You will get better results if you connect a trickle charger across C4. This is the only place to connect a power source.

Don't connect a charger across the battery while the desulphator is running. It won't be good for either the charger or the desulphator.

 

We have made a pcb for de-sulphator, i have also attached the picture of that.
Now the mosfet and c4 get hot gradually . Inductors and diode are cool to touch. There's a slight humb.
If we replace the l1 with the lower value inductor the mosfet and c4 are cool to touch but the humb noise can be heard.
what could be the solution with the higher value inductor.

 

First of all I am not expert in this topic The humming sound may tell you if the circuit works. But it will not tell you about how successful the ongoing process is. If the process works you will se see a slowly increase in the battery EMF. This because at least in theory. The battery desulfator process removes the sulfate coating the builds up on the lead plates in lead-acid batteries, and increase the battery internal resistance

Thank you very much for the advice. it was very helpful.we are assured by the humm that the device is working.how efficient?? we don't know.

 

Thank you very much for the advice. it was very helpful.we are assured by the humm that the device is working.how efficient?? we don't know.

Hope it work out for you. How did you made your circuit board. Did you use etching?

 

Hope it work out for you. How did you made your circuit board. Did you use etching?

Yeah hope for the best. We made a printed circuit board for the circuit by etching.

 

Progress! That is good.

What voltage do you measure across C1?

What toroids did you use for L1 and L2 (manufacturer and part number) and how many turns did you wind on each? (What counts is how many passes did the wire make through the center of the toroid; winding evenly is important).

I see that you did not use tape on the toroids before winding them. This can lead to rapid saturation of the toroidal core, which will result in high current in Q1 (the MOSFET).

Have you seen Ronald Dekker's page?
http://www.dos4ever.com/flyback/flyback.html
It's a really good resource for people wanting to experiment with inductors and switch-mode power supplies; the desulphator is actually very similar to a portion of a boost-type power supply.

Looks like you used aluminum electrolytic caps for both C1 and C4. They have higher ESR (equivalent series resistance) than other types. You will likely get better performance (and cooler caps) if you put several caps in parallel with C4. The exact capacitance is not terribly important; anything from 0.1uF to 100uF will help keep things cool.

Is C4 rated for at least 25v? (even that is marginal for a fully charged battery)

Standard BJT 555 timers can only get within about 1.3v of Vcc. That should be sufficient to turn off the MOSFET. However, if the voltage across C1 is lower than about 11v, the MOSFET may not be getting fully turned ON, which would lead to heating.

You can connect a trickle charger across C4. This will help a good deal to desulphate the battery more quickly.

 

Can you show the bottom of the board? The traces for the MOSFET source, drain, and L1/L2 to the battery connections should be quite wide.

 

Progress! That is good.

What voltage do you measure across C1?

What toroids did you use for L1 and L2 (manufacturer and part number) and how many turns did you wind on each? (What counts is how many passes did the wire make through the center of the toroid; winding evenly is important).

I see that you did not use tape on the toroids before winding them. This can lead to rapid saturation of the toroidal core, which will result in high current in Q1 (the MOSFET).

Have you seen Ronald Dekker's page?
http://www.dos4ever.com/flyback/flyback.html
It's a really good resource for people wanting to experiment with inductors and switch-mode power supplies; the desulphator is actually very similar to a portion of a boost-type power supply.

Looks like you used aluminum electrolytic caps for both C1 and C4. They have higher ESR (equivalent series resistance) than other types. You will likely get better performance (and cooler caps) if you put several caps in parallel with C4. The exact capacitance is not terribly important; anything from 0.1uF to 100uF will help keep things cool.

Is C4 rated for at least 25v? (even that is marginal for a fully charged battery)

Standard BJT 555 timers can only get within about 1.3v of Vcc. That should be sufficient to turn off the MOSFET. However, if the voltage across C1 is lower than about 11v, the MOSFET may not be getting fully turned ON, which would lead to heating.

You can connect a trickle charger across C4. This will help a good deal to desulphate the battery more quickly.

We have attached the bottom picture of PCB (though crude) but i think it meets the requirement.
We are using C4 100uF and 35V rating. We intend replacing it with 470uF 50V Caps.
The battery voltage is 12.6V.
Voltages measured across c1 = 10.63v and across c4 = 12.55v
We shall tape the toroids before re-winding them.

 

We have attached the bottom picture of PCB (though crude) but i think it meets the requirement.

No complaints; I have made boards much more ugly than that which worked.

However, it looks like your solder joints did not "flow" very well in a number of places, particularly around larger wires. This may be due to using a soldering iron that had a dirty tip, or was simply not hot enough (or high enough wattage) to heat the area sufficiently before applying solder, or the component leads may not have been clean enough prior to soldering.

We are using C4 100uF and 35V rating. We intend replacing it with 470uF 50V Caps.

What will help more is to have multiple small caps in parallel. For a given cap design, the larger it is (uF), the higher the ESR will be. Using multiple smaller caps will help to lower the ESR, and spread the heat around.

The battery voltage is 12.6V.
Voltages measured across c1 = 10.63v and across c4 = 12.55v
We shall tape the toroids before re-winding them.

OK.
Can you give us the manufacturer and part number of the toroids? Or are they salvaged from something?

If you can give manufacturer and part number, I can tell you just about what the value of the inductor will be when wound with a number of turns.

