Author Topic: Basic question: rectifier circuits (AC-to-DC)  (Read 654 times)

0 Members and 1 Guest are viewing this topic.

Offline raff5184

  • Regular Contributor
  • *
  • Posts: 208
  • Country: us
Basic question: rectifier circuits (AC-to-DC)
« on: June 16, 2021, 06:41:30 pm »
Hi,

this is a very basic question. What circuits can be used to rectify (I mean AC-to-DC) a 500kHz-1MHz voltage of max 20Vpp? The output DC voltage is used, after regulation, as a supply, for example to recharge a capacitor, not as a signal, so precision is not a concern.

I have used full bridge diode rectifiers, passive voltage multipliers, and I'm trying rectifier with FETs.
My bridge rectifiers are not super efficient (I tried different types of diodes), the multiplier instead has too low output current.

Are there other circuits/topologies that are more efficient?

I read about active rectifiers with one or two opamps, but they seem to be used for precision rectification and I don't see how they can be useful in my case.
 

Online bdunham7

  • Super Contributor
  • ***
  • Posts: 3424
  • Country: us
Re: Basic question: rectifier circuits (AC-to-DC)
« Reply #1 on: June 16, 2021, 06:57:28 pm »
How much current is involved?  Have you tried a low-capacitance Schottky diode?

https://www.mouser.com/datasheet/2/427/bat41-1767379.pdf
A 3.5 digit 4.5 digit 5 digit 5.5 digit 6.5 digit 7.5 digit DMM is good enough for most people.
 

Offline raff5184

  • Regular Contributor
  • *
  • Posts: 208
  • Country: us
Re: Basic question: rectifier circuits (AC-to-DC)
« Reply #2 on: June 16, 2021, 07:12:57 pm »
yes I tried the schottky diodes too, BAT85S.
The DC current that I am able to rectify is in the range 1-8mA. Mostly in the lower end (2-4mA). So, I'm trying to have higher DC current
 

Offline Manul

  • Frequent Contributor
  • **
  • Posts: 756
  • Country: lt
Re: Basic question: rectifier circuits (AC-to-DC)
« Reply #3 on: June 16, 2021, 07:39:40 pm »
yes I tried the schottky diodes too, BAT85S.
The DC current that I am able to rectify is in the range 1-8mA. Mostly in the lower end (2-4mA). So, I'm trying to have higher DC current

In that case there is something wrong with your setup. What is the source of 500kHz AC? Maybe it is too weak to provide significant current? You are also welcome to draw a schematic of what you are doing, because seeing makes it easier.

Low capacitance, fast recovery diode will happily rectify such frequencies. Hard to do better then that.
 

Online bdunham7

  • Super Contributor
  • ***
  • Posts: 3424
  • Country: us
Re: Basic question: rectifier circuits (AC-to-DC)
« Reply #4 on: June 16, 2021, 07:45:58 pm »
We'll need more details.  What is the source of the AC signal and what happens to that signal when you load down the DC output?  What DC voltage are you trying for?  How are you loading and testing the DC output?  If your source is not getting loaded down, you should be able to get power even just using plain 1N4148 diodes.

A 3.5 digit 4.5 digit 5 digit 5.5 digit 6.5 digit 7.5 digit DMM is good enough for most people.
 

Offline exe

  • Supporter
  • ****
  • Posts: 2333
  • Country: nl
  • self-educated hobbyist
Re: Basic question: rectifier circuits (AC-to-DC)
« Reply #5 on: June 17, 2021, 08:23:16 am »
Opamp rectifiers are not suitable for this because they need additional supply. If you have a supply, why bother with rectifying? :)

Bridge rectifier is a way to go, the average drop will be 1.4-1.6 or so (due to high peak current, because of poor power factor).

I suggest get an oscilloscope, it will make things much easier.
It also would help if you share details about signal source and load you want to connect.

PS may be this circuit can help reducing the voltage drop, but you'd need to find a mosfet with low-enough gate capacitance, and low-enough Rds(on): https://www.eevblog.com/forum/projects/semi-active-bridge-rectifier/ . Again, we need details to see if this a viable solution.
« Last Edit: June 17, 2021, 08:28:24 am by exe »
 

Offline raff5184

  • Regular Contributor
  • *
  • Posts: 208
  • Country: us
Re: Basic question: rectifier circuits (AC-to-DC)
« Reply #6 on: June 17, 2021, 03:13:55 pm »
Thank you all!
So here is the schematic, the general idea on the left, using a diode rectifier on the right. Maybe a step-down converter between the rectifier and the load. My primary question however, is if there are other rectifier circuits that I might not know about and that are better than 4-diode rectifiers in this case.

The source is an ultrasonic transducer.

