Shark Buck

[Greg G]

This driver works great. I finished the light that I bought this driver for. It's a pewter 2C M@g with a custom heatsink I made, a CSWOI P7, DX reflector, and a UCL lens.

The C switch was acting kind of quirky at first, but has settled down.

I'll be ordering a few more of these to keep in stock.

Edit. First Shark Buck mailed back 1/08

 

[cmacclel]

Has anyone had any success running 4x P7's with one of these drivers?

Mac

 

[gav6280]

Can i use the 3A shark buck turned down to output 2800ma to drive 4 Q5's split into 2 parallel strings like this: http://www.gavinwilkinson.com/What driver2.pdf ???

I guessed that a 3 cell battery would be the best option as the voltage input would range between 12.69 to 9. Based on that assumption I am hoping to use this type of battery pack but built into a toblerone shape:

http://cgi.ebay.co.uk/2-Li-Ion-1865...40975QQcmdZViewItemQQ_trksidZp1742.m153.l1262

Any thoughts?

 

[ambientmind]

Just one more question...which side do I attach to the heatsink? Is it the coppee C that needs it or the top side? Also, if it is the C does that need to be electrically isolated from the heatsink or not? Thanks! I just dont want to burn mine up...:green:

 

[dat2zip]

Just one more question...which side do I attach to the heatsink? Is it the coppee C that needs it or the top side? Also, if it is the C does that need to be electrically isolated from the heatsink or not? Thanks! I just dont want to burn mine up...:green:

The copper C is the part that needs to be attached to a heatsink. The Copper C is also electrically connected the Battery minus on the board. If the case is battery minus as well then it is not a problem. If the heatsink will be attached to anything other than battery minus extra care must be taken to provide electrical isolation.

Wayne

 

[ambientmind]

The copper C is the part that needs to be attached to a heatsink. The Copper C is also electrically connected the Battery minus on the board. If the case is battery minus as well then it is not a problem. If the heatsink will be attached to anything other than battery minus extra care must be taken to provide electrical isolation.

Wayne

Thanks for the fast reply wayne!!! :thumbsup:

 

[spencer]

Using the 3A version of this, about what range of output would I get with an external 20k pot? 1mA - 3000mA or is the low higher at like 100mA? I am planning a P7 Mag and want to know how low I can go.

EDIT: Also, on a one turn linear pot, will output immediately begin to drop and rise as I turn? I ask because I had a dimming light that only started to dim at about 1/3 and shut off at 2/3 so I could only adjust with the middle third.

 

[dat2zip]

Using the 3A version of this, about what range of output would I get with an external 20k pot? 1mA - 3000mA or is the low higher at like 100mA? I am planning a P7 Mag and want to know how low I can go.

EDIT: Also, on a one turn linear pot, will output immediately begin to drop and rise as I turn? I ask because I had a dimming light that only started to dim at about 1/3 and shut off at 2/3 so I could only adjust with the middle third.

The pot range is from off or nearly off to full on and adjustment is over the full range of the pot.

Wayne

 

[Techjunkie]

Wayne,

Is the 25v max input the open volt measurement of the batteries, or the actual forward voltage accounting for sag under load?

Here's what I have in mind, please let me know if this is doable or out of range for one shark buck 3A:

I want to drive a 5-series string of MC-E(p) emitters (5S4P). Vf ~17V @2.8A.

To power it, I'd like to use 8s LiFePO4 14500 batteries. Voltage at rest of each cell is 3.35v so the total pack at rest would measure ~26.5V, but under the ~2A load, the pack would settle down to or below the 25V max.

Is that running things too close to the edge, or within spec? Thanks.

 

[dat2zip]

Techjunkie,

25V is from the datasheet as the absolute Maximum input voltage. It is also true that more than likely the IC will run with input voltages higher than the absolute maximum.

Running this close to specification is a risk the end user must accept. I can only recommend running with a margin of a volt or two below absolute maximum to allow the additional switching noise that might be riding on top of the battery input voltage. So, DC voltage plus peak AC voltage should be less than absolute maximum.

In general if it works it will probably continue to work. In your case I don't think the voltage as seen by the converter board will be over 25V under the load situation you have described. But, I'm no expert here and only a fully loaded configuration will determine what voltages and currents exist.

My only caution in your case is to not put the batteries in hot off the charger if the batteries behave like regular li-ion with the off the charger voltage of 4.2V or so and at room temperature resting state of less than that. (letting the batteries rest after charging the battery voltage drops)

Being so close to absolute max may not be a problem when you factor in the small voltage drops across the switch, spring and other electrical paths that are between the battery and converter board.

Wayne

 

[Techjunkie]

Wayne,

Thanks for the well worded reply. I'm considering a modified version of what I've described, and if you don't mind, I'd like to bounce the ideas off of you.

Instead of running 5x MC-E, I'm considering running 4x MC-E and 1x XR-E (The reason being that the 5 small reflectors used here do a poor job of focusing the MC-E. I'm thinking that swapping out one MC-E and replacing it with one XR-E might allow me to fill the MC-E donut with the XR-E hotspot.)

The power supply is still 8x LiFePO4 14500 batteries (which I'm considering charging in the modamag carrier with a 7s LiFePO4 pack charger to keep maximum voltage at or below 24v), and the Shark buck 3A is still the driver of choice.

