1076.9 OTF LUMENS 5.4A SST-90, 2-D Mag build using 7135's

ti-force

Flashlight Enthusiast
Joined
Jan 14, 2009
Messages
1,266
Location
Georgia, U.S.
I really haven't seen anyone using the 7135 drivers to drive the SST-50's or 90's, so I thought I'd share my project with anyone who's interested. WARNING!! PIC HEAVY!!



My original intentions were to build this light to run on 1x26650 from Batteryspace to drive the SST-90 emitter at 5.6A using a 16x7135 driver sandwich with only one mode - BRIGHT. But during the process, I decided that it would be nice to have 3 modes, and I also got a little amp greedy:D , so I decided to throw one of my Shiningbeam boards into the mix. The Shiningbeam board added an additional 2.8A, and it also added the desired modes of brightness. All of these boards together should have given a combined current of 8.4A, but I had my doubts that the single batteryspace 26650 would supply the current because of voltage sag and a high vf SST-90 emitter, but I decided to try it anyway.

I got a reading of 7.4A at the tail-cap with a freshly topped off 26650. Now, my meter isn't a high dollar meter, so my readings could be off some, and it could be supplying the full 8.4A of current initially, but either way, it falls pretty fast. So I decided to pull two of the AMC7135 boards (8x7135 chips or 2.8A) until I can work out a parallel battery holder for both of my 26650's to run in parallel. I hope this will solve my problem and give me a longer runtime as well.

Anyway, I just thought I'd share some pictures of my mod. I added captions to some of the pics to try and help anyone who has questions:







The first thing I did was disassemble the light and mod the switch assembly and housing:

IMG_7650copy-1.jpg


IMG_7650copy.jpg

IMG_7656copy.jpg

IMG_7657mod.jpg


The second picture up from here suggests soldering the wire to the spring, but since I was only running one battery in my 2d light, I needed something to take up the space. A D sized spacer from Flashlightking does the job nicely. And this is how I implement it into the light (thanks go out to CPF users Linger and Justin Case for their help and ideas with this spacer).

Note: If you're planning to run flat top batteries and you're not planning to run a spacer that has protruding contacts, you may need to trim, saw or grind some of the plastic off of the spring side of the switch housing. If you don't do this, you may not make good contact with the switch.

IMG_7549.jpg

IMG_7544.jpg

IMG_7545.jpg



Then I put the end cap back on the spacer and run the battery positive wire through the spring on the back side of the switch housing and up to the top of the switch housing and solder the wire to the battery positive contact tab. Like this:


IMG_7677copy.jpg



Then, on the other side of the switch, I solder another piece of wire to the other positive contact tab. Like this:



IMG_7670copy.jpg




I do the same thing with the negative contact point:



IMG_7671copy.jpg




Then it's time to reassemble the switch. Like this:




IMG_7677copy.jpg



I end up with this:

NOTE: I shortened the negative contact tab after these pics were taken .




IMG_7540.jpg

IMG_7542.jpg

IMG_7539.jpg

IMG_7541.jpg

IMG_7689copy.jpg






Then I build the 7135 slave sandwich:






IMG_7596.jpg

IMG_7597.jpg

IMG_7593.jpg

IMG_7601.jpg





Then I mount the emitter to the heatsink:




IMG_7604.jpg

IMG_7607.jpg




A random pic:





IMG_7613.jpg







And the Shiningbeam driver wired to the slave sandwich and soldered to the emitter:





IMG_7666copy.jpg

IMG_7649copy.jpg






And of course, I couldn't forget the tail-cap spring wire mod:





IMG_7691copy.jpg





The battery:




IMG_7672copy.jpg






1" shedule 40 PVC pipe for the battery to sit in so there's no rattle:





IMG_7676copy.jpg






The beautiful UCL glass from flashlightlens installed:






IMG_7693copy.jpg


View from the front:


IMG_7697copy.jpg

IMG_7718copy.jpg





Next to my SSC P7 Mag:





IMG_7714copy.jpg







And a picture of the light:







IMG_7702copy.jpg








UCL Glass info:

I was recently asked about the UCL glass in this light. I think is was CPF member Bigchelis who said that OTF lumens would go up by around 10% with the install of the UCL glass. Of course, he also said that your eyes will not actually notice this difference, so it's up to you:D. I think the UCL glass is made by flashlightlens.com. I purchased mine from CPF vendor Lighthound, click here. The size of the UCL lens is 52.1mm x 1.90mm and it has a light transmittance of 98-99%.

