diy dual pos. neg. psu

Discussion in 'General Electronics Chat' started by adamclark, Aug 10, 2014.

  1. adamclark

    Thread Starter Member

    Oct 4, 2013
    472
    6
    Ok so I had a thread here awhile back on a psu I was building. I got busy with other projects and life and it got put on a shelf. It seems that I lost all of my data I had saved for it. So here is what I want to do and what ive got for it.

    1.) my transformer has the primary and 3 secondaries. The sec. HV coil is center tapped. it is the main one I want to use for my psu. it measures 22.6vac per leg so it should be around 32vdc each leg. im using an mb354-w rectifier bridge so im expecting some voltage drop from it. My intentions are to have a +/- 0-20vdc 4a adjustable supply.

    2.) Ill be using (3) LM317's and an nte996(the schematic shows an lm308 but mines the same ic) heres the schematic for that one. page 25 fig 55.
    http://www.ti.com/lit/ds/symlink/lm117.pdf
    and the schematic for the LM317 with a pass transistor is here. page 2
    http://www.ee.teihal.gr/labs/electronics/web/downloads/LM317T_Variable_Voltage_Regulator.pdf
    I need to combine these 2 schematics to achieve what I want. which is 3 LM317's using 2n3055's as pass transistors on each LM317.
    the nte 996 datasheet is here..
    http://pdf1.alldatasheet.com/datasheet-pdf/view/10931/NTE/NTE996.html.

    3.) Im using 3 nte2050's with 7seg displays for voltage and amp displays.the datasheet for the nte2050 is here..

    http://pdf1.alldatasheet.com/datasheet-pdf/view/9882/NTE/NTE2050.html..

    so this is what ive got so far.. I need opinions and help building my schematic so I can start to build.. as always,, thanks
     
  2. adamclark

    Thread Starter Member

    Oct 4, 2013
    472
    6
    I know there was more to it that ive already covered in the past but I lost all my data.. so I just covered the basics..
     
  3. MaxHeadRoom

    Expert

    Jul 18, 2013
    10,526
    2,369
    If you intend using both halves then you only need two rectifiers and use centre tap for common,
    Max.
     
  4. adamclark

    Thread Starter Member

    Oct 4, 2013
    472
    6
    heres what ive done already. its still way in testing but the mb354-w bridge rectifier I have I ran one leg of the sec. hv to one of the ac terminals and the other leg of the sec hv to the other ac terminal.. then the center tap by itself. ill attach pictures of. but I put the neg. lead of my dmm on the center wire which is clear and the pos lead to the red wire I get +22vdc and when I put the pos. lead on the orange wire I get -22vdc.. I figured that's how it worked..
     
  5. adamclark

    Thread Starter Member

    Oct 4, 2013
    472
    6
    when I multiplied the ac voltage by 1.4 I should've had 32vdc.. but I ended up with 22vdc, so I guess I lost more voltage than expected through the rectifier..
     
  6. adamclark

    Thread Starter Member

    Oct 4, 2013
    472
    6
    and I should mention that I have 2 10kuf 50v capacitors for my filter caps.. I cant remember how I came up with that value, but you MaxHeadRoom was involved in the thought process of my old thread.. and I am going to add that you were very helpful.
     
  7. adamclark

    Thread Starter Member

    Oct 4, 2013
    472
    6
    heres some pics of the wave on my o-scope.. and some of my components.
     
  8. adamclark

    Thread Starter Member

    Oct 4, 2013
    472
    6
    some more pics of the components I already have..
     
  9. adamclark

    Thread Starter Member

    Oct 4, 2013
    472
    6
    when I add the filter cap between the pos and center, and neg. and center, It goes from 20vdc to 26vdc.. is this expected?
     
  10. MrChips

    Moderator

    Oct 2, 2009
    12,440
    3,361
    Here is your starting point, a basic unregulated +ve and -ve supply:


    [​IMG]


    You can use your 10,000μF capacitor in place of a pair of 4,700μF caps.
    Watch the polarity of the caps very carefully.
     
