Hey Sgt.
Did you ever come up with a final design of a battery desulfator? I am looking to try it out myself. I have a crude version working on a test battery now and it does seem to be have some positive results, however there are some issues with it's functionality.

Reference Thread:
Battery desulphation for lead-acid types - anyone?

Thanks

i

 

Actually, I built a few of 'em. One's been running for a few months over at our Vietnam War museum, reconditioning the battery in the riding mower. They haven't seem to have had problems starting the mower since I did some work on it awhile back, but I really haven't kept track of battery capacity or electrolyte specific gravity since I plugged it in.

I just get side-tracked with too many other projects. Life gets in the way, too.

I've been thinking on and off about doing a PIC-based desulphator, but that hasn't gotten further than the "idle musing" phase.

 

Can desulphators "save" weak car batteries? Maybe give them another month or two of life, or is using it required maintenance from day 1 for longer life?

 

I did an experiment using a store bought desulphator on 2 batteries, a dump/clean/refill on two batteries and an agitate while desulphate on 2 batteries

All 8v deep cycles.

I could find no difference in charge holding or load tests between the desulphator and dump/clean/refill.

The agitate/desulphate took just as long as the desulphate without agitate.

All 6 batteries were D -E -A- D dead when I started.

All 6 batteries were installed into a golfcart when done, and it ran and recharged like norrmal for the next 2 months.

I have since left the place where I did the experiment, so I have no idea if it is still running or not.

So, desulphation does work. It would be logical to use it as part of a maintenance program rather than only to try to revive dead batteries.

 

I'm thinking the odds that a 12V battery with a terminal voltage of 10.5V or less is dead as a result of a shorted cell and probably not worth the effort of saving. On the other hand, I have several batteries that have resting voltages anywhere between 12.3V and 12.7V. The only problem is that they can not provide much on demand current. These are the ones that I would like to redeem from the grave!

@Sgt. So is the circuit in the thread I mentioned your best bet, the one I might give a try? Were there any modifications on it? Would you be so king as to post it here?

i

 

Do you have Cadsoft Eagle installed?

I did several revisions using Eagle version 4.16r2. You can view both the .sch and .brd files with that Eagle version or later.

 

Do you have Cadsoft Eagle installed?

I did several revisions using Eagle version 4.16r2. You can view both the .sch and .brd files with that Eagle version or later.

Yes I have v5.xx installed.

i

 

OK.

I hand-wound the inductors using some toroids I picked up at a local surplus store called "Skycraft Parts & Surplus". I try to not visit there very often anymore, as I'm running out of space in the garage. The place is a techo-dweeb's dream. Anyway, the toroids I picked up were 1/2" in diameter (T-50), have an Al value of ~2550; 9 turns of AWG20 gave me ~203uH, 10 turns 257uH, and 20 turns 1020uH. In another later version I used some toroids that were more cylindrical, but they were quite noisy due to the wires vibrating.
You might use these Amidon toroids:
FT-50B-77
FT-37-75
FT-37-J
FT-50-J
Amidons' website is here:
https://www.amidoncorp.com/categories/7
The material type is the suffix on the above part numbers. The center number is the approximate outer diameter in inches.

Or, just buy some pre-made inductors.
The 1mH (1000uH) inductor can have a fairly low average current rating; it's the 220uH inductor that needs a fairly high current rating, and high resistance to saturation. I used several 100uF caps to lower their effective ESR instead of going to low-ESR caps; as I had a bunch of 100uF caps around here.

I remember that I had to make some changes in the resistors after building the board; as the 220uH inductor was saturating, causing the MOSFET to heat. Reducing the ON-time took care of the heating problem, but this is one of those things you'll have to fiddle with once you connect it to a battery and test it.

You don't want to test it without it being hooked up to a battery, or you'll get REALLY high voltage spikes out of it, which will likely zortch things.

Connect it to a battery, and THEN turn on a trickle charger (if desired).

Turn OFF the trickle charger before disconnecting it from the battery.

