I really need is a good book on magnetics for SMPS supplies???)

-Dan
This is my first SMPS so i'm still learning the parmeters to quote.
The core 35mm Od, 21mm ID and 13mm high.
I'll just assume it'll work and I'll incorerate a thermal fuse just incase.
Thanks
 
fr0st said:

This is my first SMPS so i'm still learning the parmeters to quote.
The core 35mm Od, 21mm ID and 13mm high.
I'll just assume it'll work and I'll incorerate a thermal fuse just incase.
Thanks

If it's 35mm od, it'll probably do 400W (or close to it anyway). I'm guessing that you probably won't need to thermally fuse the transformer, most of your heat will probably come from the power switches (mosfets).

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