Running the new Circuit Bitoroidal Transformer

Greetings. Welcome to another video. I ended the last video saying, I think I'll get the next video out to you real soon. And so famous last words. So I'll just talk at you for a bit, kind of summarizing what was going on. I did do a lot of work. I actually reported some stuff, but I don't, you know, there's not really, I don't need to show every negative study. So to make a long story short, everything that I did made things worse, which kind of surprised me. So let's just review again, 30,000 foot view, and then we'll try and go down quickly. And then I'm going to gather some data and try to put together something useful before doing too much more basic research. So to summarize, I started off with this core here. I had some more wines on the, on the primary, but this was it. And it's just an ungapped, two ungapped E-cores. And just on a lark, I was like, okay, well, let's look at some of this bitoroidal stuff where you have two different flux paths, where the secondary can flow on a different flux path. And Thane Hines has done a lot of, out of Canada, has done a lot of work with this. And he generally uses two different materials from what I understand, so that you have a higher permeability, magnetic permeability flux path on the outside than you have for the primary. And so I looked at this, and I did notice when you load one of the secondaries, you get more out of the other secondary, indicating flux is flowing around the outside. And also, when you have the Arduino, you can get fine control of the timing, which has proven to be important. And what was the breakthrough on this one, was when I put a little piece of masking tape between here and there to provide a little bit of a gap. And then I started to get some interesting results with that. So from there, I graduated to something more like this, which is a little bit bigger Ecore. And if you can see there, there's about a one millimeter air gap, somewhere around there, that is set up for you. And so you don't bother with the masking tape, which also, you know, if you put too much masking tape, you got to worry about is, are these snug between here and there. And it worked about the same as that, with the masking tape. Maybe, maybe a tiny bit better, but not a big improvement. So then I went to, and this is, you know, things I've been working on here, is winding some of these. Here's a larger one. I just have an assortment of these, and one of the goals today is going to be to pick one. So here's one, yeah, about the same air gap. However, if we go with this secondary, and you'll notice this is 26 gauge instead of 22 gauge, so more on the, that's another variable there, is the turns ratio. Now we're back to no air gap, or if you go, let's say this one, which many fewer winds there, and we have now a pretty large air gap. And, you know, at this point it started to be a little bit worse, but I wasn't sure is it, you know, what the difference with the windings, or the difference with the air gap. So one of the things, let me put this, well actually let me use the other one. So one of the things I was looking at that, I'm not gonna do the, I won't show you all the video with it. What if you had a setup like this? Now, just pretend there's no wire on there. Pretend there's really, yeah, just pretend there's no wire on there. So now you have a primary that's magnetically coupled, but, you know, there's, now there's like no closed magnetic flux path here. There's like an infinite air gap. I thought that would work well. I thought you're not going to get much deformation here, because it'll get into here, and then this is a cul-de-sac. The problem that I ran into, and it was, it was so bad, I've done it twice, I might even do it a third time, just to make sure I didn't have a wire wrong somewhere. It's true that when you have this setup and you load these, the deformation is almost unnoticeable on the primary, the deformation of the oscillation of the LC tank circuit. But if you're using this as the primary, I'm just making these up, but it's the ballpark idea, then let's say you're getting a volt out of each of your secondaries. With this, you're getting like a tenth or two-tenths of a volt out. So I don't, I really don't understand that. I don't know why that is, because, I mean, if you, so another way, like another, another way to visualize it is, you know, consider this as the primary. So you could have it like that, or you could, if you have it like that, and then you, you like straighten this thing out, I mean, you would, you would just have a ferrite rod coming into here, and then your secondaries. And, I mean, my understanding is you can take a ferrite rod and say, you know, here's your primary, here's your secondary, and it works okay. I mean, I don't, I don't see that all that often, but I don't know why you wouldn't get a good transfer of magnetic flux. And I'm just not seeing it, so I don't know what's going on with that. But it's like, okay, back up, forget about it. So I had beginner's luck, I think is, is the answer there. And, you know, it's fun to think about, well, I'm going to go, oh, I got, where is it? I've got, I'm working on this. This is a nanocrystalline core. And so, you know, it'll be fun to, to go to different materials and whatnot, but it's like, no, let's try and go along with the thing that I had some beginner's luck with, and see if I can get something at least modestly useful. So the goal is just sort of to pick an arrangement of what I'm going to use for the primary and the secondary in here, or it wouldn't surprise me if just this one that I've had for a while turns out to work as well or better than any of the others, but I'll get started on that. And I will mention there were