Voltage Regulator with AC Source

[SewingYard]

Hi,

Beginner here again, I'm progressing to voltage regulators after making a simple full bridge rectifier that worked a treat.

I am using a TS7805 (I found the datasheet here for it: http://www.datasheetspdf.com/datasheet/download.php?id=675172)

I fed the circuit a 15v DC input and as expected I got a 5v DC output on my scope with the circuit that is shown as an example on the datasheet.

So I figured I would try and feed this thing with 12v AC from a butchered wall wart, I figured/guessed/assumed it would just oscillate from 5v to 0v and back again at a 50hz rate (UK)

Instead I plugged it in and touched the heat plate and burnt my finger (silly me for not getting out my laser thermometer)

I looked at the scope and got the attached file.

Would someone mind explaining why the TS7805 getting so hot and also allowing the voltage to drop to approx -2v

As I said I'm no genius but thought this would be an interesting experiment where I could reliably predict what would happen in advance.

Thanks.

Wayne

[MK14]

Can I check my understanding ?

You did it WITHOUT a bridge rectifier ?

I.e. You gave it both positive (ok), and negative (NOT OK) voltage polarity ?

I.e. AC.

I'm not surprised it got really hot. It is NOT designed to cope with negative incoming voltages (although I've NOT fully checked the datasheet), but it is VERY unlikely to usefully handle negative voltages.

[SewingYard]

Yep, I deliberately did it without a bridge rectifier 

I did however figure it would just flip from being at +5v to then being at 0v when the wave dropped below 0v

I didnt envisage the whole finger burning saga 

So am I to assume that all Voltage Regulators are for DC only?

Thanks

[MK14]

ALL is maybe going a bit far. (There are probably thousands of different ones, so just about no one, knows about them all).

Yes, all or almost all, are DC only.

It's conceivable that you could get one which tolerates reverse polarity.

In general they are DC ONLY!

[SewingYard]

Great, thanks for clearing that one up for me, it was a good little experiment to carry out though and always interesting to see it on the scope.

It really is amazing how fast you can learn things when you have a decent scope and an awesome forum to cross reference things.

Time to build a full bridge rectifier then slot this chip in to the circuit clip off the excess voltage and output a steady 5v

I can finally replace the batteries on my daughters LED lights with something that costs a little less to run

[MK14]

Also you need the filter capacitors, between the bridge rectifier and voltage regulator. Otherwise, the 5V will keep on dropping down to zero.

Yes the forum and scopes are a good way of learning about Electronics and stuff.

Good Luck!

EDIT:
If it is JUST LEDs, and you don't mind 50/100Hz flicker. You could leave out the filter caps.

[SewingYard]

Thanks for the advice, I just finished this on the breadboard and it looks like it works a treat with the bridge rectifier and the Voltage Regulator.

The scope looks like the attached when finished. (Blue = AC, Yellow = 12v Bridge DC, Purple = 5v Regulator Output)

I would assume this is expected behaviour?

Where would you put the filter caps in a bridge rectifier + voltage regulator circuit?

I have 3 in the circuit so far, one between positive and ground on the bridge (1000uf) and two on pins 1 and 3 on the voltage regulator going to ground (0.1uf Mylar)

Would you suggest more to stabilise it even better?

Thanks

Wayne

[MK14]

Where would you put the filter caps in a bridge rectifier + voltage regulator circuit?

I have 3 in the circuit so far, one between positive and ground on the bridge (1000uf) and two on pins 1 and 3 on the voltage regulator going to ground (0.1uf Mylar)

Would you suggest more to stabilise it even better?

That bolded section above, is EXACTLY what I meant, by filter capacitor(s).

I was worried that you had missed it out.

To properly test the circuit out, you need (if you have NOT already done this), something which uses (at least), the somewhat high current the final load will be. Or connect up the final load (LEDs ?), if you are confident enough about it.

Because the higher load, will put a strain on the circuit. You may find the voltage starts dipping, in places and/or the regulator gets too hot. It depends on how much current your load/LEDs are going to use.

You will need a heatsink (on the regulator), if the regulator is going to get too hot.

[MK14]

Would you suggest more to stabilise it even better?

The size/value depends on what the output/load current is going to be.

If it is insufficient, you will see the voltage drop out of regulation, under full load.

I can't tell you the value needed, since you have not said what current the load/LEDs uses.