My dilemma is how best to drive the one XR-E at the same time as the 4P4S MC-E. Off the top of my head, I've come up with:

a) run the XR-E in parallel with the 4P4S string of MC-E off of the Shark Buck 3A's output, using a lesser buck driver to bring the ~14.4v out of the Shark down to the 3.6V in requirement of the XR-E. This would steal some of that 3A output from the MC-E emitter's, but less than 350mA, if I'm correct, driving them at still more than 2650mA.

b) run a separate buck driver just for the XR-E, powered in parallel with the Shark Buck 3A directly from the 16-24V power supply. I'm not so interested in modes or dimming, but if I were, I think I'd be tempted to use a Shark buck 1A for the XR-E and control both the 3A and the 1A with a single pot or Remora... Would that even be possible, or would they not appear to dim to the same levels?

c) better ideas?


Thanks!

 

[dat2zip]

You could use a high voltage buck driver for the single XRE LED. The SOB max voltage rating is 16V so that is not an option. I think there are other buck drivers that would suffice for non multi-level option.

Option B is very doable with two Shark Bucks and one Remora to control both.

Offhand I can't think of any option C.

Wayne

 

[kland1234]

I did some testing and I found the converter needs about 2V of headroom to stay in regulation. So, for this configuration a min of 5.5V is needed to run full power.

I do not know why this is the case and I will look into more when I get a chance.

Wayne

Is this still the case, i.e., 2V of headroom needed? I have a current version 2A shark buck on the way, and will be driving 1-P7, ~3.5V Vf. Trying to make battery choice, and was planning on 1cell LiPo; no workie I figure? Would 2 cell LiPo, 7.4 V be too much for 1-P7?

Thanks Wayne.

 

[thegeek]

Wayne,

I sent a PM about some problems I am having with a Shark Buck, if that isn't the best way to contact you please let me know! Thanks.

 

[acgriff]

What is the voltage output for a shark buck with for (2 18650) 7.4 Volts input. I want to make sure the P7 I bin code I have will work properly. The manufacturer states the Vf is 3.25 to 3.5.

Thanks
Andrew

 

[cmacclel]

What is the voltage output for a shark buck with for (2 18650) 7.4 Volts input. I want to make sure the P7 I bin code I have will work properly. The manufacturer states the Vf is 3.25 to 3.5.

Thanks
Andrew

The Shark Buck limits current. The voltage output varies on every LED. You will be fine with 2 18650's and a single P7.

Mac

 

[thegeek]

Wayne,

I have a shark buck that seems to be malfunctioning. It has ~8V in and the Vf of the led is ~3.5V. I see no change in output when adjusting the pot over it's full range. The output remains constant and very very low. I do measure a difference in the resistance across the legs of the pot, but no change in output. Does this sounds like a faulty board or a wiring issue?

 

[wildstar87]

Hi Wayne,

Do you have an update on this Remora issue with Shark Buck? I'm interested in getting a Shark Buck + Remora, but wanted to know what's going on with this, thanks!

I did some simulations yesterday and discovered why the Remora doesn't PWM throttle to zero like it should.

It is correctable and I believe some small modifications it can be fixed.

The fix on the Shark and Blue Shark I will document soon. For the 10 Shark Bucks I have now they behave like the Shark does.

The production Shark Bucks I will incorporate the filter network onto the board and it will make the modification easier.

I need to see if the fix even works and I will test it and post on the results.

Wayne

 

[dat2zip]

Hi Wayne,

Do you have an update on this Remora issue with Shark Buck? I'm interested in getting a Shark Buck + Remora, but wanted to know what's going on with this, thanks!

Yes, the Remora design works exactly the same on the either Shark since the current implementation doesn't throttle low below 15% of full value. The three level values are ok with high being 95% of full output.

There is no issues as is.

If the Remora code was changed to lower the low then the implementation as is would not go below ~1/10 of full output and that is the way the Remora was implemented. You could use a different uP design and use the filter in the proper way to get full 0 to 100% control.

I have used the Shark buck (adding the additional filter resistor) in an implementation on a Makita Dock prototype. The uP on the base board controlled two Shark Buck converter and I had full control over the output.

The Remora board shape to to oddball and small to register properly with the programming fixture I have created or re-created and have not been able to re-program the Remora boards. The last hope is the use a dip clip from the top side to gain programming access and re-program the processor.

Wayne

 

[wildstar87]

So if I'm understanding this correctly.. The Remora as is does not drive the Shark at 100% on high, but just 95% and the low is 85%? And for it to do this, it would have to be reprogrammed? I guess there is nothing I can do on my end, since I know very basic electronics, but most of what you said there went over my head.. :thinking: I'm curious, could I use a D2Dim/D2Flex to throttle the Shark instead?

Yes, the Remora design works exactly the same on the either Shark since the current implementation doesn't throttle low below 15% of full value. The three level values are ok with high being 95% of full output.

There is no issues as is.

If the Remora code was changed to lower the low then the implementation as is would not go below ~1/10 of full output and that is the way the Remora was implemented. You could use a different uP design and use the filter in the proper way to get full 0 to 100% control.

I have used the Shark buck (adding the additional filter resistor) in an implementation on a Makita Dock prototype. The uP on the base board controlled two Shark Buck converter and I had full control over the output.

The Remora board shape to to oddball and small to register properly with the programming fixture I have created or re-created and have not been able to re-program the Remora boards. The last hope is the use a dip clip from the top side to gain programming access and re-program the processor.

Wayne