Shark Buck

[bluecrow76]

You can use the trim on the Shark and Shark Buck to specify the current to the emitter. The sense resistors set the maximum allowed output current.

Someone correct me if I'm wrong, but it seems the simpler thing to do here would be to remove the LDO (AMC7135) regulators from the 16 mode board and connect the PWM output of the microcontroller to the remora input on the Shark Buck, essentially creating your own remora. You would basically be creating your own remora, since the remora is just a microcontroller with a PWM output.

It seems to me that using the Shark Buck in front of a 7135 stack would also be inefficient in the use of your power supply. By creating your own remora you would be using the more efficient driver on the Shark Buck.

Like I said, someone please chime in if I'm dead wrong on any of this... I am an amateur after all!

As a side note, I just finished a 7135 based 2.8A driver for a Lumapower MRV P7 mod and I'm not to happy with it. It's plenty bright at first but doesn't maintain the peak brightness for long. From the graphs I've seen posted here it almost seems like I would have been better off running direct drive. I am going to be ripping it out and replacing it with a Shark Buck in January after I recover from Christmas!

 

[dat2zip]

bluecrow76, thanks for replying.

The Shark Buck would come before the other driver, that's the whole point, I need to lower the voltage getting to the 16 mode driver because it's optimum input voltage is 5V and the 2 Li Ion batteries provide 8.4V max, 8 AA batteries in series in a holder I found here... http://theledguy.chainreactionweb.com/product_info.php?cPath=48_56_73&products_id=590
Provide 12V max, less using NiMH.

My goal is to have maximum battery versatility I would use my Li Ion batteries normally, but switch to NiMH when the Li Ions could not be charged. I could use 2 of any Li Ion battery that fits, or use two aventrade holders with 14500s in them .

I saw the Assembled Shark Buck with the Remora, I considered it, but I want the other modes offered by the other board.

I think it would work I just need to know what the output voltage of the 3A Shark Buck is with 8.4V and 12V inputs respectively.

Also is the output of the 3A Shark Buck adjustable down to 2.4A using the trim pot, or do I need to swap out sense resistors? Do I even need to lower the output current to prevent overloading the 7135 boards or will the other drivers take only what they need?

The Shark Buck is a Constant current regulator. Therefore, if the 7135 doesn't draw the desired current or less than what the trim pot is set the output will rise to battery voltage as the Shark Buck will try to increase the output voltage to maintain regulation. Since it may not get there it will rise all the way till it reaches battery voltage which is no different than hooking the battery directly to the 7135!

What you need is a Voltage regulator and with a voltage regulator be able to dial down the output voltage to something close to the 7135 to reduce the power dissipated by the 7135.

I agree with bluecrow76 that it would probably be a better solution to hook the PWM of the 7135 to the Shark Buck input to get your multilevels.

Even simpler would be to use a rotary switch and use discrete resistors for the number of output levels you want.

The formula to compute the output level resistor is:

IMax = Isense max (1A, 2A, 3A for sharkbuck, 1A for BlueShark)

external_resistor
______________ * IMax = LED_Current.
20k


Basically 20K is 100%. 1/2 of 20K is 10K and that would give you 1/2 Imax. 1/10 20K would give you 1/10 IMax, etc.

If you wanted 1/20, 1/5, 1/2, 3/4 the values would be:

1/10 of 20K

1/20 of 20K

1/5 of 20k

1/2 of 20k

3/4 of 20k

100% would be any value greater than 20K

That would give you 5 plus full or 6 levels of control. With the rotary switch there is no UI and the knob tells you exactly what level you at before you turn it on and you can crank the knob in one motion to any other level. Ideal for places like a bike where a UI only hinders the rider.

Wayne

 

[Shsesc]

Under what conditions would the 7135 1400mA and 16 mode driver not draw 2.4A? Can the trim pot on the 3A Shark Buck be turned down to 2.4A?

