300w boost converter ..take 2

[oztules]

The power circuit is really only the thick tracks between the fets and the diodes, the square input pad and the longer input pad and the output pad.

5 track/pads in all are the only ones to see current above milliamps.The power handling tracks are all solder reinforced.

If you look at the pic of the board you will see a fuzzy outline (through the board) around the power tracks, this is excess flux on the outside of the tracks. Will post pics later today of the underside showing the solder on the power track area..... same as in the commercial inverters. There will usually be gobs of solder on the power tracks to aid current flow. The ones on this board could carry in excess of 70 amps if required. The solder is about 1/16 thick at least.

The inductor wires go directly to the high current pads. The pads under the transformers are only used as anchors. The 1.8mm wires go straight to the b+ pad (square one) and the long track on the edge of the board at the Fet drain. That track is 1/16th inch deep in solder the whole length. If I need nore power, that track will have a copper bus embedded into it.

..........oztules

[oztules]

Sorry for the slow response,couldn't upload as the speed had drifted down to barely usable in the last few days. So here it is:

another view of the rear end shows a single fet doing the work. The three D92's are the rectifiers. Together handle nominally 120A.

With the current clean square wave, the fet doesn't even begin to warm at 100W. As you can see, no cooling at all, so I gives us some idea how efficient these boosters can be.

You will see the leads of the coils going into the board two at the rear edge and two go to the right of the big cap into the positive pad.:

So wired like this, current is of no concern to the tracks.

Any other questions?

............oztules

[willib]

" The three D92's are the rectifiers. Together handle nominally 120A."

the D92's are paralleled?

you really need a schematic of this..

that looks cool , though

[willib]

this is the improved wave , with the 17ga. coil i believe it was,

you can see , maybe , there is still a bit of ringing as it recovers from a pulse

http://www.otherpower.com/images/scimages/2965/improvednoflash

this is the first wave ,for comparison , which wasnt good

http://www.otherpower.com/images/scimages/2965/IM002047.JPG

[oztules]

Yes Willib, the D92 are dual 20A 200V diodes.... 40A per package. If you get the wave correct the fets stay cool, but the diodes get warmer than the fets, so they are paralled to keep the heat down.

Your scope looks better, but it looks like you still have slow diodes on it. If you dont want to pull the to220 types out of a supply, use one or more of the fr155 or similar which will be on the computer psu's as well.... not the input diodes which if not bridge, will probably be 1n4004 etc.... slow ones .

Circuit..... then it might look like I know what I'm doing... can't have that...

Just line up the overlay with the board (track) diagram.... better yet, download the .pcb file and a copy of autotrax or easy trax or protel latest offer, and use it on screen then. You will then be able to use a composite to view. (tracks and component overlay concurrently.

...............oztules