Rectifier Design Help

2»

Comments

  • MoskogMoskog Posts: 536
    edited 2020-01-06 - 17:11:52
    Just an edit to my own post, thought the diodes in the second post was backwards, changed to my reading glasses, now everything looks just fine!
  • > @Moskog said:
    > Just an edit to my own post, thought the diodes in the second post was backwards, changed to my reading glasses, now everything looks just fine!

    The diodes are happy facing either direction, but then it tends to let the purple smoke out of the caps.
  • For the lazy (or board space saving) among us or don't always remember the orientation of the diodes, a bridge rectifier in a 4 pin package works fine.

    @JRoark, I like the special capacitors that come preloaded with fancy brown and silver confetti!!!
  • AwesomeCronkAwesomeCronk Posts: 899
    edited 2020-01-08 - 22:21:24
    Those are the best capacitors!
  • AwesomeCronkAwesomeCronk Posts: 899
    edited 2020-01-08 - 22:39:22
    The driver boards arrived last night. I came home from school today and decided to get busy moving ahead with this project. So, I checked the voltage at the capacitors, which had been previously discharged. I got about 0.37v. I plugged the system in to the wall and measured AC at the pack: 22v. I measured DC at the caps: 30.5v. At this point, the rectifier is not safe to use. It must be able to start up at 26VDC or less under no load and it must be able to maintain at least 16VDC under about 0.5-1amp load. I believe the best way to do this is with a 16VDC buck converter or linear regulator. Any second thoughts on this?

    These seem like viable contenders.
  • frank freedmanfrank freedman Posts: 1,542
    edited 2020-01-09 - 00:00:31
    What you have measured is correct, 22Vrms converts to 30.5 Vpk. What is the secondary rated for? A secondary of 16Vrms would give you a 24Vpk to feed the cap bank. Also do you have two or three wires on the secondary (3=center tapped) so wondering if you have a secondary with a center tap and would measure ~11Vrms lead to Center tap and 22Vrms lead to lead giving the 30Vpk you are seeing.

    If you do have a center tapped transormer, you can use only two diodes for full wave rectifier and end up with about 16Vdc. Assuming transformer secondary is 22v with 11Vrms to the center tap. See this link....

    https://eeeguide.com/two-diode-full-wave-rectifier-circuit/

    And yes, it's ok to use google to solve common household problems..... and electronics too.....

    Edited to add note for two diode w/ three lead secondary.
  • jmgjmg Posts: 14,177
    edited 2020-01-09 - 02:52:40
    ... It must be able to start up at 26VDC or less under no load and it must be able to maintain at least 16VDC under about 0.5-1amp load. I believe the best way to do this is with a 16VDC buck converter or linear regulator. Any second thoughts on this?

    What does the 30V DC sag to under load of 1A ?
    Linear regulators are simpler and make sense when there is less power loss.
    If we guess your 30V sags to maybe 24V, that's 8V drop to 16V for an 8W thermal load. On a continual basis, 8W is a lot to cool.
    So that high thermal load indicates SMPS, and there are cheap modules on eBay that may be good enough here.
  • I have a two-wire output transformer. It's rated 16V, 1A. I will try using a LM7805 linear regulator and some 47kOhm resistors to simulate a load and see how heavy I can get it.

    I have indeed been using the web, but I have no end of troubles sifting through it all. I find that the best way to get answers is to look up ideas and options, then ask on a forum for clarification.
  • I will attempt to use an LM7805 linear regulator and some not very big 47kOhm resistors to simulate a load.

    And, yes, I have been using the internet, but the only way I can sift through it is by asking real people for their input.
  • I will attempt to use an LM7805 linear regulator and some not very big 47kOhm resistors to simulate a load.

    And, yes, I have been using the internet, but the only way I can sift through it is by asking real people for their input.

    I think the problem is that you are looking and asking, but not thinking, just grabbing solutions. Take your time and don't skip the fundamentals.
    For instance "47kOhm resistors to simulate a load" means less than 1ma which is not a load of any kind. Fundamental is Ohm's law where R=V/I and I = V/R where 5/47000 or 30/47000 is very small indeed.
    Also a linear 5V regulator from 30VDC would melt trying to supply even 100ma. Fundamental is W=VI where 30-5 = 25V*0.1A = 2.5W!!!!!!!
    The list goes on.

    Definition of Fundamental is: Fun+da mental = if you da thinking, then you will have da fun :)
  • evanhevanh Posts: 8,619
    edited 2020-01-09 - 01:55:32
    I see 18 Vdc is commonly the DCC supply voltage, laptop power packs are often 18 Volt regulated. A second-hand 100 Watt one of those would give you a reliable 5 Amps without all the heat loss.

  • Alright. @Cluso99, thank you. I really need to stop and think this through. I have been scrambling back and forth trying to get this rectifier to work, on top of school and life. Not a good recipe.

