Toroid/Transformers - Better as Parallel or Series?

Thread Starter

Still Learning

Joined Jul 19, 2010
25
Looking to make a PSU as first project, to power my later projects (makes sense right?).
Wondering, if I get a Toroid/transformer with 4x secondary wires, is there a preference to connect the centre wires to convert it to a Centre Tap, or to use a full bridge rectifier? C.Taps just seem easier to me..?
I understand the difference in the Ampere outputs in the different cases, is this the only reason to connect them in this way?

Comments/Suggestions?
 

Norfindel

Joined Mar 6, 2008
326
The two diode wit center tap full-wave rectifier is only used for small voltages. It's otherwise not worth it.
You can buy an integrated bridge rectifier. That would make it look much simpler.
 

Thread Starter

Still Learning

Joined Jul 19, 2010
25
So for small voltages, look at a CT transformer. Got it..

So, what is a small voltage to you guys?
For a 0v-30v, maybe 5Amp bench PSU, I should be looking for a toroid with the 4 wires, connecting them in Parallel, otherwise I don't get the bigger Amp output right?

The toroid Im looking at is (is this overkill for a PSU?):
Certificate No.: CS 815/Q
Specifications
Primary voltage: 240V AC
Sec voltage: 30V + 30V AC
Regulation: Better than 10%
Maximum temperature rise: 75°C
Dielectric strength: 4000V for 1 Minute
Insulation class: Class B (130deg.C)
Sec. Parallel: 5.3A
Sec. Series: 2.6A
 

Norfindel

Joined Mar 6, 2008
326
So for small voltages, look at a CT transformer. Got it..

So, what is a small voltage to you guys?
Like 6vdc output, low-current. It's worth it when the size of the transformer is small, and the voltage drop of the diodes is significative when compared with the output voltage, but you will be using a 6+6v transformer instead of a 6v transformer. Not too much of an issue in this case, but if you need a 30+30v 5A transformer instead of a 30v 5A transformer, it's very significative.

For a 0v-30v, maybe 5Amp bench PSU, I should be looking for a toroid with the 4 wires, connecting them in Parallel, otherwise I don't get the bigger Amp output right?
Yes, paralleling the two secondaries would give you the higher current rating.
However, exactly how are you going to regulate the voltage? Take into account that with linear regulators, the power dissipation at voltage close to 0v 5A will be higher than 150 watts.

The toroid Im looking at is (is this overkill for a PSU?):
Certificate No.: CS 815/Q
Specifications
Primary voltage: 240V AC
Sec voltage: 30V + 30V AC
Regulation: Better than 10%
Maximum temperature rise: 75°C
Dielectric strength: 4000V for 1 Minute
Insulation class: Class B (130deg.C)
Sec. Parallel: 5.3A
Sec. Series: 2.6A
[/quote]
That depends on the circuit. If you're going to use a bridge rectifier and pure capacitive filter, the current rating of the transformer must be roughly 1.8 times the output current.
 

Potato Pudding

Joined Jun 11, 2010
688
Don't choose between 30 Volt 5 Amp and ±30 Volt 2.5 Amps.

Make it side by side 30 Volt 2.5 Amp power supply outputs with a mode switch for either configuration and also useful as two separate supplies in the same case.

You can even have a mode to put ground in the center of either or both sides and have 2 of ± 15v supplies that you could possibly still parallel for 5A.

Build variable (programmable) current limiters into all outputs is the only caution I would add to suggesting you try for all the options.

You want variable because most of the time you will want much less than 5 Amp current limits, so build it to allow 50μA to 5A settings. That will save you from at least some of the oops connections that you might make with so many ways to use your power supply.

But for starters make it work as 2 x 30V 2.5 Amps.
 

Thread Starter

Still Learning

Joined Jul 19, 2010
25
Ok, Im starting to think I don't need anywhere NEAR that amount of Amps now. Im thinking only around the 1Amp mark. So would a L317 regulator do the job with 30Volts, or do I maybe need a smaller XFormer - maybe just 18VAC which will give me around the 12-15V mark?

..Going off to do some more research now...
 

Norfindel

Joined Mar 6, 2008
326
ok... I've settled on this schematic setup using a 25V 2Amp toroid I can get hold of, and wiring it in parallel then to a bridge rectifier as in the diagram. I should have just done this to start with as it is the schematic i was looking at all along. It seems correct to me (i've been over it sooooo many times).

http://www.qsl.net/yo5ofh/projects/vps/vps.htm
Seems about right. You can improve ripple rejection by adding another capacitor in the adj pin, and a diode. Take a look at C2 and D2 in the figure 4 of this application note: http://www.national.com/an/AN/AN-181.pdf

Also, take into account that the minimum required load current for the 317 can range from a typical value of 3.5mA all the way up to 10mA.
With a 220Ω resistor from out to adj, about 5.5mA will flow right there, so there won't be any problems if the LM317 needs a minimum current lower than that, but if you happend to have a device that requires more minimum current, the regulation with very light loads (less than 4.5mA) could be worse than expected.

