trouble with bipolar powersupply

[arjepsen]

Hey.
I'm trying to make a bipolar +/- 12V power supply.
I'm using a 2x18V transformer - connecting the two secondary windings to make a center tap.
I'm using a TS7812 and a L7912CV voltage regulator. I'm attaching a schematic here of how I'm connecting it.

My trouble is, that I keep getting only about 5 volts on the positive side, though the negative side seems to output -12V.
I've tried another TS7812 to see if the chip was bad, but it's the same result.
I know I could do more filtering, but I asume I should still see 12V on both the positive and negative side?

Any help?

[tatus1969]

possibly you mixed up pinout of the 7912? Keep in mind that the two have different pinouts: 7812 in=1 out=3 gnd=2. 7912 in=2 out=3 gnd=1 (in TO220 cases). Your diagram doesn't show your pinning.

[arjepsen]

Yeah, I checked that.
(A few years back I made a powersupply for some pedals, and couldn't quite understand why the thing got so hot..... Now I know I have to check the pinouts at least 3 times...  )
(TS7812: 1-in 2-Gnd 3-out, L7912: 1-Gnd 2-in 3-out)

[tatus1969]

hmmm...do you have a load connected? how did you measure the voltages? nothing obviously wrong so far...

[arjepsen]

no - I didn't put a load on it yet - but I presumed that even without a load, the positive and negative side should be about equally close to Gnd?

[tatus1969]

sure. how did you measure the voltages?

[arjepsen]

multimeter - measured the 7812 from plus to Gnd (pin 3 to pin 2)
measured 7912 from minus to Gnd (pin 3 to pin 1)

[tatus1969]

I have no idea, except stupid things like both 7812 broken, bad china fakes, it being an 7805 instead, classic mirroring mistake when looking at the PCB's solder side, ... I assume you have checked both input voltages?

[singapol]

Please cheack transformer voltages and the diodes if you have not done so.

[arjepsen]

ok, this is making me quite annoyed now.

I just received a new order of parts, just to be certain that I had some brand new components to use, including a new 2x15V trafo, a couple of l7912's and l7812s and a couple of diode bridges.

Anyways, I'd start wondering if I'd fried something, so I went to the datasheet for the l7912 and read that it's around 30v max input. Hmm...
I've been messing around with the circuit without a load, so I knew the rectified voltage would be a bit high - and it was about 35 +/- to the centertap ground.
hmmm.. well good thing I got new components.
Anyways, I tried to put a load on the circuit by a 220k resistor from out to ground on each regulator. But the voltages was still high. Hmmm.....
finally I strapped some zener diodes from input to ground on each regulator, which brought the input voltages down to around 26V (+/-) on each input.
Put a new pair of regulators in there, just to be sure - but still no dice. Voltages on the output of the l7812 is hanging very nicely at 11,6v but the 7912 is at 17V!?!?!?
Is this problem showing up becuase of a no-load situation or ?

[bktemp]

78/79 regulators should regulate the correct voltage without any load.
Try adding diodes at the output of the regulators, because those regulators start to behave strange when the output gets pulled negative (or positive for the negative version).
If it does not work, post a picture of your circuit. It is easy to get the pinout wrong, therefor let others verify if everything is wired correctly.

[_Wim_]

Just to be sure, the ground of center tap of the transfo is connected to the gnd of the 7812 and 7912?

[janoc]

78/79 regulators should regulate the correct voltage without any load.

Err no - they don't. The datasheet specifies the output voltage for output current from 5mA up or so, with no load it is unspecified and depends on the part you get.

I have encountered a few of these regulators where the output voltage was out by almost a volt with no load and they started to regulate properly only when some minimal load was connected.

[bktemp]

78/79 regulators should regulate the correct voltage without any load.

Err no - they don't. The datasheet specifies the output voltage for output current from 5mA up or so, with no load it is unspecified and depends on the part you get.

I have encountered a few of these regulators where the output voltage was out by almost a volt with no load and they started to regulate properly only when some minimal load was connected.

You are right, there is no specification for <5mA. But I have never seen a 78/79 regulator with no load outputting a much too high voltage at room temperature (I always build the power supply first and verify the voltage before populating other ics and the voltage is spot on for most regulators). Therefore if I would encounter one outputting a significantly higher voltage, I wouldn't use it, because there is something wrong if it behaves different than most other regulators. The specification is valid for the full operating temperature (up to 125°C), therefore the minimum current is probably there because of some increased leakage currents at high junction temperature.

[arjepsen]

Yes, the Center tap is connected to the grounds of the regulators.
Here's a link to the trafo producers page:
http://www.hahn-trafo.com/english/pcb-transformers-ei30t1.php
I've connected pin 7 and 9 together to form the center tap/ground (there's no pin .

