Author Topic: Voltage multiplier  (Read 3055 times)

0 Members and 1 Guest are viewing this topic.

Offline sureshotTopic starter

  • Regular Contributor
  • *
  • Posts: 177
Voltage multiplier
« on: July 02, 2018, 08:29:28 pm »
I was looking to increase a linear transformer secondary voltage for input to linear regulators. In the image attached, would this be of use if the diodes where larger say 6 or 10 Amp diodes.
Any help appreciated.
 

Offline Zero999

  • Super Contributor
  • ***
  • Posts: 20355
  • Country: gb
  • 0999
Re: Voltage multiplier
« Reply #1 on: July 02, 2018, 09:02:27 pm »
It will work, as long as the diodes are adequately rated. The main issue with this is the capacitors would need to be huge, since it's two half wave rectifiers.

The correct solution is to simply use the correct voltage transformer, in the first place.
 

Offline sureshotTopic starter

  • Regular Contributor
  • *
  • Posts: 177
Re: Voltage multiplier
« Reply #2 on: July 02, 2018, 09:27:13 pm »
Interesting, thank you for the reply. I did wander what it would be like under load conditions. I was thinking 12 Volt transformer using 2 x 50 Volt 10000uf electrolytic capacitors and 2 x 10 Amp rectifier diodes. For up to 3 Amps output continuous from a boosted 7812 voltage regulator.
 

Offline John B

  • Frequent Contributor
  • **
  • Posts: 865
  • Country: au
Re: Voltage multiplier
« Reply #3 on: July 02, 2018, 10:15:51 pm »
The more useful application in a linear regulator is the use of 2 charge pumps to create low current power rails both above and below the full wave rectified main power rail. These 2 extra rails can be clamped and regulated to power op amps, so that they can sense down to and just below 0V. You have to tweak the impedance of the charge pumps to account for the needed current, mainly to reduce the peak voltages. You can start exceeding the maximum voltages of linear regs etc. But not with a 12V transformer, you need a higher voltage one.
 

Offline sureshotTopic starter

  • Regular Contributor
  • *
  • Posts: 177
Re: Voltage multiplier
« Reply #4 on: July 02, 2018, 10:39:43 pm »
Thank you for the info, yes I won't exceed the transformers current rating at the higher voltage level. I use series pass transistors with linear voltage regulators, it's handy to know if a transformer I already have doesn't come up to the voltage I need this is an option.
Bit of what I put together using cpu coolers.
 

Offline Zero999

  • Super Contributor
  • ***
  • Posts: 20355
  • Country: gb
  • 0999
Re: Voltage multiplier
« Reply #5 on: July 03, 2018, 07:49:55 am »
Interesting, thank you for the reply. I did wander what it would be like under load conditions. I was thinking 12 Volt transformer using 2 x 50 Volt 10000uf electrolytic capacitors and 2 x 10 Amp rectifier diodes. For up to 3 Amps output continuous from a boosted 7812 voltage regulator.
Using a 15V transformer would be better.

Better still, reduce the losses by using a Schottky diode bridge rectifier and a low drop-out regulator, such as the KA378R12, so it will work with a 12V transformer, without the need for a voltage doubler.
http://www.mouser.com/ds/2/149/KA378R12CTU-56633.pdf
 

Offline mariush

  • Super Contributor
  • ***
  • Posts: 5170
  • Country: ro
  • .
Re: Voltage multiplier
« Reply #6 on: July 03, 2018, 08:24:06 am »
Or if you don't care about money, you could replace the bridge rectifier and its ~ 1v voltage drop with a LT 4320 and 4 mosfets and then you'll have only around 0.1-0.2v voltage drop.
It would also allow you to use cheaper capacitors, let's say a couple of 6800uF 25v rated electrolytics or a single 10k uF 25v cap would probably be enough for 3A of current.

The rectified voltage will have a peak close to 16.5-17v, and you only need to get the voltage to always be over 13v or something like that, so that you'll have ~1v voltage drop on the linear regulator. 
You could use LM1085 as a low voltage drop 3A regulator, it has a maximum 1.5v voltage drop.

