Alexkor Air Core Coil Radiant Charger

[OldTimer]

Hi Cifta,

Thank you for your questions! :-)

Omission on my part, sorry: the copper wire is indeed enamelled, so that should be alright. It's 0.5mm wire, while swg 26 is 0.508mm, but I guess that should be ok in terms of diameter.

I haven't actually tried to power it with the charging batteries disconnected, so that is a good thought, will try that. And then I'll be able to tell whether or not I have a problem there :-)

I do have a multimeter, although not much else beside that. So volts and amps I can measure, although it is just a regular cheap off the shelf multimeter at that.

Will post back after I try.

Also realized that I planned on putting some arrows and labels in the pictures to identify what is actually going where, would that still be helpful to check/doublecheck? (I usuallly go with check, doublecheck, recheck, and check again before I connect everything, but I'm only human just like everyone else ;-))

Kind regards,
OT

[OldTimer]

Hi Cifta,

Quick doublecheck: how would you propose I wire the circuit up without the charging batteries? Because I'm not sure how to wire the coil up without the charging batteries. Any thoughts maybe on that?

I'd like to avoid a short-circuit because there is a lot I don't understand yet and/or have no experience with :-)

Cheers,
OT

[citfta]

To power it without the charging batteries just disconnect that jumper wire you have going from the collector of the transistor to the diode. The power going to the charging batteries goes through that diode.

I will be working outside most of the day but whenever I come in for a break I will check to see how you are doing.

Good luck,
Carroll


PS: I just looked at your terminal strip again. I gave you bad advice. I didn't see at first your coil connection going to the jumper wire. If you disconnect the jumper wire then the coil is no longer connected to the collector of the transistor. Just remove the wire from the anode end of the diode. That is the end going to the charging batteries. Or in other words just disconnect the wire going to the charging batteries at the diode end. That should keep everything separated so there is no danger of a short circuit.

[OldTimer]

Hi Carroll,


Thank you!

Based on your edit, I kept the wire jumper wire to the charging batteries from the diode disconnected, and connected the rest.
Everything else remained the same, apart from me replacing the transistor with an identical fresh one.

Outcome: the neon bulb did light! (not really brightly, but it did definitely light up) So apparently, the circuit can't be too faulty ;-)
Which leaves me to circle back to my question why the transistor gets so hot and the charging batteries apparently not getting charged. All I can think of is that the current through the transistor is too much, or that maybe my batteries are the wrong type maybe? Not wanting to lead you or anyone else in the wrong direction, just my (albeit a bit uninformed) thoughts at the moment.

Cheers,
Paul

[citfta]

It sounds like the voltage to the base of the transistor might be too high. You can check that with your meter. I will assume for the time being that using the meter may be new to you so please excuse my explanations if you already know how to use the meter.

First set the meter to read DC volts with a range of up to 20 volts. With the circuit powered up put the negative (black) or common lead of your meter to the connection on the strip where the emitter of the transistor is connected. Then put the positive lead (red) on the connection of the strip where the base of the transistor is connected. For proper operation that voltage reading should be somewhere between .5 to .9 volts or so. If it is much above 1 volt then you need to add more resistance between the coil wire and the base of the transistor. That should reduce the current going through the transistor and keep it cool. This simple circuit if working properly should not cause the transistor to heat up. Do this test with the wire connected back up to the charging batteries.

You also need to know that there are a lot of cheap Chinese transistors out there that do not meet the specs they are supposed to and don't work like they should. And this circuit is going to charge the batteries pretty slowly. You probably won't see much change until at least several minutes have passed.

Another thought. I have not taken the time to look at the pages in the PJK book. Do the instructions tell you how many feet of wire your coil should have or how many turns of wire it should have? Your coil looks like it has a pretty small amount of wire for charging batteries. The wire gauge you listed is fine but the length of wire seems pretty short to me.

You are correct this is a good circuit for learning about electronics. I will try to help all I can.

Carroll

PS: I see now on the schematic you posted it says 100 turns of wire. That still seems like a pretty small amount of wire for a battery charging circuit.

[OldTimer]

Thank you for your help, much appreciated! And I don't mind at all if you assume I know nothing, because I can only learn from that, and I indeed don't know much about electronics (yet) :-)

So. Did as you described, and found the following behavior:
Voltage diff between base and emitter starts at around .4 volts, then it has this cycle of jumping up and then stabilizing for a moment, then jumping up further. Have tried that with 500K resistor, 1M resistor, with 1.5M, but all yield the same results. I've stopped each time I reached 1.7 volts across base/emitter, because that's when I can smell the transistor heat.

Will try with higher resistor still, and come back with that. Is the stabilizing/jumping higher behavior something you recognize by any chance?