    @evanh, I do have an 18.5v laptop pack. I should have thought in the first place to use that. I thank you for that thought. It pays to collect things like power adapters!
  • Use the pack if you must, but walk yourself through the steps of building the power supply or you will miss out learning the why. Go to the sites like the one I linked for you. Lots of schools have pages like this with the theory, maths and feedback, rather than trolling the net and finding a few diamonds among the slag.

    Many of the lighter (thinner) haired among us I would bet started with this sort of supply. And built them as needed before dirt cheap switchers were out. I could weld with the unregulated section of my IMSAI 8080 way back when. Every S100 board had its own 78xx regulators on board.
  • @AwesomeCronk

    I think you are worrying for no good reason if the no load voltage is a volt or two above the recommended maximum at no load. As soon as the load starts to draw current the output voltage will become less, and the load will start drawing current as the capacitor voltage increases. Try it with a 22ohm power resistor or a pair of 12V car bulbs in series and measure the output voltage.

    I have built more of these basic analog supplies (transformer - rectifier - filter cap) than I care to think about, both with and without regulators. Definitely over a hundred for S100 computers that I built and sold eons ago. They were very reliable and easy to use, and the no load voltage was always 1.3 - 1.4V higher than when they were connected to the load they were intended for. Never had a problem that pointed to the higher no load voltage as the cause.
  • kwinn wrote: »
    @AwesomeCronk

    I think you are worrying for no good reason if the no load voltage is a volt or two above the recommended maximum at no load. As soon as the load starts to draw current the output voltage will become less, and the load will start drawing current as the capacitor voltage increases. Try it with a 22ohm power resistor or a pair of 12V car bulbs in series and measure the output voltage.

    I have built more of these basic analog supplies (transformer - rectifier - filter cap) than I care to think about, both with and without regulators. Definitely over a hundred for S100 computers that I built and sold eons ago. They were very reliable and easy to use, and the no load voltage was always 1.3 - 1.4V higher than when they were connected to the load they were intended for. Never had a problem that pointed to the higher no load voltage as the cause.

    With a bridge rectifier, when load current is flowing you will have the voltage drop across two diodes. About 1.4V together increasing with load current.
  • kwinn wrote: »
    @AwesomeCronk

    I think you are worrying for no good reason if the no load voltage is a volt or two above the recommended maximum at no load. As soon as the load starts to draw current the output voltage will become less, and the load will start drawing current as the capacitor voltage increases. Try it with a 22ohm power resistor or a pair of 12V car bulbs in series and measure the output voltage.

    I have built more of these basic analog supplies (transformer - rectifier - filter cap) than I care to think about, both with and without regulators. Definitely over a hundred for S100 computers that I built and sold eons ago. They were very reliable and easy to use, and the no load voltage was always 1.3 - 1.4V higher than when they were connected to the load they were intended for. Never had a problem that pointed to the higher no load voltage as the cause.

    The issue with all these things is that I have very little in the way of tools. I also am too young to buy things without my parents' digression, which means I cannot buy tools or parts that I do not already have.
  • Most of the tools you may want for this can be had at any home store or good hardware store. Or amazon (St. Peter don'tcha call us cause we can't go, we owe our soul to the Bezos store) <- ask the geezers if ya don't get the reference... Give the parents a list and your plan. Remind same of the educational benefits. Ask them to help you but be careful, you may end up having to share your toys with them if they get interested enough.

  • rbehm wrote: »
    kwinn wrote: »
    @AwesomeCronk

    I think you are worrying for no good reason if the no load voltage is a volt or two above the recommended maximum at no load. As soon as the load starts to draw current the output voltage will become less, and the load will start drawing current as the capacitor voltage increases. Try it with a 22ohm power resistor or a pair of 12V car bulbs in series and measure the output voltage.

    I have built more of these basic analog supplies (transformer - rectifier - filter cap) than I care to think about, both with and without regulators. Definitely over a hundred for S100 computers that I built and sold eons ago. They were very reliable and easy to use, and the no load voltage was always 1.3 - 1.4V higher than when they were connected to the load they were intended for. Never had a problem that pointed to the higher no load voltage as the cause.

    With a bridge rectifier, when load current is flowing you will have the voltage drop across two diodes. About 1.4V together increasing with load current.

    Yes, there will be a 2 diode drop across any full wave rectifier. What I meant to say in that post was that the no load voltage was always 1.3 to 1.4 times higher than when it was connected to the load.
  • Note that if you are using a two diode and center tapped transformer, you will only have one diode drop of around 0.7V.
  • Note that if you are using a two diode and center tapped transformer, you will only have one diode drop of around 0.7V.

    Good point. Typically center tapped transformers were used mainly for power supplies that provided a positive and negative voltage as required for older RS232 drivers and op amps.
Sign In or Register to comment.