Bridge rectifier is a way to go, the average drop will be 1.4-1.6 or so (due to high peak current, because of poor power factor).
That drop is what I don't like, because the source voltage can be up to 20Vpp but also much smaller, in which case I'd lose a lot   :-\

I suggest get an oscilloscope, it will make things much easier.
It also would help if you share details about signal source and load you want to connect.
I have an oscilloscope and all lab bench equipment. The load, is a supercapacitor ESR 30 \$\Omega\$


We'll need more details.  What is the source of the AC signal and what happens to that signal when you load down the DC output?  What DC voltage are you trying for?  How are you loading and testing the DC output?  If your source is not getting loaded down, you should be able to get power even just using plain 1N4148 diodes.
The source is an ultrasonic transducer.
DC voltage I need is no more than 6V DC. For DC output I tried connecting the supercapacitor or different resistors (100 \$\Omega\$ to 100 k \$\Omega\$) to see the efficiency for different loads.
Yes, if I don't load the source and connect it to only the rectifier, I can see high voltages (15-18V DC), but these are not very descriptive because when I connect a supercapacitor with 30 \$\Omega\$ ESR the DC voltage drops.

In summary the DC voltage drop is fine, because I can typically reach the 6V DC, the problem is that the DC current is too low
 

Online bdunham7

  • Super Contributor
  • ***
  • Posts: 3424
  • Country: us
Re: Basic question: rectifier circuits (AC-to-DC)
« Reply #7 on: June 17, 2021, 03:23:17 pm »
Have you examined the AC input signal with your oscilloscope when you try to get rectified DC from it?  It seems obvious that the transducer simply can't supply enough current. so there's not enough to be rectified.  As for whether there are more efficient power rectifier circuits, I think the answer in this case is 'no', the four-Schottky bridge is probably very close to ideal and I doubt there is anything readily available that would be even nominally better.  Even if you had some magic 'perfect' rectifier circuit, I don't think your situation would change much.
A 3.5 digit 4.5 digit 5 digit 5.5 digit 6.5 digit 7.5 digit DMM is good enough for most people.
 
The following users thanked this post: raff5184

Offline exe

  • Supporter
  • ****
  • Posts: 2333
  • Country: nl
  • self-educated hobbyist
Re: Basic question: rectifier circuits (AC-to-DC)
« Reply #8 on: June 17, 2021, 03:26:20 pm »
How do you limit output voltage so the supercap is not overcharged?
 

Offline raff5184

  • Regular Contributor
  • *
  • Posts: 208
  • Country: us
Re: Basic question: rectifier circuits (AC-to-DC)
« Reply #9 on: June 17, 2021, 03:48:26 pm »
Perfect, at least I know I'm on the right track and will try to optimize what I have now.


As for limiting the DC voltage, I am not actually doing it because it never goes above the rating. The moment I connect the supercap, if it is discharged, both the AC signal and the DC voltage drop (the DC voltage to about 3-4V DC), and the DC voltage slowly goes up (to 5-6V DC) while it gets recharged. But, because the current is small and decreases even more when the supercap gets charged, the DC voltage doesn't increase any more, plus I start drawing current from the supercap so it is not an issue. If I'm able to increase the charging current, then as I said, I will use a step down converter
 

Offline exe

  • Supporter
  • ****
  • Posts: 2333
  • Country: nl
  • self-educated hobbyist
Re: Basic question: rectifier circuits (AC-to-DC)
« Reply #10 on: June 18, 2021, 07:29:19 am »
I have doubts the ultrasonic transducer is designed to drive such low-impedance loads. It may not be safe for the driver. I hope it has some output protection.
 

Online langwadt

  • Super Contributor
  • ***
  • Posts: 2656
  • Country: dk
Re: Basic question: rectifier circuits (AC-to-DC)
« Reply #11 on: June 18, 2021, 07:58:26 am »
resurrecting uBeam? :)
 

Offline Manul

  • Frequent Contributor
  • **
  • Posts: 756
  • Country: lt
Re: Basic question: rectifier circuits (AC-to-DC)
« Reply #12 on: June 18, 2021, 02:47:21 pm »
resurrecting uBeam? :)

Truly great ideas never die. On a serious note, in this case I would naturally think about using some switching topology without a conventional AC to DC rectifier. Also best power transfer is achieved when load impedance matches source.
 

Offline David Hess

  • Super Contributor
  • ***
  • Posts: 13554
  • Country: us
  • DavidH
Re: Basic question: rectifier circuits (AC-to-DC)
« Reply #13 on: June 21, 2021, 01:10:09 am »
For power conversion at such high frequencies and low currents, the way to go is with switching diodes or small signal Schottky diodes.  Larger power diodes will have higher capacitance and and are usually slower.  High frequency inverters do exactly this and work fine to 10s of milliamps and could go higher if there was a need.

Fast open loop peak detectors rely on pairs of diodes to compensate for forward voltage drop and can operate to 100s of MHz but rely on an auxiliary supply so do not make sense for power conversion.  There are various current switched rectifiers using bipolar transistors but the same applies to them.

 


Share me

Digg  Facebook  SlashDot  Delicious  Technorati  Twitter  Google  Yahoo
Smf