I am using a 36V power supply to power a Meanwell LDD 700H LED driver. I am using a 1n4007 as a reverse polarity protection diode on the input side. Currently it works fine with my test setup of 3 LEDs. But can the 1n4007 handle the voltage? I try to look for it in the datasheet but I don't know which parameter I am supposed to look at.
Also, can I use a P-FET instead? I am concerned that the diode won't be fast enough for switching the PWM at the output side(I am using an arduino Uno for the PWM). For that reason I have not connected a diode at the output side. But I would really like some reverse polarity protection there too. I was looking at this P-FET - SQP90P06
Slightly off-topic but I did not want to create a new thread, will this LED driver need decoupling capacitors?
I did not mean reverse voltage. I meant the max voltage the diode can tolerate when conducting. Or does it not matter?
Sorry, I do not have a circuit diagram software installed yet. I will post a diagram as soon as I can.
However it's a trivial circuit so maybe a description will do till then?
It's just a screw terminal for input, a screw terminal for output, and another one for PWM pin. The arduino and LED driver are common grounded via another screw terminal. The input side has the 1n4007 on the positive side.
The 0.7volt forward drop of that diode, with the 700mA of that LED driver (=0.5watt), could make that diode smoking hot.
Not sure why you think you need reverse protecton. Shouldn't this be a permanent connection.
Nothing is going to happen to the LED driver if you short it's output or if you don't connect the LEDs, or connect them the wrong way.
The LEDs, with a max reverse voltage of ~5volt (each), might not like it.
Connecting a 10-LED string the wrong way shouldn't be a problem.
Leo..
Not saying it can't cope with 700mA.
This is ofcourse a switching LED driver, so it will draw less than 700mA, depending on Vf of the LED string.
Leo..
Wawa:
The 0.7volt forward drop of that diode, with the 700mA of that LED driver (=0.5watt), could make that diode smoking hot.
Yes, that's another reason why I asked for a P-FET recommendation. I could use that for reverse polarity protection. It should have lower power dissipation, right?
Not sure why you think you need reverse protecton. Shouldn't this be a permanent connection.
Nothing is going to happen to the LED driver if you short it's output or if you don't connect the LEDs, or connect them the wrong way.
The LEDs, with a max reverse voltage of ~5volt (each), might not like it.
Connecting a 10-LED string the wrong way shouldn't be a problem.
Leo..
Because I get careless and have blown a few things due to connecting them the wrong way
Also because these drivers and LEDs are expensive and hard to get where I live.
The LEDs are 3W 700ma LEDs. Their Vf varies with the color of the LED in the range 2.6 to 3.5V as far as I can remember. I thought connecting them the wrong way would damage them but if an experienced person tells me that nothing will happen, I will take their word for it
I meant the max voltage the diode can tolerate when conducting. Or does it not matter?
I suspect from that question that you are considering the voltage across the diode incorrectly. The voltage across a diode (any diode) while conducting is not something that is applied externally by the circuit it is in, it is a characteristic of the diode. You pass current through the diode and it responds by dropping a voltage, typically about 0.6V or 0.7V for a simple PN silicon diode. Other diode technologies drop different voltages. Of note are LEDs, which have different typical voltage drops depending on their colour, and Schottky diodes, which have low voltage drops.
Your question:
Can the 1n4007 handle the voltage?
Does not therefore make sense in the context of the diode's forward voltage drop as that voltage is a characteristic of the diode, not something your circuit is applying to it. That's why I gave you the reverse voltage because the reverse voltage is the only voltage you need to consider in the context of a question about a diode's ability to handle an applied voltage.
Because I get careless and have blown a few things due to connecting them the wrong way
And so does everyone else, including me. Sometimes you get lucky and components get hot but still work when you connect them the right way. Sometimes the magic smoke escapes and you have to buy new ones.
Well I just connected something backwards again, lol. Thankfully, it did not blow up. I guess it has reverse polarity protection. I'm not kidding when I say I need it.
Thank you Perry, that clears things up for me.
That answers the question that I created the thread for. For the other questions(the P-FET and decoupling) should I create a new thread?
For the other questions(the P-FET and decoupling) should I create a new thread?
I didn't answer about using a P-FET because as far as I know there is no way to use a FET as a diode. I didn't want to say this only for someone to explain how to use a FET in place of a diode.
I answered as best I could with the available information for decoupling. Provide a diagram. Don't worry about not having software for circuit diagrams, draw a diagram with a pencil and, preferably, a ruler and post a photo of it. My answer remains what it was, everything needs decoupling capacitors. Or, if you prefer, if in doubt use decoupling capacitors. Extra ones won't do any harm and they don't cost much.