 

I tried to zoom in the best i could, and it appears you have at least one, but it could be two solder-bridges. I circled them in black in the attached photo. With the size of the cap so near the leftmost bridge, you could be discharging to the next pin.

 

I don't know that they're solder bridges, but the excess lead lengths should be trimmed.

I've marked a few areas of the board to re-flow. Generally, the surface of the solder joint should look bright and shiny, and like it was still molten. The component leads should look like the solder is adhering to it all around evenly. This is hard to do when you have big areas that are copper clad, as the copper conducts the heat away from the area you're trying to solder - particularly hard if you have a low-wattage iron.

 

Thank you for sparing time to look at that ugly contraption . the photogrphy is even worse. I checked the connections on the two points suggested by you with a magnfier.they are safe.one is in the 'air', the other is spreading flux

 

Actually, I circled three joints to reflow, and two leads to clip/trim.

It really isn't so ugly or bad. It's just hard to solder due to the large amount of copper around those areas.

The photography isn't bad, either - better than many other such images that have been posted. It's really hard to take a good image of the reverse side of a board; you really have to use bounce/diffused flash, or better yet take the photo outdoors on an overcast day.

 

If the flux is connecting two solder points, (depending on the flux) you could have a short there also. After re-flowing to Wookies specs, you should clean the board to remove flux spread.

 

Rosin flux does have a small amount of conductivity, but it's quite low. Still, it may be enough to affect the timing on the 555, as high value resistors have been used.

Get yourself some 90% Isopropyl Alcohol (or better) - don't use the 70%; too much water in it. Ace Hardware sells 99%, but it's expensive. Wal-Mart sells 90% for a couple of bucks for a quart.

Get some acid brushes. These have black nylon bristles and a rolled sheet-metal handle. Auto parts stores and hardware stores carry them. Apply the isopropyl alcohol with the acid brush and scrub it around. It will all clean off.

It's a good idea to wear rubber gloves when using isopropyl alcohol, as it will dry out the skin. It is also flammable, and burns with a nearly invisible blue flame. However, it is one of the more readily available cleaners that works pretty well, and is relatively safe.

Don't use lacquer thinner; it will take the markings off of your components and is not good to breathe.

 

Actually, I circled three joints to reflow, and two leads to clip/trim.

It really isn't so ugly or bad. It's just hard to solder due to the large amount of copper around those areas.

The photography isn't bad, either - better than many other such images that have been posted. It's really hard to take a good image of the reverse side of a board; you really have to use bounce/diffused flash, or better yet take the photo outdoors on an overcast day.

The Three joints you circled have been cleaned of flux after reflowing. We have added another 100uF 35v capacitor in parallel with c4. The problem persists.(Mosfet and c4 get hot).

 

We don't know the manufacturer of toroids because they have been salvaged. However their outer diameter is 1inch, they are black in color and they are attracted by magnet.

 

OK.
Your C4 may be defective; excessive leakage current. This is difficult to test unless you have a variable output DC voltage supply that is capable of putting out at most 35V.

The PWM time from the 555 might be incorrect. However, if your values are close to what's shown, they should be OK.

What is more likely is that the inductors are saturating due to the unknown toroid material and the way they were wound; no tape. If you look at Ronald Dekker's "Flyback Converters for Dummies" page, you will see what happens to current through an inductor when the saturation point is reached; a sharp increase in current.

 

OK.
Your C4 may be defective; excessive leakage current. This is difficult to test unless you have a variable output DC voltage supply that is capable of putting out at most 35V.

The PWM time from the 555 might be incorrect. However, if your values are close to what's shown, they should be OK.

What is more likely is that the inductors are saturating due to the unknown toroid material and the way they were wound; no tape. If you look at Ronald Dekker's "Flyback Converters for Dummies" page, you will see what happens to current through an inductor when the saturation point is reached; a sharp increase in current.

After having checked every thing we reached the conclusin probably our PWM needs adjustment. We, therefore replaced the R2 (68kohm) with a potentiometer (0~100kohm). We tried many settings finally at 30kohm result seems to be satisfactory.
Now the observations are
1. Lively spark while connecting or disconnecting the battery.
2. Humming sound is heard.
3. The parts(Mosfet, L1, L2, D1 and C4) are warm but not hot.
How long should we wait to see the results.

 

The change in PWM duty cycle percent is a good fix. You should replace the pot with a fixed resistor for reliability's sake.

It didn't occur to me before to recommend a power switch, which it needs. The easiest way would be to simply break the circuit at R3.

Don't expect quick results. Depending on the size of the battery, the output of your desulphator and the extent of battery sulphation, It may take a month or more before the battery is restored.

One good way to track the process is to measure and record the specific gravity of each cell during the process. Heavily sulphated cells will have a much lower specific gravity, even down to 1.0 - which is plain water. As the desulphator works on it, you should slowly see the specific gravity increase to around 1.255-1.265 - that is, if you have a trickle charger connected across C4.

Another way to track the progress is the peak voltage output of the desulphator. You can measure the peak voltage across capacitor C6. When the battery is heavily sulphated, the peak voltage output will be rather high. As the plate sulphation is removed, the peak voltage output will decrease, depending on how short/large your wires from the board to the battery terminals are.

Note that during the desulpation process you should subject the battery to a heavy discharge cycle, and then re-charge it - every four to seven days. Otherwise, dendrites of lead may form between the plates creating a short, and make the battery unusable. The heavy discharge cycle will burn up dendrites that are forming.