Відео

Hello! Questions are always good to ask and thank you for your kind words :). What ID would you prefer?

@@ivanrodionov9724 ur mail id would be fine😇

Great question. R1 is there to reduce the RC time constant of the mosfet and make the switching more elegant reducing current spikes and preventing undesired secondary oscillations. You can try with and without it to see how the circuit will behave. Alternatively, you could consider using a (very small) ferrite bead as is done in SMPS modules.

Thank you for your kind words! With enough motivation anything is doable, the problem is, you need circuitry that can switch quicker and deliver more current in a shorter time as I = dq / dt , your total charge stays the same on the mosfet gare, but your dt decreases resulting in many times larger current draw which makes it harder to switch and limits the size of the mosfet and creates inefficiencies in switching. Depending on your output goals it can be achieved, but if it makes sense to achieve it is the question.

I have a slayer excitor that self resonates at 3.5 megahertz with a TIP35C transistor. Its over its "transition frequency" of 3 mhz. I need a better transistor that can switch in resonant mode as high as possible up to 10 or 50 mhz.

@@TravisTellsTruths thank you for your reply! Is there a reason you are using a BJT transistor at all and not a field effect transistor here? These tend to be supperior for the purpose, especially if you want to push more power through the circuit.

​@@ivanrodionov9724 only beause i cant yet learn the best way to trigger the mosfet. I have some really nice 50$ mosfets I just got but still learning exactly how to trigger it in self Oscillation Mode without any driver circuitry. Only maybe.... something like a joule theif coil arrangement to pulse the mosfet trigger? Maybe you know how to do this best? I'm looking for a teacher I can be friends with, or who I can pay to help me with a few things, quickly.

Because I'm so new to building anything electronically. But I already do have some very special things I believe. Like true wireless power across my whole house and a device that makes 3/4 inch plus LOUD white sparks from nothing (12 volts 7 milliamps input). And I'm trying to finish my hydrogen car project.

Hey! Sorry for the late reply, that is a very good question why slayer exciter circuits are mostly run off of low voltage on youtube and at quite low powers. It's not so much about the volts as it is about the current and power the circuit draws. One major problem of single transistor designs is that they just do not scale very well. You can forget using 2 nine volts batteries to power something that will draw circa 1000 watts unless you want to drive it pulsed, but that will drain the batteries in like 5 pulses.

@@ivanrodionov9724 thx. ye i thought bout charging a cap bank first nd then dumping it into the thing. but now i have a problem with mosfets and transistors. bc the transistors i can buy online have shit current rating nad the mosfets need a driver nd i actually wanted the circuit to be as compact as posible so idk if i should try somehow drive a mosfet, or if I should use like 2 or even more transistors in parallel. idk btw the application is "hand held tesla coil" i gues. but I think it's not possible with just a slayer exciter if I want really big arcs. I think I might need to just get a supper well tuned spark gap coil. idk. If you have suggestions that would be very nice.

Thank you for your kind words :) The schematic in this one could use some serious improvements, see the later uploads, if you tweak the diode arrangement you can get maybe 1.5 to 2x more output at the same input power due to a more spectrally dense driving waveform and the circuit running as a switch and not an amplifier.

That's an idea!

@@ivanrodionov9724 I have you the blue print all you have to do is execute.

@@caseyosborne3148 Sure! I'd love to have a look at what you got! Do you use a discord/email/whatever else?

@@ivanrodionov9724 Instead of this, move on to a push pull feedback topology similar to the Skori/Tefatronix. Channel Magneticitist has good examples of these circuits

@@sFeral hey! Thanks for your reply! TBH, much preffer bridged designes over a push pull design as these tend to be easier to build. I might however eventually tr making a push pull design using a center tap coil. Thanks for the recommended channel's I'll check them out.

Thank you for watching. I do believe this is the first digitally modulated Slayer Exciter like circuit on youtube. Maybe sometime in the future (ie after exam season at Uni) I'll soup it up some more.

I'm glad you like it! I will experiment further with this design, perhaps making a resonant half bridge with irfp460s which should in theorhy double the output power for a given voltage as the high and low states will now be driven instead of just the high state and I will be able to run it from mains. I'll keep the modulator circuit though and use maybe an igbt to modulate the circuit. I'll make a video about it eventually too when I get some free time.

@@ivanrodionov9724 the only set back with the half bridge is the need for higher input voltage. symmetric power supply. But yes, then you can use AC from the wall, rectify it to positive and negative DC.