some like simple things related to timing, and not even like making the capacitor smaller or something like that, but just some like simple sort of obvious things that I had overlooked before that I was seeing do make a difference. So we'll discuss those as we move along here, but let's start with this one. And if this one works well, just run with it. We'll start off where the efficiency probably won't be that great initially, and then we'll make changes that will gradually improve it. So this is the circuit that I built for you in the last video, and the only difference from the last video is I've also added a switch there, and a switch there. And their purpose is you're going to be gathering stuff off of the secondaries, and you might want to get rid of that after a set amount of time so that, you know, it doesn't go too high in voltage, and just so that you can make standardized comparisons. And not to rehash all of this, but what you're doing is you're taking the cap and you're charging it from the bus lines, and then you have your primary goes to one leg of the cap. The secondary goes to one leg of this switch. When that switch turns on, the connection is made to the other leg of the cap. The cap discharges through the inductor. So you have an LC tank circuit, and because you have an LC tank circuit, this cap will discharge, and then it will go negative, zero, positive, and it will ring back and forth. And depending on how good the Q factor is will determine how long the cap rings. So the nice thing about this setup is for the cost of a couple extra switches, what you do is instead of having the cap go negative, and then as far back positive as it will go, you let the cap go negative. And then when it's negative is when you shut everything off. When it goes as negative as it's going to get, and then you charge the cap negative, and you discharge it the opposite way, and it'll go positive. And because you don't have a perfect Q factor, then it's going to go further negative than it will go as far back positive. So in other words, you know, maybe this will go from 10 to negative 8 to positive 7 to negative 6, and so now you can catch it at the negative 8 and just put in 2 volts instead of catching it at the positive 7 and having to put in 3 volts. And you know, of course, when you're running it, this will be here, and your secondaries, the secondary lines will go to this bridge rectifier and the other full wave bridge rectifier, and so that's where you'll gather your output out and see what you get. So let's just go step by step, make sure everything's working. Slow and steady wins the race. So we're at 2 volts, and we're just discharging the cap through here and then charging it again. And what do we see it doing? Goes up to 2 volts, discharges, and then charges the opposite way and discharges. So that's what we want. So the next thing we want to do is to know, figure out the inductance of this coil, which if it's just sitting there like that, we have a reading of 1.91, but if we put this on here, then you now have a complete magnetic flux path through that tiny air gap, and let's make sure those are where we want them, and we're seeing 6.23, or after I line it up a little bit better, 6.34 millihenries. So let's jot that down in our spreadsheet, 6.34. So when we put that in, we see that after that many milliseconds, or 791 microseconds, is the place where this cap should be most negative. So nothing is loaded yet, but this is, we've gone up to 10 volts here, and we discharge it for the time where it goes most negative. So here we're charging it up to negative 10, and then it swings back just about to 9. Charge it up to 10, swings back just about to negative 9. Charge it up to negative 10. So it's, we're putting in about 1 volt in order to get back to the initial condition. Now when we start to load it, there'll be some deformation of that. It's, the Q factor will be worsened, if worsened is even a word. But that's the game here, is that you want to preserve as much of that swing of the cap as you can, while getting as much off the secondaries as you can. And there's just a lot of variables going on, but as I said, through dumb luck, I had something that's not too shabby. So now let's hook up the secondaries. Okay, so now let's just hook up one of the secondaries. The workspace is starting to get a little busier, so I'm taping things down. But we have the first secondary hooked up, and what we see, it's a zero. Now this charges up, 0.9, and then swings back the other way, 1.3, and then we discharge it. And you'll notice that it's not making it all the way back to 9, but it's still, you know, it's doing pretty well. Now what we're going to do is hook up the other secondary. I know I've showed this plenty of times before, but I'll show it again. And this is how you know that you have something interesting going on. So this is with only one secondary, and we see, you know, it gets to there with the two things. Now I'm going to put in the other wire, and okay, and now we're getting 2.89 out of there. And so we're getting, you know, more than twice as much out. And we're also, I haven't hooked the meter up, but we're also getting the same out of here, so about four times as much. And that's how you know that some of the magnetic flux is recycling in a beneficial way, instead of all of the Lenz law interaction going here. But because this is just, you know, same material cores, we also have a much greater dampening, you know, it's not getting back towards, it's not getting back towards 10 anymore, or nine was where it was at. But it's still doing okay, and we got four times out more there. So what we're going to do now is we're going to load it, and we're going to figure out efficiency. So what I've done here now