Maker Pro
Maker Pro

Sony Ericsson CST-60 Charger

55pilot

Feb 23, 2010
434
Joined
Feb 23, 2010
Messages
434
An internal fuse is not required to pass UL or CE. As far as keeping costs down, this is one of the more innovative designs I have seen. I do not like designing this far into the bare bones territory, I do not think it is good idea doing it, but if that is what you are into, this is about as bare bones as it gets.

Adding to Mitchekj's explanation, lets ignore the optoisolator for a moment. R5 turns Q1 on, which allows current to flow through the primary. If the current increases more than 70mA, Q2 turns on and starts turning off Q1 regulating the current. Basic two transistor current regulator.

As the voltage in the control winding increases C2/R6 turn off Q2, which causes the current to decrease, which eventually allows Q1 to turn on and the cycle continues. R7/D2/C4 form the flyback for the transformer.

D3 and C3 form a half wave rectifier from the control winding, creating some DC voltage on C3. If this voltage gets high enough, R4 will turn on Q2, turning off Q1. This control the maximum output power. That is where the "short circuit protected" claims comes from.

As the output voltage increases, the current through U1's input increases and its output transistor turns on which turns on Q2, which keeps Q1 from turning on. That is how the output voltage is regulated.

ZD1 would make sense if it were backwards. That would cause the U1 to turn on quite abruptly at a set voltage, which is what you need.

All that said, I would not recommend fixing this. Far too many parts are dead and you may still not have identified the real culprit. One of the features of bare-bones designs is that when they fail, lots and lots of things fail because by design, everything is on the edge of not working. The risk/reward of fixing simply does not make sense. Buy a used one or even a new one and be happy. The voltages here are dangerous and if you screw up you can cause some serious hurt to yourself or others or to your possessions.

---55p
 
Last edited:

TimeManx

May 11, 2010
31
Joined
May 11, 2010
Messages
31
All that said, I would not recommend fixing this. The risk/reward simply does not make sense. Buy a used one or even a new one and be happy. The voltages here are dangerous and if you screw up you can cause some serious hurt to yourself or others or to your possessions.
I'm doing this for more of a learning experience. I'm in second year of college & I know very little:(.

Unfortunately I brought home a 2.5 ohm in place of a 2.5 Kohm for R3 (Yea I'm a moron:eek:). Will it be ok If I connect two 1Ks and a 680 ohm in series? The 1Ks are 1/4 W and the 680 ohm is a 1/8 W (I'm gonna take em out from another board:D).
Originally, it was a 1/8 W 2.5 K.
And how bout I up R2 to 1/2 W from 1/4 W.
 

55pilot

Feb 23, 2010
434
Joined
Feb 23, 2010
Messages
434
Unfortunately I brought home a 2.5 ohm in place of a 2.5 Kohm for R3 (Yea I'm a moron:eek:). Will it be ok If I connect two 1Ks and a 680 ohm in series? The 1Ks are 1/4 W and the 680 ohm is a 1/8 W (I'm gonna take em out from another board:D).
Originally, it was a 1/8 W 2.5 K.
And how bout I up R2 to 1/2 W from 1/4 W.
All that sows that it is not a good idea for you to be using this for learning. As you connect 3 resistors to get your 2.5K you will have an unwieldy setup that can short to things. In a high voltage circuit like this, the resulting sparks and failures will be more spectacular and the you are likely to get hurt.

Therefore I would reiterate my recommendation: DO NOT TRY AND ACTUALLY FIX IT. Leave it as the academic exercise.

---55p
 

Tesla

May 10, 2010
168
Joined
May 10, 2010
Messages
168
Pilot,

More circuit explainations ... very nice.

When I showed TimeManx that he can get a new one delivered for $5, and he still wanted to try to fix it, I knew it was an acedemic excercise. I already provided the warnings. After all, it's junk and can't get any worse. If he can fix it with a few parts from his junk box and learn something in the process ... why not?

What's kinda cool about it is that it's simple enough for us beginners to understand (provided the experts like yourself provide the explainations). I'm still trying to get over that schematic "out of thin air".

I've fixed a lot of electronics over the years, but I'm always interested in theory and design. Every time I fix something I get a little more confident, learn something new, eager to try again, etc. You have to start somewhere, right?

If we can't fix something simple like this, what can we fix? :)
 

TimeManx

May 11, 2010
31
Joined
May 11, 2010
Messages
31
Unfortunately I brought home a 2.5 ohm in place of a 2.5 Kohm for R3. Will it be ok If I connect two 1Ks and a 680 ohm in series? The 1Ks are 1/4 W and the 680 ohm is a 1/8 W.
Originally, it was a 1/8 W 2.5 K.
And how bout I up R2 to 1/2 W from 1/4 W.

I tried it out anyway. Both the transistors blew up right after plugging it in. WOW...
 

Tesla

May 10, 2010
168
Joined
May 10, 2010
Messages
168
I'm doing this for more of a learning experience. I'm in second year of college & I know very little:(.

