Need (lots of) help on DIY light

[Packhorse]

I think you will need to have the driver after the switch. So it will have to be in the light head. But that is simple enough just bore out the heat sink and mount them inside. 5 will only be about 20mm high. I did say 5 before not 7. 5 will give you 7 amps. I dont think there is much benefit running more than 7 amps. Maybe even go with 4 at 5.6 amp. Still plenty of lumen and less heat and more run time.

You could have 2 rows of 3 cells then 2 cells on their side for a 8 cell pack that is shorter than a 9 cell pack.

However, to run your double LED, you'll need twice as many boards (12 or 14 total)

No you dont. there is a tricky way to wire AMC 7135 drivers for twin LED setups.
You will have to up the voltage though which is a good thing.
Run a 2P7S pack giving 8.4 volt. or just a 7S pack
Check out this pic. Ignore the paralleled LEDs. The AMC7135 should have a total of 4 or 5 four chip boards. You could have 1 board wired to one stage of the switch for low then switch in the next 3/4 for high.

 

[bradsab]

I think you will need to have the driver after the switch. So it will have to be in the light head. But that is simple enough just bore out the heat sink and mount them inside. 5 will only be about 20mm high. I did say 5 before not 7. 5 will give you 7 amps. I dont think there is much benefit running more than 7 amps. Maybe even go with 4 at 5.6 amp. Still plenty of lumen and less heat and more run time.

You could have 2 rows of 3 cells then 2 cells on their side for a 8 cell pack that is shorter than a 9 cell pack.



No you dont. there is a tricky way to wire AMC 7135 drivers for twin LED setups.
You will have to up the voltage though which is a good thing.
Run a 2P7S pack giving 8.4 volt. or just a 7S pack
Check out this pic. Ignore the paralleled LEDs. The AMC7135 should have a total of 4 or 5 four chip boards. You could have 1 board wired to one stage of the switch for low then switch in the next 3/4 for high.

I think I'm going to have to order the drivers and have them in hand before I try to figure this out. I've never even seen a driver, and the pics on DX don't help much for someone who does't know what they're doing. I do want to regulate the current, though, after playing around a little today. 3s CTA 12,000mAh batteries fresh out of the charger showed 4.24v on the Fluke. Hooked to one SST-90, it dropped to 3.45. I suppose having 3s3p wouldn't drop so much, but I'd rather do the 4s3p & drivers. I tried to check current, but I don't have a clamp, and running the test thru the meter made the LED noticably dimmer.

 

[DIWdiver]

Brad -

Two LEDs at 7A each in a small light head WILL fail if run very long in air. I've heard of other equipment in a dive bag being damaged by a light that turned on accidentally. There are ways to prevent overheating electronically, but they involve sensing temperature and shutting down the driver. With the driver in a separate package, you'd need some additional conductors to connect the temp sensor(s) to the drivers. However, a mechanical thermal switch like this B12 Thermal Cutouts from Cantherm should work to protect the LEDs, while being small enough to fit in the head. I found this one in about 1 minute. I'm sure there are many others. Okay, so I spent a few more minutes and found if you go to DigiKey Corp. | Electronic Components Distributor | United States Home Page and search on 'thermostat' you will find a large selection in stock.

18 ga. is okay for 14A. You will loose a little voltage, which is a big deal if direct driving, but not so much with CC. You would be fine to put 2x7A drivers in the canister and use 18 ga. 3-wire cable, with switches and thermal cutouts in the head.

Now for those drivers...

D

 

[DIWdiver]

Now for those drivers...

Unfortunately if you run a 7S pack, you can't run the single LED very well using these drivers. The recommended max input voltage is 4.5V. Packhorse's suggestion cleverly gets around this by dropping the battery voltage into the acceptable range by using an LED in series with the driver but with only one LED this doesn't work very well. Even if they survive the massive overvoltage, you'd be using just as much power to drive one LED as two, and the extra all goes into the driver as heat. Can you spell S_M_O_K_E ?

Also, you need to be careful if you want to stack several of these drivers on top of each other. With either the 4s or 7s setup, they will dissipate nearly 2 watts each, and will quickly overheat unless there's somewhere for the heat to go. There needs to be a heat path out of the center of the stack (actually, you need a path from each board in the stack). Packhorse, what do you recommend?

Is there any possibility of the top or bottom of the canister being aluminum? I could easily build you a prototype dual 7A high/low driver for the 4s pack, but it would need to bolt to a substantial heatsink, and it wouldn't be as compact (or pretty) as the commercial ones.

D

 

[bradsab]

So the AMC7135 1400mA drivers from DX won't work? Shoot, I thought I had a possible solution drawn in the attched .pdf file. I didn't show any of the negative wiring for clarity.

I can make the lid of the cannister from aluminum. The PVC I have for it is .75" thick; that's a hunk of aluminum I don't have, but I can scrounge some up somewhere.

 

[Packhorse]

Pot the driver boards in a thermal conductive potting. Then slip them into an aluminum tube the same ID as the OD of the driver. Should work fine using the twin LED serial method I posted above.
It will also work with the setup you have on the PDF but I think that is not the best way to do it. You need twice the drivers and you are dealing with twice the current.

I also think you are over complicating things with 3 SST-90's. If you want a spot light have a separate light.

You could just forget about the AMC 7135s and wait for a decent driver to come out. There will probably be one before you finish the build.

