In theory it could be simplified to that but it will probably don't work in real life. If you want to experiment I recommend that you simulate first with switchercad.
http://www.linear.com/software/
Also search the internet for information. Some interesting stuff can be found at:
International rectifier
Also at:
ESP
And in the power supply section of:
Epanorama
http://www.linear.com/software/
Also search the internet for information. Some interesting stuff can be found at:
International rectifier
Also at:
ESP
And in the power supply section of:
Epanorama
Hi,
I saw a schemeatic on the net for a 24V@40A (!!!) bench PSU that was switched mode...I would steer clear...All the inductors had to be self wound and moulded using epoxy etc etc...
It was no beginners project - that's for sure (I'm not suggesting you're a beginner...Just stating a point)
Good luck though,
Gaz
I saw a schemeatic on the net for a 24V@40A (!!!) bench PSU that was switched mode...I would steer clear...All the inductors had to be self wound and moulded using epoxy etc etc...
It was no beginners project - that's for sure (I'm not suggesting you're a beginner...Just stating a point)
Good luck though,
Gaz
hacknet said:
There is a lot more to it than that!
Few switch mode power supplies use air core indutors, for one, and you frequently need a custom designed and built inductor, unless you are using one of the special switchers manufactured by National or Linear Tech that are designed around standard inductors specially made for these devices. If you are trying to directly convert AC rather than first convert it to DC with a rectifier and step down transformer, these devices won't work.
You need to design what is called an "offline" power supply, which is short for "off of the line", where the AC mains are directly switched by a high voltage MOSFET. Do a search at companies like Texas Instruments (who bought out Unitrode, a major manufacturer of power supply controllers) and ON semiconductor for offlline power supply controllers for some design ideas.
Note that in order to control the output voltage of the supply, the period that the MOSFETs are conducting has to be controlled precisely. For lower input voltages (low line), the ON period is low, and it has be ON longer for higher line voltages. A simple oscillator driving a MOSFET will be unable to regulate the output voltage. So, you need to use some sort of pulse width modulator with feedback. Also, as the output load varies, the output voltage will vary in relation to losses in the switching transformer, MOSFETs and wiring. Voltage feedback is needed, therefore, to compensate for both line and load variations. I don't think a 555 could give very good performance.
You have a very ambitious project to get +/-35 volts at 10A directly of the mains without an AC transformer. I really think this is not a project for the average DIY'er, but if you want to go for it, good luck - just do your research and be VERY SAFE!!!
Having said all that, and reading numerous warnings about how EMI mucks up the sound, there are advantages...
Chord (UK company) swears by them. They have 3 PSUs that that put into all of their products depending on size. They have great regulation and are cheap to build...at least on large scale...
As the advert says...This is what happens when aerospace meets audio (They guy who started Chord used to work in Aerospace)
Gaz
Chord (UK company) swears by them. They have 3 PSUs that that put into all of their products depending on size. They have great regulation and are cheap to build...at least on large scale...
As the advert says...This is what happens when aerospace meets audio (They guy who started Chord used to work in Aerospace)
Gaz
Modern, high quality switch mode power supplies can be very efficient and relatively low noise. I have designed them for military applications where the general standard is for 1% line and load regulation and less than 1% peak ripple voltage. To get noise even lower, you can use additional inductive filtering and, as always, a ton of low ESR capacitance, but using low noise design techniques (i.e. ground plane, minimal loop area, etc) is paramount as it is easy to radiate a whole lot of stuff.
For an audio application, I would certainly use toroidal inductors, soft switching techniques and lots of shielding. With typical 200 KHz switching frequencies, a lot of the switching noise will be above the closed loop response of the power amplifier. I still have my doubts about how well it would work for an audiophile application, though. Maybe I am just old fashioned.
I have never designed an offline power supply, though I have seen some designs where the AC mains are directly switched. The thought of doing that as a DIY project makes my hair stand on end due to the danger, though I have little doubt that there are those who can do this at home.
For an audio application, I would certainly use toroidal inductors, soft switching techniques and lots of shielding. With typical 200 KHz switching frequencies, a lot of the switching noise will be above the closed loop response of the power amplifier. I still have my doubts about how well it would work for an audiophile application, though. Maybe I am just old fashioned.
I have never designed an offline power supply, though I have seen some designs where the AC mains are directly switched. The thought of doing that as a DIY project makes my hair stand on end due to the danger, though I have little doubt that there are those who can do this at home.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.