is put a 100 ohm resistor across each of the meters, so there's a 100 ohm load across each of the gap outputs there. Now I've seen that this thing in the past generally worked a good bit better with a higher resistance load. So as I said, we'll start off with what is not the best setup, and just see how much we can improve things in different ways. Now that we have this, we can also, we can calculate milliwatts out and milliwatts in, and I went over this in some considerable detail in the video some time back, running the bitoroidal transformer the first time I did this. But just to try to recap that briefly, you know the resistance here, we'll know what voltage shows up. Ohm's law, V equals IR, I equals V divided by R. We know the resistance, we know the voltage, we can calculate the amps. So now you have the volts and the amps, and volts times amps is watts. So that's your watts out, one from one side, one from the other side, and add them both together, you get your watts out. The watts in is a little bit more complicated, but not difficult to understand. We know that we're going to have this thing going from say 10, let's say to 5, and then you're going and putting 5 volts back in. So you're putting 5 volts into a cap of known size, say these are 10 microfarads. So you know the capacitance, you know the voltage. So charge in a capacitor is defined as capacitance times voltage, so you know 5 times the 10 microfarads. Amps are of course also defined as flow of charge per second, so flow of coulombs of charge per second. And you know from here how many times in a second, once you add up all the timing, you know how many times in a second you threw the voltage into the cap and it discharged. So you know how many coulombs of charge are flowing per second. So you have amps, and you also know if you had to put 5 volts in, you have your voltage. So then you have your milliwatts in. So again, not difficult to understand. If you know the voltage change that you made in a cap of a set size, you know how many times you discharged the cap per second. You know how many, you know the amps that flowed and you know the milliwatts that flowed. So let's just run this thing first and see if we're in the ballpark. That's what I'm looking for here, just in the ballpark. And then there are some changes that I'm going to try. Okay, so just as the very first thing, you know we're just running it here and it flips back to 6.8 something, 8.6. And we're getting that voltage out from one, that voltage out from the other. So when we put that in the spreadsheet, it's just the first one here. There's your total milliwatts out, and that's your total milliwatts in. And so that's your efficiency. So it kind of stinks. Is for whatever reason, and I don't know how it happens, but the loaded and unloaded resonant frequencies are different. So in other words, we should be able to find a spot where this swings back a little bit better and we don't lose much off of the secondaries. And so we kind of have to poke around for it. So what we do, what we do, we go down to the bottom here. We run it once and then we take the on time and decrement it by 10. And then run it again. And we're looking for what's the best spot with this load for this. Okay, forgive me. I had forgotten to update the timing for calculating the watts and reverse time. So I've just, I got rid of everything else. And you know, it charges for 400 microseconds and 200 off. And then the on time, which was 820 and then 200 off. And then it does it a second time there. So if we charge that, if we, I'm sorry, sum that, you have that many microseconds. So every 1620 microseconds, you charge the cap. And so we put that in there and there's your frequency. And we calculate watts in again. And so that is actually watts out, watts in efficiency. So better. And now we have an easy way to keep track of this. But let me go back to fishing around for the best frequency, the best on time. So after fishing around, I found the best timing is in the ballpark of 700 microseconds, as opposed to the calculated 791 for this particular load. And when we run it here, let me recompile. Now when we run it, we see this flipping instead of 6.8, whatever it was back to about 7.31. And this is what we're getting out from our secondaries. So let us now put that in the spreadsheet. So we're actually getting a teeny tiny bit less out from the secondaries and we're going a little bit faster. So that's all bad. But the big thing is instead of going from 10 to 6.86, we're going from 10 to 7.31. And when we put that in, now that's our milliwatts in and that's our milliwatts out. That's the efficiency. So the easiest thing to do is to go up on voltage. But before we do that, there's another thing that I want to check out, which I think will work. So one thing that occurred to me is we're charging the capacitor for 400 microseconds. Now let's just pretend we charge it for like four seconds. The whole time that's charging, you have a 100 ohm load across the secondary and the secondaries are just discharging. So in other words, you want to discharge this thing as soon as it's charged up because otherwise you're just wasting voltage across your resistor. So is that too long? And I think we can get away with going a little faster. If we go over to, I love Digi-Key. Here's their calculator for this. If you say you have 10 UF and the resistance is going to be one ohm or less across your solid state relay. So the time constant, you want five time constants for full charge. The time constant is that. So we want five of those. This, we got to times by a million, 0.1231 times 1, 2, 3, 1, 2, 3 equals 10 microseconds. Now we got to times that by five for full charge. So you'd need about 50 microseconds. Now you also have to factor