 

Offline sureshotTopic starter

  • Regular Contributor
  • *
  • Posts: 177
Re: Voltage multiplier
« Reply #7 on: July 03, 2018, 09:29:12 am »
Thanks for the ideas guys. Yes if i need a specific transformer I've found a reasonably priced source for them. Some time ago i put together a "MOT" power supply, and rewound the secondary. But was limited to 14.7 Volts by the time I'd finished winding it. I used a heavy silicon covered wire and even added shrink wrap to insure good insulation. Anyway i used a low drop out voltage regulator for that, can't remeber the part number. It worked fine, i ended up using 90000uf of capacitance to avoid drops in voltage under heavy loads. It worked, but those capacitors where expensive though. I'm wandering if i could have used this type of voltage multiplier to lift that MOT voltage at the time. I've wanted to try a mosfet bridge rectifier for a while, that looks really cool idea. Thanks again for the help and links.
 

Offline sureshotTopic starter

  • Regular Contributor
  • *
  • Posts: 177
Re: Voltage multiplier
« Reply #8 on: July 03, 2018, 09:20:13 pm »
After some practical having a go.... I can see quite quickly what you gain on the swings you lose on the roundabouts. I put some 25 Volt 6800uf in parallel and series combination, I did manage to drag 3 Amps out of a 3 Amp 18 Volt transformer. But with the load I put on was dropping from 57 Volts down to 30 Volts, that was divided up between 4 x halogen lamps. It was a daft load for that transformer, and I only had connected to gather a couple of readings. My conclusion is its useful with in limitations. You would need huge capacitance for even 3 Amps continuous and limited voltage drop. My load was unreasonable, but it's all I had to share that voltage to see. If the load is reasonable it's of some use I think. Thanks again for the help.  :)
 

Offline Zero999

  • Super Contributor
  • ***
  • Posts: 20355
  • Country: gb
  • 0999
Re: Voltage multiplier
« Reply #9 on: July 03, 2018, 10:14:53 pm »
Or if you don't care about money, you could replace the bridge rectifier and its ~ 1v voltage drop with a LT 4320 and 4 mosfets and then you'll have only around 0.1-0.2v voltage drop.
It would also allow you to use cheaper capacitors, let's say a couple of 6800uF 25v rated electrolytics or a single 10k uF 25v cap would probably be enough for 3A of current.

The rectified voltage will have a peak close to 16.5-17v, and you only need to get the voltage to always be over 13v or something like that, so that you'll have ~1v voltage drop on the linear regulator. 
You could use LM1085 as a low voltage drop 3A regulator, it has a maximum 1.5v voltage drop.
A more sensible solution, would be to use the voltage doubler circuit for the circuit driving the pass transistor, which can be connected to the usual bridge rectifier voltage. That will make a low drop-out regulator which will be very stable and the voltage doubler circuit will only need to give a few 100mA to bias the output transistor. A good IC for this is the LM723. If no one beats me to it, I will post a schematic an a day or so, when I have the time.
 

Offline innkeeper

  • Supporter
  • ****
  • Posts: 615
  • Country: us
Re: Voltage multiplier
« Reply #10 on: July 03, 2018, 10:30:32 pm »
hi sureshot.

Some random thoughts here..

transformers are typically (but not always) rated for their output voltage at 100% load
so its not unusual for a 12V transformer to be outputting much higher with low and 0 loads.
you can use this to your advantage and is something to keep in mind.

rectified output voltage - fill bridge that's gonna be roughly  1.414 x the ac voltage - the diode loss so 12V ~= 15.7 and the lm7812 dropout is ~14.5 depending on the model, so your close but should be ok. there are lower dropout voltage models of 7812's. by the time you get under 12v on the transformer under load, you likely passed the transfomers VA rating.  that being said, i prefer to use 15V transformers for 12V regulated circuits.

doubling the voltage through a doubler is gonna get u around 30V and higher unloaded that is close to, if not over(unloaded) for the working voltage of an LM7812
even if you are under, its gonna get really hot.  not sure id feel safe with that myself.

voltage multipliers with heavy load need a significant capacitance in each stage to handle the current draw to maintain voltage and have low ripple




Hobbyist and a retired engineer and possibly a test equipment addict, though, searching for the equipment to test for that.
 

Offline sureshotTopic starter

  • Regular Contributor
  • *
  • Posts: 177
Re: Voltage multiplier
« Reply #11 on: July 04, 2018, 12:13:19 am »
As a hobbyist I glean as much info as I can find, I'm much more proficient at putting circuits together than working out the maths. With the maths I scrape by just. Thank you for the info you've all posted, much appreciated. I've got reams of data sheets, circuit schematics stored for references and such. The lm723 is the one I've not done yet, although I have collected a few variants of schematics from good sources. I'm mainly interested in power electronics, AC and DC motor control, and some relay stuff. Short story...
A friend asked me to repair a clay pigeon machine some years ago. It's wiring was a mess, it needed stripping right out and rewiring. I was going to see if I could work some of the 5 motors circuits out, that was before another friend tore out the wiring with side cutters whilst I went to get the tea. Started from scratch, took me two weeks to rebuild it. The owner then as me to adapt it to RC control for remote firing. Gained a few free trips to the range with my Daughter as a thanks for repairing it.
Very interested in anything power electronics. I'd like to attempt a smps in the near future, but will tackle the lm723 circuit first. Thanks again for the help and information you've all posted.  :)
 

