+-12V Problem

[minya]

hello,

as described in my fm-threat i built the psu unit today. but i thought it wouldn't be good to put this question into it

i build scheme 2 of +-12v exactly as told on the page (with 2Diodes and 1x15V transformer), now i've got the half part working :) when i measure 12v+ vs ground the power is ok at 12v, when i measure 12- vs ground there are about5v more, when i change the pinning of the 7912 i/o there is 24v ... i don't understand this, exspecially because the 12+v is exactly the right amount. but now there is something else which confused me completely, when i measure at ac scale, i don't get zero. Measuring - vs ground is ok, and +vs- is also ok, so it seems as a ac circuit exists át 12v+ ... i don't understand that, i have changed the diodes and the 7912, checked the circuit nearly 10 times and i don't have an idea what can cause that.

thanks for your help,

markus

[bugfight]

is the decoupling electrolytic cap backward on the 7912 input?

the + side should go to gnd (which may be confusing) and the - side to the i 7912 input...

[minya]

hey thanks for your fast reply,

the - of the small elko behind the 7912 is connected to the input of the 7912, and the output of 7912 to the ceramic 100nf and also the - of the big 4700uF elko

i guess that should be ok

[minya]

i also thought about an 79 pin issue, but also changing input and output does not change much, but the above mentioned volt change at 12v-, the ground pin must be correct!

i made some photos and tried to mark where the circuit is, but i'm not sure if you can recognize all details. it looks real bad because i just changed the diodes on the back quickly and the i/o pnning of 79xx to see if they cause the error

the solderings in the middle rectangle is the 9v transformator

[minya]

i checked the circuit a few times but i have no idea where the problem could be. so i'll buy a 2x15v transformer today and build the circuit 1 with 2 tranform  ::)

[nILS]

7912s have a weird tendency to give you 17V  ::) Try to get a hold of new ones from a different store :-)

[doc]

7912s have a weird tendency to give you 17V  Roll Eyes Try to get a hold of new ones from a different store :-)

hehe. nILS.... I remember it.

Mostly they have the tendence to do 17V after they are wired once incorrect.

greets

Doc

[MTE]

My first wrong soldered 7912 was bringing -23 V ...  ;D

[nILS]

Mostly they have the tendence to do 17V after they are wired once incorrect.

True as well, but that's not alkways the case - I had 5 and only 1 of them was ever plugged in wrong (on purpose) :-)

[Imp]

I once had a 7812 wich gave 15V while measuring, but proper 12V under load...first i thought it might be labeled wrong, because i would not expect such a round voltage from a malfunction...

[nILS]

Well 15.0V is just as likely as 14.67V or 15.32V or anything else :-)

[minya]

hey guys,

thank you very much for your replies, that could be the issue i was looking for ... so can you tell me if the input or the ouput goes to the 12v- line? so that i wan't solder it wrong the next time :D

markus

[nILS]

http://hamradio.lakki.iki.fi/new/Datasheets/transistor_pinouts/

Make sure to check the datasheet of the manufacturer of your 7912, too.

[minya]

it is incredible! i rebuilt he hole circuit working excact and not so dirty as in the first one ... i changed the 78xx and 79xx, 10 and 100 caps, the diodes and i've got exactly the same problem! the 12+ V vs ground is ok, bt the 12v- vs ground gives me about 24v

i gues the 7912 vom reichel are not working ... i don't know what else can be the problem. i used 3 different of them but nothing changed. i took vacation for today, that i can do that and the parts for building the 2 transformer version have not jet arrived  :'(

[minya]

[stryd_one]

i've got exactly the same problem! the 12+ V vs ground is ok, bt the 12v- vs ground gives me about 24v

I thought It gave you 17V last time?

Edit: Gee you like gigantic images huh?

[minya]

hi stryde_one,

it gave me 17v  after i changed the pinning of i/o in 7912 at the first time

"when i measure 12- vs ground there are about5v more, when i change the pinning of the 7912 i/o there is 24v"

after i read the threats i thought i crashed the 7912 when i changed the pinning, because doc told me, that he got also 17v after he wired the 7912 wrong.

now i checked the pinning of the 7912 before carefully and did not rewire it ... i have following pinnings at 7912

ground to ground,

input to 12v- output (- of the small elko)

output through the small cap to - of the 4700uF elko

bugfight told: "is the decoupling electrolytic cap backward on the 7912 input?

the + side should go to gnd (which may be confusing) and the - side to the i 7912 input..."

ps: i deleted the first pics, because of thir size, i'm not a fan of big pictures, but i thought you wouldn't see anything if i make them smaller ... :) i'll try to mak them more shape and smaller next time

thanks,

markus

[JohnnieBee]

Markus

The diagram that you have drawn over your circuit board shows the middle pin of the 7912 as the output. The middle pin is the input.

Please check your board again!

[minya]

thank you guys for helping myself in this noob problems ... so i took my last 7912 which i had laying here without any usage before and switched the i/o as you told me johnybee (as in the drawing but i/o switched)

the consequences are not far away from the issues before

when measuring DC:

0 V vs -12v    ->  -17,8v

0´V vs +12v   ->   +12v

-12v vs +12v   ->   30v

measuring AC:

-12v Vs 0v    -> 0V

+12 vs 0v       -> 26v

+12v vs -12v    -> 65V

shouldn't all AC measuremens be 0V?

[minya]

i think i'll try to get others ... maybe these are all silly weired ones :D

[JohnnieBee]

Markus

What are the diodes that you are using for the rectifier?

[minya]

i use the second version with just 1 transformer, so i don't use a rectifier, just the diodes ... i tried it wirh SCHOTTKY-DIODE 1A/40V  and 1N 5401 ... but the effect was the same ...

but i ordered a double transformer for further times

[JohnnieBee]

Yes, those diodes should be fine.

[stryd_one]

shouldn't all AC measuremens be 0V?

Yes. Something's not right there man. Perhaps you have some of the diodes in backwards?

[minya]

i don't think so ... the line on the diode at the top of the circuit (as on the scheme in the wiki) is on the right side, the lower one is on the left side ... so it should be a circuit ...

i bought about 10 new 7912 but i tried 6 ones of these they are all the same. now i got a little bit more ... about 18v ...