Isolation from Mains power

Discussion in 'The Projects Forum' started by jerseyguy1996, Apr 5, 2011.

  1. jerseyguy1996

    Thread Starter Active Member

    Feb 2, 2008
    206
    9
  2. debjit625

    Well-Known Member

    Apr 17, 2010
    790
    186
    Here the power transformer will do sufficient isolation.You just need to isolate once if you do it twice or more the effect will be just power loss for the system.

    Any transformer that is using a separate primary winding and a secondary winding is an isolation transformer.I think you are thinking about 1:1 isolation transformer as the only isolation transformer.

    Good Luck
     
    Last edited: Apr 5, 2011
    jerseyguy1996 likes this.
  3. SgtWookie

    Expert

    Jul 17, 2007
    22,182
    1,728
    The 24vac transformer you're looking at will require a bridge rectifier and filter capacitors (around 10,000uF) on the output before you can use it to supply the charger with it.

    That will also put it a bit too close to the input-to-output limits for the 7805L regulator that's built into the charger. When you rectify and filter AC, you wind up getting ~1.4 times the voltage that the AC is rated for, less the drop across the bridge rectifier (2 diodes' worth) so 24VAC becomes around 32vdc - however, when the transformer is lightly loaded (charger not charging) the DC voltage could be a good bit higher.

    Have a look at this laptop supply:
    http://www.mpja.com/prodinfo.asp?number=18506+PS
    You'll need to order the power cord like it says near the bottom of the description. The 18.5VDC @ 3.5A output is regulated, and has over-voltage protection.

    This supply would be operating at about 71% of its' rating with your charger on maximum, so that's a reasonable amount of "headroom" - you don't want to operate things at 100% of their rating or you'd have a short service life.
     
    jerseyguy1996 likes this.
  4. jerseyguy1996

    Thread Starter Active Member

    Feb 2, 2008
    206
    9
    Thanks! This was exactly what I was looking for!
     
  5. jerseyguy1996

    Thread Starter Active Member

    Feb 2, 2008
    206
    9
    Just saw your response here. I'm glad you brought up the filter capacitor issue. Is there a size that is too big for filtering? I put one of these down for my rectifier:

    http://www.goldmine-elec-products.com/prodinfo.asp?number=G14525

    and I was going to buy a few of these for filtering:

    http://www.goldmine-elec-products.com/prodinfo.asp?number=G9114

    I didn't realize I would need as much as 10,000uF but I can link a bunch of these up in parallel to get that value. Is there such a thing as too much capacitance when it comes to filtering?

    I'm also glad that you brought up the computer power supply example. I built a dumb charger using an LM317 and a computer power supply (using the +12V and -12V outputs to get 24V total). Is there a big issue with doing something like that long term?

    Also in terms of the 5V regulator, could I just choose one with a higher max input voltage? I think the one that I have has a max input voltage of 40V.
     
    Last edited: Apr 5, 2011
  6. SgtWookie

    Expert

    Jul 17, 2007
    22,182
    1,728
    Those 1500uF caps won't do, as they are only rated for 35v. When the transformer operating with no load (charge complete), the voltage rating of the caps will be exceeded, resulting in heat, boiling electrolyte, and blown up capacitors.

    10,000uF will give you about 2v ripple on the filter caps with a 2.5A load. You don't want a lot of ripple, as that implies lots of current in and out of the filter capacitors, which will heat them up.

    Well, the -12v output from the computer power supply is probably limited to under 0.8A, so not terribly useful as a charger. Your efficiency won't be very good, as you have to load the 5v output in order to get regulated output, and your LM317 is limited as to how much power it can dissipate, and current that it can pass.

    If you're charging 12v batteries using 24v in, you're using about half the power in heating up the LM317.
     
  7. jerseyguy1996

    Thread Starter Active Member

    Feb 2, 2008
    206
    9
    Ya my computer power supply was converted using these instructions which includes a power resistor in between the +5 and ground:

    http://www.wikihow.com/Convert-a-Computer-ATX-Power-Supply-to-a-Lab-Power-Supply

    The LM317 definitely gets very hot ( I have a big heat sink on it). My dumb charger was supposed to be temporary until I could build a smart one. I am a bit of a diy freak or else I would just buy a charger.

    So I could always just purchase a transformer with a lower output voltage. I don't understand why a rectified and filtered AC output results in a higher voltage. Is there an easy layman's explanation for why that is?
     
  8. jerseyguy1996

    Thread Starter Active Member

    Feb 2, 2008
    206
    9
    Oh....just read up on RMS versus peak voltage. I think I answered my question.
     
  9. jerseyguy1996

    Thread Starter Active Member

    Feb 2, 2008
    206
    9
  10. debjit625

    Well-Known Member

    Apr 17, 2010
    790
    186
    Small power transformers have voltage and current rating on them but in case of large power transformer it will have voltage and VA rating on them.

