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.

Varying current

Status
Not open for further replies.
Here's my suggesion for a PWM-ed system with current limiting.
Pot1 sets the PWM duty cycle and hence perceived LED brightness. Pot2 sets the current limit.
The PWM generator operation is described in my Article.
PWM-ed LEDs.gif
 

Attachments

  • PWM-ed LEDs.asc
    6.4 KB · Views: 191
The current is the problem I need to reduce the current will the 0.2 ohm 15 w resistor in series be enough to drop the current 2 amps as I keep destroying the led lightes.
Thank you
Ok, hold on. I was under the impression you just needed to drop a bit of the 37V to below the led's rating of 36V and then vary the brightness from there. Although you did say vary current as well, but I assumed varying brightness is the end game.

Flyback, although a buckboost is probably a good idea, if it was me personally I wouldn't want to go that route, seems a little complex. So I'm not disagreeing, but I would personally do it differently. I do like the whole idea of using a transistor with negative feedback, although I think the OP needs to clarify a few things first.

Why do you want to drop the current by 2A, more specifically, how did you come up with that number? Previously you've just said 36V with a max 7A but you want to vary the current.

Can you answer the following for us to help you further:
1) What is the absolute max current you want?
2) Find out what voltage applied to the LED's gives you that current. This is critical.
3) Is your heatsinking adequate? Have you done any calcs? I know you said it'll be in water, however that may not be enough and thats why they're blowing, not due to overdriving them, but because you've assumed your heatsinking is enough, but haven't actually calculated anything.
 
good points Cicero, good to hear answers to any of them from OP.
anyway, if a bukboost sounds too complicated then I would just go for a really simple hysteretic mode buck boost, you know, one of those ones that just monitors the peak current of a low side sense resistor and does it "rough" like that....look at eg the HV9910B datasheet.....ill give you the sizing of the sense resistor/inductor if you want.
I assume led voltage = 36v
led max current = 7a
you want to dim down to 5a?

(remember Overvoltage protection at the output.)
 
Gregory, let me ask a few different, but important, questions.

1) Where is your 36Volt coming from? Is it a stable, regulated voltage? Or is it from a series string of 3 12 volt batteries?
--- Can that source voltage be increased? If so, you could use a far simpler buck converter instead of a buck-boost.
2) How far from the LEDs will that power source be? What wire size? Or more importantly, what total loop wire resistance.
 
36 volt is what OP says is the led string voltage when it has 7 amps in it...the OP's psu delivers 37V I believe?
 
Information about the project.
The battery is 10/ 3.7 volt lithium polymer cells (battery space supplied) rated 37volts 6.5 amp hr with a 30 amp max draw.
High Power Li-Ion Battery: 37V 6.4Ah (8048168-5C, 236 Wh, 30A rate)
Controller is the 50 volt model Si14HyATPC1-50V-20A Adjustable Ramp-Time
Led are rated 30 - 36 volt 9 amp

Cable is approx 1metre, cable is rated 16 amp 600v.
Higher voltage would maybe 2/ 37 volt batteries
Have larger canister to accomodate them.
Remember we are underwater using wet connectors!!!!
If you go to this web address you will find the controller that is being used and specks.
**broken link removed**.
 
Last edited:
In that case...
bukboost like page 46 of this....
**broken link removed**
it has an adj pin for your current adjusting pot
remember overvoltage cut-off if using buckboost.

or do a rough bukboost with hv9910b
**broken link removed**
or, if you don't like bukboost, then do a boost followe by a buck
do the buck with the HV9910B (eg page 1 scematic)
Do the boost bit again with hv9910b, and just connect it in boost and boost up to say 50V.
uSE constant off time connection (as in datasheet)
the hv9910b has an LD pin for your current adjust pot
 
Information about the project.
The battery is 10/ 3.7 volt lithium polymer cells (battery space supplied) rated 37volts 6.5 amp hr with a 30 amp max draw.
High Power Li-Ion Battery: 37V 6.4Ah (8048168-5C, 236 Wh, 30A rate)
Controller is the 50 volt model Si14HyATPC1-50V-20A Adjustable Ramp-Time
Led are rated 30 - 36 volt 9 amp

Cable is approx 1metre, cable is rated 16 amp 600v.
Higher voltage would maybe 2/ 37 volt batteries
Have larger canister to accomodate them.
Remember we are underwater using wet connectors!!!!
If you go to this web address you will find the controller that is being used and specks.
**broken link removed**.
Doesn't quite answer my questions but lets work with what you've given. And you've already got a PWM controller for brightness as well, I presume you're already using this Si14HyATPC1-50V-20A for dimming?

Anyway, I'm just gonna assume that if you apply 30-36V to the LED's that it'll work fine and the LED's can handle that kind of current. Adding to this, lets assume at 36V its gonna pull 9A because thats how you've specced it.

Further to that, lets say max input voltage would be 37+2=39V. That means you need to drop 3V somehow before you hit 9A. 3V at 9A gives you a series resistor of 0.33Ω. So if you slap in a 0.5Ω resistor in there you should be fine. One component, job done. Only thing to worry about is the power rating of that resistor. As a worst case lets assume the full 9A will be flowing, which it wont because the resistor is 0.5Ω, not 0.3Ω. Either way, 9*9*0.5=40W potentially. Can get those off the shelf for less than a few dollars.

http://uk.rs-online.com/web/c/passi...cl_4294956056,cl_4294956957,cl_4294956191&m=1
 
Remember we are underwater using wet connectors!!!!
If that's salt water, things could get interesting ;). Methinks the connectors wouldl have a short life.
 