in how long it's going to take the switch to turn on, which supposedly is like 200. But if you leave the resistors out, then you're hitting it with a lot more current. And so it's probably like 50 or a hundred microseconds to turn on. So you could probably get away with a hundred to 150 microseconds of charge time for it and still get a full charge. We're at 400. So let's just first drop it to 300 and then see if we can get away with 200, but see if that makes things better or worse. So we can see the thing's still swinging back to 7.32, 7.34. Now we're getting 1.78 out from one at 1.44 as compared to 7.1 and 3.9, but it's going a little faster. So we have to change that 400 to a 300, put in the new numbers and see what it, what it all calculates out to. Okay. So it did make better. That's now millawatts in millawatts out in the ratio. Let's see if we can get away with 200. I don't think we can go faster than that, but we might be able to get away with 200 and still have the cap charging fully. So now I've gone to 200. We're getting still more out from each of the secondaries and you know, I'm realizing that this cap is draining out as well when nothing is going on. So we're seeing this looks a little bit better too. So let's put those numbers in. Okay. So when we put those numbers in now, now that's a ratio. Um, the next efficiency improvement here at 10 volts will be what about the off time for the switches? So let me shut that down. If we go back here now we could, you know, if you put it on the oscilloscope, you could see if it's still plateauing, you might be able to go even a little bit faster, but that that's enough now. I mean, that was a, that was a fairly significant improvement. Now you have the off time where the switches turn off and I put each of those at 200 microseconds. Now, if we go, this is for the Commiss AC37. The typical off time is 50 microseconds, but it might be as long as 500. So I had it at 200, but it sure would be nice to take it down to 150. I'm not going to go all the way down to that because you don't want, you don't want the thing stepping on its toes. Now for the other, the other ones, which there are actually more of the typical turnoff time is 140 microseconds. So there's a lot of those off times. So I'm going to see if we can get away with going from 200 down to 150 without anything stepping on its toes. Okay. When I did that, we again got a nice little bump in our ratio there. So 1.7 to, or whatever it was to this now. So now the thing to do is to go up to 20 volts. So I know it reads 20.3 there, but if we slow everything down, 20.01. Sue me, I'm off by one two thousandth. That's good enough for me. Actually don't sue me. I don't, I don't want the headache. Please don't sue me. So now we'll get rid of the delay, run it and see what, see what happens here now. Okay. That's, that's what it's springing back to. And you can see that we're getting more than a doubling out of the secondaries, despite the input voltage doubling. And that will be reflected when we put our numbers in. So I've left the timing all the same. And I am, I'm pretty sure that it, it's independent of voltage, but I'm going to, I'm going to diddle around and make sure that, but that's now the ratio. There's in and there's milliwatts out. So let's take it up to 30 and then I'll diddle around with the timing, make sure the timing's still optimal. And then we're going to change the load. So here we are at 30 volts and it's swinging back to 21.1 something, which actually percentage-wise isn't as good as at the lower voltage. It's, it's more damped, but this is moving up faster than it should, the voltage there. So I'm going to put, I've left all the same. I'll put this in for 30 volts. So that's the ratio that we have at, at 30 volts. And so we have this trend going on from 10 to 20 to 30. I think the trend continues. Now, the problem I have right now, these Comus AC37s, they're 60 volt alternating current relays. And what I'm starting to think, because I keep frying them, is that when you go from positive 30 and it swings down to negative, you know, 28, that's counted as 58 volts. So I'm trying to, I keep trying to put it up to 40 volts and 50 volts and I keep frying them and I'm either running too much current through or 60 volts alternating current means you get 30 volts positive to 30 volts negative. And I suspect it's the latter, but I'm running out of these. I might have one or two more. It's not like they're the perfect relays. It's just, I've used them a lot and I kind of like standardized things on them. So in other words, in this video, we're not going up to 40 or 50 volts, but you know, there are plenty of 400 volt relays. So the thing to do would probably be to run it at 200 volts and see if this trend continues or at least run it at, as I'm saying, I want to run it at 36 volts with the battery test. But instead of a hundred ohm load, let's do the same thing with a thousand ohm load. So a bit more resistance and I'm not going to go through all of them. Well, I probably will, but I won't film it. I'll just work it out, the timing out at 10 volts because you don't want to mess something up and blow something up when you're at 30 volts. And we'll see what this ratio is with a little bit higher resistance load. So here I'm running it at 30 volts and the main difference with it less loaded, it really springs back well. And so when I ran it at 10 volts, that was our ratio. And when we plug in the numbers here at 30 volts, that's our ratio of out to in or in to out, whatever. And, you know, consistent with being less loaded, the optimal timing did slow down from 700 to 750 microseconds. So, I mean, this starts to be where it doesn't look as much like, well, maybe it's a measurement error. But I, you know, I