Unfortunately I brought home a 2.5 ohm in place of a 2.5 Kohm for R3 (Yea I'm a moron:eek:). Will it be ok If I connect two 1Ks and a 680 ohm in series? The 1Ks are 1/4 W and the 680 ohm is a 1/8 W (I'm gonna take em out from another board:D).
Originally, it was a 1/8 W 2.5 K.
And how bout I up R2 to 1/2 W from 1/4 W.

This is a good site for Ohms Law and other formulas.

http://www.allaboutcircuits.com/vol_1/index.html

I personally would use the right part (2.5K Ohm) and up it to 1/4 watt.
 

(*steve*)

¡sǝpodᴉʇuɐ ǝɥʇ ɹɐǝɥd
Moderator
Jan 21, 2010
25,510
Joined
Jan 21, 2010
Messages
25,510
Yeah, I definitely agree that any repairs on this should be of a theoretical nature only :)

I suspect that you'd end up replacing a large number of the components, and without specs for the transistors you'd be left guessing.

Still, as an exercise it is very interesting, and I suggest it is interesting at different levels for those of us with differing skill levels.

For anyone who knows enough that they understand linear regulators, they should know enough that they don't understand switch mode regulators :)

For anyone with an understanding of switch mode regulators, this is interesting because it does indicate what can be done at the bare bones end of the design spectrum.

I'm certain that the circuit diagram is still not quite right, however I found 55p's explanation enlightening.

edit: if you need to be pointed to Ohm's law, then you are especially advised to do the repairs in theory only. If you do want to press on, be very careful and buy several of each component, because you're sure to blow some of them up. It also may be worth operating this in series with a low wattage light bulb (say 15W). This may delay damage -- possibly long enough to see the order in which the smoke starts to appear. Oh, and don't kill yourself.
 
Last edited:

Tesla

May 10, 2010
168
Joined
May 10, 2010
Messages
168
I tried it out anyway. Both the transistors blew up right after plugging it in. WOW...

Sounds like a variac would come in handly right about now.

A heavy gauge extension cord comes in handy so you don't have to holding it when the AC is applied. Working in the garage comes to mind.

This thing really needs a fuse, even if it's only installed externally while testing.

I guess it's time to start over and recheck all the parts. Are you sure those were the proper transistors?
 

(*steve*)

¡sǝpodᴉʇuɐ ǝɥʇ ɹɐǝɥd
Moderator
Jan 21, 2010
25,510
Joined
Jan 21, 2010
Messages
25,510
Sounds like a variac would come in handly right about now.

Unfortunately not :( A SMPS draws heavier current at low input voltages. The problem is that you may actually stress components *more* and actually INCREASE the current if you reduce the voltage.

A heavy gauge extension cord comes in handy so you don't have to holding it when the AC is applied. Working in the garage comes to mind.

Working with one hand in your pocket and with a friend watching you (to resuscitate you) would not be unreasonable precautions to take. Remember that if it takes longer than a couple of minutes for emergency services to reach you...

This thing really needs a fuse, even if it's only installed externally while testing.

Light bulbs are really good -- not as a fuse, but as a current limit.

I guess it's time to start over and recheck all the parts. Are you sure those were the proper transistors?

And connected the correct way around, etc. Changing things willy nilly may easily have damaged more components. It may be fun, but eventually there won't be anything left to make smoke.
 

Mitchekj

Jan 24, 2010
288
Joined
Jan 24, 2010
Messages
288
I'm certain that the circuit diagram is still not quite right.

I never claimed it was perfect, lol! But, I can't find anything wrong... I've traced through net to net, pin to pin a couple times now. The more eyes we get on it though, the better chance we have of getting an accurate schematic. I have no problem admitting something's buggered, so please do peer review. :)

Agreed here that this thing is not worth repairing, though I do get a kick out of trying to understanding something. Especially something like this... like 55p said, this is one hell of a minimal design.
 

(*steve*)

¡sǝpodᴉʇuɐ ǝɥʇ ɹɐǝɥd
Moderator
Jan 21, 2010
25,510
Joined
Jan 21, 2010
Messages
25,510
I never claimed it was perfect, lol!

Gee, don't take my remarks as critical of your efforts. You've done a great job.

You recall my effort is in the round file...

But, I can't find anything wrong... I've traced through net to net, pin to pin a couple times now. The more eyes we get on it though, the better chance we have of getting an accurate schematic. I have no problem admitting something's buggered, so please do peer review. :)

Looking at the low voltage side of the circuit, it appears that the cathodes of the 2 diodes are connected together. and also labelled + on the component side of the board. It looks to me that the stuff to the right of the low voltage filter capacitor is upside-down.

Last time I looked (when I wrote what you replied to), I could swear I was going on the impression that something was wrong on the output side of the optocoupler, but now I look again, it seems right again :)

The stuff that's happening around the base of Q1 looks correct, but the way it's drawn makes the optocoupler's effect difficult to fathom. However I'm not going to criticise you for that either as I can't quickly come up with anything better.

