Maybe.

I would ask around before spending.

These bulb SGMs' were commonplace in many peoples toolbox.

My father and grandfather had one or two in their toolboxes.

Desulphating does work. The thing is does YOURs work. That is the tricky part.

You can follow the gravity plot day to day (week to week) to let yourself know that you are not wasting your time/energy.

 

Even the SGM might be throwing money at it at present.

My Auto Repair shop did a "battery Test" while changing my tires. The results were:

Rated CCA = 500A

Voltage - 14.4V
CCA - 224A

I've been "Desulfating" for a couple of weeks now and plan to have my car inspected in a few weeks. I will have them test the battery again. I'd think that this will give we a great indication of effectiveness after a months worth of "Desulfating."


i

Wal-Mart will do free battery tests that give you a nice printout of open terminal voltage, load voltage, CCA, alternator ripple, and all that. Advance Auto Parts does this as well with no charge.

I think if you explain it to one of the counter guys, they'd have no problem printing one a week for you. At least where I am they are very helpful on things like that. I guess larger cities aren't as "friendly".

 

Wal-Mart will do free battery tests that give you a nice printout of open terminal voltage, load voltage, CCA, alternator ripple, and all that. Advance Auto Parts does this as well with no charge.

I think if you explain it to one of the counter guys, they'd have no problem printing one a week for you. At least where I am they are very helpful on things like that. I guess larger cities aren't as "friendly".

The auto shop will do this for free, perhaps not every week, but after a month+ of desulfating I ought to know if my circuit is working...

i

 

Large batteries can take a lot longer to desulfate.

I did a small 12v Yuasa motorcycle-type battery; ran the desulfator circuit on it for a month or so. Three of the cells were completely dead when I started; specific gravity was that of water. After a couple of weeks, the specific gravity started to come up near normal, and after about 5-6 weeks, they all read normal.

 

Large batteries can take a lot longer to desulfate.

I did a small 12v Yuasa motorcycle-type battery; ran the desulfator circuit on it for a month or so. Three of the cells were completely dead when I started; specific gravity was that of water. After a couple of weeks, the specific gravity started to come up near normal, and after about 5-6 weeks, they all read normal.

Understood.
The battery in question, as mentioned previously, was resting about 12.5V prior to starting the desulfating process. Either some cells were overcharged to compensate but I doubt it. It's CCA was 224 and is rated for 500. The car started so I am guessing that it's a case of well distributed sulfation. This is just the sixth sense working here so I'm not putting ALL my eggs in one basket...i'll keep one for to "save face" if I am wrong.

I'll take a look at the SGM meeter if the local Hardware store has one. I think I could be using one more frequently than I may think. I have several candidates for desulfation and It would be handy.

Q: Has anyone ever removed the liquid contents from one cell of one battery and placed it in a bad cell of another, making a better battery???

 

The battery in question, as mentioned previously, was resting about 12.5V prior to starting the desulfating process.

You did not mention the battery internal temperature, which is very important. The preferred method is to take the internal temperature reading from the positive terminal, as it's physically connected to the plates. The negative terminal could also be used, but the positive terminal is preferred.

Specific gravity readings are also very important. Of course, if the battery is sealed or an AGM type, you'll have a hard time getting SG readings.

I have a precision optical specific gravity instrument that's temperature compensated and very accurate. It wasn't cheap, either.

Either some cells were overcharged to compensate but I doubt it.

Here's where specific gravity readings for each cell would come in handy. Use distilled water to raise the electrolyte level to just above the plates, but well below the opening. You don't want it filled to the top.

The electrolyte specific gravity was correct when the battery was made. During times of overcharge, the water in the electrolyte can emit hydrogen and oxygen gases, thus making the solution more acidic and lowering the level of electrolyte in the cell(s). Having the plates exposed to atmosphere will shorten the battery life considerably, as well as damage/corrode the plates very rapidly.

It's CCA was 224 and is rated for 500. The car started so I am guessing that it's a case of well distributed sulfation. This is just the sixth sense working here so I'm not putting ALL my eggs in one basket...i'll keep one for to "save face" if I am wrong.

It seems as cells in the middle of a multi-cell battery tend to sulphate more quickly than the ends, and the cells near the ends tend to short more often than the middle. The center of the battery will have the greatest temperature stability.

I'll take a look at the SGM meter if the local Hardware store has one. I think I could be using one more frequently than I may think. I have several candidates for desulfation and It would be handy.

It will help a great deal if you also have an accurate way to determine the battery temperature.

Q: Has anyone ever removed the liquid contents from one cell of one battery and placed it in a bad cell of another, making a better battery???

You'll upset the SG of the cell.

When the plates become sulfated, the SG of the cell decreases because of the sulfur deposits on the plates. When you desulfate the cells, the sulfur recombines with the electrolyte, and the SG increases until all of the sulfation is removed.

Periodically overcharging (known as "equalization charge") the battery helps to stir the electrolyte; otherwise it can vary in SG at different places within the cell. If it's an AGM type, it won't need the equalization.

