shark low output problem

[gillestugan]

My shark gives only 0,56A at maximum. It is regulated and stable at different input voltages. I can adjust the output with the potentiometer, but the board only gives 0,56A at maximum. It's the old green version and I have mounted an external 20k pot.

Does anyone have an idea of what can be wrong?

 

[LED Zeppelin]

Have you verified the actual resistance of the pot? Some I have measured are 17k, but even 16k falls within the 20% manufaccturing tolerance.

 

[dat2zip]

You didn't mention your battery configuration or the number of LEDs.

If the number of LEDs exceed the max vout which is around 22V for the regular Shark it could be going into voltage regulation and not able to go higher to reach the full 1A of regulation.

That could happen with 5+ number of LEDs in the string depending on the Vf of the LEDs.

There may be resistance in the switch, tail or Power supply limiting the current that can be drawn.

Verify R5 is on the board. There were a few boards that R5 was either not loaded or it came off when the input wires were soldered on. It's too close to the input connection and I have accidentally removed R5 when wiring up the Shark and Blue Shark.

Wayne

 

[gillestugan]

Thank you for your answers! I have made some more measurements.

First some more info of my setup:
The shark is cooled by a copper plate from the backside of the board.
I have the shark connected to 5V from a PC powersupply (max 16A). I have put a 100uF capacitor with low impedance on the input, close to the shark.
On the output 2 leds in series. Vf=6,6V@0,55A.
I measure I-in, I-out and V-out simultaneously.

When I turn the pot from low to high, the output current first increases relatively linear up to 0,45A, then It starts to flatten out to 0,56A
The input current continues to rise though, and is 3A when pot fully turned.
The Shark eats all extra power fed to the board. Fortunately the heatsink works well.
I find this behaviour strange.

I have measured the resistance of the pot and it is 18.9k.

I also put the board on top of a flashlight to see if any of the tracks were broken, or components missing.

I'm not willing to test the components on the board individually, as they are so hard to solder with the iron. Even if I use my smallest tip...

 

[dat2zip]

Can you perform one test for me.

Disconnect the wiper of the trip pot and short the wiper pad on the board to the upper pad similar to the mod when putting on the Remora board.

This should apply full power control to the Shark and you should get very close to the 1A to the LED.

Your configuration should be fine.

Something doesn't sound right. Your configuration should draw less than 2A when regulating 1A to the LEDs.

Another test would be to measure the temperature of the converter IC under full power. It could be there is too thick of a epoxy layer between the converter board and the heatsink. This would be even worse if the Shark is of the old style and doesn't have the copper C on the bottom of the board.

Did the Shark have a copper C on the bottomside?

If that doesn't give you 1A to the LED then it there might be something wrong with the Shark.

Wayne

 

[gillestugan]

The shark is 1 1/2 years old and didn't come with the copper C on the backside.
I bought it for a project, but choose a buck solution in the end. (Made the mistake of choosing driver before battery)

When I short the wiper pad to the upper pad I still only get Iout 0.56A and Vout 6.6V. I-in is 3,05A.

With 3A in, the IC gets very hot. I have not driven it at this current for more than 10 sec as it probably woud gett way too hot. When I use my finger as a probe the top of the IC gets warm in about 3 sec, hot in 5 and burning hot in 7-8.

So the IC is probably damaged in some way and there is probably nothing I can do. A pity as it is a very nice driver.

Thanks a lot for your help.
/Anders

 

[gillestugan]

I have found the problem. Sorry for taking your time Wayne, it was not caused by the shark.
There was something wrong with the meter I used for measuring the output current. It has way too high internal resistance, (even if rated 10A). Unfortunately I was measuring the Vout AFTER the faulty meter, so I didn't see the Vout from the cirquit bottom out at 22V, only the 6.6V over the LEDs.
Stupid. And one extra sorry for you, as I no longer have to buy a new Shark.
/Anders

 

[dat2zip]

That's great news. I'm glad you were able to come to a resolution. The Shark has been a very stable product and I worry when someone is not able to get proper functionality.

The new Blue Shark with the copper C moves the heat from the IC to the copper C very efficiently and allows the IC to run at cooler temperatures and more efficiently by keeping it cooler.

Thanks for the update.

Wayne