Hi,
I just build a K8071 Power Led Driver. Apparently it works fine but the IRF9520 transitor (T1) becomes too hot, above 100°C. I use this driver to power a single 5 W led with a 12 V DC supply. BTW, as my power source is DC, I removed the rectifier bridge.
Is there anyone experiencing this problem? To cure it for now I put an heathsink but it’s not very smart…
Kind regards,
What is the led current ?
Did you modify the circuit for use with a 5W led ?
Hi the LED current is 700 mA, I didn’t modify the circuit. By the way I use a Prolight Power LED warm white 5 Watt. The nominal current is 700 mA @ 7.5 V. I guess this LED should be ok?
You can try putting an 100n capacitor over R2.
Hi,
I have try but still T1 is over heathing…
Sorry, but I’m unable to solve this problem by mail.
We can check out your assembly if you return it to:
Velleman Projects Tech. Dept.
Legen Heirweg 33
9890 Gavere
Belgium
Repair is always less than price of a new kit (shipping not incl.)
[quote=“dbodart”]Hi,
I just build a K8071 Power Led Driver. Apparently it works fine but the IRF9520 transitor (T1) becomes too hot, above 100°C. I use this driver to power a single 5 W led with a 12 V DC supply. BTW, as my power source is DC, I removed the rectifier bridge.
Is there anyone experiencing this problem? To cure it for now I put an heathsink but it’s not very smart…
Kind regards,[/quote]
it is not neccesary to remove the rect. bridge, i use this kit on 12 v.DC, and i even have a bipolar input, i don’t need to care about the polarity. I use a recycled powerled spot, i removed the defective print and i connected the 2 wires so i have 3x1 watt in series, perfectly cooled, and no excessive heat emitted from the board; originally the kit is designed for 2x3 watts or 4x1 watt, both in series, placing back the rect. bridge may solve the heat problem, but i didn’t tested or even found a powerled from 5 watts, can’t help you with this issue.
[quote=“dbodart”]Hi,
I just build a K8071 Power Led Driver. Apparently it works fine but the IRF9520 transitor (T1) becomes too hot, above 100°C. I use this driver to power a single 5 W led with a 12 V DC supply. BTW, as my power source is DC, I removed the rectifier bridge.
Is there anyone experiencing this problem? To cure it for now I put an heathsink but it’s not very smart…
Kind regards,[/quote]
Is there anykind of regulator board attached on your prolight powerled, if yes, remove it until you obtain the powerled pcb; without any other part on it, only 2 connection points ; vcc and gnd called. Then connect it, but ONE SINGLE powerled runs on 3,7 v. not on 7,5v. ( test this to be sure, even i can make mistakes)
I’ve built a couple of these, and both have this same problem.
In my case, this is because the unit is switching at a rather high frequency (about 400kHz). Although the MOSFET is normally in either an ON or OFF state in this design, each time it switches, it passes through a transition where it’s between on and off and has enough resistance to dissipate significant power. Normally this transition time is so short in a switching regulator that it is an insignificant source of power dissipation. However, it’s happening so frequently that the transition time becomes significant, and hence the power dissipation is significant. On a scope, the square wave has become completely rounded off.
Replacing C3 (68pF) with 10000pF brings the operating frequency down to about 50kHz, and power dissipation in the MOSFET drops dramatically, so it doesn’t need a heatsink. This does increase the triangular wave ripple on the output though, but that probably doesn’t matter for the LEDs.
I would be interested to know what frequency the designer intended it to run at.
Originally, it was designed for sub 100KHz operation.
Thanks for that, so I have it running in the correct range then.
I’ve modified one of these kits slightly to drive 3 x 3W LEDs from 24VDC. To do this, I changed the following:
C6, increased to 35V rating, and reduced value to 220uF so it still fits on the board (and as I’m running from DC, I don’t need it for smoothing out ripple anyway);
VR1 (5V regulator), replaced with different make which is rated up to 30V input;
R4 (comparator output pull-up), increased from 1k to 1k2 to maintain similar max current.
The MOSFET, schottky diode, transistors, and comparator IC all seem adequately rated for 24V operation anyway.
Given I’m using a DC supply, I didn’t bother to fit the bridge rectifier diodes as they would have lost me an extra 0.5-1 watt in heat.
It seems to work fine - a nice little kit and easily adapted. Just ordered another one…
Thanks.
Thanks for your feedback.