Hi, I'm pretty green, and had a question that I didn't even know how to search for.

I'm trying to PWM dim an LED circuit with a customizable voltage between (48v-250v) and up to a 650w total load.

The arduino board outputs @ 5v.

I'm attempting to use the arduino to drive an IRLZ44 switching a seperate 10v 500mA PS which drives an IR644 to PWM switch the final circuit (of up to 650w).

10V PS...


IRZL44...

IRF644...

 

Post a diagram of your proposed concept.

Hard to tell by what you wrote, but it sounds like you want to PWM the input of a power supply?
That's not going to work.

 

Welcome to AAC!

I'm trying to PWM dim an LED circuit with a customizable voltage between (48v-250v)

LEDs are better dimmed by controlling the current through them rather then their supply voltage. Or does that voltage range refer to the time-averaged supply voltage, dependent on the PWM duty cycle?

 

Hi, I'm pretty green, and had a question that I didn't even know how to search for.

I'm trying to PWM dim an LED circuit with a customizable voltage between (48v-250v) and up to a 650w total load.

The arduino board outputs @ 5v.

I'm attempting to use the arduino to drive an IRLZ44 switching a seperate 10v 500mA PS which drives an IR644 to PWM switch the final circuit (of up to 650w).

10V PS...
View attachment 170792

IRZL44...
View attachment 170789 View attachment 170790
IRF644...
View attachment 170788 View attachment 170791

Huh? You are controlling a PSU to dim the LEDs? No need. Just apply PWM to your MOSFET that runs the LED.


Here is an example from a recent project. The MCU is connected to LED0- LED3. The cathode of the LEDs are connected to J*. Their anodes are tied together and connected to the supply voltage. PWNM is applied to LED0-4

 

Post a diagram of your proposed concept.

Hard to tell by what you wrote, but it sounds like you want to PWM the input of a power supply?
That's not going to work.

Firstly, thanks for the replies.

I'm trying to make a circuit that can PWM dim different power supplies and circuit loads via a wide ranging MOSFET. Idk if this is correct thinking or not. :/

Big picture = way to PWM dim different. static load LED circuits with possibilities of 250v or 14A max, or a combination of some mix but a max continuous power consumption of 650w.

This my first schematic attempt ever lol..

Welcome to AAC!

LEDs are better dimmed by controlling the current through them rather then their supply voltage. Or does that voltage range refer to the time-averaged supply voltage, dependent on the PWM duty cycle?

Perhaps you may know a better way to dim the LED arrays?

I currently use "a" style MW drivers and adjust current for appropriate brightness, but would like to PWM dim between the max current set via the "a" style built-in POT. So I could dim within a set range. Id also like to do this wirelessly, via phone. I'm new to E-circuits and don't know my options aside from analog POT or MOSFET PWM.

Id like to make a few "adapters" that can be used universally between different drivers and LED arrays to PWM dim via Arduino MC. I figured PWM was the most translatable for my "universal" intentions.

Ordering one fat slice of knowledge pie please lol

 

Huh? You are controlling a PSU to dim the LEDs? No need. Just apply PWM to your MOSFET that runs the LED.


Here is an example from a recent project. The MCU is connected to LED0- LED3. The cathode of the LEDs are connected to J*. Their anodes are tied together and connected to the supply voltage. PWNM is applied to LED0-4

View attachment 170814

The circuit voltage could be 250v, and I didn't think a 5v Arduino could drive a 10v Vgs. So... I'm trying to use a 5v Vgs Logic n-MOSFET to switch a 10v PS to then drive the 10v Vgs of the n-MOSFET capable of handling 250v. ?? Idk if this is correct thinking or not though???

Welcome to AAC!

LEDs are better dimmed by controlling the current through them rather then their supply voltage. Or does that voltage range refer to the time-averaged supply voltage, dependent on the PWM duty cycle?

Max duty cycle would be 100% at 650w continuous. 250v max. Idk if this answers your question or not though lol

 

The circuit voltage could be 250v, and I didn't think a 5v Arduino could drive a 10v Vgs. So... I'm trying to use a 5v Vgs Logic n-MOSFET to switch a 10v PS to then drive the 10v Vgs of the n-MOSFET capable of handling 250v. ?? Idk if this is correct thinking or not though???


Max duty cycle would be 100% at 650w continuous. 250v max. Idk if this answers your question or not though lol

You want to use a logic level MOSFET.


Why on earth are you trying to switch 250V to light a LED? If you are trying to switch 250V @ 650w and you don't know what you are doing then STOP NOW! Before you kill yourself.

 

You want to use a logic level MOSFET.