As for the turns: it stated 2x 100 turns bifilar around a 32mm pvc pipe in the drawings, which is what I did and what you see in the pic (about 10cm/4inches of coil). I have no objections to increasing the wire length/number of turns, I just thought I'd start with as much of an exact copy of the circuit as I can.

Kind regards,
Paul

[citfta]

That jumping up in voltage across the base to emitter junction is pretty strange. Can you try that again with the charging batteries disconnected and watch the neon bulb and see at what point it starts to light up. I have not built that particular circuit but have built many similar and it seems like something is not right.

And after testing with the charging batteries disconnected please try another test with the coil wire going to the cap disconnected. With that wire disconnected you should not see the neon bulb light and the transistor should not turn on at all so it should stay cool. And the voltage should remain very close to zero at the base. It may show a little voltage as leakage current in the transistor but not enough to turn on the transistor.

Carroll

[OldTimer]

Ha, this is getting more and more strange it seems.

Ok. First: with charging batteries disconnected, the neon lights up immediately, and the voltage across base/emitter rises steadily, without jumping up. I cut it off above 1v.

With the coil wire to the cap disconnected the neon does not light, and the voltage across base/emitter is zero. (both with charging batteries connected and without)

And I hear you, because something doesn't feel right about it from the start, but I can't put my finger on it. Just a gut feeling, but still. What it might be, however, is beyond my understanding at this point in time.

Because as far as I can tell from your questions and the results of the tests so far, the transistor is behaving as expected (more or less), and the circuit is also behaving as expected (more or less), apart from the jumping up in voltage and I guess current that doesn't seem to want to be restricted by a resistor. Although the resistors I use are quite a lot bigger in Ohms than the one from the original circuit.

Is it possible that the coil as described is actually too small? I'm thinking that this circuit should provide a frequency-driven charging action, where the frequency is determined by the size of the cap as far as I understand. Is it possible that the coil is off by for example a factor 10, and therefore gets into some kind of 'overdrive' mode where it becomes self-amplifying? Or am I talking rubbish here?

Just realized something else which may or may not be relevant: the resistance of the coil wire with the 100 turns bifilar (so 200 total) was listed as approx 2 Ohms per strand of wire in the original diagram, where mine are 1.5 Ohm. Don't know if that would make a difference that might add up to the jumping behavior?

Cheers,
Paul

[citfta]

OK. Some more testing for you. Try going back to the 2K resistance going to the base but remove the cap. I am getting suspicious there might be a problem with the cap leaking current. I have seen that before. With the cap removed again measure the voltage and watch the neon bulb. Also try this with the charging batteries connected and disconnected. You can also try this with another cap if you have one after you have done these tests.

If you have the patience to continue with this and don't mind waiting for my replies we WILL figure out what is going on. I have worked in electronics for over 50 years so I know we can figure it out. Your clear explanations of what you have found are also a big help. Some people just don't know how to describe what they are seeing, but your explanations are very clear.

Take care,
Carroll

[OldTimer]

Yep, will test, no worries at all there. Will come back with results, although that might be Monday (I work a lot, and this I have to do sorta through/inbetween that ;-)

And I'm glad I'm making sense!

As to me minding waiting for your reactions: Are you kidding me?? This is absolutely brilliant, because I get to follow along with someone who knows a lot more than me, AND I get the hands-on doing it :-)

Still and once again: thank you!

Cheers,
Paul

[OldTimer]

Right. Now I have even less idea what the heck is going on than before. The results are:

Without the cap and without charging batteries:
Neon on straight away, voltage base/emitter 1.3, then settled slowly down to 0.59v, with neon going off at around 1.2v

Without the cap and with charging batteries hooked up: neon does not light up, voltage base/emitter 2.6, then settled slowly to 0.59v

With new cap and without charging batteries: eehhhmmm. I got -1.3 volts, with neon coming on sometimes and sometimes not, and then the voltage across base/emitter flipped from negative to postive and settled on 0.2v... ???

With new cap and with charging batteries: I got 0.2 volts stable, but it seemed like the transistor was heating up, and before I could do anything else, the transistor starting smoking so I yanked the cables loose....

And that's where I'm at now... no clue...

Cheers,
Paul

[citfta]

Hi Paul,

Well now things are looking more like a bad transistor. The voltage at the base should be stable. The fact it keeps dropping usually indicates the transistor is heating up which is what you have also confirmed. I think you have already changed the transistor once. Do you have any more you can try?