@@MasterIvo I was gonna get back to you, but I was under a rock for the past month cause of Uni, so sorry I am replying late. For a DC load, a half bridge supply will give half the voltage per cycle instead of full and full voltage, so the power of the signal will indeed be lower as it is proportional to the square. I am unsure if this is true with an AC load close to resonance however, as weird stuff begins to happen around resonance. Going with mains voltage certainly is a good idea. The main problem with a bridge circuit is throughshoot which frankly is a pain in the ass to deal with especially at high frequencies. If you go with a half bridge circuit, the main advantage is simplicity and only throughshoot can occur, whereas with a full bridge you need to adjust the phasing just right so that you don't create a short and cross shoot. Trouble is, the frequency of the circuit is dynamic so some fancy tricks will need to be used. TBH, if you have any ideas on how to create a system that can dynamically adjust phasing, I am all ears.

@@ivanrodionov9724 almost missed your comment. but saw it! phasing is still a mystery to me. I also have weird phasing effects around resonance. as if the fields push eachother away. so balancing the field strength would be key (maybe) the through shoot is what I am using in my experiments, as it sets up displacement current from the rapid change in (high) voltage

Hey! I'm glad the schematic works well for you! It has a few major problems though. I recommend you use a modified version of this schematic where you instead use a fixed gate voltage and clamp it with two shottkies and a low ESR source is used for the gate bias such as a LiPo. You get a more square waveform and thus higher efficiency as well as a more stable and modulatable circuit. I made a video testing the circuit last week but still have to upload the video with the schematic analysis.

Nice work with the audio modulation, easily the most powerful audio modulated slayer on YT however I believe the most powerful slayer exciter may belong to myself.

@@sandman1567 Hey, thank you for your comment! That may well be the case! I've been super busy with uni and work the past little bit so I haven't been active on youtube. I'd love to see what setup you are running! The thing with these slayer exciter type circuits is, they are not always behaving optimally and under certain conditions there can be several limit cycles some of which are efficient, some of which are not and blow transistors. I'll eventually redo this circuit but with a superior design that will have only one stable limit cycle and should have a faster convergence due to a much higher loop gain, see this is another point why the slayer exciter circuit is only marginally useful, as higher power mosfets tend to have lower gain factors so ironically, you are better off using a weaker mosfet.

The diode used to prevent negative voltage.

@@ivanrodionov9724 IK BUT WHAT IS IT? sorry for caps

@@nguyendesign4517 ohh, no problems! Honestly any shottky type diode will be fine for this, even works with an LED. I think I used a uf4007 for this.

@@ivanrodionov9724 ohhh okay great thanks man

and can i use tip31c instead?

You are right, I should make a clearer v2. The basic idea is to measure the step responce of the coil and from that measure the resonance frequency via the oscilloscope probe. Now that I think about it, I should have probably done the setup a bit differently, and subtracted the effects of the probe capacitance(which caused a slight under estimation of the resonance frequency).

@@ivanrodionov9724 Thanks for your reply and for taking my comments as the constructive criticism I intended them to be. Your other videos are really good. Thank you.

Its not that there is anything wrong with the slayer exciter circuit, its more that it does not scale well. As they say, better late than never eh :P?

@@ivanrodionov9724 Yes thats what I mean, I was trying to throw 140V at it expecting it to work... Although it was modified version with IGBT and a driver... Had to use 10 1,5kW TVS diodes to prevent the IGBTs from exploding. There should be a better way tho, if you use RCD snubber across the primary winding, it would work as a flyback PSU. But it was much better to build a half bridge SSTC at that point, since I was using IGBTs and driver already, all I had to do was to add a GDT.

RMS or peak to peak? I wouldn't push it beyond 10 watts RMS.

yup.

It should not matter

Antonio, thanks for your detailed reply! Please correct me if I am wrong, but as I understand it R1 is used for two things: Firstly, if the transistor is assumed to be an ideal linear amplifier, it is used to provide an initial condition to the circuit so as to prime it from the unstable equilibrium point about 0 current 0 voltage and move it to a stable limit cycle in which it will oscillate. Secondly, as you are well aware the BJT exhibits some nonlinearities. with this circuit, we want to get maximum power output from it, and this goes down to having more current flow through the primary coils. Now sure, as you have pointed out, in your comment, this circuit can indeed be removed from its unstable equilibrium point by any nudge "even touching". However, without a fixed bias, if too much load is applied to the circuit, the dynamic equilibirum point will shift, eventually converging towards 0,0 as there wil not be enough energy going back to the transistor to maintain the oscillation, and due to aforementioned non linearities, as the equilibrium shifts, efficiency I am assuming will also decrease to due a distorted waveform, hence causing heating and other problems such as voltage spikes. Here, the R1 helps avoid these problems by maintaining a fixed initial condition, that is time invariant! As for the L2, fair enough, the transistor indeed does not truly get turned off and on in this circuit due to it being in the linear region, and R1 maintaining a bias current. As for the simulation, fair enough, a lower coupling value would yield more realistic results, but I doubt that would change much in terms of conceptual behavior, as as far as I know LTspice uses linear coupling loss models. Lastly, I included the parasitic capacitance in the coils in the simulation, however, I am unsure about a load capacitance being mandatory for the simulation to run, as it ran with and without one.