Agreed here that this thing is not worth repairing, though I do get a kick out of trying to understanding something. Especially something like this... like 55p said, this is one hell of a minimal design.

I agree, this is personally one of the most interesting threads I've had the pleasure to be involved in.
 

Mitchekj

Jan 24, 2010
288
Joined
Jan 24, 2010
Messages
288
Looking at the low voltage side of the circuit, it appears that the cathodes of the 2 diodes are connected together. and also labelled + on the component side of the board. It looks to me that the stuff to the right of the low voltage filter capacitor is upside-down.

Last time I looked (when I wrote what you replied to), I could swear I was going on the impression that something was wrong on the output side of the optocoupler, but now I look again, it seems right again.

If I had my druthers, I'd say they screwed up the layout. On the DC output side, notice that the output cap's stripe (-) is on the same copper as the output marked (+) on the silkscreen. That would be bad. If they just swapped the wires (instead of getting new boards spun to fix it) it would at least work, though I believe the zener/resistor would just be shunting some small percent of the output current, not acting as any kind of OVP or clamp. In my second schematic, I labeled "pos" and "neg" as they should be, not according to the silkscreen on the board. (The output rectifier and positive side of the output cap should always be on the positive output, afterall.)

It would make MUCH more sense if the zener's cathode was on the positive output, with the resistor underneath it (just like I had drawn in the first attempt at the schematic.) That would be an effective OVP scheme. Though the way they have the board laid out does not jive. All that just makes me think they screwed up the layout.

IF that were the case, with the zener/resistor in correctly, then when the voltage on the output went above the zener voltage, the opto would start to turn on. That would apply the pulses from the aux coil (after being rectified and smoothed by D3/C3, then it's a straight DC level) right to Q2's base, which would immediately shut off Q1 since it'd be applying 'ground' directly to Q1's base..
 
Last edited:

Tesla

May 10, 2010
168
Joined
May 10, 2010
Messages
168
Unfortunately not :( A SMPS draws heavier current at low input voltages. The problem is that you may actually stress components *more* and actually INCREASE the current if you reduce the voltage.

So a variac is really only for use with non-SMPS? ... older equipment?
 

(*steve*)

¡sǝpodᴉʇuɐ ǝɥʇ ɹɐǝɥd
Moderator
Jan 21, 2010
25,510
Joined
Jan 21, 2010
Messages
25,510
So a variac is really only for use with non-SMPS? ... older equipment?

It's a tool that is best used judiciously, but certainly not on SMPS unless you're wanting to test how they operate at lower input voltages.

I've had PCs continue to operate as the power line voltage sagged so much that incandescent bulbs were a dim red glow (around 60V, down from a nominal 240). I also had experience with another type of PC that was guaranteed to blow the power supply if the input voltage sagged at all. The former one was probably drawing 4 times the typical current and operating without obvious problems, the latter one (and it was from HP in the very late 80's) was obviously operating very close to some limit and the extra current tipped some component over the edge. Operating the latter one using a variac would be a sure way to kill it.

In contrast, a light bulb limits the available power to around half the rating of the bulb, and short circuit current to the bulb's normal current. It also provides a form of feedback that doesn't require you to keep one eye on a meter. It's slightly "rough and ready" but the advantages include the non-linear resistance, the extremely low chance that it will over-temp and burn out, the easy availability in various power ratings, and the very low cost. On the negative side, it doesn't actually vary the voltage to the device under test in the same way as a variac, and it does present it with a varying input voltage.
 

Mitchekj

Jan 24, 2010
288
Joined
Jan 24, 2010
Messages
288
Ok, well... call me crazy. I just ordered one of these things. I don't know why, it's just so novel to me. I just have to know how it works, in-and-out. Hehe. In a few days maybe I'll have some scope screen caps, etc. if anyone was as interested as I.

Should be a good use for my new isolation transformer which comes soon, too.
 

TimeManx

May 11, 2010
31
Joined
May 11, 2010
Messages
31
Ok, well... call me crazy. I just ordered one of these things. I don't know why, it's just so novel to me. I just have to know how it works, in-and-out. Hehe. In a few days maybe I'll have some scope screen caps, etc. if anyone was as interested as I.

Should be a good use for my new isolation transformer which comes soon, too.

You are crazy.
 

55pilot

Feb 23, 2010
434
Joined
Feb 23, 2010
Messages
434
In a few days maybe I'll have some scope screen caps, etc. if anyone was as interested as I.
For someone who has the experience to safely work with high voltage, it would be a very fun and educational exercise. Thanks for volunteering!

For a fleeting moment I considered doing the same myself but then the realist in me reminded me of the several similar projects scatter on and under the workbench....

I am looking forward to seeing your scope plots!

---55p
 

TimeManx

May 11, 2010
31
Joined
May 11, 2010
Messages
31
Guess what, I replaced the damaged components again & this time it worked. 5V at the output.
 
Top