With only .2 volts on the base the transistor should not have been heating up. One thing I did notice was that your neon has a resistor in series with it. That will cause the current to be limited going through the neon. That is normally what we would want if we were using the neon as an indicator light. But in this case we want the neon to act as a safety valve to catch the spikes in case the charging batteries were accidentally disconnected. By leaving the resistor connected to the neon you may be getting spikes with enough current to damage the transistor. I would remove the resistor. You might blow the neon if you get some really big spikes but with the way your circuit is working right now I don't think you will get spikes that big.

Also when the neon is lit can you tell if only one of the electrodes inside is lighting the gas around it or are both electrodes lighting up. Only one electrode should be lit. That means the power going to the neon is DC. If both are lit then the power is AC. That may give us another clue as to what is going on.

Do you have any other electronic parts? If you have a 10k pot we could try to modify the circuit a little to see if that gets it working properly. I guess I should have cautioned you before now that a lot of what you find on the internet is not very accurate and sometimes even very misleading. Also the page you got this from claims that you can keep 4 batteries charged up while only using one to do that. I don't know of anyone that has found that to actually be true. Eventually they will all be run down. But this circuit of using pulses to charge with is very good for rejuvenating batteries that are sulphated. And getting it to work can be a great learning experience.

One other thing you can try is to insert some kind of core into the coil. It should be some kind of iron or steel. Some people have used the sleeves that come on concrete anchors and they seem to work pretty well. My coils for charging are made by winding the bifilar wire on to an empty one pound solder spool. This give me an inside core of about 1/2 inch which I fill with a bunch of pieces of electric fence wire cut to the length of the spool and packed inside the center of the spool until they are tight. Having a core and a lot more wire will give you much better charging. This may be another project for you after you get this one working properly.

I also will be rather busy over the weekend but will try to check at least a couple of times to see if you have made any new discoveries.


Take care,
Carroll

[OldTimer]

Hi Carroll,

Been quite hectic, and still are. So I'm going to have to ask you for a bit of patience, as I will not have time to work on this project for several days. But... I will come back after I've had the chance to test what you suggest, so bear with me :-)

Regards,
Paul

PS what I can say, is that I don't just trust stuff I read ;-) I am curious to see how this behaves, as I'm no stranger to research. So I will not believe nor exclude anything upfront, and am trying to see with as neutral a view as I can what this circuit actually does. I must say that I've seen a lot of things in other areas/specializations that theoretically shouldn't be possible, but my take on it is that I then try to find out how the observations I have collected can exist (or if I am wrong). And so far, I've succeeded, at least for my own understanding. (can't put everything in an expensive double-blinded randomized trial, now can we :-) From that perspective, I am curious what your view is on why you'd use coils for charging at all. Why not without a coil? What is it that makes it beneficial to use a coil? Not trying to bundle a load of questions, because for me they are all the same: what is the value of that specific component in a charging circuit.

[citfta]

Hi again Paul,

You are actually asking a pretty good question. Why indeed do we need a coil to charge a battery? We don't actually need a coil to charge the battery. A battery can be charged just fine with a constant voltage charger that is set a couple of volts above the normal charge level of the battery. So the question is why use a circuit with a coil in it to charge a battery?

Most of the circuits that use a coil for battery charging are using the discharge of the coil to charge the battery. When a coil discharges the interesting thing about that is the coil discharge voltage will rise to a level higher than the original voltage used to charge the coil if the discharge current meets a resistance in the discharge circuit high enough to cause the voltage to rise. Why is this helpful?

If a battery has not been used regularly for a while it will become sulphated. That means crystals of sulphation have built up on the plates of the battery. This causes the capacity of the battery to be lower and also raises the internal resistance of the battery. This limits the current the battery is able to deliver and the amount of power we can get back from the battery.

As I just posted above when a coil discharges it has the interesting property of raising the output voltage to a high enough level to overcome the resistance of the discharge circuit. So if we use a coil circuit to charge a battery the effect is we can charge a battery that has become sulphated. The side benefit is that as we do that we remove the sulphation. I have personally seen many many batteries restored to a usable condition by using the Bedini SSG (simple school girl) circuit which uses the discharge of the coil to charge the battery. If the battery is in good condition pulse charging helps to keep the battery in good condition. There are now commercially available battery chargers that use pulse charging because of this benefit.

Sorry for the long-winded post but I hope this has helped a little.

Since I check this forum usually every day I will watch for when you can get back to your project. No hurry on my part. I have been at this for a long time.

Take care,
Carroll

[doGalot]

Hello,

I am a newbie and am very interested in this circuit. The one I am looking to make is the 'more advanced' one without the neon. One question I have is it seems to me (by the diagram) it would be short-circuiting from the last connection leaving the 'batteries being charged' (would be '-') and it is connected to the '+' going to the drive battery. Or am I missing something? sorry, wrong image. One I'm looking at has 2SC3552 and UF5408.

Alexkor advanced.png