Site Logo

 
It is currently Thu Jul 31, 2014 9:47 pm

All times are UTC - 6 hours




Post new topic Reply to topic  [ 75 posts ]  Go to page Previous  1 ... 4, 5, 6, 7, 8  Next
Author Message
 Post subject: Re: Hot Wire Post Mortem
PostPosted: Tue Mar 10, 2009 7:37 am 
Offline
Global Moderator
User avatar

Joined: Mon Jan 12, 2009 10:24 pm
Posts: 634
uptownmaker wrote:
You keep talking about learning- maybe the lesson here is when to abandon a solution that will never work and move on to one that will?

You won't get any argument from me there, but I don't feel like I've exhausted the possibilities with this yet, and analyzing why it is failing and perhaps will never work is helping me understand some fundamentals. I'm at the bottom of a very steep learning curve (and am not always that great of a climber), so I need to take any opportunity I can to understand.

Quote:
What is the amperage of your fuse? Have you calculated the delivered power to your wire at that current? If it's a 1A fuse, your maximum delivered power will be 2W, which isn't going to cut butter.

2A fuse, that's what the transformer is rated at. How do you arrive at your delivered power number of 2W?

Quote:
You HAVE to look at max power. You have a center-tap 120V-24V transformer, which means you're going to get either 12V or 24V out of the thing. You have a two-ohm load which means that you're either going to be pushing 6A or 12A through the load- which implies either 288 or 144W of delivered power. If you put the rectifier in there, you can reduce the power by 1/2, but you're going to need to find a rectifier that can handle a minimum of 6A of instantaneous current, which is close to half a watt of instantaneous power (6A x .7V). Your average power will be much lower, but 6A is still a lot of current.

I still don't understand this, won't it only be at max power if the input to the transformer is at 120V?

_________________
Make it happen.


Top
 Profile  
 
 Post subject: Re: Hot Wire Post Mortem
PostPosted: Tue Mar 10, 2009 8:30 am 
Offline

Joined: Tue Jan 13, 2009 8:26 am
Posts: 378
Here's the power delivery calculation method:

First we start with Ohm's law:
Voltage = Current x Resistance (V=IR).

Now we add in the power calculation:
Power = Current x Voltage (P=IV).

If we shuffle things around with algebra a bit, we can write
Power = Current x Current x Resistance (P=I^2*R) or
Power = Voltage x Voltage / Resistance (P=V^2/R).

You have a 2A fuse- that fixes the maximum current. Your load is about 2 ohms. Using the equation above, the maximum deliverable power to your wire is
P= I^2*R= 2A * 2A * 2 ohms = 8W.

That's non-negotiable- you aren't going to deliver more than 8W to that wire if you have that fuse in the circuit. Bump it up to a 3A fuse, and the equation changes:
P=3A*3A*2 ohms = 18W.

Now we're talking! Notice also that a 50% increase in current resulted in a 225% increase in power delivered, because of the square term. Since this is a purely resistive load, we can now calculate how much voltage we need to dump across that resistor to provide it with a 3A current, because Ohm's law is in full force:
V= IR = 3A * 2 ohms = 6V.

Of course, that's cutting it close. Go up a little and you blow the fuse, so you might consider either upping the current rating on the fuse to 3.5A or dropping the current by a smidge- say to 2.5A.
P = 2.5A * 2.5A * 2 ohms = 12.5W
V = 2.5A * 2 ohms = 5V

So, how do we now GET a supply like that? Well, now we start to run afoul of the difference between AC and DC. With a DC source, voltage adjustment is fairly straightforward to understand- voltage goes up or down. But with an AC voltage, it becomes more complex. That dimmer switch is probably (though not certainly) chopping up the waveform it gets on its input before it sends it to the output. I don't know what happens if you feed that chopped up waveform into a transformer, but I'm betting it's not exactly what you have in mind. I don't know what happens if you feed a 24V AC signal to that dimmer switch, but I'm betting it's not exactly what you have in mind- although I'd guess it's closer to what you'd have in mind than connecting the dimmer BEFORE the transformer.

