Maker Pro
Maker Pro

High frequency PWM dimming of LEDs?

eem2am

Aug 3, 2009
429
Joined
Aug 3, 2009
Messages
429
Hello,

We are wanting to do high frequency PWM dimming of the output of Resonant Inductive Couplers (which feed LED lamps) hooked up to a twisted pair bus.
Do you think the “gap toothy” resulting waveform (see below) will be a cause of any problems?

(This is a different question on a subject discussed elsewhere here, needing a separate thread because of the diagramatic info hereby included)

Block diagram………………..
http://i46.tinypic.com/28151kz.jpg

….you can see the “shorting FETs which dim by stopping current flowing to the LEDs when the FETs are ON.
(please ignore the number of secondary turns....as you know, these are resonant inductive couplers)

We do NOT want to do low frequency PWM dimming, (~200Hz) because the low frequency of this dimming causes problems for the current source. (-it means the current source is switching from high power to low power output which causes problems)

Therefore, we wish to do dimming at a much higher frequency (83.3KHz). But do you think that the resulting current waveform (as below, when D is not equal to 1) is going to mean problems?


LED lamp current with no dimming (D = 1)
http://i47.tinypic.com/o9iv.jpg


LED lamp current with dimming with D = 0.083
http://i48.tinypic.com/3ctas.jpg


LED lamp current with dimming with D = 0.33
http://i50.tinypic.com/1zofdzp.jpg

LED lamp current with dimming with D = 0.92
http://i49.tinypic.com/np5d91.jpg

(Also, how would you reduce dissipation in the current source by correcting the power factor in the twisted pair?)

Resonant inductove coupling…………
http://en.wikipedia.org/wiki/Resonant_inductive_coupling
 
Last edited:

KrisBlueNZ

Sadly passed away in 2015
Nov 28, 2011
8,393
Joined
Nov 28, 2011
Messages
8,393
I'm no expert on this subject, but I think your approach has several serious problems.

I don't see how you can use a twisted pair of wires (with equal currents flowing in opposite directions) as the primary of an inductive coupling system. The whole idea of twisted pairs is to reduce susceptibility to external magnetic fields, and this characteristic will also minimise the radiated field.

I've had a quick look through the Wikipedia article (thanks for the link) and as far as I can see, it only deals with inductive coupling between two coils.

Also, the major feature of resonant couplers is that they're resonant. I don't see any capacitors involved with the primary or secondary inductances.

Connecting MOSFETs across the diodes will certainly prevent the LEDs from lighting when the MOSFETs are turned ON. Where will you get the voltage to keep the MOSFETs energised, though?

What is the reasoning behind trying to PWM the LEDs at 83.3 kHz? Your twisted pair "bus" is operating at 50 kHz. Is there some relationship between these frequencies that I don't see?

How much of your system did you simulate in order to get the waveforms that you posted? What practical testing have you done so far?

I may be wrong on all of this; as I said, I'm no expert and I haven't done any experimentation with resonant inductive coupling. But I would like to see the reasoning behind the various design decisions you seem to have made. Maybe an explanation would help me understand how it would be workable.
 

davenn

Moderator
Sep 5, 2009
14,011
Joined
Sep 5, 2009
Messages
14,011
I'm no expert on this subject, but I think your approach has several serious problems.

I don't see how you can use a twisted pair of wires (with equal currents flowing in opposite directions) as the primary of an inductive coupling system. The whole idea of twisted pairs is to reduce susceptibility to external magnetic fields, and this characteristic will also minimise the radiated field.

.

agreed, a twisted feed like that isnt going to work
you are going to need multiple parallel windings to get any significant induction from one coil to the other

you have also said nothing about the voltage in that 12ft long "coil" or what sort of current

I suspect you are seriously overestimating the amount of voltage you are likely to transfer from that coil to the ones with the LEDs

and as Kris has said, there's absolutely nothing resonant about this system you have shown, there are NO tuned circuits present
I have to also assume that you havent considered how long a wavelength is at 83 kHz ( VERY, VERY Long) ...
that is going to be an important factor in how many turns there are on your primary and secondary coils

Dave
 
Last edited:

eem2am

Aug 3, 2009
429
Joined
Aug 3, 2009
Messages
429
I don't see how you can use a twisted pair of wires (with equal currents flowing in opposite directions) as the primary of an inductive coupling system

.thats what i thought before i saw this system working...you just "pull out" one of the twists, and hook in the R.I.C.
There is ferrite in the R.I.C., and its gets hooked over the single bit of twisted pair wire which gets kind of pulled out of the twist so the R.I.C. can be hooked over it......so surely now you know there is ferrtie inside the R.I.C., you no longer have this doubt?

