Center tapped transformer rectifier question

[Dannyx]

Re: Center tapped transformer rectifier question

There's no need for a bridge for both supplies. You can full wave rectify both outputs with just 4 diodes, and still have a common earth. The attached diagram is similar to the one mariushm posted, except my one has the smothing cap -ve for the higher voltage connected to earth rather than the half-voltage output. (either configuration will work).

The 48Vdc output uses the 4 diodes as a bridge. The 23Vdc output uses D3 & D4 with reference to the centre tap to full wave rectify the output.

D3 & D4 will therefore be passing the combined current taken from both supplies, so will need to be rated accordingly.

In my case D3 and D4 are built into the bridge rectifier itself, though it's a 50a bridge, so I doubt it will complain.

[Radio Fox]

Re: Center tapped transformer rectifier question

What do you think of the one I posted then ? Would that work ? It's like starting off by rectifying just one of the windings (with respect to the center tap) and then adding an additional bridge for the other one too so in theory nothing should go boom.

There's no need for a bridge for both supplies. You can full wave rectify both outputs with just 4 diodes, and still have a common earth. The attached diagram is similar to the one mariushm posted, except my one has the smothing cap -ve for the higher voltage connected to earth rather than the half-voltage output. (either configuration will work).

The 48Vdc output uses the 4 diodes as a bridge. The 23Vdc output uses D3 & D4 with reference to the centre tap to full wave rectify the output.

D3 & D4 will therefore be passing the combined current taken from both supplies, so will need to be rated accordingly.

[Dannyx]

Re: Center tapped transformer rectifier question

As long as someone doesn't go and think "oh, this GND is black. That GND is black. It'll be just fine if I hook them up together..."

Nobody will tinker with the insides of this power supply and I'll label them Fan GND and supply GND

[eccerr0r]

Re: Center tapped transformer rectifier question

They don't: one runs the supply, the other runs the fan - I'd say that's pretty separate

As long as someone doesn't go and think "oh, this GND is black. That GND is black. It'll be just fine if I hook them up together..."

[goontron]

Re: Center tapped transformer rectifier question

Your diagram shows 2 diodes in series directly across the outer winding of the transformer.

Something's gonna blow!

Oh shit.... I need to pay attention better when designing schematics...

[Dannyx]

Re: Center tapped transformer rectifier question

The double bridge solution requires that you keep the two circuits isolated from each other - they must never touch.

They don't: one runs the supply, the other runs the fan - I'd say that's pretty separate

[eccerr0r]

Re: Center tapped transformer rectifier question

Go with mariushm's solution if you want a common 0V point.

None of the other ones will work trying to have a common ground point, at best. The halfwave version will work with a common ground point but has awful PF, awful regulation, and will require double sized transformer; and the double bridge solution requires that you keep the two circuits isolated from each other - they must never touch.

[budm]

Re: Center tapped transformer rectifier question

Your diagram shows 2 diodes in series directly across the outer winding of the transformer.

Something's gonna blow!

Two more diodes are also connected across the center tap and the top and bottom windings.

[Dannyx]

Re: Center tapped transformer rectifier question

What do you think of the one I posted then ? Would that work ? It's like starting off by rectifying just one of the windings (with respect to the center tap) and then adding an additional bridge for the other one too so in theory nothing should go boom.

[Radio Fox]

Re: Center tapped transformer rectifier question

Keep it simple?

Your diagram shows 2 diodes in series directly across the outer winding of the transformer.

Something's gonna blow!

[eccerr0r]

Re: Center tapped transformer rectifier question

mariushm has the only workable solution as far as I can tell - only 4 diodes needed and though his drawing only has 7.5V windings, you can use full 12V (24V/CT). A lot of these other "solutions" don't quite work because the two windings are not isolated from each other. If there is a need for isolation of the two GNDs (i.e., 0V potentials), then you'll need two isolating switching converters to get this done or another transformer that has isolated secondaries...

Also note that the capacitance on the output legs will need to be fairly large to filter the higher voltage output...

[Dannyx]

Re: Center tapped transformer rectifier question

How does THIS looks ? It's actually the original idea I had and I then thought I could simplify by omitting two diodes. One of you guys posted something similar, but the values were off, so THIS is what I believe it should look like.

[goontron]

Re: Center tapped transformer rectifier question

It seems i forgot C3 between the 18v rail and 0v rail.... Ooops.

[goontron]

Re: Center tapped transformer rectifier question

^ Positive.

[petehall347]

Re: Center tapped transformer rectifier question

Keep it simple?

is the 9v plus or minus ?

[goontron]

Re: Center tapped transformer rectifier question

Keep it simple?

[Dannyx]

Re: Center tapped transformer rectifier question

big clive would run the fan from mains with a capacitive dropper!!

That would work too, but the amount of components required to do so outweighs the benefits of using the unused winding like I want to do, plus it's isolated and "to standard" so to say

[stj]

Re: Center tapped transformer rectifier question

big clive would run the fan from mains with a capacitive dropper!!

[Dannyx]

Re: Center tapped transformer rectifier question

The "high" output of the transformer powers one of those ebay power supply modules and I also need a lower output voltage to power a DC-DC converter which runs a cooling fan for the thing...won't go into too many details on how I built this, but I tested it extensively with a variable laptop power brick which goes up to 24v and it works as it should, regulating the output to the fan based on temperature (nothing WOW there), so the voltage, while not too critical, I wouldn't dare let it go beyond 24v either. Current draw ? I'd say around 1A absolute maximum, if that (perhaps if the fan stalls or jams, worst case scenario).

[mariushm]

Re: Center tapped transformer rectifier question

Something like this would probably work :



Note that with two 18v windings you get 36v AC and that's rectified to DC voltage by the bridge rectifier and you get 36 x 1.414 - 2 x voltage drop so actually you'll have a DC voltage with a peak of ~ 50v
Also note that it's probably a 230v to 36v transformer, but here at around 3-4 am where there's small load on the network, i get up to 245-248v at the mains socket which is still legal (it's 230v +/- 10% in Romania and most of Europe, so up to 253v is legal)

The point is with 245v at the input, you're not gonna have 36v you may have 48-40v ac , and you may have 50-55v peak dc voltage. So for example, use capacitors rated for 63v or with higher voltage rating.

As for your 18v requirement ... you'll have 18v rms , if you rectify it to dc you'll have much more. what voltage you actually need? And more to the point, how much current you expect you're gonna need?

If it's just let's say 50-100mA for some opamps or something, you could just use a LM317 HV (which works up to 60v input voltage) adjustable linear regulator to and get it over with .. 40-50v in (since i doubt you're gonna install 100000 uF) , 18v DC out at 100mA means it would only waste (~40v-18v) x 0.1A = 2.2 watts of heat, which can be doable with a heatsink. Or, you can use a 3-5w resistor to drop some voltage before the regulator and then some power would be dissipated on the resistor and some on the regulator ... the regulator would only need around 2v above the output voltage to work.

For lower currents you could even use a zener diode... see App Note AN954 page 6/14 Resistive Transformerless Power Supply : https://cdn.badcaps-static.com/pdfs/...5df14516be.pdf
... change the zener diode to 15v one or 18v or whatever you're happy with, change the resistor to the value matching your peak current (keeping in mind that you'll have ~ 36v AC and not 110v/230v AC) and everyone should be happy.