@@ivanrodionov9724 The transistor operates linearly just in the first cycles. In steady-state in operates as a switch, open when the current through L2 is going through the LED and saturated when the current goes to its base. What heats it is essentially the interruption of the collector current when the current in L2 starts to go "up". Without the resistor the circuit easily stops when the load is excessive. With the resistor the oscillator restarts when the load is removed. In a simulation a capacitor from the open top of L2 must be added, chosen so the oscillation frequency is correct. K=1 produces very unrealistic waveforms.

Barely any heating? I am impressed, usually these types of driverless circuits are quite inefficient, sure they are better in terms of efficiency than bjt designs but there is much improvements that can be made, you should try out the schematic from part two. Or better yet build either a bridge/push pull schematic using several fets or transistors for that matter, works like a charm, Ill eventually make a video on it, right now uni sucks up all my free time :D. Glad it worked though!

Well, the mosfet (infineon 07N60C3) has a low R_DS_on at 0.54Ω, and it's FAST, so that helps. I also swapped R1 for 40Ω and VR1 for 10K, and the zener is 6.5V. It shoots about 2.5cm long streamers at 24V and even at that voltage the heatsink is just decoration. So, great work mosfet-optimising the venerable Slayer Exciter, and thanks again for posting it with a readable schematic!

@@TinkerTom Nice! 2.5 cm breakout to air or to an object by the way? I doubt the internal resistance of the mosfet plays a significant role here because the mosfet is not driven by the circuit in the digital modes but rather the mosfet is close to hysteresis, which is great for having a stable oscillator but can lead to tremendous heating of the mosfet as the current voltage characteristics of the mosfet will emulate a higher value resistor than the Rds value (another problem of this circuit). What could however be of help is the gate capacitance as that could theorhetically leed to a more efficient coupling at the cost of more heating. What coils are you using btw?

Open-air streamers don't shot longer than about 1.7 - 2 cm, but they're very fluffy. It arcs to objects 2.5cm away. My primary coil is 5 turns, 1.5cm tall 6cm diameter. Secondary is approx 500turns 36awg (80Ω dc), 4cm diameter, 23cm long. And to be honest, mosfet physics go a bit over my head. I think I might have just won the silicon lottery with this one.¯\_(ツ)_/¯ it works

​@@TinkerTom I guess the streamer lenght is simply a factor of how much juice you give the coil as well as what exactly consitutes a streamer, but yes they do tend to be under 2 cm for lower power systems. What frequency does your circuit operate at and are you using a topload? Seems very similar to the secondary I used, but a bit larger. As for the silicon lottery quite possibly :D ! The funny thing in this circuit is that the mosfet physics are actually rather straightforward for the most part.

Glad you like it! It's a shame it crashed but crap happens. I'll probably build my own new drone as there are quite a few things missing in my opinion from the phantom 3 series i.e. remote controller direct click to mode and the telemetry and video should not be sent on the same carrier. Might make a video on that too. You stay safe too!

Platinum Boi cheers

Both very true observations, however in the first circuit I do want the mosfet to operate in the linear region, while in the second circuit I try to avoid this and drive it in the saturation region to reduce losses, hence why I do not need a monster heatsink. It would actually be an interresting side test to see how the coupling coefficient affects the behaviour of both circuits. Ill try it out when I have some spare time from uni...

What exactly do you mean by lowering the resonance frequency? The operating frequency of the circuit, the resonance frequency of the primary or secondary coil? I will assume you mean lower the resonance frequency of the secondary coil. The simplest way to do this is to wind some more wire on it, no seriously larger coil means lower resonance frequency ceteris paribus, or you could stack a few cans on top of it.

To lower the resonance frequency increase the capacitance, I.e larger top load or toroid capacitor. Or increase the number of windings on your secondary. Larger inductor or capacitor means lower resonance frequenxy

@@ivanrodionov9724 thnx man

@@tf3confirmedbuthv54 i see you everywhere lol

Juntendo lmao

Especially NE5532 LM358 or TL071

@@privatepage7670 yep, im about to get some of those timers.

​@@Uraimhuh

my secondary coil consists of 110 meters of 0.2mm wire wound around a 40mm acrylic tube, while my primary coil consists of 4.5 turns of 0.7mm multistrand wire wound around a 5cm PVC tube.

@@ivanrodionov9724 Translation:About 32AWG secondary and 21AWG primary

@@tf3confirmedbuthv54 I guess your right lol