Why on earth are you trying to switch 250V to light a LED? If you are trying to switch 250V @ 650w and you don't know what you are doing then STOP NOW! Before you kill yourself.

That's why I'm here asking questions first, and thanks for looking out!

(2) QB304+ wired in series is 222v. A QB304+ is (8) parallel strings of (38) LM561c in series. (4) QB288 V2 (LM301b, 18s:16p) wired in parallel is 11.2A, ect, and then there's COB's, ect. MW drivers don't go bigger than 600h, but are underrated, thus my 650w max load. Plenty of different configurations to come up with given those variables. Im trying to build a universal dimming circuit that will work with any arrangent within the maximum bounds of 650w, 250v, and 14A.

Would be nice to be able to PWM dim. A logic level MOSFET is determined by how great the difference between Vgs(th) and Vgs is and how low its On resistance is, right? Or am I incorrect?

I figured the IRF644 w/ a Vgs(th) between 2v-4v might be sufficient to PWM 10v? It has an Rds(on) of 280mΩ.

Is there a common logic level MOSFET capable of 250v that you'd recommend?

 

You do know that LEDs are DC not AC devices? Where are you getting your high voltage DC from? Or are you thinking you can do this from your mains? ( in which case I'm not sure it's allowed to even be discussed here) If working with mains voltage you need a step down transformer to a much lower voltage than your mains, then it has to be rectified(made into DC) by diodes.

And no you don't have the logic level mosfet correct. First off just get the notion of threshold out of your thought process, it is the point where the mosfet is considered off. It is only relevant if your making an amplifier like for audio. Logic level has to do with the Vgs, Voltage gate to source. Logic level mosfets are fully on in the 5V range. A standard mosfet is fully on at around 10V.

 

Im trying to build a universal dimming circuit that will work with any arrangent within the maximum bounds of 650w, 250v, and 14A.

I can't see that working out if you want to use it in conjunction with existing off-the-shelf LED drivers, which will have their own idiosyncrasies.

 

To help the TS continue this project is simply irresponsible. TS is looking at switch 650 watts! One slip and zap no more TS. This thread should be shutdown.

 

To help the TS continue this project is simply irresponsible. TS is looking at switch 650 watts! One slip and zap no more TS. This thread should be shutdown.

Thanks for looking out!

Im an engineer by profession. I'm not going to work unsafe, and understand the basic electrical fundamentals from my engineering background. I work with these voltages and amperages daily.

"B" style Meanwell drivers come with 10v PWM control leads. I'm trying to be able to wirelessly dim "A" style Meanwell drivers, and thought I might be able to do that via PWM.

I can't see that working out if you want to use it in conjunction with existing off-the-shelf LED drivers, which will have their own idiosyncrasies.

That's what I was wondering.

It seemed to take a second to power on the LEDs after plugging in the driver, and was wondering if this "rise time" or turn on delay time would effect how I wanted to dim (PWM) :/

Perhaps I'm looking for a digital POT? I'd like the LEDs (driven via drivers w/o PWM leads) to be dimmable, without mechanical movement or physical touch. IE wirelessly...

Here's 2 LED arrays wired in parallel with a 320h-2800a...


Working on a new LED PCB design and generating the Gerber via EAGLE as we speak but would like to have some universality in my junction box design so that it could also control my old LED arrays utilizing "a" style CC MW drivers as well as my new light utilizing a "b" style MW driver (built-in 10v PWM dimming leads).

 

Thanks for looking out!

Im an engineer by profession.

Why do I have a hard time believing that?

 

You do know that LEDs are DC not AC devices? Where are you getting your high voltage DC from?

Yes, external Meanwell driver.

And no you don't have the logic level mosfet correct. First off just get the notion of threshold out of your thought process, it is the point where the mosfet is considered off. It is only relevant if your making an amplifier like for audio. Logic level has to do with the Vgs, Voltage gate to source. Logic level mosfets are fully on in the 5V range. A standard mosfet is fully on at around 10V.

So the only difference between logic and non is the Vgs voltage? Other than that they operate the same? (That's what I was counting on)

I was thinking that you could switch a 10v circuit on and off via a logic level MOSFET and 5v PWM signal coming from a MC, and that this resultant switching 10v circuit, could then drive the Vgs for the bigger MOSFET which would be in charge of switching possibly 250v. In essence, duplicating the 5v PWM signal from the MC to a 10v PWM signal. That was my goal. To be able to increase the 5v PWM MC signal to a 10v PWM MC signal so that I could PWM dim an external CC driver via traditional MOSFET (as it seems the logic level are too small for my possible circuit voltage).

 

Why do I have a hard time believing that?

