rhd
0
- Joined
- Dec 7, 2010
- Messages
- 8,475
- Points
- 0
I'll nutshell one of the big takeaways: A 1085 based LDO DDL driver is not sufficient for 2-cell driving of a high current 12x 405nm diode.
Here's the story:
A few days ago I listed this laser for sale in the BST section:
http://laserpointerforums.com/f39/no-longer-fs-thin-silver-600mw-405nm-12x-violet-laser-68011.html
It was a laser that I had built for my own collection 4 or 5 months ago. I took some photos, gave it a few LPMings, and posted the sales thread.
With a BLUE coated standard 3-element glass lens, it was LPMing at 600mW. That seemed pretty par for the course. There was some interest, and a few potential buyers had PM'd me, so I figured it was time for a bit more of a thorough testing. Generally, if I'm selling something that I haven't used in a while, or built some time previously, I like to give the item an updated test and thorough once-over before I actually do anything with it.
This is where the strange story begins -
As part of my updated testing, I took measurements of the current at the tailcap. I knew that the driver I had used was a custom 1085-based LDO driver, and that I had probably set this laser at a fairly ambitious 612.5 mA. But when I took a tailcap measurement, there was only ~400mA of current flowing. I was using good AW IMR 14500 cells, so what was going on?
I took the modular driver/diode capsule out, and connected it to my bench supply. I set that voltage at 8.2V to approximate two fully charged lithium ions, and I increased the current. What I observed, was that the current still topped out at ~400mA. This didn't immediately make sense to me. I was 95% certain that I hadn't built this driver at 400mA. So what was the issue?
In a nutshell: 1085 based drivers are not appropriate for two-cell, high current, 12x 405nm builds. The dropout on a 1085 based driver is not as low as we often think it is!
Sure enough, when I increased the bench supply's voltage to slightly above 9V, the current maxed out at the ~612.5 mA that I had expected this driver to be regulating it at. In other words, at just the voltage of 2x lithium ion cells, there wasn't enough voltage left for the diode, after the driver's dropout, to allow the diode to hit the 612.5mA on it's IV curve that I had been targeting.
Ok, so what about the previous 600mW power output reading?
Turns out, that power output reading was good. At 400mA, I had been correctly reading ~600mW of output with a standard aixiz glass lens. This left me with three questions:
1) What would the power output be at 612.5mA with an Aixiz 3-element lens?
2) What would the power output be at 612.5mA with a 405-G-1 lens?
3) Could the 1085 based driver actually sustain the power dissipation of supplying the diode from a source of 3x lithium ions, or would it shut down before I could get a power reading?
The first obstacle was finding a way to fit 3x lithium ions in a host made for 2x 14500s. That wasn't an easy task. There aren't a lot of battery options that are as thin as a 14500, yet shorter. But there was one combination that would work, and I just so happened to have a few on hand:
AW 14250 Unprotected Rechargeable Lithium Battery
14250s are half the size of a 14500. Using three, as well as some magnetic spacers, I was able to load up the host with 3x cells. Those 14250s are tiny, and wouldn't be ideal for a high current build. But relatively speaking, this wasn't really a high current build. At 300mAh, I was really only tasking the cells with a 2C discharge rate to supply 612.5mA into my linear driver.
So I carefully setup the host with 3x 14250s, and started testing. The figures?
- With a standard 3-element glass lens: 800mW
- With a 405-G-1 lens: 935mW
In terms of the 1085 dissipating the extra heat, it was just fine. I use TO-263 1085s, which can handle a lot more heat dissipation than the 1085s used on some pre-built drivers. Even though the IC was dropping 2x the voltage that it might otherwise in a 445 setup, at 612.5mA, it was only contending with 1/3rd of the current of a 1.8A 445 build. So watts for watts, even with 3 cells, I was ahead of the game here.
The 405-G-1 was the second test I ran, so the cells were slightly depleted. I'm going to top them up and LPM again, and I'm thinking I'll hit 950.
Here's the story:
A few days ago I listed this laser for sale in the BST section:
http://laserpointerforums.com/f39/no-longer-fs-thin-silver-600mw-405nm-12x-violet-laser-68011.html
It was a laser that I had built for my own collection 4 or 5 months ago. I took some photos, gave it a few LPMings, and posted the sales thread.
With a BLUE coated standard 3-element glass lens, it was LPMing at 600mW. That seemed pretty par for the course. There was some interest, and a few potential buyers had PM'd me, so I figured it was time for a bit more of a thorough testing. Generally, if I'm selling something that I haven't used in a while, or built some time previously, I like to give the item an updated test and thorough once-over before I actually do anything with it.
This is where the strange story begins -
As part of my updated testing, I took measurements of the current at the tailcap. I knew that the driver I had used was a custom 1085-based LDO driver, and that I had probably set this laser at a fairly ambitious 612.5 mA. But when I took a tailcap measurement, there was only ~400mA of current flowing. I was using good AW IMR 14500 cells, so what was going on?
I took the modular driver/diode capsule out, and connected it to my bench supply. I set that voltage at 8.2V to approximate two fully charged lithium ions, and I increased the current. What I observed, was that the current still topped out at ~400mA. This didn't immediately make sense to me. I was 95% certain that I hadn't built this driver at 400mA. So what was the issue?
In a nutshell: 1085 based drivers are not appropriate for two-cell, high current, 12x 405nm builds. The dropout on a 1085 based driver is not as low as we often think it is!
Sure enough, when I increased the bench supply's voltage to slightly above 9V, the current maxed out at the ~612.5 mA that I had expected this driver to be regulating it at. In other words, at just the voltage of 2x lithium ion cells, there wasn't enough voltage left for the diode, after the driver's dropout, to allow the diode to hit the 612.5mA on it's IV curve that I had been targeting.
Ok, so what about the previous 600mW power output reading?
Turns out, that power output reading was good. At 400mA, I had been correctly reading ~600mW of output with a standard aixiz glass lens. This left me with three questions:
1) What would the power output be at 612.5mA with an Aixiz 3-element lens?
2) What would the power output be at 612.5mA with a 405-G-1 lens?
3) Could the 1085 based driver actually sustain the power dissipation of supplying the diode from a source of 3x lithium ions, or would it shut down before I could get a power reading?
The first obstacle was finding a way to fit 3x lithium ions in a host made for 2x 14500s. That wasn't an easy task. There aren't a lot of battery options that are as thin as a 14500, yet shorter. But there was one combination that would work, and I just so happened to have a few on hand:
AW 14250 Unprotected Rechargeable Lithium Battery
14250s are half the size of a 14500. Using three, as well as some magnetic spacers, I was able to load up the host with 3x cells. Those 14250s are tiny, and wouldn't be ideal for a high current build. But relatively speaking, this wasn't really a high current build. At 300mAh, I was really only tasking the cells with a 2C discharge rate to supply 612.5mA into my linear driver.
So I carefully setup the host with 3x 14250s, and started testing. The figures?
- With a standard 3-element glass lens: 800mW
- With a 405-G-1 lens: 935mW
In terms of the 1085 dissipating the extra heat, it was just fine. I use TO-263 1085s, which can handle a lot more heat dissipation than the 1085s used on some pre-built drivers. Even though the IC was dropping 2x the voltage that it might otherwise in a 445 setup, at 612.5mA, it was only contending with 1/3rd of the current of a 1.8A 445 build. So watts for watts, even with 3 cells, I was ahead of the game here.
The 405-G-1 was the second test I ran, so the cells were slightly depleted. I'm going to top them up and LPM again, and I'm thinking I'll hit 950.
Last edited:
