Voltage regulator max current

[SaleB]

Hello,

this is my first regular post. I am novice in electronics.

I am working on a timer relay that will disconnect power to three separate devices for roughly 10 seconds, and activate them again.

http://www.kpsec.freeuk.com/555timer.htm#monostable

I was using that circuit and created the test circuit with R1=33K and C1=220micro. I intend to connect a NC relay on the pin 2 output. Then I intend to connect three separate NC relays in parallel for the three devices (in the circuit where the first relay is the switch).

Then I thought why don't I use stronger power supply for my device and power all three devices from my circuit (instead using their original power regulators and cutting their cables).

Questions:
1. Do I need a diode on exit of NE555 to protect it from the coil in the relay or should I connect the diode across the coil of the relay?
2. I have three devices using factory power supplies 9V@1.5A, 5V@1A and 7.5V@1.5A. If I use voltage stabilizers with nominal current of 1.5A (TO220), is it enough current for my devices? (I intent to use 7809, 7805 and LM350 for 7.5V)
3. Is 12V@5A power supply enough for all devices, assuming that relays will not be energized at the same time when the devices. In devices mode I calculated 4.1A consumption and in reset state I assumed roughly 250mA for relays and a few hundred mA for the NE555. Is 20% margin enough?
4. Do I need any fuses anywhere in the circuit?
5. If I would want an signaling LED, where shall I put it? I would not want to add more load to the NE555 output, but in the second circuit behind the first relay I can add a LED with appropriate resistor, right?

Thank you all for your input, it will be much appreciated
SaleB

[randtek]

Re: Voltage regulator max current

I would not recommend using the output of the NE555 as a relay drive. It probably cannot supply that much current. You should use it to control a transistor based drive circuit, possibly a darlington pair, depending upon the current draw of the relays.

If the maximum voltage needed is 9v, I would use a 9v power supply as the source. It will be more efficient, and generate less heat, as every volt you regulate down wastes a certain amount of energy in heat dissipation. If all devices are to be switched at the same time, you can probably do it with just one relay, switching the output of the 9v source to the 9v device and the other 2 regulators. A 5 amp, 9v supply should be sufficient.

[Dgtech]

Re: Voltage regulator max current

If current is your goal, I wouldn't use small electronics. There are such things as timer relays with normally open/closed contacts. SCR's handle higher currents fine as well.

[mariushm]

Re: Voltage regulator max current

You shouldn't connect relays to the output of 555 - you connect the output to a npn transistor or a darlington and go from there.

See this page: http://www.kpsec.freeuk.com/trancirc.htm
It includes info about using it directly or to power a relay and how to calculate the resistor required to not burn up the npn transistor.

As for what to use, consider

You calculate how much current the three relays need to stay closed , add about 10% extra
You need a diode to prevent current from going back into circuit when relay disconnects.

Something like TIP112 or TIP122 would be suitable.

Go with the 12v single power supply, something like a 65w 12v power brick for a laptop or a lcd monitor springs to mind - that's about 4A. If you have a 9v power supply handy, or it's cheaper than a 10-15$ adapter I mentioned, use that instead.

No matter what regulators you choose, they'll still dissipate the difference in power as heat, so you'll have to use a bit of heatsink for them.

The 555 shouldn't use 100mA, should use much less. Looking at the 12v relays in an online store, most use about 33-50mA @ 12v to keep the contacts closed, so your relays and 555 probably use 250-500mA overall.

If not all three devices run at the same time, you'll be safe with 4A of current in total, but it wouldn't hurt to go a bit higher.

[smason]

Re: Voltage regulator max current

Yes you need a diode across the relay coil

[PeteS in CA]

Re: Voltage regulator max current

Agree with others. Diode across the relay coils as close to the relays as possible. Use BJTs or MOSFETs to buffer the 555 and drive the relays. A darlington or MOSFET will minimize the current supplied by the 555. Heatsink your 3-terminal regulators really well. They have circuits that will cause the IC to self-protect if they get too hot. You might lookk into an LM317, an adjustable linear regulator, for your 7.5V output. LM317s are inexpensive, very available and very decent.

[SaleB]

Re: Voltage regulator max current

Thank you all for your answers. I have read it all.

Unfortunately I do have a nice 12V power supply, rated 5A; so I will use it. Fortunately I do have no space limitations, so I can add some heat sinks, I have some salvaged from ATX power supply.

I have already used NE555 in circuits powering relay (automotive relay, with integrated protection diode) and it worked fine (one relay powered by NE555, and another stronger relay or relays powered by that relay). I understood that it is better with a transistor in place of the first relay, but I have never used transistors.

I also have here laying around a few LM350's so I'll use them; they are similar to LM317.

I have two new questions:
This suggestion about using a fet in place of the first relay. I like it, I have a few fet's here. Do I need a current limiting resistor between NE555 and the gate of the fet? Should I connect the relays between + and fet or between fet and -?

If I use the fet for powering of the second row of relays, do I still need the diode? How should I orient the diode, cathode to +, anode to - or should I use it in reference to the component that I am protecting?

Can I use fet's instead of all the relays, or is it not advisable?

Thank you all again, I will soon try some bench project with transistors and learn how they work, but I think it's not advisable to use them for the first time for a project that is not a test project.