    What is VA rating? its the product of the winding voltage and current for example if I have a transformer of voltage rating at primary winding 220VAC and 24VAC at secondary winding with a VA rating of 1000VA (1KVA) for the transformer we could calculate the current for both of the winding ...

    max primary winding current will be 1000/220 = 4.546 Amp
    max secondary winding current will be 1000/24 = 41.667 Amp

    Good Luck
     
    jerseyguy1996 likes this.
  11. jerseyguy1996

    Thread Starter Active Member

    Feb 2, 2008
    206
    9
    Oh well that wasn't near as complicated as I was making it out in my head. Thanks!
     
  12. SgtWookie

    Expert

    Jul 17, 2007
    22,182
    1,728
    I'm wondering why you are still looking at transformers. Is it because you're really rather build the mains to low voltage portion, or is there something you don't like about the laptop supply I suggested in my first reply?

    Don't forget, you'll still need to buy or build an enclosure for your transformer, bridge and capacitors, along with a switch, fuseholder & fuse, a power cord and socket for it. It's going to wind up being rather bulky and heavy. Shipping costs for the transformer will probably add 50% to it's price. They're heavy.

    You could probably get by with a 12.6v transformer that output 4A or more.
     
  13. jerseyguy1996

    Thread Starter Active Member

    Feb 2, 2008
    206
    9
    I guess it was just due to an unreasonable desire to build every aspect of it. You are right of course that I should just purchase the laptop power supply as it would be about 1/4th the cost and a million times less hassle. That is what I will do. I actually have an old laptop power supply that outputs 19.2 volts at 3.4 amps but after plugging it in I was disappointed to see that it was not working. Come to think about it that was one of the reasons I stopped using that laptop in the first place.

    I don't think I can use any power supply less than 16 to 18 volts due to the bulk charge of the charging algorithm charging at a constant voltage of 15 volts and some assumed losses in the circuit (I think there should be a voltage drop across diode D6B). Is this a correct statement?
     
  14. SgtWookie

    Expert

    Jul 17, 2007
    22,182
    1,728
    I understand the desire to build something, but the supply I pointed out really makes a great deal of economic sense; it's cheap, you hardly have to do anything to make it work except connect it, and it will be efficient along with protecting your charger. (See below; that laptop supply will not put out enough voltage).

    There will be some voltage drop across D6A and D6B (probably around 0.42v), but D6B is rated for 4 Amperes; D6A is rated for 1A.

    I haven't looked at the algorithm. I guess it would've been helpful to read this paragraph at the outset:
    Hmm - six cells as in six cells per battery, as in 12v battery? Sounds like it to me... so the laptop supply I pointed out earlier is too low, and the transformer you were originally looking at would have been way too high.

    So, need a different supply; minimum 22v, max 30v, >2.5 Amperes.

    Have a look at this supply; 24V, 4.2A, ~$18:
    http://www.mpja.com/prodinfo.asp?number=17428+PS
    That'll be running at 60% capacity; no sweat.

    Here's a 24v @4.5a supply for ~$35:
    http://www.mpja.com/prodinfo.asp?number=17825+PS

    Now, they do have a 24v @ 2.5A for ~$27, but that's running at 100% capacity:
    http://www.mpja.com/prodinfo.asp?number=16008+PS

    If you need a supply for your laptop, MPJA has this one: http://www.mpja.com/prodinfo.asp?number=18507+PS
    19.5V@4.62A, ~$16.
     
  15. jerseyguy1996

    Thread Starter Active Member

    Feb 2, 2008
    206
    9
    I am absolutely amazed at all of the fun stuff available on that site. Looks like i will have a power supply soon. Thanks for helping me out with all of this! Oh and I think diode D6A and D6B is a one or the other type of arrangement. I am not sure what the dashed line around the 2nd diode means but there is a footnote at the bottom of the schematic that unfortunately doesn't tell you anything and I can't find anything in the datasheet or appnote that explains it. I have a question on TI's e2e forum but no one has responded to it yet (posed the question on sunday).
     
  16. SgtWookie

    Expert

    Jul 17, 2007
    22,182
    1,728
    MPJA is one of my favorites.

    I'm not sure about the D6A/D6B either. Looks like the diode they specified for D6B is being discontinued; Digikey is out of stock on it, went to the manufacturer's site and found the datasheet; "Not recommended for new designs".

    So, might double-check on what version of D6 is installed on the board.
    The 1N5821 is actually rated for 3A, 30V instead of 1A - so you should be OK (I'd downloaded the datasheet for it before, but forgot about it). The 5821 will drop around 0.5v with 3A current flow.

    Yeah, I noticed the footnote, and that D6 wasn't discussed anywhere else.
     
Loading...