I don't see any capacitors involved with the primary or secondary inductances
......Well spotted, i forgot to add the cap across the secondary..........but in all truth, i have never been invited to see a schematic, and i "suspect" that the R.I.C. secondary is buffered from the load by an adjacent current transformer.....i dont know this though.......what do you think?

Where will you get the voltage to keep the MOSFETs energised, though?
...good question...getting bits of phantom power to do housekeeping stuff like this seems to be a bit of a pain from what i've overheard....but i think they just use another little coupler

What is the reasoning behind trying to PWM the LEDs at 83.3 kHz?
.....how can i explain it succinctly......we could do phase cut dimming at 100KHz, but then we need a way of phase locking to the 100KHz half sine pulse train......so i just chose a period slightly more than the half sine pulse train, and duty cycle it......it works, and is simple.................dont ask me why we're not just doing low frequency PWM dimming (~200Hz)...i think that the constant ON-OFF switching that this creates is a problem for the big current source that puts the 2.9Amp constant current in the twisted pair bus....but i am not too sure?......its secret stuff, and i never saw this bit.

What practical testing have you done so far?
......i havent really, others have....but i have seen the lights shining, and they appear to work well.

Its all about waterproof lighting system......there is no making metal to metal contact in connectors...meaning that water can seep into those contact surfaces too.....this is contactless, and via magnetic field...so its waterproof.....................youre guess is as good as mine as to who needs waterproof lighting.

agreed, a twisted feed like that isnt going to work
again, thats exactly what i thought...till i saw the lights shining...they've even installed a working system with a potential customer.

By the way, the sine wave of current in the twisted pair is 50KHz....the rectified pulse train is 100KHz as you know, and my PWM dimming frequency is 83KHz.....................................there is also a low frequency PWM dimming available but needs a separate card.
 
Last edited:

KrisBlueNZ

Sadly passed away in 2015
Nov 28, 2011
8,393
Joined
Nov 28, 2011
Messages
8,393
Sounds like you're trying to reverse-engineer a system that you don't have direct access to, and don't know much about. Every guess you make is potentially a reason why your version won't work. Good luck; you'll need it!
 

eem2am

Aug 3, 2009
429
Joined
Aug 3, 2009
Messages
429
...OK, but the dimmed circuit that i'm working on does work......i just wondered if the waveforms presented may give problems?..the waveformes are very irregular
 

KrisBlueNZ

Sadly passed away in 2015
Nov 28, 2011
8,393
Joined
Nov 28, 2011
Messages
8,393
Oh, so you HAVE built it. You answered one of my questions by accident LOL.

What kind of problem do you think an irregular waveform would cause? I don't think it will make any difference to the LEDs; they'll only draw current on the positive peaks of the waveform anyway.
 

eem2am

Aug 3, 2009
429
Joined
Aug 3, 2009
Messages
429
I don't see how you can use a twisted pair of wires (with equal currents flowing in opposite directions) as the primary of an inductive coupling system

this is still troubling me, because i agree with it.

.....but ive seen it with my own eyes...at least i saw him take one led luminaire, pull apart a bit of the twisted pair wires, and hook the coupler into the twisted pair piece, and the attached led luminaire lit...............i am now wondering if he discreetly clicked a switch and lit it from an internal battery that i didnt see.

.............But the thing is, an ex-boss of mine came in as a rep for a big lighting company, and he was talking to our apps guy about installing it at his HQ...so it must work........but how?

They claim to have a 240W system.................thats a ~12 foot length of twisted pair, and six 40W led luminaires "hanging" off it.

................how does 40W's get out of the twisted pair and into the LEDs?

I am trying to understand it becasue i am doing a dimming rectifier for it, and id like to try and understand it first....................maybe they have a current transformer buffer between the resonant inductive coupler and the load?

Ive heard them saying that the "couplers" are 90 something percent efficient.
 

(*steve*)

¡sǝpodᴉʇuɐ ǝɥʇ ɹɐǝɥd
Moderator
Jan 21, 2010
25,505
Joined
Jan 21, 2010
Messages
25,505
Got any pictures? A web site where they sell it?
 

KrisBlueNZ

Sadly passed away in 2015
Nov 28, 2011
8,393
Joined
Nov 28, 2011
Messages
8,393
Are you sure the LED module doesn't actually make contact with the wires inside the insulation, like an insulation displacement connector?
 
Top