_________________
Mike H.
My maker blog


Top
 Profile  
 
 Post subject: Re: Hot Wire Post Mortem
PostPosted: Tue Mar 10, 2009 8:49 am 
Offline
Global Moderator
User avatar

Joined: Mon Jan 12, 2009 10:24 pm
Posts: 634
OK, now we're getting somewhere, it's starting to make sense to me, thanks for being patient.

So really my only options are more current or more resistance, right?

So I wonder how sensitive the transformer is to running current a bit higher than it's rated. My understanding is that they are pretty sensitive, but I expect it was running at much higher than 2A to get any heat in the galvanized wire and when I was first testing I had it glowing red (and it still worked after that).

I even have proof. :)
Image

_________________
Make it happen.


Top
 Profile  
 
 Post subject: Re: Hot Wire Post Mortem
PostPosted: Tue Mar 10, 2009 9:07 am 
Offline
Global Moderator
User avatar

Joined: Mon Jan 12, 2009 10:24 pm
Posts: 634
Also, what is the cause of failure on a transformer, is it purely an overheating issue? If I'm overloading it am I just melting the insulation and shorting it out?

Could I add a couple of heat sinks and some decent airflow to get more life out of it, or perhaps remove the metal enclosure altogether to improve cooling?

Edit: I just realized I have a dead transformer sitting here, time for an autopsy. *dives for Dremel*

_________________
Make it happen.


Top
 Profile  
 
 Post subject: Re: Hot Wire Post Mortem
PostPosted: Tue Mar 10, 2009 11:52 am 
Offline
Global Moderator
User avatar

Joined: Mon Jan 12, 2009 10:24 pm
Posts: 634
Didn't need a Dremel, barely needed a tool to take the thing apart, it was just glued together.

Definitely a heat issue, not sure cooling would help much, there is actually cardboard lining the inside of the metal case, by the time the heat gets to the surface the damage has probably already been done.

Image

_________________
Make it happen.


Top
 Profile  
 
 Post subject: Re: Hot Wire Post Mortem
PostPosted: Tue Mar 10, 2009 12:17 pm 
Offline

Joined: Tue Jan 13, 2009 8:26 am
Posts: 378
Side note- you don't want your wire for the cutter glowing red. Bad gases are released by expanded polystyrene foam when it burns.

On to the main event: any engineer worth his salt puts in at a minimum a 50% safety margin. If Radio Shack stamped "2A limit" on your transformer, you can bet that it'd be good to 3A. At least. Possibly more, with forced air cooling. When it fails, it fails by burning through insulation on the wires and shorting out, which causes MORE current to flow which causes more heat which causes...POOF!

To understand the practical limit, let's consider how a transformer works. Imagine an iron picture frame (no glass or anything, just the frame). Take a wire, wrap it around one side of the frame, say, 10,000 times. Now take another wire, go to the opposite side of the frame, and wrap that wire around the frame, say, 1,000 times. Voila! You have a transformer, with a 10:1 ratio. Put 120V on the 10k turn side, you get 12V on the 1k turn side. The flip side is that the actual power has to remain the same, so the equation becomes
Pin = Pout
Vin * Iin = Vout * Iout

So if voltage goes down, current goes up. We'll use your transformer as an example: It's essentially a 5:1 ratio- 120V in, 24V out. It has a 2A limit- so let's consider what the input current is when the output current is at 2 amps:
Vin * Iin = Vout * Iout
120V * Iin = 24V * 2A
Iin = .4A

This is called a step-down transformer. You COULD flip it around, put your 120V line across the 24V terminals, and step up the output voltage to 600V, but I wouldn't recommend it, because the insulation on the transformer is probably not up to blocking 600V, which means you'd have a punch-through of the insulation somewhere, which leads to arcing, current flow, and POOF!

So why the limit? Why can't I just increase the load and draw more and more current from the line voltage side, until I trip my house's breaker? After all, the wall outlet probably has a 15 or 20A breaker, right?