Remember we are underwater using wet connectors!!!!
If that's salt water, things could get interesting ;). Methinks the connectors wouldl have a short life.

But not if you use connectors which are designed to be used under (sea) water.
Have a look here:
**broken link removed**

JimB
 
Interesting. Is 'wet connector' a term of the art for that type?
 
What you are saying is to put a .5 ome 40 wat resist in series on the output side of the controller

No.

Your operating conditions vary to widely to be able to do adequate current control with pure resistance. Your battery voltage will be as high as 42 on peak charge (unless you never run the LEDs while charging) down to 30Volts at the low end of the charge cycle. As such, the voltage dropped across your resistor varies from zeroV to 6V. With a half ohm resistor, your current will range from zero to 12 Amps. And it won't max out at 40 Watts. Even if the LED forward voltage stayed at 36, you could have as much as 72 watts. But led voltages decline with temperature. I don't know the exact ratio for your LEDs, but let's suppose 2V. You now have a 8 volts across your half ohm resistor giving 16 amps of LED current and dissipating 128 watts in the resistor.

If you had a higher input voltage you could get away with a linear control circuit, or a buck only switcher. But with the varying voltage that you have, about the only practical choice you have is a buck/boost switcher that provides some intelligent control to compensate for the operating variables.
 
No.

Your operating conditions vary to widely to be able to do adequate current control with pure resistance. Your battery voltage will be as high as 42 on peak charge (unless you never run the LEDs while charging) down to 30Volts at the low end of the charge cycle. As such, the voltage dropped across your resistor varies from zeroV to 6V. With a half ohm resistor, your current will range from zero to 12 Amps. And it won't max out at 40 Watts. Even if the LED forward voltage stayed at 36, you could have as much as 72 watts. But led voltages decline with temperature. I don't know the exact ratio for your LEDs, but let's suppose 2V. You now have a 8 volts across your half ohm resistor giving 16 amps of LED current and dissipating 128 watts in the resistor.

If you had a higher input voltage you could get away with a linear control circuit, or a buck only switcher. But with the varying voltage that you have, about the only practical choice you have is a buck/boost switcher that provides some intelligent control to compensate for the operating variables.
I thought Gregory said his max input voltage was 39V. Either way, I explained all the steps and results with regards to HIS specs and he could easily go through it again with a higher max input voltage.

At 42V, he'll needs 0.66Ω @ 50W max. He could put in two 390mΩ 50W's in series. Still only a few dollars solution instead of having to design a whole 9A variable buck boost.
 
Last edited:
Unless Gregory uses two batteries as he implied, and his vin is then about 70V or so, and then Gregory can use a hysteretic buck. EG the HV9861A or the HV9910B say.

https://ww1.microchip.com/downloads/en/DeviceDoc/20005333A.pdf

or even ZXLD1370, take your pick. (anyone know any others?)

give me a shout if you want the inductor sizings etc.
Buck is easier then bukboost as you imply.

With a buck at 70vin, then when you take the load off, the output cap will get charge up to vin (70v)....now, the output cap will only need to be about 150nf (Max) and so it may not be a big problem, ie, if you were to suddenly connect the load to a powered up driver.. the leds would have a short surge current from the 70v output cap......probably be allright but you may need Over voltage protection circuit.....just a simple zener shutdown. -But its up to you, its not guaranteed to kill the leds so start without it and see.
 
Last edited:
Yeah, if he's going 70V, thats a whole other ballgame. But I presume he meant in parallel for the extra battery time, not series surely.
 
I have red all the posts and would like to know the general concenious of every one . I need to know what way to go to fix the problem of high current and voltage and make the current start at 6 Amps and adjustable and the votage at 36 volts Dc with what I have in place know or replace the unit with another controller. I thank you for your time and help but I need to resolve this
In the post that Flyback posted. Should I use this circuit in conjunction with the PWM.
Hv9910B high Brightness Led Driver.
We need to come up with a combined answer for one fix
Greg
 
Last edited:
Cicero
To answer you question in a earlier post a want a maximum current of 6 amps at a max voltage of 36 volts dc.on the LEDs at this current
There are plenty of heatsinking.
Hope this helped you make your decision
 
Gregory, please forgive me but the fact that you are asking this question tells me that you are not very knowledgaeable about switch mode led drivers...however, due to your low vin and high vout, you will need a switch mode led driver...since I believe you don't have a good understanding of switchers, I believe you must get a higher vin and use a hysteretic buck type of buck converter, with the controllers that I have mentioned earlier...you will need your vin to be 50v at least, so you will need 16 lithiums in series, since they drop to around 3v at the end of the discharge cycle.
I looked at your controller datasheet, sorry I don't recognise what its supposed to be doing, it doesn't look like a controller that you would use to drive leds with...wasn't it a motor controller?
 
Status
Not open for further replies.

Latest threads

New Articles From Microcontroller Tips

Back
Top