Hi, thanks for all replying. 
-Why am i trying to make it myself?
There are two reason.
First, if i make it, i can learn how power supplies works.
Second, in Turkey, lab power supplies are too expensive. I can make it cheaper than buying.
So, first of all, i want to learn how they works and i would like to make it. Just, i need some information about power supplies.
Best regards,
Karamel 
Hi
I would be very careful about the "to expensive here" part of the argument. The same rules / taxes / shipping / hassle that make a $20 used lab supply expensive also run up the cost of getting *all* of the right parts.
=====
The first part in any design is to decide what you want to end up with. That certainly includes voltage and current. It also includes issues like the enclosure, heatsinks, fans, fuses, metering, and RFI filtering.
Next step is to take a rough look at this and evaluate a couple of design approaches. This lets you evaluate your decisions in the first section and likely you find gaps (problems) with your specifications.
After that you loop back to the start and adjust your specifications a bit to line up better with what a practical design can accomplish. Then you head back through the rough design phase and see how that stacks up. You may loop several times.
Once you have that part fairly well thought out, you can start looking at pc board layout software and the various tools for analyzing a design. You may also start investigating surface mount construction techniques and the test gear needed to check out your design. All of these details are as much part of the design as anything else above.
Note that there is no money spent so far.
You should have enough information at this point to make a guess at what all the stuff needed to do this will cost. Tools to fabricate sheet metal, reflow soldering gear, DVM's, a scope probably all are part of it. They are not directly a cost of the supply, they are a cost of the project. You may or may not be able to guess how long it all will take. Try to do so and then keep track of how you do against your estimate. It will help you a *lot* in the future when you have to estimate how long a job will take.
This is the point you actually have a decision to make. So far it's all been talk and emails. You now will be spending money. That's not a reversible thing. Check over all of your assumptions. Re-check your design. Get in the first (small) group of parts. Get enough gear to check them out. Validate a small portion of the design, see how it works, make sure you have it right. Move on to the next section.
Once you have all of the prototype assemblies worked out, the result will be a giant bunch of wires all over your bench and a few parts buried under them. The result is sort of a power supply, but not anything you can use. You then take what you have learned on each little assembly and roll it up into a final design. You do the various pc boards, you start on the sheet metal. You now order the final parts.
Once all the parts come in, you put it all together and see how it works. If you are like me, it will be <90% correct. The decision then is if you do another pass at the finished product.
Bob
Home
Help
Search
Login
Register