How would I use only the PWM part of the 16 mode controller? Would I still need a voltage regulator to keep the input under 5V? Where would I find such a voltage regulator (2.4A 5V)?

 

[dat2zip]

Under what conditions would the 7135 1400mA and 16 mode driver not draw 2.4A? Can the trim pot on the 3A Shark Buck be turned down to 2.4A?

How would I use only the PWM part of the 16 mode controller? Would I still need a voltage regulator to keep the input under 5V? Where would I find such a voltage regulator (2.4A 5V)?

Assume you dial the Shark Buck down to 2.4A. If you are not drawing 2.4A or more the Shark Buck output will float to battery voltage which is not what you want if I understand you correctly.

Even if... Since you want 16 levels the other 15 levels will probably be less than 2.4A and all of the other 15 levels the Shark Buck will float to full battery voltage. I don't think that is what you want.

I believe you really need a Buck voltage converter that will step down to say 5V or something that the other board can handle.

I am not familiar with the the other board to advise you how to power it. I would suggest using the resistor method and rotary switch if you can go that route. It would be the easiest method to wire up. If you need mode switching via the power switch the resistor method won't work.

You can probably find a good buck regulator over at National.com , linear.com or ti.com and many of them you can buy an eval board which might work for you if the size will fit your project.

Wayne

 

[1-3-2-4]

Help!!

So my Shark Buck arrived today.. I wired in the P7's in aeries and turned on the power LED's worked so I I set everything in my fixture and went to snap some pictures of the output and now the shark buck is not working at all...

I also noticed the LED's are flickering but not a bright flicker.

 

[dat2zip]

Re: Help!!

So my Shark Buck arrived today.. I wired in the P7's in aeries and turned on the power LED's worked so I I set everything in my fixture and went to snap some pictures of the output and now the shark buck is not working at all...

I also noticed the LED's are flickering but not a bright flicker.

Did you mount the Shark Buck to a heatsink. The heatsink is required or the Shark Buck will burn itself up.

Wayne

 

[1-3-2-4]

Re: Help!!

Did you mount the Shark Buck to a heatsink. The heatsink is required or the Shark Buck will burn itself up.

Wayne

no because it was never on longer then 5 seconds and the first time I turned it off I made sure to feel how hot the shark buck was and it was not hot at all.

 

[dat2zip]

The Shark Buck requires an external heatsink of some form. It will not run without it.

Wayne

 

[1-3-2-4]

so what are you saying it will work again with a heatsink attached? I think someone asked you if it was safe to run for just testing.

What I don't get is the shark still is outputting power... more then enough to drive the two P7's like 11V worth.

also checked to make sure the LED's were not blown and they both work with my power puck 1A.

 

[Shsesc]

Would that use an external input resistor?


Where would I put the rotary switch on a Maglite? I mean, I can see the simplicity of the design on paper, but physically I'm not sure there is room.

 

[dat2zip]

so what are you saying it will work again with a heatsink attached? I think someone asked you if it was safe to run for just testing.

What I don't get is the shark still is outputting power... more then enough to drive the two P7's like 11V worth.

also checked to make sure the LED's were not blown and they both work with my power puck 1A.

For more than a few seconds the heatsink is required if you are driving the LEDs with more than a few hundred millamps. For quick verification of board a bench test can be performed. I would setup the bench setup and when powered is applied verify if you have a trim pot version that you can turn it down. Once you get it turned down to a low level this is sustainable without a heatsink.

When it flickers are you not moving anything or does it only flicker when you move wires or something?

The LED+ and LED- for the Shark Buck can not touch the copper C or GND.

Always double check and triple check your wiring before turning on for the first time. Use an ohmeter if possible and make sure the LED+ is not shorted to LED- or GND or any other connections and do the same for LED-, battery plus and battery minus.

