the STPS16L45CT in the data sheet doesnt say the forward voltage drop so why do you choose 600mv?
Mostly because I had to choose something. It seemed a reasonable value.
As to why I thought that, and whether my conclusions are valid... that's another story. It also might be one we need to investigate further later on.
does 2 X 8 A mean 16 amps total?
Yes. Each individual diode is rated at 8A and they are used in parallel, so you can pass 16A through them in totsl.
the reverse voltage I think is the vrrm which is 45 V right? also how much can I exceed here?
The actual value you require is hard to determine. However knowing that the part being replaced is rated for this, and knowing that this is a limit, beyond which the device may behave unexpectedly (or be destroyed), it is wise to select a replacement with the same or higher rated reverse voltage.
In essence there is no limit as to how high you might go, however some characteristics of the device will suffer as you go in this direction. The one which is possibly most important here is the forward voltage drop. So if you go too far in reverse voltage rating you may find that the forward voltage drop (that you normally want to be as small as possible) starts to get too high.
you said you selected forward currents 8a-15a, would like to know how u got that from the datasheet, also how much can I exceed.
Again, this is a limit. The original part's datasheet specified this as 2 x 8A (so 8A per diode). As a limit, I can go higher, but not lower. And also as you go higher other characteristics may suffer (reverse voltage, forward voltage drop) so you don't want to go too far. I decided that up to about double was OK.
also the isolated tab you mean ontop where u put the screw to a heatsink right?
Exactly.
If that heatsink is connected to ground, or if there is another device attached to it then the isolated tab means the cathodes of these diodes will not be shorted to them.
Something I didn't check (but you should) is that the thermal resistance (junction to case) is similar for the devices. It is either 1.3 or 3.5 deg C per watt depending on the case style. Given that manufacturers don't put oversized heatsinks in power supplies, you need to get something with a similar rating so it doesn't risk overheating.
In this case, with a higher Vf, the dissipation will be higher and it might be wise to determine if the junction temperature is going to be a limiting factor.
metal tag at the top is not connected to the diodes. didnt understand this
Yes, and that can be a really big trap.
Knowing it has a metal tab, I now know you have the TO-220AB version, not the TO-220FPAB version (I probably picked that up earlier though)
Assuming these run at 16A, with a 0.5V drop, they are dissipating 8W. If the heatsink is 10 degC/w, then the junction rise will be around 8 * (10 + 1.3) = 90 degC. If ambient is 30 degC then the junction temp may be as high as 120C. (most of these figures come out of thin air). You would need to repeat these calculations with the old and the new device, with figures based on reality rather than guesswork to see if the replacement will operate at a safe temperature.
Thank for giving me some pointers, sorry for all the question
As you can see, it's a non-trivial problem, so it is good that you ask questions.