am kind of disappointed. I was seeing considerably better than this previously. And so this is, for me, it's sort of, there's a lot of art involved because I don't really know exactly everything that's going on. There's one thing I'm almost certain would improve it. And perhaps this is what I was using before. If we go with a 4.7, I looked at this more formally and it improves down to about one UF. But a 4.7 is about all I'll be able to get a reading on here. I'm almost certain that would improve things a bit. I'm not sure how much. Now, the other questions. I'm starting to think that, you know, the other E-cords that I have are a little bit larger. That might help. You might want a little bit larger air gap. You don't want too large an air gap because, as I said, when I just made an infinite air gap, it didn't seem to want to transfer out to the secondaries. But the main thing that I think, the main thing I'm thinking, I mean, right now, it looks like I might even have more winds on the primary than I have on the secondaries. Now, I realize that this isn't, this isn't, you know, your standard AC turns ratio thing going on. But the turns ratio is still important. And you might want to get some of that flyback transformer stuff going on. So, in other words, if I take some winds off here, it will transform voltage better. And I actually, I want to nail that down, that can you just keep removing winds here, because they're going to be, they'll probably, like everything else with this, they'll probably be a sweet spot. But can you just get away with removing winds, change the turns ratio, and not make things worse, and ideally make things better, at least up to a point. Because the other thing here, I mean, you know, we're putting in 147 milliwatts, and that's, that's out, it'd be nice to actually be putting like a watt in, or two watts in. And you do that by getting rid of some of these winds to let it run faster. So I don't know, I'm going to work on that now. And I'll either use that to end the video, or that'll be the start of the next video. So let me just work on that, and especially if it improves things, I'll probably throw it on the end of this video. But this is enough to test it with some nickel metal hydride nine volt rechargeables, and just see. Because, you know, the other question, when you're going batteries to batteries, of course you're not going to, you're not going to have a resistor in, but what, you know, the idea here is you could run as I was saying, say four nine volts, 36 volts in, and then run it to two nine volts on one side, two nine volts on the other side. And it would probably be, if you put the two in parallel, then that would probably be like similar to having a thousand ohm load. And if you put them where you're going from 36 volts down to 18, I'm sorry, the nine volts would be like a hundred ohm load. The 18 would be closer to a thousand ohm load. And if you really, I might do this, if you really wanted to try and nail it down, you could look at the waveform with a hundred, a thousand ohm load, look at the waveform when it's going into a battery, you know, to try and match it to something like this. I've also, I've never looked at it with like a 10k ohm load. But at that point, you're going to be, it's going to be, I don't know, I don't know what it would do. But this is, this is good enough. And this is good enough to go with the batteries, but I'm going to work on it more. And as I said, I'll either update you here or in the next video. Okay, now we're cooking with gas. So I just, I just went to a different set of E-cords here. And these, the windings are better, the inductances are matched better on the two secondaries. It's thicker wire here, blah, blah, blah. I don't really know what's going on. There's still, this is actually a bigger air gap. Bigger air gap, but now you got a turns ratio between the primary and your secondary. You know, there's more winds on the secondary. To make a long story short, when you plug everything in, we're at 10 volts. And that's, that's our ratio now between milliwatts in, milliwatts out. Let's take her up to 30 volts. Well, one step forward and one quarter step forward. I mean, it's better than at 10 volts, but it's not the big jump I was, I mean, each of these is its own animal. And I don't know exactly, like I said, it's like an arc, but we're moving in the right direction. I would have liked to have seen two for the ratio, but that's out and that's in. I'm going to fiddle around with a few more things and then I'm going to end this. This is a good place to wrap it up. But this is, you know, I mean, this is with that circuit that I showed how to build in the last video. And with a couple of E-cores, it's better to have a little bit of an air gap that you can get that kind of ratio of into to out. So I'll confirm that with some nine volt batteries. And then, you know, maybe buy a solid state relay that handles five times as much power. And you have something where you're able to run a cell phone and your laptop or something like that, or at least like keep them charged in an emergency. But I'll be looking for better ways to do this that are less unpredictable and less finicky. And ideally that have a much better ratio. I think this one will be unpredictable and finicky, but we'll have a much better ratio. If I had to guess, I have no idea what's Tom Bearden thing is going to show. And then, you know, can you get to doing this, this same idea of breaking the lens effect away from the primary when you're drawing power off of the secondary? Can you do that without a core? And that would be really cool if you could. So those are the things that will be worked on going forward. Thank you for your time. Stay healthy and talk to you later. Bye.