Continue to Site

Welcome to our site!

Electro Tech is an online community (with over 170,000 members) who enjoy talking about and building electronic circuits, projects and gadgets. To participate you need to register. Registration is free. Click here to register now.

  • Welcome to our site! Electro Tech is an online community (with over 170,000 members) who enjoy talking about and building electronic circuits, projects and gadgets. To participate you need to register. Registration is free. Click here to register now.

50KHz Power bus is OK..?...Yes 50 KiloHertz

Status
Not open for further replies.
I don't think it works as drawn in the simulation, but I have no idea what you are trying to do.
 
I would be rather concerned about the potential for radiating a high level of interference from a power bus running at 50kHz. I trust adequate screening/suppression is in place?
 
Why do we need an LC tank circuit?

Knowing that you are not doing a self resonant circuit, but instead supplying the frequency external to this circuit. Then you really don't need the LC tank. In theory you can cancel out the secondary's inductance altogether if the secondary section is resonant. But that is not the case with the values given, nor is it required with this application.

Do you agree the cap where i put it corrects the power factor?

Sorry no, not really. As you can see by the first picture I posted in post #7 in that configuration C1 will pass more than 1 Amp and do a whole lot of not much else. By law of conservation of energy, the vast majority of that energy is returned back to the inductors, as it is not really being dissipated in anything else. So, not a whole lot has changed. It cancels out some of the secondary's inductive reactance. Leading to a slightly lower secondary impedance at the target frequency.

628.319 Ohms (secondary inductive reactance at 50Khz)
Minus
31.831 Ohms (C1 capacitive reactance at 50Khz)
Equals
596.488 Ohms (Total reactance at 50Khz.)

If you take C1 down to 5n it will (more or less) totally cancel out the secondary inductance, then it would be mostly resonant. Not seeing much happening in the sim though if you do this, so there is no real point. I would do as I have shown the the last picture of post #7 and just put a 10μf cap in parallel with the output and an inductor in series before it. This gives you the cleanest output.


Edit: Alec_t brings up a very valid point as well.
 
Last edited:
I wouldn´t take the values in that schematic seriously.

True enough, but as long as the science carry's over I'm not to worried about it. He can figure out the correct values for his actual part if needed, bit only so long as he knows the process. Honestly I'm not taking this thread too seriously, just playing around for my own amusement. I don't want to be negative or pick a fight with anyone, but when the OP played the "senior engineer" card, I lost most of my hope for this being a reasonable and honest project. Maybe I'm just pessimistic, but it seems like a fictions entity to (A) shift blame on to someone else, and (B) Restrain project flexibility to what the OP wants, regardless of what is suggested to him.

But I'm open to the possibility of being proven wrong. :p
 
but when the OP played the "senior engineer" card, I lost most of my hope for this being a reasonable and honest project. Maybe I'm just pessimistic, but it seems like a fictions entity to (A) shift blame on to someone else, and (B) Restrain project flexibility to what the OP wants, regardless

..sorry i forgot to mention that the senior engineer concerned is probably one of the top engineer/scientists in the world in this subject.........just thought it relevant to bring this in, to kind of instill that this isnt some start-up staffed by clueless jacks.....the only clueless jack there is me.

Here is the system.....your commects on power factor are very good all of you.

Here is a block diagram to remind us:
https://i46.tinypic.com/28151kz.jpg


.......The twisted pair bus carries AC volts and current........i am wondering....

.....what is the power factor likely to be?
......do you agree that the fundamental of V & I are in phase?
.....How would you correct the power factor?
......why would you think correcting the power factor was wothwhile?

........Do you think that correcting the power factor will significantly reduce dissipation and thermal heating in the 50KHz current source?


(BTW, the 50KHz current source is mains fed, and , as you know, its INPUT is power factor corrected, as demanded by the regulatory authorities....though i here, am speaking of the power factor in the OUTPUT of the current source, and the regulators dont care about this as you know)

This system is waterproof and thus good for "remote operations in wet places".......and the regulatory bodies are not going to be concerned about emissions from the twisted pair.......(and if they were bothered, they'd be told to "butt out")....



just put a 10μf cap in parallel with the output and an inductor in series before it. This gives you the cleanest output.

....that was the first thing i did.....put the said L and C as you describe.....but the senior engineer told me to get rid of the LC filter, because he said that the current waveform was "very smooth"...........his very words..."things are very smooth here, get rid of your inductor"
 
Last edited:
Ahh, Power factor correction for the led load. I think you can see how it works if you increase the voltage in your simulation until the leds conduct (also add some resistance to the source). If you then look at the current and voltage waveforms in the primary with and without the cap you can see that with the 100n the the waveform returns everything to a nice sine wave with 90 degrees of phase.
 

Attachments

  • pwr fact.png
    pwr fact.png
    118 KB · Views: 115
  • power factor.asc
    1.9 KB · Views: 120
()blivion.............thankyou....................so that's what's happening in this system...."resonant inductive coupling"...that explains why they call then "couplers" and not "current transformers".