    Last edited: Aug 10, 2014
  11. adamclark

    Thread Starter Member

    Oct 4, 2013
    472
    6
    In the pics above you will notice 2 small rectifiers,, I believe they are 2a max.. the big one is 20a max.. I read that the 2050 meter ic needs its own power supply away from the main one, so I planned on using one of the other sec. coils that's 8.1vac it should be 12vdc after the rectifier (but after seeing the voltage drop from the other rectifier I have to question my math again) to power the 2050's, and the datasheet for the 2050 says I need a 1vdc supply to one of the pins to calibrate it.. anyhow, I plan on using the other sec. coils to power the rest of the circuit not directly affiliated with the main psu. the third sec. coil measures 38.5vac/ 54.05vdc. im not sure what im going to do with that one,but I have plenty of lm317's so I can pretty much make it any voltage I want as long as I don't demand to much current from it, right? I have plenty of heatsink also before that comes in question.. I have a few heatsinks from some old car amplifiers. the whole body of these amps are just massive heatsinks and I have the tools and capability of cutting and forming any size heatsink I need.lol
     
  12. adamclark

    Thread Starter Member

    Oct 4, 2013
    472
    6
    so the way I have it is wrong then? lol.. I mean I have no problem changing it. I just thought that's how it worked from what I gathered from various schematics ive searched on google images.
     
  13. MrChips

    Moderator

    Oct 2, 2009
    12,440
    3,361
    I don't know. I cannot find your circuit diagram.
     
  14. adamclark

    Thread Starter Member

    Oct 4, 2013
    472
    6
    wait a minute.. that is kinda how I do have it, right? if the center tap goes to ground and the trace between the + and - goes to ground then they are connected anyways correct? I just have to add the caps. and I see your point about the polarity. I would have got it right on the + side but would probably have put it backwards on the - side.
     
  15. adamclark

    Thread Starter Member

    Oct 4, 2013
    472
    6
    sorry..like I said, I lost most my data.. the only schematic I have is in eagle. ill upload it if that helps
     
  16. to3metalcan

    Member

    Jul 20, 2014
    228
    23
    Yes...you've turned more of the ripple (AC voltage) into DC voltage by filtering it.
     
  17. adamclark

    Thread Starter Member

    Oct 4, 2013
    472
    6
    here is a crude drawing I did in paint. hope its understandable for those who cant see the eagle one. remember I said crude..lol.. I just did it in like 5 minutes.
     
  18. adamclark

    Thread Starter Member

    Oct 4, 2013
    472
    6
    here is the scope shot after the filter caps have been put in my circuit.. I notice that im now back to 30vdc +/-.. Also notice the freq. is now 0 instead of 120hz before.. I am guessing this is normal also? and does this wave look ok? like clean enough?
     
    Last edited: Aug 10, 2014
  19. to3metalcan

    Member

    Jul 20, 2014
    228
    23
    0Hz = DC = mission accomplished!

    20V/Div is not a very good level of resolution to check AC ripple. For an unloaded supply, the caps you're using should be plenty big enough, but if you want to make sure, use a smaller V/Div setting so we can actually get some idea how much ripple there is.
     
  20. adamclark

    Thread Starter Member

    Oct 4, 2013
    472
    6
    ok..ill do a better scope tomorrow.. but glad to hear im going in the right direction.. I believe it only gets harder from here on.. Im learning a lot of this as I go. ive got a solid understanding though so I should be ok.. I guess the next step is to make a schematic of my regulator circuit. I have a good idea of how it needs to be done, I just don't want to overlook something dumb and cost me a complete redraw..lol.. So im just going to go at it piece at a time.. one thing ive noticed is when I turn the transformer off, Ive still got voltage because of the filter caps. will this go away as the circuit builds? I was thinking of putting a power on led across the caps to bleed them off, but im not sure if it will be necessary when the regulator circuit is built. I posted 2 schematics earlier in this thread. one showing the 3 lm317's and the nte996(not sure what this ic does though). and the other is a basic lm317 with a pass transistor(im using the 2n3055 for this). So I need to draw up a schematic of both of these combined. next step.
     
Loading...