SuzukiSavage.com - Print Page

SuzukiSavage.com
/cgi-bin/YaBB.pl
General Category >> The Cafe >> Kind of a random topic
/cgi-bin/YaBB.pl?num=1342069771

Message started by vineman on 07/11/12 at 22:09:31

Title: Kind of a random topic
Post by vineman on 07/11/12 at 22:09:31

Ok, so I know this may be a random question but are there any electrical engineers on here? I have a fairly technical question to ask and thought I'd probe around here for answer. (of course I'm also asking else where)

Title: Re: Kind of a random topic
Post by rfw2003 on 07/11/12 at 22:17:32

I'm no electrical engineer, but I do know my way around electronics and other electrical projects fairly well.

Just depends on what your asking.

Title: Re: Kind of a random topic
Post by thumperclone on 07/11/12 at 22:18:49

sure you need an engineer??

ask your question you never know ;)

Title: Re: Kind of a random topic
Post by vineman on 07/11/12 at 22:38:44

OK, bear with me here, i am in no way very electrically adept but I'm always willing to try something once.

I need help in designing a switch that when a device ceases to receive electrical current from a standard 110 wall plug (AC yes?) that device then switches to running off of an internal battery source (DC right?). I know these exist obviously, laptops for example, but if this switch type already has a name I would love to know what it is.

Also, this is for a small scale build. Say utilizing a board about 1" square.

Any questions to clarify what I'm aiming for please ask! Though I may only divulge only so much info as to elude to what I am designing.

Funny that I ask for an electrical engineer when really most anyone with any electrical experience could answer my question and any others that follow. :P

Title: Re: Kind of a random topic
Post by rfw2003 on 07/11/12 at 22:42:44

Your talking about an Auto transfer switch, similar to what is used for generators. They also have the smaller ones that are used in computer UPS's as well that may fit the size requirements in what you are looking for.

R.F.

Title: Re: Kind of a random topic
Post by vineman on 07/11/12 at 22:53:34

697D6C292B2B281B0 wrote:

So this switch could be engineered to the scale of say...the size of a 1" cubed space? How so? maybe a little bigger at a 2" cubed space?

I guess my biggest question would be what components are involved?

Title: Re: Kind of a random topic
Post by rfw2003 on 07/11/12 at 23:18:39

The simplest way would be the use of a relay for the switch over, but to get it as small as possible you could use TRIAC's so that it could pass the AC currents and be in a smaller package. Alot depends on the amount of current you are needing to pass as to what options you have for building this transfer switch.

R.F.

Title: Re: Kind of a random topic
Post by Cavi Mike on 07/12/12 at 02:54:16

The easiest way to do this to make sure the power supply is truly uninterrupted is to get a power-supply that's the same voltage as what you want and then get a battery that's the same voltage. When you unplug the power-supply, the battery just keeps powering everything.

Laptops don't "switch" from AC power to DC power, they're always DC power. Laptops just will dim the screen and slow the speed of the processors and even the hard-drives when they're unplugged to conserve battery power. They're always running off the same circuit the battery is on.

Uninterruptible power supplies don't switch from AC to DC either. They have power inverters that run off DC and the AC line supplies a power supply that powers that inverter and charges the batteries. When the AC gets cut, nothing happens except that the batteries are no longer getting charged.

It's pretty simple really, it's no different than the charging system in a car except instead of an engine spinning an alternator, it's a 120v AC line powering a DC power supply. Turn off the engine, the lights don't turn off. No switches flick to change from alternator to battery power. It's just up to the battery to keep the lights on, the alternator can't do it anymore.

Title: Re: Kind of a random topic
Post by Paraquat on 07/12/12 at 09:03:47

A normally-open relay wouldn't work?
I recall seeing a few on www.hackaday.com but some searching is required.

--Steve

Title: Re: Kind of a random topic
Post by 360k+ on 07/12/12 at 09:53:54

Is the 1 inch gadget operating from AC or DC; i.e., is there an onboard power supply? If it is a DC gadget then it's pretty easy to use one or more diodes to isolate the battery and power supply DC voltage. The diodes prevent the DC supply from trying to back charge the battery (assuming non rechargeable), and/or the battery from trying to dump current into the DC supply when it goes nonop. If the battery is rechargeable, then there are other ways to wire the diodes so that charge the battery where there is input power. There's no need to engineer anything as this is well understood technology and there are lots of examples of this on the web. For instance, automatic emergency lighting that lights up when power is removed have the circuit you might be looking for. You can build or buy these units that plug into individual AC sockets.

BTW, I'm an MSEE.