One other way to test the Shark is to measure the resistance of the wiper of the trim pot to GND and turn the trim pot so this resistance is nearly zero. The wiper feedback controls the output brightness and when the wiper is close to GND means it will put out the least amount of current. When the wiper resistance is high (20K) to GND then the control voltage will try to regulate maximum or higher LED current. Adjusting the wiper prior to turning on will ensure the regulator will not generate a lot of heat and you won't have to worry as much about heat issues with the converter or LED.

Once you have the conveter powered up and the output is low you can turn the trim pot and verify the brightness varies as you turn the trim pot. If this is successful then it is most likely a good converter board and your wiring is correct.

If there is no light, excess battery current turn off the power and double check your wiring.

Heat from the converter board is not a sign of poor efficiency. It is a direct result of not being 100% efficient. It's impossible to have a 100% efficient converter. For a simple analogy lets say the converter is 85% efficient. That's 15% loss across the converter. If you are driving 1 P7 at 3A then the power to the LED is approximately 3.3V @ 3A or 9.9W. Let's call this 10W of LED power. That means the converter loss is 15% of that at 85% efficiency or 10W * 0.15 = 1.5W. That's 1.5W the converter must dissipate. If the converter is less efficient like 78% which I believe is the lowest I have measured then the power loss on the converter is more than what was just computed above.

If you have two P7 then the power is twice that or ~20W and the converter power dissipation is now 1.5W * 2 or 3W of power. The board in free air space is good for 0.5W-0.75 or so without a heatsink.

Hope this helps.

Wayne

 

[dat2zip]

Would that use an external input resistor?


Where would I put the rotary switch on a Maglite? I mean, I can see the simplicity of the design on paper, but physically I'm not sure there is room.

The resistors replace the trim pot and have no major current flowing through them.

If you are using a hotlips style heatsink and the converter is mounted to the backside you probably have removed the long tube that the bulb screwed into. That cavity is fairly large and can host the heatsink and still have space to drill a hole in the wall and mount a switch, potentiometer or something else.

Try searching Mouser.com or digikey.com for rotary switches. There are many from small to large. Smaller ones generally will cost more.

Here's one I think I used on the Irakongi board over at digikey. The pin stops allow you make it any number of positions you want.

Try to get a make before break rotary switch. This type of switch will connect the next position before it releases the current one and will provide glitch switch operation. Getting a break before make will mean the switch as it changes will break or be open before the next connection. The open state of the switch may cause the converter to go full brightness or full dim or anywhere in between and could cause the light to glitch when switching positions.

Wayne

 

[1-3-2-4]

For more than a few seconds the heatsink is required if you are driving the LEDs with more than a few hundred millamps. For quick verification of board a bench test can be performed. I would setup the bench setup and when powered is applied verify if you have a trim pot version that you can turn it down. Once you get it turned down to a low level this is sustainable without a heatsink.

When it flickers are you not moving anything or does it only flicker when you move wires or something?

The LED+ and LED- for the Shark Buck can not touch the copper C or GND.

Always double check and triple check your wiring before turning on for the first time. Use an ohmeter if possible and make sure the LED+ is not shorted to LED- or GND or any other connections and do the same for LED-, battery plus and battery minus.

One other way to test the Shark is to measure the resistance of the wiper of the trim pot to GND and turn the trim pot so this resistance is nearly zero. The wiper feedback controls the output brightness and when the wiper is close to GND means it will put out the least amount of current. When the wiper resistance is high (20K) to GND then the control voltage will try to regulate maximum or higher LED current. Adjusting the wiper prior to turning on will ensure the regulator will not generate a lot of heat and you won't have to worry as much about heat issues with the converter or LED.

Once you have the conveter powered up and the output is low you can turn the trim pot and verify the brightness varies as you turn the trim pot. If this is successful then it is most likely a good converter board and your wiring is correct.

If there is no light, excess battery current turn off the power and double check your wiring.