For anyone who's skeptical about the UCL glass because of durability reasons. I'd recommend the acrylic lens from flashlightlens. Click here. It's a little bit thinner than the UCL lens at 52.1mm x 1.68mm, but it's such a small difference; I think it's negligible. It's also lacking about 3-4% in light transmittance (95%) vs the UCL lens (98-99%), but you would never notice this difference with your eyes, again negligible.

Borofloat has also been recommended as an option; it doesn't scratch as easily as the UCL glass. It's manufactured by Schott glass, out of Germany. Most of the high end optics manufacturers use Schott glass, so if you're familiar with good quality rifle scopes and binoculars, you know how nice the upper end glass is.




Heatsink info:

The heatsink I used in this build is an SSC P7 heatsink, with HA111 anodizing. The HA111 anodizing provides electrical insulation for the heatsink, so in the event that a bare wire comes in contact with the heatsink, you don't have to worry about a short. This heatsink was purchased from CPF member H22A, and since it's actually made for an SSC P7 emitter, if you choose to use this heatsink, you will actually have to take a little bit of aluminum out at the top of the pedestal (where the wires are soldered to the bottom of the emitter).


You will also have to center the emitter on your own; the tabs you see on the pedestal are made for a round P7, and as you know, the base on a bare SST-90 is square. I think I read that Setapong (H22A) has some prototype heatsinks for this emitter, but I'm not sure if they're available at the moment. Here's the link to his sales thread: Click here.


CPF member Britelumens also offers heatsinks, and his heatsinks are actually designed to work with the SST-90. Click here for his sales thread.
 
Last edited:
Re: SST-90, 2-D Mag build using 7135's

Now turn them on and take a picture.:p




:D Yeah, I'd like to do some beamshots, and I might try some sooner or later, but I've never really taken beamshots before so it'll probably take some time to get the settings right.
 
Re: SST-90, 2-D Mag build using 7135's

Hey tri-force nice thread. When I stack AMC7135 boards I put a little dap of Arctic Alumina epoxy on each of the 4 chips between each board and then mount the last board either directly on the emitter heatsink or on a copper (you could use aluminum from a pop or hopefully beer can :grin2:) "L" shaped bit that is also epoxied to the emitter heatsink or inside the flashlight body. Just a little bit of thermal protection. Heat should not be an issue with 3 D NiMH cells, single lithium, or parallel lithiums.
 
Re: SST-90, 2-D Mag build using 7135's

Hey tri-force nice thread. When I stack AMC7135 boards I put a little dap of Arctic Alumina epoxy on each of the 4 chips between each board and then mount the last board either directly on the emitter heatsink or on a copper (you could use aluminum from a pop or hopefully beer can :grin2:) "L" shaped bit that is also epoxied to the emitter heatsink or inside the flashlight body. Just a little bit of thermal protection. Heat should not be an issue with 3 D NiMH cells, single lithium, or parallel lithiums.


That's a great idea, thanks for sharing:thumbsup:.

298233.jpg
 
Re: SST-90, 2-D Mag build using 7135's

Nice build, and really nice pictures!
 
Re: SST-90, 2-D Mag build using 7135's

Sweet build... Whats the current draw on various modes? What heatsink are you using?
 
Re: SST-90, 2-D Mag build using 7135's

Sweet build... Whats the current draw on various modes? What heatsink are you using?


I haven't actually taken any current readings on the low or medium levels yet, but that's a good idea, I'll do that when I have some time. My readings at the tailcap on high are 5.7A with 8x7135 chips removed from the original sandwich (see explaination in first post).

I've updated my first post to include the heatsink information. Just scroll all the way to the bottom of the first post.

Casey
 
Last edited:
I sent this light to CPF member Bigchelis, and he kindly measured out the front lumens of this light in his sphere for me. Please see the update in the first post if you're interested in the results.

Casey
 
If you have a bench power supply that can deliver 10A, you should measure the emitter Vf at the relevant drive current. jtr1962's measurements show that you can expect about 3.65V at 5.4A drive. Add in the 120mV overhead for a 7135, and you need at least about 3.8V to run in full regulation. If your SST-90 has a higher Vf, and/or if there are some other voltage drops in the system, such that you need say 3.9V, then you could be in trouble trying to use 1xLi-ion. The Batteryspace discharge curves for the 26650 shows 4A and 8A data. If you need 3.9V min, run time in full regulation might be quite short even with two 26650 cells in parallel. Hopefully, you don't need 4.0V. Also, you can try powering the Mag mod with the bench supply to see where Vin is for full regulation. You might connect the power supply to the switch center spring and to the bare Mag tube edge. Although you bypass the tailcap and any associated voltage drop, this might be offset by the contact resistances of the bench supply leads that you use to hook up to the Mag.
 