Sorry, I did not post a drawing earlier because my drawing skills are horrible.
I am posting the drawing as well as the perfboard circuit I made(I'm not particularly proud of either of them).
The yellow wire is the PWM signal wire. The long black wire is for connecting the arduino's ground.
The portion with the diode is the input side. The extra terminals are for future use.
Regarding the FET, it is supposed to be possible to use a P-FET for reverse polarity protection. Here's a video explaining it - link. The FET part starts at 1:40 .
Edit: I had a brainfart and labelled the diode as IN4007. It should be 1N4007.
nova_IN:
I am using a 36V power supply to power a Meanwell LDD 700H LED driver. I am using a 1n4007 as a reverse polarity protection diode on the input side. Currently it works fine with my test setup of 3 LEDs. But can the 1n4007 handle the voltage? I try to look for it in the datasheet but I don't know which parameter I am supposed to look at.
As has been pointed out its the current or power rating you should be concerned with - using a schottky
diode is a much better idea if heat dissipation is an issue, and better still a lot of schottkys are available
as a TO220 package so can be easily heat-sinked and finding one rated for 30A or more is quite possible.
Also, can I use a P-FET instead? I am concerned that the diode won't be fast enough for switching the PWM at the output side
The input side doesn't see the PWM so much as the converter has internal decoupling, but a schottky diode is very fast anyway.
Sorry to be a pain Nova_IN but please can you upload the drawings to this site, not provide a link to somewhere else. This is explained in How to use this forum - please read
Please don't worry about not being good at drawing, no one minds that.
Regarding the FET, it is supposed to be possible to use a P-FET for reverse polarity protection. Here's a video explaining it - link. The FET part starts at 1:40 .
Thanks for that! Makes sense, not something I would have thought of. I have learned something today. ++Karma to you for teaching me.
That seems like a reasonable approach to me, try it.
MarkT:
As has been pointed out its the current or power rating you should be concerned with - using a schottky
diode is a much better idea if heat dissipation is an issue, and better still a lot of schottkys are available
as a TO220 package so can be easily heat-sinked and finding one rated for 30A or more is quite possible.The input side doesn't see the PWM so much as the converter has internal decoupling, but a schottky diode is very fast anyway.
That's a good idea. I will just have to find a schottky diode in TO-220 package with a low reverse leakage current at a good price(I would only need 3A current rating max). Do you have any recommendations so I could get a head start?
PerryBebbington:
Thanks for that! Makes sense, not something I would have thought of. I have learned something today. ++Karma to you for teaching me.
That seems like a reasonable approach to me, try it.
I had a feeling you would like it. Regardless, I am humbled. : )
Please call me nova. I have updated my post - removed the links and uploaded the images as... well, images, to the site.
I would just draw a big red + and a big black - on the board with permanent marker pens.
Some day you have to learn to triple-check what you have done before applying power.
Not wise to use a 36volt supply for a string of three LEDs.
That's going to drop efficiency of that driver.
Use 7-10 LEDs in series, or drop the supply to ~15volt.
An old 19volt laptop supply is perfect for strings of 3-5 LEDs.
Leo..
Wawa:
I would just draw a big red + and a big black - on the board with permanent marker pens.
Some day you have to learn to triple-check what you have done before applying power.
Not wise to use a 36volt supply for a string of three LEDs.
That's going to drop efficiency of that driver.
Use 7-10 LEDs in series, or drop the supply to ~15volt.
An old 19volt laptop supply is perfect for strings of 3-5 LEDs.
Leo..
That's just my prototype setup right now for testing. My final build will have about 20 LEDs split among 2 or 3 drivers.
The efficiency is something I didn't think about before buying the 36V smps. I checked the datasheet and the efficiency at 36V 6 LEDs is ~93%, 10 LEDs ~ 96% and at 24V DC 7 LEDs is ~96%. So I will be losing some efficiency with 6 or 7 LEDs per driver at 36V but not that much. That's assuming I interpreted this correctly. Let me know if I'm wrong.
TomGeorge:
Hi,
How are you powering the Arduino?
What model Arduino?
Thanks. Tom..
Hello,
currently I'm using an Uno powered using the USB port of my laptop.
In the final build I will use a Nano and a buck converter module that will step down the 36V to 5V, and feed the output to the Nano's mini usb. I made a test perfboard circuit for that module with a 1N4007, a 0.1uf and a 100uf cap on the output and an LED to indicate power. It works fine.