SaleB

[SaleB]

Re: Voltage regulator max current

I have just found the answer about the protection diode, I suppose a 1n4006 can be used for that purpose.
http://www.electronics-tutorials.ws/...tor/tran21.gif
I also found that load, when using N-fet is between + and drain.

What I did not find is how to calculate the source resistor?

SaleB

[mariushm]

Re: Voltage regulator max current

1n4006 would work, but so would 1n4003 probably. The difference in the last digit is the amount of reverse voltage they can handle before they pop.
These are a few cents each, really basic parts so yeah, it makes no difference, just go for the more powerful one.

Watch these below when you have time.

This guy is very good at explaining how to use a transistor as a switch, I really recommend it: http://www.youtube.com/watch?v=8DMZS...ature=g-user-u

He does all the calculations and explains each step, should be easy to understand. He turns off and on a solenoid but that's basically the same thing as the relay, so it's almost exactly what you need.
Darlington transistors are just like the npn in the video above, just the gain is much higher (the one he uses in video has gain 100, darlingtons usually have 250-1000 and higher) and instead of that 0.7 constant there's actually 1.2-1.5v.. just see the video.


This other guy is bit quirky, but the explanations are sound and I think you would understand him.

Diodes and Transistors : http://www.youtube.com/watch?v=w9cd7...ure=plpp_video

Mosfets: http://www.youtube.com/watch?v=CFt8h...ure=plpp_video

There's also a video about Inductors and relays , about 555, opamps by the same guy, and they're quite good, might be worth a watch: http://www.youtube.com/playlist?list=PL9EF3C374FD903ACE


It should make all this transistor/fet business clear to you.

[budm]

Re: Voltage regulator max current

More learning here:
http://www.kpsec.freeuk.com/trancirc.htm

I myself stock 1N4007 (1000V) since it the about same price (pennies diff.) for the lower voltage version.

[SaleB]

Re: Voltage regulator max current

I have relearned a lot of things that I have forgotten over the years. Most important of them, that a transistor is a current controlled device, and that a fet is voltage controlled device. That simple sentence gives answers to many of my questions about diodes and resistors.

I have also found the answers about the clamping diode.

I have made a test device using a BUZ11 fet and connected on relay on it. Now, I will assemble the other parts of the circuit (voltage stabilizers) and if everything goes according to plans, I will make the final product.

Thank you all for your suggestions, I will feel free to come back with more questions about some other projects that I have on my mind when I finish this one.

SaleB

[lti]

Re: Voltage regulator max current

How can you tell what voltage rating the diode needs to have?

[larrymoencurly]

Re: Voltage regulator max current

Why does a voltage regulator with built in current limiting and thermal protection fail when you connect a 10,000uF directly across its output?

[tom66]

Re: Voltage regulator max current

Why do you need such large capacitance?

The reason it may fail is if the input is shorted, the internal diode which goes in reverse from input to output (not normally conducting) may begin conducting, which will blow a nice hole in the junction, as that cap is easily capable of 20-30A discharges through the junction.

Fun fact -- I tried to blow up a 7805 by reversing the polarity, and it just drew 3A with no concern (it got hot but it would work afterwards); I decided, therefore, to pulse it with a high capacitive source, and fried it on the first pulse (shorted both directions.) Pulse current can damage semiconductors easily.

[randtek]

Re: Voltage regulator max current

A switching regulator uses frequency for regulation. Adding a huge capacitance like that drastically alters the impedance of the LCR network that is basically what the secondary circuit is. With such a large capacitance, the regulator is unable to regulate itself properly, and burns itself up trying.

[SaleB]

Re: Voltage regulator max current

I have another, not really related question about printed boards design. I tried to find that information on the forum and on google, but had no luck.

Can anybody give me some formula that presents a relation between width/length of a trace on PCB and current that should flow through? Or simpler how wide should be a trace for my relays if it should be rated for 1.5A?

But, I would more like a formula that I can reuse when needed

SaleB

[mariushm]

Re: Voltage regulator max current

http://circuitcalculator.com/wordpre...th-calculator/

About 25-30 mils for 1.5A if the traces are not inside the pcb.

It's based on IPC-2221, a standard for pcb's : https://cdn.badcaps-static.com/pdfs/...5d7751670b.pdf

And yeah, you don't need such high capacitors after regulators. The 10k uF will act as a short and the regulator might enable its short circuit protection. More than 3300uF or so shouldn't be needed.

[tom66]

Re: Voltage regulator max current

I'd say any more than 100uF and You're Doing It Wrong!

[larrymoencurly]

Re: Voltage regulator max current

Why do you need such large capacitance?

The reason it may fail is if the input is shorted, the internal diode which goes in reverse from input to output (not normally conducting) may begin conducting, which will blow a nice hole in the junction, as that cap is easily capable of 20-30A discharges through the junction.

Fun fact -- I tried to blow up a 7805 by reversing the polarity, and it just drew 3A with no concern (it got hot but it would work afterwards); I decided, therefore, to pulse it with a high capacitive source, and fried it on the first pulse (shorted both directions.) Pulse current can damage semiconductors easily.

I was trying to zap whiskers out of nicad batteries (worked for only one cell, unless I always left the battery on trickle charge), and I did discharge the cap first.

[tom66]

Re: Voltage regulator max current

I have done the same but I used a 35V, 3A power supply with 20000uF of external capacitance. That did it, but each cell took at least 100 zaps, and a good charge cycle, before performance returned to normal.