Never overcharge gel cells; as bubbles will form in the gel, and its' capacity will be permanently reduced.

 

Q: Has anyone ever removed the liquid contents from one cell of one battery and placed it in a bad cell of another, making a better battery???

I dumped the electrolyte out of a motorcycle battery and replaced it with new electrolyte (some batteries are sold "dry", you can buy the electrolyte).

It made the battery a bit worse overall.

 

All good information guys...thanks!

 

Back around 1970 when I was a teen, I had a bad battery in a car I'd bought for next to nothing. I drained out the existing electrolyte, and used baking soda and water to neutralize any remaining acid inside the battery; filling/agitating/dumping the battery several times. The first time I did this to my surprise, the baking soda/water solution turned a frothy pink; I have no idea why this happened.

After I flushed all the cells out repeatedly with fresh water and allowed it to drain, I refilled it with fresh electrolyte, and was able to use the battery for at least several more months. I no longer even remember what time of year I did this, but it was sometime between spring and fall.

This might work with deep-cycle batteries, but I doubt much good would happen with more modern automotive batteries. The plates were thicker and fewer in the automotive batteries of the 70's, and the chemistry was a bit different.

 

Sgt.
Since I am going right to a proto board and not a bread board, does this mean that I should be using a 91K resistor as apposed to a 1.2M?

Made the Inductors and am ready to get a general parts layout.

 

Let's see - from memory, R1 controls the ON time, R2 controls the OFF time.

Do you have an O-scope? That'll help a lot.

Basically, you want to turn the MOSFET on long enough so that the 220uH inductor saturates, but not so long that Vgs rises appreciably, or the power dissipation in the MOSFET will become excessive and it'll get hot.

Then you want the MOSFET off long enough so that the caps recharge pretty close to full. If the MOSFET starts getting warm, you can increase the OFF time by increasing R2.

On my proto, I used 3.3k for R1 and 91k for R2. Your mileage may vary, due to the use of somewhat similar, yet different toroids. The AL value is close, but how quickly they will saturate depends somewhat on how you've wound them (evenly-spaced windings and taped toroids helps to delay the onset of saturation). If you scrunch the windings together on one side, you kill the Q of the inductor, and it will saturate much more quickly than if the windings are spread evenly.

Note that the point at which the inductor will saturate changes depending on the battery load, which I found interesting. Some folks use an NTC thermistor thermally connected to the MOSFET tab to control the on-time, which does work; as the resistor heats up, the ON-time decreases, which lets the MOSFET cool down.

[eta]
The original author used a physically larger inductor that was more resistant to saturation. These smaller inductors saturate more quickly, but the PRT (pulse repetition time) is much less. You'll still get a big voltage spike out of it.

 

I used the Amidon FT-50B-77 toroids
10 turns ~=240uH; About 1'3" of wire.
21 turns ~=1mH, about 2'3" of wire.
Use AWG-20

I have a friend who has an O-Scope and can experiment at his house.

Basically, you want to turn the MOSFET on long enough so that the 220uH inductor saturates, but not so long that Vgs rises appreciably, or the power dissipation in the MOSFET will become excessive and it'll get hot.

Then you want the MOSFET off long enough so that the caps recharge pretty close to full. If the MOSFET starts getting warm, you can increase the OFF time by increasing R2.

Thus the MOSFET saturation is controlled, to some extent by the R2. Would a 50K in series with a 100K Trim make a good method for tweaking? I take it we do want saturation, but not continous, yes?

Some folks use an NTC thermistor thermally connected to the MOSFET tab to control the on-time, which does work; as the resistor heats up, the ON-time decreases, which lets the MOSFET cool down.

Any special tape to use on the inductors...??

The termister is taking the place of R1 or R2, or neither. Or perhaps as input to a comparator to turn off the 555?

The circuit I am using now does saturate enough for the MOSFET to get hot. I have a 555 PWM fan turning on and off to keep the temp stable. Sounds like this is not the best way to solve the problem.


i

 

I used the Amidon FT-50B-77 toroids
10 turns ~=240uH; About 1'3" of wire.
21 turns ~=1mH, about 2'3" of wire.
Use AWG-20

I have a friend who has an O-Scope and can experiment at his house.i

OK, good.

Thus the MOSFET saturation is controlled, to some extent by the R2.

No. R1 controls the ON time, R2 controls the OFF time.

Would a 50K in series with a 100K Trim make a good method for tweaking?

Well, that would work. Trimpots have a limited life as far as the number of adjustments go. Better to tweak R1 so that the inductor just starts going into saturation (Vds starts increasing sharply just before the MOSFET turns off, indicating saturation.)

I take it we do want saturation, but not continous, yes?

You want to get the inductor saturated. You'll see the "knee" when it hits saturation; Vds will rise sharply. A layer of tape will help to soften the knee.

Any special tape to use on the inductors...??

You could use masking tape. I use a yellow polyester tape that's very thin, tough, high insulation value, and resistant to heat. 3M56 and Permacel P256 are examples.

The NTC thermistor is taking the place of R1 or R2, or neither. Or perhaps as input to a comparator to turn off the 555?

