The Inverter Works!!!.... Well Sort Of

[zking786]

I have finally gotten the inverter to work. It seems that there was a minor wiring error. I'm not completely sure what was wrong, since I had everything wired correctly, but when I resoldered a couple of wires (which still had solid connectivity), it worked. Maybe it must have been a minute resistance error?

 

Anyway, there's a new problem .

The described circuit is meant to handle 300W, but it seems way to weak. I've tried a laptop charger plug (nearly 90-100W) and it barely powers it. When I connect a 100W soldering iron, it doesn't even heat. Why?

 

Also, I notice the resistors seem to heat up pretty quickly. Sometimes it's the small 180 ohm, 1W resistors, other times (and more rarely) it's the 10 ohm, 5W resistors. I'm wondering why, and whether I should replace these with higher power rated resistors. I've fused the battery-circuit line at 7.5A and the fuse hasn't blown, so it can't be drawing too much current. If I'm not mistaken, the transistors (and the whole circuit) should be able to handle 15A current draw. Any idea what's the problem?

 

For reference, here's the site for the original schematics:

http://www.aaroncake.net/circuits/inverter.htm

[YT2095]

remember that the 15A quoted for a 2n3055 is a MAX limit and requires a heatsink, free air disipation is considerably Less than this!

secondly, the 300W quoted is a technical impossibility!

a 2n3055 will output 115W max!

you can`t get something for nothing.

[5614]

If something is heating up quickly then if you get any faults that component should be the first one you check because, well, from experience I'd say a resistor which heats up quickly will die quickly!

[zking786]

I'm thinking of putting on better resistors. Maybe 180 ohm, 5W and 10 ohm 10W. I don't think it could hurt to increase the power the resistors can handle.

 

Also, I have recently purchased and setup heatsinks for each of the transistors.

 

How do you know the max power is 115W, though. Wouldn't it be 12V x 15A = 180W?

 

When I hooked up the 100W soldering iron, it didn't even light up. The circuit did, however, draw about 4A from the 12V battery. Wouldn't it be able to draw 15A before skimping out? Any idea why the iron didn't heat?

 

Also, will 200 ohm, 10W resistors do if I can't find the 180ohm ones? (This is, of course, assuming you recommend increasing the power rating of the resistors.)

[zking786]

No advice?

[5614]

No prob with increasing the power of the resistors.

 

180Ω to 200Ω is no problem, although 180Ω would be preferable.

 

The 2n3055 will only output 115W that's a fact. If you want info on a component like then do a search like this:

http://www.google.co.uk/search?hl=en&q=data+sheet+2n3055+&meta=

and that's where YT got the figure from (possibly) and it is where I double checked it, it's correct.

 

Heatsinks are good.

 

Remember just because a circuit can handle 15A doesn't mean it will draw it.

 

You def using tantalum capacitors for C1 and C2?

 

And you cheked out his forum here:

http://www.aaroncake.net/forum/topic.asp?TOPIC_ID=2996&CAT_ID=10&FORUM_ID=18&Forum_Title=Power+Supply&Topic_Title=12%2F120V+inverter+again

it's linked to on the page with the schematic.

[zking786]

Definitely Tantalum. I have everything on the circuit.

 

I tried the 200 ohm (219 to be exact) resistors and I got significantly less AC voltage. I have purchased 2160 ohm 1W resistors to hook up in parallel and these should get me closer to 180 ohms. I'm wondering whether the power rating on these is high enough, though. If I analyzing the circuit, I find that

 

i=I(219)/(219+2160)

 

where:

i=the current drawn from the branch with the 2160 ohm resistor

I=the current entering the branch (of both the 2160 ohm and the 200 ohm)

 

Since the resistor is 1W, then should 1=i^2*(2160) ?

 

What do you think of this modification? Should it allow me to draw more current?

 

When a circuit like this outputs less power, could it be the resistors that are limiting it? Or would they just heat up and explode?

[YT2095]

5614, I`ve been using the 2n3055 for decades, I know the specs off the top of my head, but I`m pleased to see this data can be verified on the net too

 

upping the wattage of the resistors will Never be a problem while Space on the board allows, changing the Values WILL however (as you`ve seen) this particular circuit is what I`de call a "Cheap and Nasty", run on a razors edge of breakdown and barely within tollerances of the parts involved, altering ONE can (and probably Will) bring the others down

 

with resistors in series you simply ADD them, in parallel you simple take the reciprocal value of each resistor, add them and then take the reciprocal of the result.

(capacitors are the same but the other way around for parallel/series).

[zking786]

What's odd is all my tinkering must have done something to one of the tantalum capacitors. Now, if I plug the circuit in (with the correct resistors), I get about 12VAC out of the transformer (instead of 120VAC). If I do a DC test on the capacitors, I get one charging to about 17.5V and the other to 2.5V. Is it blown, or gradually deteriorating?

[YT2095]

if you test these OUT of Circiut (pointless otherwise) and there`s a fantastic difference, then yes, one is certainly FUBAR

 

anyway, I`ve told you this before, get a High impedance speaker with approriate resistor and LISTEN!

if you can`t hear Oscillation from the cct, it`s NOT WORKING!

please do as I told you, and use this simple but effective test method

[zking786]

Will try that test and get back to you.

 

What would you clasify as "high impedence"? Would 10 ohm do?

 

BTW, when I have the transformer hooked up to the circuit, I do hear oscillation. There's a soft ticking noise from the transistors. When I disconnect the transformer, the ticking noise stops, though. Is this what you were talking about?

[YT2095]

yes, that`s the idea

it just gives Some indication that it`s working, even if it`s at a low power, for High Power/voltage tests using a little neon pannel lamp with a suitable resistor`s a good idea too (put this across the transformer output) if it lights up, you know you have a min of 90volts coming out

[zking786]

Not high enough to power a bulb, but high enough to cause the transistors to tick. Must be a capacitor that could have blown or been damaged. Could this cause the resistors to heat up?

[YT2095]

look at the Circuit diagram, what does it say to YOU?

it`s ok Building these things blindly with no understanding of it, but along the way with experimentation you should also LEARN things for yourself TOO.

 

so, You tell Me, is a faulty cap likely to be causing problems here for that resistor (you Have in your Other threads about this been given ample info and links to work this out yourself), Think ASTABLE Multivibrator circuits here

think also of WHY these resistor have a high(ish) power rating each, Are they suposed to get hot! even?

 

 

failing that, spend the money, send the thing to me and I`ll do it for you

[zking786]

Haha...

 

I acually intend(ed) to learn something from this project. From my experimentation, I have found that as I increase the resistance of the 180 ohm resistors, the AC current generated decreases (by about half). I don't get the mathematics behind why this happens, but conceptually it sort of makes sense.

 

Here's how I think the circuit works:

There are two transistors with bases controled by the two charging and discharging capacitors. These bases act as gates allowing the collector current to be amplified through the emitter. I'm guessing that the frequency of the current is generated by the charging and discharging of the capacitors. Thenceforth, I don't understand.

 

I'm not sure exactly how these capacitors are charged to create the frequency. From my limited experience, it seems a capacitor, when connected to a power source, will only charge--not discharge and then recharge. How, specifically, is the frequency created?

 

Maybe I should try using capacitors of different capacitances? Maybe this will alter the frequency? Can someone explain, mathematically, how the circuit works?

 

As for the heat issue, I would think that the resistors would heat, but not too much. I mean, they chose high watt resistors to avoid the heat, right? I've more than doubled the power ratings, so I shouldn't expect any significant heating.