Believe what you'd like, my concentration isn't electrical but I understand the fundementals. I could talk days about subsurface fluid migration or the principles of geosteering or down hole pressure management but I'm interested in learning something I don't know.

 

Yes, external Meanwell driver.


So the only difference between logic and non is the Vgs voltage? Other than that they operate the same? (That's what I was counting on)

I was thinking that you could switch a 10v circuit on and off via a logic level MOSFET and 5v PWM signal coming from a MC, and that this resultant switching 10v circuit, could then drive the Vgs for the bigger MOSFET which would be in charge of switching possibly 250v. In essence, duplicating the 5v PWM signal from the MC to a 10v PWM signal. That was my goal. To be able to increase the 5v PWM MC signal to a 10v PWM MC signal so that I could PWM dim an external CC driver via traditional MOSFET (as it seems the logic level are too small for my possible circuit voltage).

Yes they are commonly called "logic Level" because of their ability to switch at logic levels.
http://www.nteinc.com/Web_pgs/LL_MOSFET.html

 

Well it looks like there's a name for what I'm trying to create. A (IGBT) gate driver?..

Soaking up as much as I can, once I think I've got it figured out Ill post an ammended/new schematic to have you guys look it over. Thanks!

 

These might help -

https://toshiba.semicon-storage.com/info/docget.jsp?did=59460

http://www.ti.com/lit/ml/slua618a/slua618a.pdf

https://www.infineon.com/dgdl/an-937.pdf?fileId=5546d462533600a40153559ea1481181


Regards, Dana.

 

Yes, external Meanwell driver.

I was thinking that you could switch a 10v circuit on and off via a logic level MOSFET and 5v PWM signal coming from a MC, and that this resultant switching 10v circuit, could then drive the Vgs for the bigger MOSFET which would be in charge of switching possibly 250v. In essence, duplicating the 5v PWM signal from the MC to a 10v PWM signal. That was my goal. To be able to increase the 5v PWM MC signal to a 10v PWM MC signal so that I could PWM dim an external CC driver via traditional MOSFET

Maybe we can get to what you"re seeming to want to do by another way. As I understand it now, you want to control the power supply on and off with a micro controller, is that correct? If it is you could do that with something off the shelf and easily available it's called a solid state relay or SSR. https://en.wikipedia.org/wiki/Solid-state_relay By using one of these you don't need to make a PCB, choose a mosfet and circuit or burn the house down or hurt yourself.

You would though need to know what the output voltage of the micro controller is and the amperage and voltage your power supply needs. There are many places to buy them after making your selection so that is also for you to decide. But I would choose one of the big name brands like Crydom or Omron for reliability.

If this isn't what you're wanting to do maybe try again to describe it. Or give the model number of the Meanwell so we might determine what/how to do what you want.

 

Maybe we can get to what you"re seeming to want to do by another way. As I understand it now, you want to control the power supply on and off with a micro controller, is that correct? If it is you could do that with something off the shelf and easily available it's called a solid state relay or SSR. https://en.wikipedia.org/wiki/Solid-state_relay By using one of these you don't need to make a PCB, choose a mosfet and circuit or burn the house down of hurt yourself.

You would though need to know what the output voltage of the micro controller is and the amperage and voltage your power supply needs. There are many places to buy them after making your selection so that is also for you to decide. But I would choose one of the big name brands like Crydom or Omron for reliability.

If this isn't what you're wanting to do maybe try again to describe it. Or give the model number of the Meanwell so we might determine what/how to do what you want.

Yes! Lol I'm not the best at describing things.

I'm ultimately trying to make an external circuit thats universally attachable to any AC-DC driver (CC or CV, up to 650w continuous and 250v) on the output side, so that I can dim the LEDs attached.

I'm finding that dimming on the output side of CC drivers is going to be difficult given the voltage is variable. I'm guessing at low duty cycles the driver voltage will try to adjust (speaking about the averaged low current over time) and not be able to exceed the max rated voltage, in turn failing to power the lights? Is this correct thinking?

If so, I'm wondering if I can design a cct with some sort of slow discharge capacitor in front of the MOSFET, SSR, Digital POT (or whatever dimming component, although like the SSR), so the driver gets a more constant load but the LEDs can still be dimmed due to a constant PS?

What about doing anything on the AC side of the CC driver?



* I know it will be more complex than just adding a cap, I'm just talking generalizations, is there merit to this logic?

Can I store the bursts of output supplied by the driver when it exceeds its max rated output voltage and flashes instead of dims, and then discharge that power more evenly with some sort of continuous disharge cap cct? How hard on the driver would this be?