Offline Zero999

  • Super Contributor
  • ***
  • Posts: 20355
  • Country: gb
  • 0999
Re: Voltage multiplier
« Reply #12 on: July 04, 2018, 07:58:52 am »
rectified output voltage - fill bridge that's gonna be roughly  1.414 x the ac voltage - the diode loss so 12V ~= 15.7 and the lm7812 dropout is ~14.5 depending on the model, so your close but should be ok. there are lower dropout voltage models of 7812's. by the time you get under 12v on the transformer under load, you likely passed the transfomers VA rating.  that being said, i prefer to use 15V transformers for 12V regulated circuits.
For 1A, with Schottky diodes and monster filtering capacitors, you might just about get away with a 12V transformer and the LM7812. However, I think you've forgotten that the original poster wants to add a booster stage to get 3A, which will require an extra 1V to 2V of headroom.
 

Offline sureshotTopic starter

  • Regular Contributor
  • *
  • Posts: 177
Re: Voltage multiplier
« Reply #13 on: July 04, 2018, 08:28:18 am »
For 3 Amps continuous output i would use 6 Amp diodes, and 4 x capacitors of 35 Volt 22000uf each. Two parallel pairs, then put them in series. I think the capacitance would be 22000uf total of the two parallel series pair capacitors. With a 12 Volt 5 Amp transformer that should work.... I think. With the use of a full wave voltage doubler there's enough headroom to not worry about drops across the components.
 

Offline Zero999

  • Super Contributor
  • ***
  • Posts: 20355
  • Country: gb
  • 0999
Re: Voltage multiplier
« Reply #14 on: July 04, 2018, 10:46:02 am »
It makes no sense to double the voltage, with enormous capacitors, only to burn it off again in a power transistor. Here's a simplified schematic, showing what I was saying before about powering only the bias circuitry off the voltage doubler. This configuration has a lower drop-out, than the classic LM723 circuit, because Tr2's base voltage can go higher, than the output voltage, so the drop-out voltage can be as low as Tr2's saturation voltage, which will be about 1V, for a typical power transistor. D1 should use Schottky diodes, for a lower voltage loss.

The same principle can be applied to other circuits, such as the LM7812 with a booster transistor or an LM431 and a pass transistor. It will not work with the plain LM7812 or a fully integrated 3A regulator. The pass transistor must be accessible for it to work.


The original circuit can be found here, which is for a variable 3V to 30V, 2.5A PSU and will need some of the component values changing for a fixed 12V, 3A output.
http://powersupply33.com/adjustable-regulated-power-supply-3-30-v-2-5-a.html
« Last Edit: July 04, 2018, 11:03:20 am by Hero999 »
 
The following users thanked this post: sureshot

Offline sureshotTopic starter

  • Regular Contributor
  • *
  • Posts: 177
Re: Voltage multiplier
« Reply #15 on: July 04, 2018, 12:00:27 pm »
Thank you for posting that circuit. It looks quite good. Worth having a go at it.  ;)
 

Offline Zero999

  • Super Contributor
  • ***
  • Posts: 20355
  • Country: gb
  • 0999
Re: Voltage multiplier
« Reply #16 on: July 04, 2018, 01:36:02 pm »
Here's the LM317 version. The principle is the same: feed the biasing circuitry from the voltage doubler and the pass transistor, straight from the rectifier. The LM317 passes 50mA maximum, so it could be replaced with the cheaper LM317L, in a plastic TO-92 package. It doesn't have any current limiting, but that can be added, if needs be.

EDIT: I got the VRectifier and VDoubler labels wrong.
« Last Edit: July 10, 2018, 06:57:54 pm by Hero999 »
 
The following users thanked this post: sureshot

Offline sureshotTopic starter

  • Regular Contributor
  • *
  • Posts: 177
Re: Voltage multiplier
« Reply #17 on: July 04, 2018, 06:48:36 pm »
Thank you for that, yes I'm yet to use the lower current lm317L device. Thanks again for the schematics. :)
 


Share me

Digg  Facebook  SlashDot  Delicious  Technorati  Twitter  Google  Yahoo
Smf