If you have a bench power supply that can deliver 10A, you should measure the emitter Vf at the relevant drive current. jtr1962's measurements show that you can expect about 3.65V at 5.4A drive. Add in the 120mV overhead for a 7135, and you need at least about 3.8V to run in full regulation. If your SST-90 has a higher Vf, and/or if there are some other voltage drops in the system, such that you need say 3.9V, then you could be in trouble trying to use 1xLi-ion. The Batteryspace discharge curves for the 26650 shows 4A and 8A data. If you need 3.9V min, run time in full regulation might be quite short even with two 26650 cells in parallel. Hopefully, you don't need 4.0V. Also, you can try powering the Mag mod with the bench supply to see where Vin is for full regulation. You might connect the power supply to the switch center spring and to the bare Mag tube edge. Although you bypass the tailcap and any associated voltage drop, this might be offset by the contact resistances of the bench supply leads that you use to hook up to the Mag.


I need to purchase a bench power supply; the importance of having these tools for diagnostic purposes, is rapidly unfolding in front of my eyes:candle:.

I think this definitely explains why the light was visually dropping so rapidly in output (lumens) when I was trying to run 1x26650 at 8A using 7135's. I'd click the switch on my mag, and for about 1 second it looked like the trees were going to catch on fire :poof::crackup: and then brightness would fall off, WAY off:mecry:.
 
Last edited:
I thought 7135 is a linear buck driver, which means you'll only benefit
if you need to reduce the voltage/current.

Since you are running 1x26650, I am not sure what's the benefit of using this driver at all.
It'll only add resistance & reduce your current to the emitter.

Where you'll get higher current w/o the driver. With IMR 26650/SST-90, you won't have any risk of drive it over spec,
as you won't be able to get to 9A even w/o any additional resistance.
 
I thought 7135 is a linear buck driver, which means you'll only benefit
if you need to reduce the voltage/current.

Since you are running 1x26650, I am not sure what's the benefit of using this driver at all.
It'll only add resistance & reduce your current to the emitter.

Where you'll get higher current w/o the driver. With IMR 26650/SST-90, you won't have any risk of drive it over spec,
as you won't be able to get to 9A even w/o any additional resistance.


The 7135 is a linear regulator, which means it changes the current to keep it constant. My reason for running this type of driver, is to get a regulated runtime instead of a constant drop in output from when the light is turned on. Although, if I can't get a battery solution that will perform, I might end up with a direct drive setup:devil:.

Casey
 
The 7135 is a linear regulator, which means it changes the current to keep it constant.
Casey

This is true only if your voltage after sag can drive led above spec.
In your case, you are driving the led under spec & the 7135 will default into direct drive & you won't get "constant current" regulation, you'll see lumen drop with battery voltage go down. The net result is introduction of extra resistance.
 
This is true only if your voltage after sag can drive led above spec.
In your case, you are driving the led under spec & the 7135 will default into direct drive & you won't get "constant current" regulation, you'll see lumen drop with battery voltage go down. The net result is introduction of extra resistance.


Yeah, 1x26650 just can't perform for this setup. I'm planing to run 2x26650's in parallel to see if that solves the problem. Maybe the voltage won't sag as bad with two of these batteries in parallel.
 
A linear regulator acts like an automatic potentiometer (essentially, like a smart resistor) and adjusts resistance to maintain a constant output voltage. Kirchoff's Current Law also shows that current in equals current out.
 
Yeah, 1x26650 just can't perform for this setup. I'm planing to run 2x26650's in parallel to see if that solves the problem. Maybe the voltage won't sag as bad with two of these batteries in parallel.


Techjunkie had a SSR-90 build with 2x26650 IMR in parallel, DD.
I don't think he got it up to 9AMP with that.
So you still won't get constant current regulation with 7135.

If you are good at stacking drivers, and you are willing to use 2xIMR26650
You can use 4 DX P7 Driver I used in this thread to get 10Amp constant current regulation,
the only problem is to make 4 drivers fit, as it's pretty big.
http://www.candlepowerforums.com/vb/showthread.php?t=254481
.
 
Last edited:
Top