Not that simple. I ignored a little detail earlier- efficiency of the transformer (usually given as the greek letter eta, but frequently shorthanded to a n for people who don't want to find the unicode value for eta. I'll use n). The REAL equation is
Pout = n * Pin

n is usually on the order of 80-90%, or .8 to .9. That means that whatever your input power is, 10-20% of it going to get bled away as heat in your transformer. Let's look at your transformer again:
Pout(max) = 2A * 24V = 48W
Pin = 48W / .85 = 56.5W input power
Ptransformer = 56.5-48 = 8.5W

So, using a fairly unremarkable value of .85 as the efficiency of the transformer, we find that at 2A, your transformer is sucking up 8.5W of power, just to warm itself up. Not too bad, but you'll notice it if you hold the thing in your hand- it'll be quite warm.

In your case, you maxed it out, got WAY too much current flowing in and out of your transformer, and that efficiency killed it. Unlike silicon based electrical elements, something like a transformer can handle quite a lot of current for some time. We just have no way of knowing HOW long...

That 2A rating is undoubtedly for continuous operation, in a somewhat enclosed environment, without cooling. If you put it on a table with a fan blowing across it, that number will go up. Wrap it in a blanket, that number will go down.

_________________
Mike H.
My maker blog


Top
 Profile  
 
 Post subject: Re: Hot Wire Post Mortem
PostPosted: Tue Mar 10, 2009 12:19 pm 
Offline
User avatar

Joined: Wed Jan 14, 2009 10:06 am
Posts: 1219
Location: NE Minneapolis
well... poping the case adn ventilating the thing can't hurt.... provided thigns don't arc....


Top
 Profile  
 
 Post subject: Re: Hot Wire Post Mortem
PostPosted: Tue Mar 10, 2009 10:52 pm 
Offline
User avatar

Joined: Wed Jan 28, 2009 10:43 am
Posts: 118
Location: Minneapolis, MN
this has both been enlightening, AND entertaining.
it seems far simpler [via Occams razor = more robust] to make if your objective was to simply make a working hotwirecutter.

but here at tcmaker - were all in love with the learning!

also: brought to you by the letter ohm.


Top
 Profile  
 
 Post subject: Re: Hot Wire Post Mortem
PostPosted: Wed Mar 11, 2009 6:43 am 
Offline

Joined: Tue Jan 13, 2009 8:26 am
Posts: 378
Nicholas Lee wrote:
also: brought to you by the letter ohm.


Greek capital omega, actually, which is NORMALLY unicode ALT + 937 but on the font used for this forum's postings, ALT + 937 gives me a weird little shape (⌐).

Oddly enough, ALT + 0937 should give the same symbol (i.e., in other text editors I see omega for either of them) but here it gives me the copyright symbol (©).

And if you don't know what I'm talking about, try it: open a text editor, hold the ALT key, and start punching in four-number codes. You'll find a whole new set of characters to play with. You can figure out what's what by using the windows character map (Start>Programs>Accessories>System Tools, or Start>Run... charmap). Obviously there are a limited number of truly regularly useful characters available, but it's nice to have access to things like the degree symbol (°)(0176).

Note that this is NOT 100% portable unless you are using one of the Windows system fonts, as those display the same pretty much everywhere (Times New Roman, Courier, or Arial, for instance). And even in that case, you may have mixed results with people in foreign lands.

_________________
Mike H.
My maker blog


Top
 Profile  
 
 Post subject: Re: Hot Wire Post Mortem
PostPosted: Wed Mar 11, 2009 6:48 am 
Offline
Global Moderator
User avatar

Joined: Mon Jan 12, 2009 10:24 pm
Posts: 634
˙˙˙˙˙ɯɯɯɯɯɯɯɯɯɯɯɯɯɯɥo

_________________
Make it happen.


Top
 Profile  
 
Display posts from previous:  Sort by  
Post new topic Reply to topic  [ 75 posts ]  Go to page Previous  1 ... 4, 5, 6, 7, 8  Next

All times are UTC - 6 hours


Who is online

Users browsing this forum: No registered users and 2 guests


You cannot post new topics in this forum
You cannot reply to topics in this forum
You cannot edit your posts in this forum
You cannot delete your posts in this forum
You cannot post attachments in this forum

Search for:
Jump to:  
cron
Powered by phpBB © 2000, 2002, 2005, 2007 phpBB Group
Template made by DEVPPL Flash Games