It's probably not something to worry much about, as National says that "Usually the 5 mA programming current is sufficient" in their application note from 1975, just know that it can happend.

National recommends solid tantalum capacitors for this IC, as they have a low impedance at high frequencies.

EDIT: I almost forget, that the current limit on the LM317 varies with the IC's input-to-output voltage difference. It will be lower than 1A if the input-to-output differential is higher than 15v. Take a look at the "Current Limit" graphic in page 6 of the datasheet: http://www.national.com/ds/LM/LM117.pdf

Do you already know what heatsink to use? (the IC has thermal protection, anyways, so it isn't likely to just explode, but you need a big enough heatsink for the device to work normally).
 
Last edited:

Thread Starter

Still Learning

Joined Jul 19, 2010
25
Thanks for the tips! I'm just gathering all my components at the moment, going for the Solid Tatalum Caps, and the extra 10uF over the ADJ pin...

Also, say I wanted to put an extra set of outputs with the straight AC from the transformer, is it as simple as bypassing the Rectifier of the same Sec output wires and connecting them direct to front of case, or should I be looking for some kind of regulation on them inside the box instead?
I'm just considering this; so to have the extra ability to go direct @ (about) 25Vac for testing prototypes later on. I would probably then use a transformer inside the final design in the end, but the practice at converting AC/DC would be useful, yes?
 

Norfindel

Joined Mar 6, 2008
326
Yes, you can do that. Get the AC voltage after the fuses, so that they're also protected by them.
Take care at not exceeding the tranformer's current rating. The regulated part has current limiting, but fuses can take a long time to blow, unless the current is very high.
Also, i think that the sine wave is likely to be distorted, as the load isn't pure resistive. To rectify it, would be good enough, however.
 

marshallf3

Joined Jul 26, 2010
2,358
If you can find a dual secondary (not just center tapped) transformer you can hook up a DPDT switch and make the basis of a dual voltage/current range supply.

I sure wish toroids weren't so hard to wind, I've got a lot of oddball voltage but high power ones I'd love to reconfigure. Even a simple 100 turns of #20 AWG by hand is a pain to wind.

It isn't too bad if you just want to go down in voltage and the secondary wire is thick enough to provide the needed current, you just remove turns and that goes fast enough. A digital micrometer to measure the diameter of the secondary wire along with a very accurate ohmmeter will get you in the ballpark of how many turns is already on it.

[HINT:] May not be many left, but if you can find an old satellite shop (back in the 10' dish days) that doesn't throw anything away and has accumulated a pile of the really old antenna arm actuators most every one I've taken apart has a 24 VAC 4A toroid in it along with some pretty decent relays.

As they slowly disappear, it pays to hit up old electronic repair shops. Most of these places are still barely surviving and cash often speaks pretty well to a hungry owner.

If you ever find a half-decent vacuum tube tester grab it, they can bring high $$$ on eBay. I'd love to still have my old B&K 747 and my 470 CRT tester. Any old stash of vacuum tubes can also be a bonanza. While usually more than half of what they'll have left on the shelf will be too TV specific to be of any value to ham radio operators but things like forgotten 6L6GTs and 12AX7s are like gold anymore if they're from the days when they were still made by the reputable firms.

Old equpment too. I once scarfed up a very old HH Scott stereo receiver that had a bad power transformer. Whille it really wouldn't have been practical to fix since it wasn't one of the highly sought after models it was loaded with Telefunken and USA made RCA tubes. Some of these tubes, not by number but more by certain build types, can be worth a fortune and I got $300 for a set of 4 of what to most people would have just been a plain 12AX7.
 

Thread Starter

Still Learning

Joined Jul 19, 2010
25
Took on all your tips, after a long time between planning and getting to the creating part! I'll post the schematic in a few for any more tips or hints..!

Thanks again guys/(girls?)..!
 

Thread Starter

Still Learning

Joined Jul 19, 2010
25
Just checking if the bottom schematics look ok before I continue. Have found no netlist errors within Express SCH, but that doesn't mean I'm on target...

Also, the second schematic with the Bypass Transistor should give me more Amp Output, up to 4A as long as I dont go over 40v yeah???

Any help appreciated..!
 

Attachments

Top