There are online guides about smoothing capacitors, such as this one:
http://www.radio-electronics.com/info/circuits/diode-rectifier/rectifier-filtering-smoothing-capacitor-circuits.php

[SewingYard]

Thanks for the confirmation and extra details.

I will be driving a string of 50 Micro LED's, they are currently driven by 3 x 1.5v AA Batteries with a resistor (they are 50 miles away at the moment so I don't know the exact value of the resistor until tomorrow when I take a peek)

I am however just doing all of this simply for learning though so proof of concept is fine for me to move on to the next challenge

Thanks again for all your help and input, Ill have another silly question before you know it, lol

[mariush]

Here's some info for you:

7805 regulators need to always have the input voltage about 1.5v higher than the output voltage if you want the output voltage to be stable in time. For LEDs it doesn't matter that much, but it's good to remember for other projects.
Since the output voltage is 5v, this means you should pick the capacitor to be big enough so that the regulator will always see at least 7.5v

You have a 12v AC transformer, which means that after you rectify it to DC using a bridge rectifier, you get a peak DC voltage of 12v x 1.414 = 17v, and then you also have to take out the voltage drop on the two diodes that are always conducting inside the bridge rectifier, so you're left with a peak DC voltage of about 15v.

Now you need to think how much current those leds would use and then you can choose a capacitor to keep the minimum DC voltage always above 7.5v ... you can figure out how much you need with this formula that approximates how much is needed:

C = Current / [ 2 x Mains Frequency x ( Vdc peak - Vdc min) ]

Let's go with 0.5A (i'm sure those 50 leds use much less) and let's say you want at least 8v and for safety, let's say you'll have up to 14v DC ... in which case, you'd need to use C = 0.5A / ( 2 x 50Hz x (14v-8v) ) = 0.5 / 100 x 6 =  0.000833 Farads or 833 uF .. so the next value up of 1000uF would be good enough for up to 0.5A in all cases.

You could use a very large capacitor but then the minimum voltage will always stay closer to 15v and you'd only dissipate more energy in the linear regulator ...

Remember, all above 5v is dissipated on the regulator as heat, and the IC can't dissipate more than about 1w without a heatsink... so for example if you have 15v in and 5v out , at 0.5A you'll have (15-5) x 0.5) = 5 watts of power dissipated as heat.

[SewingYard]

Thanks mariush,

The math is certainly very useful and great for me to work out where the current is going.

I have put a multimeter on the 12v AC input to the Bridge Rectifier and im getting 60mA

I have then put another multimeter on the output of the voltage rectifier in with a 100 Ohm resistor to create a dummy load of 49.4mA on the output.

I can see the voltage drop on both the 12v AC and on the bridge rectifier when I apply the load but I notice that because I have enough of a voltage going through the regulator the output remains a steady 5v DC

I think I feel comfortable enough to move onto the next project in Learning The Art Of Electronics now.

Thanks again.

Wayne

[Audioguru]

Your 'scope shows half-wave rectification with only a single rectifier diode. We are talking about using a BRIDGE rectifier that has 4 diodes and it doubles the ripple frequency so that the 1000uF capacitor can smooth the ripple much better.

[Delta]

Thanks mariush,

The math is certainly very useful and great for me to work out where the current is going.


Wayne

Current goes through the load, there is nowhere else it can go. "Math" or otherwise.

You have already reverse polarised a 78xx and got a burnt finger to prove it.  Pease be careful in your future endeavours.

[alsetalokin4017]

Thanks for the advice, I just finished this on the breadboard and it looks like it works a treat with the bridge rectifier and the Voltage Regulator.

The scope looks like the attached when finished. (Blue = AC, Yellow = 12v Bridge DC, Purple = 5v Regulator Output)

I would assume this is expected behaviour?

Where would you put the filter caps in a bridge rectifier + voltage regulator circuit?

I have 3 in the circuit so far, one between positive and ground on the bridge (1000uf) and two on pins 1 and 3 on the voltage regulator going to ground (0.1uf Mylar)

Would you suggest more to stabilise it even better?

Thanks

Wayne

Not that it matters very much at this point.. but your scopeshot actually appears to show:

Blue trace with rectified pulses of 12V peak at 50 Hz (half-wave bridge output unfiltered?)
Yellow trace at 5 Volts DC with very little ripple  (filtered Regulator output)
Purple trace at just under 12VDC (Filtered bridge output?)

The complete output (unfiltered) of a full bridge rectifier should appear as below, with peaks at twice the frequency of the input AC. Your scopetrace appears to show only a half-bridge type of output.