Voltage doubler no center tap

[xtian]

I have a question about loaded vs. unloaded voltage. I created a voltage doubler from my 5VAC, 6amp tap on my Hammond 272JX, using the Greinacher schematic <http://en.wikipedia.org/wiki/Voltage_doubler>. I used a pair of 10uF/25v caps.

When I measure the UNLOADED voltage, it reads 15vDC. But this supply is failing to power my LED string. To test my supply, I used a 47R/25W resistor as load, and I see that voltage drops to a piddling 0.5vDC under this load.

In contrast, I have a "9vDC" wall wart that reads 15vDC unloaded, and using the 47R resistor as load, it drops to 11.7vDC.

So, what's wrong with my supply? Why won't it supply enough current? Is it because my caps are undersized?

[xtian]

Back to the main point, here you have the ripple calculation formula:

http://www.voltagemultipliers.com/html/multripple.html

Thanks, but can't figure out how to use the formula. Did I create a TWO stage multiplier? If so, my ripple is 0.375v, which seems like it wouldn't be a problem.

[John_P_WI]

Xtian,

Do you have a cap in series with the winding first? If so, I expect you are limited by the impedance of the cap. With a 47r resistor you would be dropping roughly 1/3 amp IF you could hold the rail at 15v, that is a fair amount of current.

[xtian]

My circuit is just like the attached schematic. Polarized 10uF caps, with + ends as shown. 1N4007 diodes.

So, while there is one series cap, I'm not sure if that's what you were asking.

[xtian]

With a 47r resistor you would be dropping roughly 1/3 amp IF you could hold the rail at 15v, that is a fair amount of current.

I was able to deliver enough current with the wall wart, but not with my own circuit. As i mentioned, voltaged sagged to 0.5v.

[John_P_WI]

Xtian,

I'm afraid that you are limited by the cap in series with the winding.

May I suggest that you try something similar to this:

[img:592:220]http://www.electronic-circuits-for-hobb ... bridge.JPG[/img]

Theory says P=VI, therefore double the voltage - halve the current.

[xtian]

John, I built the FWB as you suggested. I see 15v unloaded, but it still sags to just 4 volts under load. Do I just need larger caps than the 10uF i'm using?

[darryl_h]

I would suggest using this circuit, but with capacitors of 1000mfd or greater, rated at 10volts or greater.

[img:280:215]http://www.daenotes.com/images/full-wav ... oubler.png[/img]

The theoretical output under load is approximately 5 x 1.4 x 2 = 14VDC but at low voltages such as this, the approximate 0.7V forward voltage drop of the silicon diodes must also be considered.

[John_P_WI]

Xtian,

Good news, changing to a bridge increased the "under load" voltage 800 %, going from 0.5v to 4v - a step in the correct direction.

Next, your cap values are extremely low, it is not uncommon to see caps in the thousands of micro-farads in this position.

As Darryl pointed out, the diode drops should be considered, as this is a low voltage application and FYI IF you need a little more voltage Schottky diodes may be used as their voltage drop is less than silicon (1n4001).

I would suggest sussing out the cap values first, in fact, clip lead a couple of 1000 uf of higher caps across the caps already in place, maybe you have some left over from pedal building?.

[John_P_WI]

Delete

[martin manning]

Note that the schematic posted by Darryl above is effectively the same as the one with the four diode bridge. The extra diodes aren't really doing anything.

[cxx]

What's wrong with the standard bridge? I don't imagine it's the .7 V drop or a couple diodes.

What's the split cap wire on the bridge shown? Most I've seen are pretty straightforward, bridge to cap.

[xtian]

Problem solved. I replaced my 10uF caps with 1000uF. Now the voltage drops only to 11v under load, and this is powering the LED string properly. Excellent! Thanks, all y'all.

[JMFahey]

Good

It was clearly a problem of WAY too little caps

You still have 2 diodes, both "bottom" ones in the "bridge" rectifier, doing nothing.

You may (should) remove them and edit the schematic.

Check the classic schematic which only uses 2 diodes.

[LOUDthud]

Good

It was clearly a problem of WAY too little caps

You still have 2 diodes, both "bottom" ones in the "bridge" rectifier, doing nothing.

You may (should) remove them and edit the schematic.

Check the classic schematic which only uses 2 diodes.

Not exactly nothing. Depending on which polarity comes out of the transformer on the first half cycle, one of the capacitors can see a big reverse voltage. Can the caps you are using handle it? Feeling lucky?