Title: Re: Kind of a random topic
Post by ZAR on 07/12/12 at 10:40:16

quoting 360K+ "BTW I'm an MSEE"

But are you radioactive??? de KG4ZAR

Title: Re: Kind of a random topic
Post by 360k+ on 07/12/12 at 11:14:39

4C5744160 wrote:

quoting 360K+ "BTW I'm an MSEE"

But are you radioactive??? de KG4ZAR

Funny you should ask, because I might be a little bit. I used to work for a company that manufactured "fill height detectors" that we installed in breweries throughout the world. These machines were used on can lines to detect under filled sealed cans of beer. They used an Americium 241 alpha emitter source and scintillation detectors (Geiger counters). There were many occasions where I had to stick my hand in the machine while it was running to clear a toppled can of beer and probably got a dose of alpha radiation (we wore dosimeters, tho).

Title: Re: Kind of a random topic
Post by vineman on 07/12/12 at 20:40:14

4D4B5853131F1918132A0 wrote:

Well imagine the device i running when plugged into the wall and then when the power i cut, or turned off, the device continues to run but on a small 3 volt battery. 3 volts is really all the power this device needs.

That is essentially what I need to wire. If I have to have it where the power from the wall just charges the battery and the battery runs the device then that's fine as well. I would just imagine that the former method would result in a much smaller package.

Title: Re: Kind of a random topic
Post by 360k+ on 07/13/12 at 08:37:01

Well, you're mostly right. Usually it depends on how much power the device requires and/or how long it is expected to operate without power (outage)? For instance, emergency lighting may require substantial power, but may only need to run for 20 minutes until everyone has exited. If your device is a tiny nano power type gadget that runs for years on a 2032 coin type power cell, then there's no need to rig a charging system. On the other hand, if it's a lighting, noise making, or motorized device then it probably requires larger AAA or AA cell power, and rechargeables might make sense.

Most engineering projects go thru 4 distinct phases...

1) a clear problem definition (don't underestimate this; it sets your starting point for the rest of your solution)

2) a proposed solution with goals it will solve & achieve (i.e., a gadget)

3) a REQUIREMENTS phase where you decide the physical dimensions, power requirements, color, noise level, etc.

4) the DESIGN phase where you build one or more prototypes that meet or exceed the REQUIREMENTS you have defined. These units are used for testing. If they fail, then you may to restart step 4 or even go back to step 3 and rethink the project from a different approach. Note, this is called empirical scientific method.

Keep in mind if you're applying AC directly to this gadget (rather than using a DC wall PS) it could be LETHAL! Most power supplies use a small transformer which steps down the voltage, but also electrically isolates the circuit from the AC line. You can apply direct AC (rectified, of course) to a circuit, but it can be tricky and DANGEROUS if you don't have a good understanding of AC/DC principles.

Title: Re: Kind of a random topic
Post by vineman on 07/13/12 at 21:34:01

Ok so i got to taching to a tech savvy friend of mine and he told me of a double-pole double-throw (dpdt) relay that I could use to switch from ac (wall) to dc (battery). this would enable me to not have to set it up with the battery charging off the wall. This relay would need to be a 3v relay and if at all possible a reed relay as well.

Also, I need to find a transformer of some kind that will convert 110v to 3v in a fairly small package. Im thinking possibly the kind of transformer that resides in a cell phone charger, small and compact and able to convert to such a small voltage. This same friend of mine gave that type of transformer a name but I cannot for the life of me remember what it was!

Title: Re: Kind of a random topic
Post by 360k+ on 07/14/12 at 13:32:19

You're making this much harder than it is. First, a relay is not needed, and in fact using a relay can cause a momentary power loss and some glitches as the contacts switch. Second, you do not need to use a 3V transformer. Using a 3.3V DC regulator will allow input of up to 25VDC to the regulator. This means a 5:1 step down transformer will suffice (assuming 120VAC). This transformer will be much more common, but you are still going to be hard pressed to find one that will fit into a 1 cubic inch gadget (was your size choice arbitrary?). Also, note that the battery does not have to "charge". In fact, trying to charge a lithium 2032 button cell, may cause it to explode. Besides, a nonchargeable backup battery is even easier to implement anyway.

Here is the basic circuit you are looking for. Note that a regulator can be chosen for battery backup circuits that prevents back flow from the battery thru the regulator. Diode D2 prevents the battery from charging while under power, but allows the circuit to instantly switch to battery power when AC is removed. Keep in mind that silicon diodes drop approx 0.7V so a 3V battery would supply 2.3V to the load. If this is insufficient then using more than 1 battery and different voltage regulator can be utilized. It depends on how voltage critical your circuit is and/or needs to be? Remember the design steps I outlined earlier. Voltages and dimensions are chosen based on real requirements, and not simply because you would like it to be that way.

http://1266.photobucket.com/albums/jj534/gary95329/Misc/BatPS2.jpg