I wonder if the chip is instantly ruined, if the voltage on the input pin goes above/below 30V? Doesn't seem like it with the l7812 which just happily keeps putting out the correct voltage.
And yeah, this is off by much more than a few volts.

A 220k resistor from output pin to ground should work ok as a load, right?

[bktemp]

Try connecting 1kohms to make sure there is a load drawing >5mA.
The absolute max voltage is 35V, so 30V shouldn't do any damage.

[Zero999]

That's odd. I couldn't view the PDF in Firefox, so I had to use LibreOffice.


It should work. Have you mounted the L7912 & L7812 on a heat sink? Are both of the tabs electrically isolated from one another?

What load are you planning to draw? The main filter capacitors may need to be significantly larger than 470µF.

The maximum rated input voltage for the L7812 40V and the L7912 is 35V so the input voltage shouldn't be a problem.

http://www.st.com/content/ccc/resource/technical/document/datasheet/c9/16/86/41/c7/2b/45/f2/CD00000450.pdf/files/CD00000450.pdf/jcr:content/translations/en.CD00000450.pdf
http://www.hobbytronics.co.uk/datasheets/L7812.pdf

[arjepsen]

 ok, here's a few  picks of the current setup

[Audioguru]

2 problems:
1) Intermittent contacts on the solderless breadboard. Solder everything together properly!
2) Who is TS? It is Taiwan Semiconductor Company. Use a reputable manufacturer like Texas Instruments.

[tatus1969]

did you try with different output capacitors? maybe the ones you use are all old and dryed out?

[tatus1969]

why have you chosen a 2x18v transformer? a 2x12 would have been better in terms of efficiency. but you still should never have more than +-26v (with load)after your rectifier. if you measure that, then the problem already starts before the regulators. is your transformer very small?

EDIT.. your transformer *is* very small. This means that its no-load voltage will be way above its rated voltage. This is because transformer makers normally specify the output voltage at nominal output current. But these tiny transformers have significant winding resistance. What the manufacturer does to compensate this, is to add more turns to the secondary. At full load, your new transformer will deliver +-15V, but at no load it will be way above that. And that is what is probably killing your regulators. My suggestion here would be to
- use a transformer with a smaller output voltage. +-12V is more than enough, remember that the rectified voltage is +-12V * 1.41 = +-16.9V. This is more than enough drop for the regulators.
- add zener diodes to both regulator inputs, 1W / 27V types should be fine.

[arjepsen]

Thanks for the info :-)

I wonder though - wouldn't a 2x12V trafo be to low on the output to supply the regulators with high enough voltage, when under load?

What confuses me is that the l7812 consistantly seems to be fine, and delivers about 11,5 volts.
However, even when I used zener diodes on the inputs and made sure that there only was about 26 volts going in on both regulators, AND put in a brand new 7912, the output voltage was around 17 volts - no difference between a brand new, and an "used" regulator.
I can understand that there can be less trust in the taiwanese products, but I presume that would more be one defective chip here and there - not ALL bad...?

[CJay]

11.5V from a 7812 off load is not fine, it really should be 12V.

For the life of me I cannot see how you're getting more than 30V input to either regulator if you're measuring from the centre tap of the transformer and yikes, mains on a breadboard??

Power it all off and measure continuity from the centre tap to the ground pin of both regulators, measure continuity from centre tap to the shared connection of the main capacitors.

Are you sure you've got a connection between the secondaries, it looks like you've got the link between them in adjacent holes and both wires in one hole would be a tight fit.

 

[Lee Leduc]

Does your circuit have protection diodes from the output rails to
ground? If the negative regulator "comes up" quicker than the
positive regulator and pulls the positive regulator's output below ground, the
regulator can latch up. It's rare nowadays but I have seen it happen with some brands.

See Fig.16 from the attached Fairchild data sheet for a dual supply with output protection diodes.

[janoc]

You are right, there is no specification for <5mA. But I have never seen a 78/79 regulator with no load outputting a much too high voltage at room temperature (I always build the power supply first and verify the voltage before populating other ics and the voltage is spot on for most regulators). Therefore if I would encounter one outputting a significantly higher voltage, I wouldn't use it, because there is something wrong if it behaves different than most other regulators. The specification is valid for the full operating temperature (up to 125°C), therefore the minimum current is probably there because of some increased leakage currents at high junction temperature.

I have seen this with some breadboard power supplies (this style: http://artofcircuits.com/wp-content/uploads/2014/04/MB102-3V3-5V-3.jpg ). Those are most often using Chinese clones of the regulators, so it is well possible the circuit inside isn't exactly the same as the original LM78xx. The supply I have will give 3.5-3.8V without load, if I recall right, once loaded the output voltage is correct.

I have seen this also on another ocassion in a non-Chinese device - again the regulator wasn't working right until loaded properly. So it isn't behaviour I would rely on.