R1, as R1 controls the MOSFET ON-time. You'd need to start off with a thermistor that measures a bit too high at room temp; as the MOSFET heats up, the resistance decreases, so the ON-time decreases.

The circuit I am using now does saturate enough for the MOSFET to get hot. I have a 555 PWM fan turning on and off to keep the temp stable. Sounds like this is not the best way to solve the problem.

It's one way, but arguably not the best way.

When the inductor saturates, it does no good to keep running yet more current through it; it's not going to result in a higher-intensity pulse once the MOSFET cuts off.

 

Would a 50K in series with a 100K Trim make a good method for tweaking?

Well, that would work. Trimpots have a limited life as far as the number of adjustments go. Better to tweak R1 so that the inductor just starts going into saturation (Vds starts increasing sharply just before the MOSFET turns off, indicating saturation.)
You want to get the inductor saturated. You'll see the "knee" when it hits saturation; Vds will rise sharply. A layer of tape will help to soften the knee.

I was thinking that it would allow for exact tweaking prior to fixed values.

R1, as R1 controls the MOSFET ON-time. You'd need to start off with a thermistor that measures a bit too high at room temp; as the MOSFET heats up, the resistance decreases, so the ON-time decreases.

I think I'll forgo this option, It would delay he completion. It's a bit sketchy for me to figure out.

Is the current spike in any way proportional to the batter size?

The circuit I am using now does saturate enough for the MOSFET to get hot. I have a 555 PWM fan turning on and off to keep the temp stable. Sounds like this is not the best way to solve the problem.

It's one way, but arguably not the best way.

When the inductor saturates, it does no good to keep running yet more current through it; it's not going to result in a higher-intensity pulse once the MOSFET cuts off.[/QUOTE]

I am assuming it does still get a high-intensity pulse as the inductor is still singing...

 

@ Sgt. Wookie
In the schematic you posted on page 1 (Link below) you used 3 100uF, 50V caps in parallel. If I am correct this adds up to 300uF, but also reduces the ESR of the combined capacitors (a positive feature in this circuit. I have seen many posts/internet links, including the original designers circuit, that use different values.
I've seen 100uF 16V, 25V, 50V and even 100V. Additionally I have seen recommended cap values from 200uF - 10,000uF. Am I to assume that the capacitance is not so critical? What about the voltage, is this also not so critical?

Eagle CAD Schematic

 

The voltage isn't critical; 16v or more is fine. However, the higher the cap voltage rating, the greater the leakage current - something worth considering if you are not planning on using a trickle charger/battery maintainer while desulfating a battery or batteries.

And yes, I used multiple caps to lower the ESR, and also to spread any heat around.

The capacitance value isn't particularly critical either - as long as there is enough of a charge stored to get the inductor saturated, it's OK. The voltage across the caps should remain pretty constant; if it is not, then you don't have enough capacitance.
With the small inductors I was using, I didn't need a lot of capacitance to get them saturated. If one wanted to use really large inductors that can store a lot more energy, they would need much more capacitance - and a better MOSFET.

 

The voltage isn't critical; 16v or more is fine. However, the higher the cap voltage rating, the greater the leakage current - something worth considering if you are not planning on using a trickle charger/battery maintainer while desulfating a battery or batteries.

In other words, the "leakage current is draining the battery", especially without the trickle charger. The circuit I have been using seems to be a power hog as a trickle charger is an absolute, the reason being, I think, is that the MOSFET is pretty hot.

And yes, I used multiple caps to lower the ESR, and also to spread any heat around.

The capacitance value isn't particularly critical either - as long as there is enough of a charge stored to get the inductor saturated, it's OK. The voltage across the caps should remain pretty constant; if it is not, then you don't have enough capacitance.
With the small inductors I was using, I didn't need a lot of capacitance to get them saturated. If one wanted to use really large inductors that can store a lot more energy, they would need much more capacitance - and a better MOSFET.

Thanks for the additional insight.

 

I tuned the MOSFET on-time so that the inductor was just beginning to saturate, and my average current draw was approximately 7mA to 8mA. Even a draw that small will eventually drain a battery. However, if you are letting the inductor saturate, the energy not stored in the inductor; it is dissipated as heat in the MOSFET and associated circuitry - your current usage increases drastically. Don't remember offhand how I measured the average current.

 

Do you think the circuit characteristics will change dramatically from Breadbord/Protoboard to Perfboard to printed circuit board? I assume they might be between Breadboard/Protoboard to Perfboard, but suspect not so much with the transition from Perfboard to Printed circuit board.... with respect to this particular circuit.

 

I didn't breadboard it; I designed the PCB, and did the fine tuning on the PCB itself.

I didn't want to have to re-compensate for the lack of the breadboard parasitics.

Your mileage is going to be different anyway, as with the FT-50B-77 toroid with 10 turns on it, you'll have around 250uH instead of ~220uH, and the 1,000uH will actually be around 1,100uH. If you reduce each by 1 turn, you'll get a good bit closer to what they are supposed to be; right around 210uH and 1,000uH.