I used the IRF730 MOSFET because I had a bunch of 'em. As long as the MOSFET's Vdss is >=60, it should be OK.

The 1N5401 diode was a poor choice, but it was what I had on hand. It should be a fast-recovery type diode.

The 1N645-1 diode can be replaced with a 1N914/1N4148. I just had some of them kicking around left over from a project I worked on some 35 years ago; I figured it was about time I did something with them.

I didn't bother to check the dimensions of project boxes I had on hand before I laid out the board; as a result they are still not enclosed.

Pay no attention to the WX2NJ prefix; this schematic/board bears little resemblance to what they posted on the 'net a year or so ago.

I'm not going to do anything more with the 555 design. Next stop: a PIC version.

 

I'm not going to do anything more with the 555 design. Next stop: a PIC version.

What advantages do you forsee in the PIC design over the 555 designs?


i

 

Fewer parts, more functionality, the ability to sweep over a range of PRT's, readout of info (like voltages, etc.) using a blinking LED, possible high-current discharge mode to burn off dendrites, stuff like that.

Don't expect anything soon.

 

Let me know what you want a PIC to be doing and I'll post it here, it'll be a neat software only project for me to help my kid with.

 

Fewer parts, more functionality, the ability to sweep over a range of PRT's, readout of info (like voltages, etc.) using a blinking LED, possible high-current discharge mode to burn off dendrites, stuff like that.

Don't expect anything soon.

Sounds great, bells and whistles are much needed on this type of circuit.

 

Well, not a lot of bells & whistles.

Knowing the peak voltage will help. If there is a really good broadband connection between the desulpator board and the battery, the peak voltage will decrease as the sulphation is removed from the plates due to the lowering of the batterys' internal resistance - so that provides an indicator of the battery condition without having to perform a load test.

I'd originally planned on using fairly wide copper strips to connect the boards to the battery terminals. As you're probably aware, a wire of a given diameter and length has inductance, but an infinite plane has virtually no measurable inductance. Wide strips would not be the same as an infinite plane, but would have far lower inductance than the AWG-14 stranded wire I used to hook mine up.

 

Thanks Sgt. for the circuits. I will look over the circuit and see if I can find all the parts and place an order.

i

 

Sgt., I'm could use a bit of help finding a suitable replacement part for the 1N645-1.
Also: If it not be too much trouble, the 220uH and 1mH inductors are a problem for me as well. I generally use mouser, but can order from digikey as well. Your help would be most appreciated.

iOnic

 

Just use a suitable fast-recovery diode. I used the 1N645-1's because I had them on hand - they were made in the 60s and have been obsolete for quite some time.

I wound the toroidal inductors myself, like I said in this post:
http://forum.allaboutcircuits.com/showpost.php?p=299679&postcount=8

Get a couple of the Amidon FT-50B-77 toroids on this page:
https://www.amidoncorp.com/items/18
Order some extra to have on hand. If you break one or two, you'll be glad you ordered spares.

10 turns ~=240uH; close enough. About 1'3" of wire.
21 turns ~=1mH, about 2'3" of wire.
Use AWG-20

 

Thanks for the tips.

i

 

My mouser parts should be here today, thus I can begin building the Desulfator.
The Inductor parts should be soon to follow.


Q: Would the following circuit be a decent indication of the desulfators effectiveness?

Describes as:
this part of the circuit detects the Pulsed Current Spike, "Directly At the Battery Terminals". This signal is than amplified and gives a reading on a 200 uA Meter. And this reading will reflect on the Internal Resistance of the battery, So the "LOWER the meter reading the better".

Desulfator Test Indicator

 

The best test a (SGM)specific gravity meter/tester and a multi-meter.

The SGM will show the electrolyte change as the desulphator changes the crystals back into sulfuric acid.

The multi-meter to show the full charge voltage and to plot the charge over time.

http://www.toolfetch.com/Category/Automotive/Battery_Tools/Battery_Testers/THE117.htm

or if you want to throw some money at it:
http://www.americantechsupply.com/sbs2002.htm

 

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