Creating a stable 5V DC supply out of 6 V (RMS) AC

[davenn]

I've never seen a 7805 data sheet specifying 100 uF on the output.

maybe not, but it's standard practice when using regulators whose initial source is AC rather than a battery supply

 

[shjacks45]

I need to create a 5 V stable DC supply out of 6 V AC from a transformer. By "stable", I mean that I do not want the ripples in the voltage that one gets from the output of a full-wave bridge rectifier. I have to feed this to two ATmega 8A μCs.

I am going to create the following circuit:

View attachment 275252

6 V RMS means ≈ 8.5 V peak. The capacitor will hopefully smooth out the ripples. The LM 7805 voltage regulator will bring down the voltage to 5 V. I can attach a heat sink to it if it tends to get heated up too much.

I don't yet have all the materials; once I get them, I can use the Arduino analogRead() function to have a look at the waveform of the output of the circuit because I don't have an oscilloscope.

Does this circuit look okay? Any suggestion for improvement?

"two ATmega 8A μCs." So a 16 Ampere supply? The 1N4007 are 1 Amp max and LM7805 versions are 1 Amp max. Transformer standard voltage 6.3 VAC typically max out at 10 Amps. Or pay for custom transformer. It will weigh a lot and cost $$$.
8 volts is typical computer power supply from 6.3 volt transformer, rem diode forward voltage drop which increases with current. Actually old time computer power supplies used a center tapped 12.6 volt transformer and 2 diodes (not a bridge as you show) so only one diode drop not two.
I'm puzzled by your "8A" requirement as Atmega products are usually low power.
8 Amps is more 5V power used than desktop computers with a billion times more computing capability than Atmega chips.

 

[Wrichik Basu]

"two ATmega 8A μCs." So a 16 Ampere supply?

Where did I say 16 A?

I'm puzzled by your "8A" requirement as Atmega products are usually low power.

ATmega 8A is a uC. It is not the same as ATmega 8 ampere (which doesn't make sense anyway).

 

[Wrichik Basu]

You need another 100uf capacitor on the output to reduce ripple, I would use 330uf or 500uf on the output.

If you see this post, where I have posted the updated circuit diagram, you will find that the transformer has been changed to 9V, and the electrolytic capacitor has been increased to 4700uF.

 

[shjacks45]

Where did I say 16 A?

ATmega 8A is a uC. It is not the same as ATmega 8 ampere (which doesn't make sense anyway).

ATmega8A is micro, "ATmega 8A" is just more bad spelling.

 

[DaveE]

If you see this post, where I have posted the updated circuit diagram, you will find that the transformer has been changed to 9V, and the electrolytic capacitor has been increased to 4700uF.

I think he meant after the regulator, but I wouldn't bother. It wouldn't hurt, but shouldn't be unnecessary.

I would do this if I had a load that intermittently drew more current than the regulator could source, but for a duration that exceeds the ability of the other downstream caps (bypass caps, mostly) to supply. I don't think 2 uPs are that sort of load. If they were, then the caps need to be fast (low esr, ceramic, etc.) and very close to the uPs. When digital circuits want lots of current the di/dt can be huge (but usually isn't); they want that current faster than an electrolytic located back at the power supply can deliver. If your uPs came on a PCB designed by someone else then they have already solved that for you with local storage.

 

[Averagesupernova]

...they want that current faster than an electrolytic located back at the power supply can deliver.

This was my whole point. I'm not saying a single 1uF across the complete 5 volt supply rail is adequate. Capacitors that exist to dump current on sudden demand need to be placed as close as possible to the device that demands it. This means often times several in parallel as @Baluncore described.

 

[Wrichik Basu]

After a really long wait, I could finally gather all the required components (including the oscilloscope proposed earlier). But trying to put the Schottky diodes into a breadboard drained away my enthusiasm. Man, I just couldn't push them in — the leads were so thick! Even the large capacitor was creating trouble: the positive lead was so long, that when I pushed it in, the shorter negative lead couldn't be pushed in. Finally, made a compromise:

That diode got bent when I was trying to put it into the breadboard. Probably one of the ugliest circuits I have ever built.

Will test it tomorrow morning, after rechecking the diode bridge connections.

 

[DaveE]

After a really long wait, I could finally gather all the required components (including the oscilloscope proposed earlier). But trying to put the Schottky diodes into a breadboard drained away my enthusiasm. Man, I just couldn't push them in — the leads were so thick! Even the large capacitor was creating trouble: the positive lead was so long, that when I pushed it in, the shorter negative lead couldn't be pushed in. Finally, made a compromise:

View attachment 283086View attachment 283087​

That diode got bent when I was trying to put it into the breadboard. Probably one of the ugliest circuits I have ever built.

Will test it tomorrow morning, after rechecking the diode bridge connections.

Be careful about forcing fat leads into the breadboard, you can stretch the springs and cause connection problems later with really thin leads.

Honestly, I hate those things and will only use them for something I know for sure I won't be using tomorrow. OTOH, most everyone else doesn't agree with me!

95% of my breadboards look more like this:

 

[Wrichik Basu]

95% of my breadboards look more like this:

With experience, you can directly solder components. But I don't have that, and have to depend on breadboards.

 

[Wrichik Basu]

First of all, I would like to thank @Borek who, in post #2, mentioned the DSO138 oscilloscope. The device has really been very helpful.

Next, I have performed the experiment, based on the circuit I had posted here. These are the results from the oscilloscope:

Output of full wave rectifier bridge

After passing the output through the voltage regulator
with necessary filter capacitors
​

I can see that the average voltage is 4.43 V, the maximum being 4.83 V. 5 V is not reached. Is this an accuracy error of the oscilloscope?

 

[pbuk]

Does the first picture look like full wave rectification? Check your diode orientation.

 

[Wrichik Basu]

Does the first picture look like full wave rectification?

Yeah, I have seen that. Trying to figure out the issue.

 

[Wrichik Basu]

Update: Fixed it.