So the capacitor is there to give resonance......so i am now wondering why the designer told me it was to correct the power factor.?

ronv....i think the 100n is causing a problem in your schem.......it seems to be "stealing" current away from the LEDs...also, the source should be a current source.......anyway, as siad, the 100n you show makes for less current in the LEDs.......
 
Last edited:
Resonance between the primary and secondary does more or less correct power factor. It improves coupling of air core transformers by allowing the LC circuits to build up power over many cycles. Improved coupling = improved power factor. This is because more of the power is transferred across the transformer gap rather than rejected at the primary do to high reactive impedance.

That's the theory anyway.
 
Last edited:
There is another thread with some more information that would "maybe" support your theory except the frequency is wrong. It works great as power factor correction. My guess is (based on the other thread) that the coupling may be capacitive because he talks about just throwing the twisted pair over the primary wire. Might also explain why the worry over power factor (unless there are more leds)
 
Thanks, yes i did run it.....i found so much current flowing in the cap that less current was flowing in the leds.....and this cant be good, or have i missed something......this "resonant inductive coupling" business.......i have never seen the schem of whats really inside the coupler
 
i found so much current flowing in the cap that less current was flowing in the leds

I believe any current flowing in the cap will by law of conservation of energy be returned to the system via the inductors regardless of other factors. Energy bounces back and forth from the cap and inductor. If its not getting used up then it's only dissipated in the very small ESR of the cap + inductor. This is so small a power loss that there is almost nothing lost from cycle to cycle.

However, Ronv isn't wrong to poopoo the resonant inductive coupling idea per se. The cap is not the correct value to entirely support the hypothesis. Though this of course depends on the real inductance of the secondary, which is sort of a variable as far as I have seen. With 100nF the inductance should be more like 100μH, in order to resonate at 50Khz. I think the effective secondary inductance may change based on how much the LC tank is loaded.

It would be nice to know the exact specs of the transformer, or its construction and dimensions at least.
 
Last edited:
()blivion,
I think you are right it is similar to a wire follower. Or a little like a wireless charger, but I think in those cases both sides are resonant at the same frequency.

Like this:

**broken link removed**
 
Yes they are. And that is pretty much how I would do it too.

I think this particular project gets around that by forcing current through the inductor with an "AC current source". As I understand it this means the primary is driven by a supply such that the waveform is not a sine voltage, but rather a sine current, peaking at ~2.7 Amps, and rotating at 50Khz. So in effect, the primary is resonating at 50Khz. One simply has to remember to include the current source as part of that primary circuit. I suspect it is done the way the OP has shown so that the primary inductor can be any arbitrary value without effecting the resonant frequency. Though with a high inductance you would run into our old friend "reactive impedance" and once again need some hundreds of volts to cut through it. I suspect the true primary inductance is far far far lower than what was shown by the OP.

As I said, I would not do it that way I don't think. I would more than likely make some form of tuned power oscillator out of the primary.
 
Last edited:
This starts to resemble how RFID tags get their power. But these use resonant tanks on the primary.
Also, how can a twisted pair cable exhibit a significant magnetic field?
 
how can a twisted pair cable exhibit a significant magnetic field?

...dont forget, the twisted pair has the coupler ferrite hooked over it when the lamp is connected up.

I suspect the true primary inductance is far far far lower than what was shown by the OP.
...i believe you are right....this is all secret stuff to which i am not invited to see schemtics etc.......ive just been asked to look into the dimming of the leds with high frequency pwm dimming.........i dont know, but i think that the resonant inductive coupler feeds into a current transformer, and then the CT feeds the load....i think it has to be like this otherwise the load would load down the resonant inductive coupler's secondary ?

I would not do it that way I don't think. I would more than likely make some form of tuned power oscillator out of the primary

..well, for all i know, maybe in the coupler they are doing just that?
 
how can a twisted pair cable exhibit a significant magnetic field?

I was wondering the same thing myself. I expected that it was twisted until a point, then untwisted for some kind of receptacle, or taped for the coupling. That, or some convoluted crap.

The twisted pair has the coupler ferrite hooked over it when the lamp is connected up.

Ugh (-_-), so then it's not wireless lighting? God I hate compartmentalized proprietary engineering. Tell them to give me a damn NDA already, I'll do the whole project, and then they can pay me for it. Telephoning in 20 questions is going to figure out the "secret engineering" anyway, So I don't see the point. Waste of everyone's time.

I've just been asked to look into the dimming of the LEDs with high frequency PWM dimming

I wouldn't necessarily resort to using PWM. If it was my task, I think I would try a similar trick as the Thyristor dimmer. The human eye can't see anywhere near 50Khz flicker, so it would be perfectly comfortable missing entire cycles. If it needs to be computer/centrally controlled, well... I would think of something.

For example, one could use phantom powered MCUs and piggy back address/brightness information on the 50Khz carrier. Then you could independently control all the lights wirelessly, from a central location.
 
Status
Not open for further replies.

Latest threads

New Articles From Microcontroller Tips

Back
Top