Not overkill 
There are many choices,
GSIB2560-E3/45 - VISHAY GENERAL SEMICONDUCTOR - BRIDGE RECTIFIER, 25A, 600V | CPC
or this style,
GBPC3504 - FAIRCHILD SEMICONDUCTOR - BRIDGE RECTIFIER, 35A, 400V | CPC
Everyone has there own ideas... I would suggest somthing like two 4700uf to 6800uf reservoir caps and also a simple snubber network of a 0.1uf 400vac poly cap in series with a 4.7 ohm 1 watt resistor connected across the combined secondary winding of the transformer (so that is across the input to the bridge)
There are many choices,
GSIB2560-E3/45 - VISHAY GENERAL SEMICONDUCTOR - BRIDGE RECTIFIER, 25A, 600V | CPC
or this style,
GBPC3504 - FAIRCHILD SEMICONDUCTOR - BRIDGE RECTIFIER, 35A, 400V | CPC
Everyone has there own ideas... I would suggest somthing like two 4700uf to 6800uf reservoir caps and also a simple snubber network of a 0.1uf 400vac poly cap in series with a 4.7 ohm 1 watt resistor connected across the combined secondary winding of the transformer (so that is across the input to the bridge)
heh... diode bridges cost about as much as a diode well that means i dont have to make one myself
http://www.datasheetcatalog.org/datasheets/166/250526_DS.pdf
it says max current with 300cm^2 fin is 10A... if i devide 300cm^2 by the 10A, i get about how much cm^2 i need to cool that current, dont i? (i know it would be really rough, but if i round it up, i think there shouldent be a problem...)
well that means i dont have to make one myself http://www.datasheetcatalog.org/datasheets/166/250526_DS.pdf
it says max current with 300cm^2 fin is 10A... if i devide 300cm^2 by the 10A, i get about how much cm^2 i need to cool that current, dont i? (i know it would be really rough, but if i round it up, i think there shouldent be a problem...)
Lets just go back a bit...
Designing an amp for 100 watts or even just 40 watts RMS and designing it so that it can supply that on a continuos basis is a whole different ball game to a nominally specified 40 or 100 watt amp that would be used for normal domestic listening. Heatsinks, transformers etc need to be much larger. Most amps you buy fall into the latter category in that they are for domestic use. Its a big big difference you have to understand.
So for "normal" use your transformer and bridge are fine. The bridge (if its a block type) can be bolted to the chassis if you wish.
Caps on or before the bridge are to stop "ringing" caused by the commutation of the bridge as it comes in and out of conduction... remember I said earlier that it was non conducting for most of the time... it is that switching each cycle that can cause the ringing in combination with the inductance of the transformer.
One cap and one resistor is all you need across the input to the bridge.
There's nothing to draw really. It's just the basic transformer/rectifier reservoir cap arrangment.
Designing an amp for 100 watts or even just 40 watts RMS and designing it so that it can supply that on a continuos basis is a whole different ball game to a nominally specified 40 or 100 watt amp that would be used for normal domestic listening. Heatsinks, transformers etc need to be much larger. Most amps you buy fall into the latter category in that they are for domestic use. Its a big big difference you have to understand.
So for "normal" use your transformer and bridge are fine. The bridge (if its a block type) can be bolted to the chassis if you wish.
Caps on or before the bridge are to stop "ringing" caused by the commutation of the bridge as it comes in and out of conduction... remember I said earlier that it was non conducting for most of the time... it is that switching each cycle that can cause the ringing in combination with the inductance of the transformer.
One cap and one resistor is all you need across the input to the bridge.
There's nothing to draw really. It's just the basic transformer/rectifier reservoir cap arrangment.
Like this. Use around the values suggested earlier such as a 0.1uf and 4.7 ohm.
Grounding is SUPER SUPER important and makes or breaks an amplifier.
Read this
http://www.diyaudio.com/forums/solid-state/101321-3-stage-lin-topology-nfb-tappings.html#post1623053
http://www.diyaudio.com/forums/soli...-lin-topology-nfb-tappings-2.html#post1624677
Grounding is SUPER SUPER important and makes or breaks an amplifier.
Read this
http://www.diyaudio.com/forums/solid-state/101321-3-stage-lin-topology-nfb-tappings.html#post1623053
http://www.diyaudio.com/forums/soli...-lin-topology-nfb-tappings-2.html#post1624677
Attachments
Not really
I mean the way you arrange and connect all the grounds electrically because although they are essentially the same point, the resistance of the wire and PCB print means that electrically they are not. This is the number one cause of amps that are unstable and amps that have noise and hum and buzzes. A passive equalizer like that is BAD
Don't even think about it...i knew that it wouldent be good in the beginning, but few moments later, (when i was going to sleep) i realized, that eaven if it would not have noises, then with lower hz tuners i would also adjust the higher pitches
later today i realized, that i can make an equalizer as another project and use the 5v of pc usb port as supply voltage for an active equalizer... but thats for another time
later today i realized, that i can make an equalizer as another project and use the 5v of pc usb port as supply voltage for an active equalizer... but thats for another time
http://www.datasheetcatalog.org/datasheet2/9/0ozxelyhtgfl5jsx9108fj3sqq3y.pdf
so i wanted to choose a heatsink for the chip... i know that the formula is
T(Max)=P(dissipated)*(R(chip)+R(from chip to heatsink)+R(heatsink))+T(room)
so from the datasheet i get it like this
70=35(1.5+0.25+X)+25 (the room temperature is about 25, and with termopaste R(from chip to heatsink) is about 0.25)
so i will decrease the max temperature by 15 (just for safety)
55=35(1.5+0.25+X)+25 --> 30/35=1.75+x
0.86=1.75+X --> X=-0.89 ... my question is... why is it negative... the rezistance cant be negative...
so i wanted to choose a heatsink for the chip... i know that the formula is
T(Max)=P(dissipated)*(R(chip)+R(from chip to heatsink)+R(heatsink))+T(room)
so from the datasheet i get it like this
70=35(1.5+0.25+X)+25 (the room temperature is about 25, and with termopaste R(from chip to heatsink) is about 0.25)
so i will decrease the max temperature by 15 (just for safety)
55=35(1.5+0.25+X)+25 --> 30/35=1.75+x
0.86=1.75+X --> X=-0.89 ... my question is... why is it negative... the rezistance cant be negative...
You need to fully understand the problem and its one whole big subject in itself.
The main points are that the connection between the two reservoir caps is not to be used as a ground point itself. You take a small (any length) spur off that connection and the end of that becomes our reference point or audio ground. Its electrically clean and has no current from the PSU lowing in it. This point is used as a star that the other signal grounds return too.
Heatsinks... I've never done thermal calculations and always choose empirically with an eye on the degrees C/per watt rating
Again this is where the difference between pro and domestic applications come to the surface. The TDA7293 type chips might be rated at 100 watts RMS but in practice you are never going to get that heat away from the chip fast enough. So you compromise and choose something thats acceptable for your use.Something like this,
350AB1000B - ABL HEATSINKS - HEAT SINK, 0.67°C/W | CPC
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.