Heat from the converter board is not a sign of poor efficiency. It is a direct result of not being 100% efficient. It's impossible to have a 100% efficient converter. For a simple analogy lets say the converter is 85% efficient. That's 15% loss across the converter. If you are driving 1 P7 at 3A then the power to the LED is approximately 3.3V @ 3A or 9.9W. Let's call this 10W of LED power. That means the converter loss is 15% of that at 85% efficiency or 10W * 0.15 = 1.5W. That's 1.5W the converter must dissipate. If the converter is less efficient like 78% which I believe is the lowest I have measured then the power loss on the converter is more than what was just computed above.

If you have two P7 then the power is twice that or ~20W and the converter power dissipation is now 1.5W * 2 or 3W of power. The board in free air space is good for 0.5W-0.75 or so without a heatsink.

Hope this helps.

Wayne

only thing I think might of happened was LED + might of touched the Copper C however with it corrected the board is not providing the right current anymore.. my guess is that it got fried I was running 2 P7's in series.

I did test the trim pot yesterday and it was showing the correct numbers.

also the LED will only flicker if one of the two things happen.. I move the LED + lead or what I found out yesterday trying to read the voltage by touching the LED + with my probe each time the LED would flicker a tiny bit but never came on.

I have about 3 projects I'm working on now so I can't really buy another shark to test.

Also I'm a little concerned.. not sure how many others are using this for an auto application but where this shark will be mounted will be on the rear gate of a Subaru legacy so it's going to be slammed shut a few times over it's lifetime and that might ask for issues later on.

 

[Shsesc]

The resistors replace the trim pot and have no major current flowing through them.

If you are using a hotlips style heatsink and the converter is mounted to the backside you probably have removed the long tube that the bulb screwed into. That cavity is fairly large and can host the heatsink and still have space to drill a hole in the wall and mount a switch, potentiometer or something else.

Would a 20k pot be more or less practical than a switch? It seems like it would be even simpler than a switch. Would I just replace the trim pot with a full size potentiometer?

 

[dat2zip]

Shsesc,

Each method is a matter of personal preference. Some people would prefer one over the other.

What type of light it is and how you plan on using it may have a overall effect on which method will suit you better.

The rotary switch may be more applicable on a bicycle knowing fixed levels and how many clicks can be done without looking at the control mechanism and having a fixed number of light levels may be important in that situation.

A pot may be more applicable in a handheld light application. In actual usage the light output requirements may vary from near off to full on and everything in between.

My personal experience is when I use a pot I tend to drift towards more light since I never really know what current is being driven to the LED. With a three position switch for low, med, high I can do a rough calculation from the battery amp/hr rating to determine for each setting what the approximate runtime is. With this I can then moderate how much high I use and can somewhat predict when I will need to change the batteries or put them on the charger. I tend to try to make my Makita lights such that medium is adequate for 90% of the tasks I plan on using the light for and use low and high occasionally as needed.

Wayne

 

[Ms_SS]

Luckily only nine 1oz SHARK Bucks were shipped. Please return them to the Shoppe for a swap out with the 2oz Shark Buck. Orders 15369, 15371, 15372, 15380 may be affected. Send an email please. Thank you.

RECALL

Some of you (15372 and 15380) still have not returned the 1oz copper Shark Bucks. Please do so ASAP. We do not want them circulated with the 2 oz batch.

 

[Greg G]

Mine is boxed and ready to mail. I'm slammed here at work and with a sick family member, and have not had time to go to the post office.

 

[Greg G]

Thanks Wayne and Cindy. I got the replacement Shark Buck today in the mail. I was so surprised, because I haven't mailed the original back yet.

The original is still in the box wrapped in bubble wrap as you guys sent it. I never unwrapped it. I'll make sure and get to the post office in a day or two.

Thanks for the great service! :twothumbs

 

[gillestugan]

What is the lowest output you can dim the shark buck to with a pot?

Im looking for a low mode of about 30mA and can replace parts on the board if needed.

 

[DaFABRICATA]

I wanted to let you know the replacements came and I will be sending back the old ones this week.

Thank again!! For EVERYTHING!!