I personally would avoid putting the driver in the cannister. I have a very short Maglite build going on at the mo with the driver in the head. I use a constant voltage driver with a PTC thermisitor giving thermal feed back.
When I start it up cold it runs at 8 amp and as it heats up the voltage drops and so does the current. If I leave it running out of water the head gets to aboout45-50 deg in 5 minutes. Current drops to about 3 amps. In water I think the current will sit at about 6 amps.

the driver I use is a LDO10C. takes an input of upto 13.8 volt and is about 90% efficient. There is a good post on CPF about it but I cant find it at the mo.

 

[bradsab]

Pot the driver boards in a thermal conductive potting. Then slip them into an aluminum tube the same ID as the OD of the driver. Should work fine using the twin LED serial method I posted above.
It will also work with the setup you have on the PDF but I think that is not the best way to do it. You need twice the drivers and you are dealing with twice the current.

I also think you are over complicating things with 3 SST-90's. If you want a spot light have a separate light.

Yeah, I see that now. Switching from one to two LEDs is screwing it up. But the whole purpose of this build is to put it all in one package. I suppose I could add another spot LED, but my old light is 800 lumens, and that was enough. A single SST-90 will give me more than that.

You could just forget about the AMC 7135s and wait for a decent driver to come out. There will probably be one before you finish the build.

I think that's a good idea. The cannister is the easiest for me to build, I can do one in an evening. I think I'll get working on the head and leave as much room as I can for a driver. If nothing becomes available before the trip, I'll just build a 3s3p can and direct drive it for now.

I personally would avoid putting the driver in the cannister. I have a very short Maglite build going on at the mo with the driver in the head. I use a constant voltage driver with a PTC thermisitor giving thermal feed back.
When I start it up cold it runs at 8 amp and as it heats up the voltage drops and so does the current. If I leave it running out of water the head gets to aboout45-50 deg in 5 minutes. Current drops to about 3 amps. In water I think the current will sit at about 6 amps.

the driver I use is a LDO10C. takes an input of upto 13.8 volt and is about 90% efficient. There is a good post on CPF about it but I cant find it at the mo.

Packhorse, looking at your lights I see you have some with aspheric lenses that have a very defined hot spot. Every light I've used has had some spill. I kind of like the spill to light up a broader area at night so maybe something out there can catch my eye. How is it diving with the crisp spot of an aspheric? Is it pretty dark outside the hotspot? All my diving is in the clear waters of the Carribbean.

 

[Packhorse]

I dont do alot of night dives. Infact probably 1 % would be pretty accurate.

But I would agree my aspheric based lights are probably not the best for night. There are ways to compensate for this though,. I did a build with 3 aspherics and a small reflector for the 4th LED. Worked out great.
My W300 based can light is well suited to night dives too.
But if you dive in clear water then a wider angle aspheric may be best. One large even spot of light.

 

[bradsab]

New question:
I redesigned my canister to hold 12 Ni-MH D cells. I have three options, one being 3s4p for 3.6v direct drive if I can't get drivers in time. Option two is 6s2p for 7.2v, option three is all 12 in a series for 14.2v. I'm building this so the batteries can be easily removed and put in my Maha charger, and I can quickly change between the three options. There are a couple of drivers being developed on CPF, both will take higher voltage than my max 14.2v. Does 14.2v have any advantage over 7.2v? Does DC behave the same as AC; i.e. higher volts= lower amps= lighter wire? Does battery life differ between the two?

 

[DIWdiver]

New question:
I redesigned my canister to hold 12 Ni-MH D cells. I have three options, one being 3s4p for 3.6v direct drive if I can't get drivers in time. Option two is 6s2p for 7.2v, option three is all 12 in a series for 14.2v. I'm building this so the batteries can be easily removed and put in my Maha charger, and I can quickly change between the three options. There are a couple of drivers being developed on CPF, both will take higher voltage than my max 14.2v. Does 14.2v have any advantage over 7.2v? Does DC behave the same as AC; i.e. higher volts= lower amps= lighter wire? Does battery life differ between the two?

Yes, higher voltage means lower current = lighter wire (assuming you have proper switchmode drivers, definitely NOT with linear drivers). Nominally, 12s gives twice the voltage and half the current of 6s. The drain on each cell is identical, so battery life should be identical. The wire losses are lower, so you'd think that's an improvement...

But it ain't necessarily so. It turns out that most drivers are most efficient if the input and output voltages are close. This is really important in linear drivers like the AMC7135 based ones, where efficiency = Vout/Vin. It's less important in switching drivers, but still nothing to sneer at. Doubling the input voltage decreases wire losses, but increases driver losses. Since wire is cheap compared to drivers, the rule of thumb is to optimize for the driver, which for most drivers means keeping Vin/Vout as close to 1 as possible, without violating min/max input voltages.

There are times when you want to break the rule of thumb. To know when to do that requires thorough understanding of the tradeoffs. That means knowing not only the wire losses, but the driver efficiencies at the appropriate operating points. Since this is information you probably don't have access to, I'd go with the rule of thumb. Do you have an option for 4s3p configuration? That's the config that follows the rule, for most drivers.

In a week or so I might be able to get you two prototype 7A drivers. See this link if you're interested High current (SST-90) driver - CandlePowerForums It would work best with 4s configuration.

D