NIMH cells, PWM modulation of high loads?

krutzbeuazen

Newly Enlightened
Joined
Jan 31, 2004
Messages
69
Location
germany
Hi!

again my LED-Cluster-Project.
i´ll use between 50 and 100 LEDs, maximum 10 watts.
i hope to be able to use 2x4 NIMH cells, but perhaps just 4 will fit. these 2300mAh Ansmann cells can deliver up to 4.6A maximum.
my question is now, how "fast" will they give me high currents? if i use a PWM, it would try to get 1.5A or 3A in fractions of a second. i´m quite sure thats way too fast.

did anyone try something similar yet? perhaps PWM and a LS?

if it doesnt work, i could use a huge capacitor between PWM and cells. but would be large and dangerous, with that huge energy stored.. would prefer to avoid that..

any hints?

thanks

manuel / krutz
 

Ginseng

Flashlight Enthusiast
Joined
Feb 27, 2003
Messages
3,734
Willie Hunt of SF who builds PWM circuits for lightbulbs makes ones for 10A and beyond. I would assume that means it is possible to get that fast of a rise time from good cells.

I'm trying to get 100W out of a PWM circuit driven by a bank of 1.8Ah high-current nimh. I'll let you know how that goes. /ubbthreads/images/graemlins/ooo.gif

Wilkey
 

CM

Flashlight Enthusiast
Joined
Sep 11, 2002
Messages
3,454
Location
Mesa, AZ
PWM is suitable for light bulbs (like the LVR and A2 circuits) but not really the way to go with LED's. LED's are current mode devices and should be biased as such. The Vf of a LED varies with temperature and if one is overdriven and not current biased, it will go into thermal runaway. If you do use PWM, pay attention to the pulsed current rating. You can probably go higher than that but wouldn't push it too far. A cap is generally not used with PWM also since you're relying on the on/off duty cycle to drive the bulb to an "average" value. Also PWM is used for the highest power efficiency and I would imagine a cap would diminish the efficiency somewhat since real caps are lossy.
 

krutzbeuazen

Newly Enlightened
Joined
Jan 31, 2004
Messages
69
Location
germany
Ginseng:
oh, wow! that sounds like a nice project! obviously more like a regular bulb or motor and no semiconductor.. hehe

CM:
actually i will use a FET circuit for the LEDs (several FETs in that circuit, to have a smaller number of LEDs on one current supply). i will then, if possible at all, use a PWM between the cells and the FET circuit. so i have no problems with pulsed high voltage or anything, all regulated by the FET.
again, only if NIMH are capable of these high loads at these short timings.

thank you all for your hints!

manuel
 

INRETECH

Flashlight Enthusiast
Joined
Aug 22, 2002
Messages
1,318
Location
HILLSBORO, OR
If you are using PAR FETs to drive your loads, make sure you have very small resistors in series with the GATEs of the FETs to avoid FET "Ping-Pong"
 
Top