DTR
0
- Joined
- Jun 24, 2010
- Messages
- 5,680
- Points
- 113
2W SKYRAY SR-5 With Modded Single 1.8A Microboost Driver
Ok so this build started out with me thinking I was going to use a Moh 1.8A driver and two 18350's as this host looks like a clone of the Ultrafire U-501B. What I came to find as they look very alike there are a few differences and when I finished building it I realized that it was not going to fit the batteries. I then tried 16340's and no joy. So it is an 18650 only host. That unfortunately means I need a boost driver and I did not have any.:cryyy:
Well right now boost drivers are still scarce while we wait for flexdrives or some of the new designs to come rolling out.
Anyway so here is the Failed build. This is just part of this hobby. Sometimes things don't work out so I felt I would show the whole progression of events that led up to what may be a fluke or could be a great mod for the microboost driver.:beer:
So I started out with the host. I have had this sitting around for a while waiting for time to build it. Well I got some time last week and wanted to do a review on it as it really is an awesome host. It has a lot of mass for great thermal properties and a good size head to get a good sized heatsink in.
Flaminpyro made the heatsink for this host and also an awesome copper pill with threads that is for a driver heatsink.
The threads on the pill will allow very efficiently heat transfer into the massive host body. Great design.
The tolerances on the heatsink are amazing. It fits so tight into the head of the host. That entire smaller diameter ring area on the back is in perfect contact with the inner walls of the host for excellent electrical and heat transfer.
Next I got one of Moh's drivers out to set up and sink to the pill.
I started by adding leads to it.
Then I got some two part thermal adhesive ready to sink the driver to the copper plate.
Next I set it and used a pair of helping hands to keep pressure on it as it hardened. Give about 15-20 minutes to be sure it is fully hardened.
Now I needed to set up the pills battery contact. So I ran the wire through the pill and striped the end.
Ran it though the contact board.
Then I just soldered it in place.
Next a little more of the thermal adhesive to secure the contact board so it would not come out of place and again used the helping hands to keep it where I wanted it while it hardened.
Next I made a little heatsink for the resistors. It is shaped like this so I didn't bypass the resistors by making contact with them on both sides with the aluminum.
And again I used the helping hands to keep pressure on the heatsink while it hardened.
Now that the pill is all set up I screwed it in to the host. I used a small Allen wrench with a drilled hole that is there to help screw the pill in tight.
Next I prepared a diode in a module.
Added heatshrink and soldered the leads to the diode.
Next I slide the module up through the heatsink and the heatsink into the host body. I could tell how well the tolerances are with this host as you put the heatsink in it shoots the module out like a geyser from the air pressure in the host.
Then I set the module to the height I wanted and tightened down the set screws.
Screwed the head on and thought I was done.
Oh wait. Now I tried to but the batteries in. Here is how tow 16340's fit. Not even close for getting this tail cap on and something I should have checked before getting started.:banghead:
Here is the extra room you get with an unprotected 18650 and it is a nice tight fit. Which is good. This is not a host you have to worry about not holding the battery tight enough.
Alright now I have to go back to scratch. I just wasted an hour and a driver as in my frustration I pulled it out with a pair of pliers destroying it.:evil:
So I regrouped and came up with another game plan. I did not have any boost drivers and did not at first know how I was going to proceed. Then I remembered I had an old microboost that I had thrown in my parts bin because I figured it was toast.
When setting this driver up I accidentally pulled one of the resistors off and lost it in the copper mesh that I clean solder off my iron's tip with. So I proceeded to try the driver anyway and it was giving 1.8A and heating up very quickly. I figured this driver no longer had the ablility to regulate and thew it to in my parts box and moved on as I did not have time to mess with it at that time.
This is the resistor that came off.
This is how the remaining resistors are bridged.
EDIT after another test I believe this is the correct way to bridge the resistors. EDIT
Ok so I decided that I would pull it back out and see if it could maintain 1.8A with good heatsinking. As Morgan discovered in his dual microboost skyray build the flat package here is the one that produces the most heat and is where you should heatsink this driver.
So I followed that and made a nice little aluminum mounting sink and sinked the microboost to the pill. I used the contact wire that I had set up in this pill earlier so I did not have to reapeat that step.
Next I proceeded to test the driver on my test load and it held relatively constant considering how over driven the microboost was being run. The pill did get pretty warm so I stopped about 2:30 into it but in a host the heat can escape with great efficiency into the massive body of this unit. Although the pill was getting hot I had done a run with my finger on the driver to feel for any hot spots and It seems that all the heat was being pushed into the pill which was a very good sign.
Next I decided to get a little perspective on how it would react as the battery drains. This was odd to me because I already noted the unit was pulling 3A with 4.2V but it seems as I lowered the voltage from the power supply the current draw stayed constant and the current output dropped. Down to 3.6V the unit was still producing over 1.5A. Which I though was pretty awesome considering this driver is only rated for 1.1A. It seems this mod kicks this driver into more of a constant voltage mode like the NJG-18 driver.
I felt pretty comfortable at this point so I paired the unit with a decaned M140 diode that I would not cry to bad if I lost and the results were very good.
Just to note I used my Ophir head which gives virtually instantaneous readings. All units have the initial peak and drop in the first few seconds of about 200mW that you would not even detect on a slower response LPM like my Laserbee.
On my laserbee II 3.2W it's peak reading would probably be about 2090mW. I always use the Laserbee II when quoting people what a laser peaks at as saying this is a 2.4W laser would not be accurate but a 2W would be.:beer:
Anyway here is a three plus minute run with the unit driven by an 18650.
Ok so I have been playing with this unit for almost a week. I have run it between 3-5 minutes at least 8 times and 30second to 2 minuet runs many many times to see if it would run into any issues. None yet.
So I do want to say this if you try this mod you do at your own risk as this was one driver being tested. This needs to be tested more to see if it really is safe and reproducible.
I don't have any more drivers at the moment so when I do get some more I will be trying this again and will report my findings.
Anyway if this turns out to work consistently this could be a good middle of the road option between a NJG driver and a dual flex setup.
More data will be added as it comes and now no review is complete withought some glorified beam shots.
Ok so this build started out with me thinking I was going to use a Moh 1.8A driver and two 18350's as this host looks like a clone of the Ultrafire U-501B. What I came to find as they look very alike there are a few differences and when I finished building it I realized that it was not going to fit the batteries. I then tried 16340's and no joy. So it is an 18650 only host. That unfortunately means I need a boost driver and I did not have any.:cryyy:
Well right now boost drivers are still scarce while we wait for flexdrives or some of the new designs to come rolling out.
Anyway so here is the Failed build. This is just part of this hobby. Sometimes things don't work out so I felt I would show the whole progression of events that led up to what may be a fluke or could be a great mod for the microboost driver.:beer:
So I started out with the host. I have had this sitting around for a while waiting for time to build it. Well I got some time last week and wanted to do a review on it as it really is an awesome host. It has a lot of mass for great thermal properties and a good size head to get a good sized heatsink in.
Flaminpyro made the heatsink for this host and also an awesome copper pill with threads that is for a driver heatsink.
The threads on the pill will allow very efficiently heat transfer into the massive host body. Great design.
The tolerances on the heatsink are amazing. It fits so tight into the head of the host. That entire smaller diameter ring area on the back is in perfect contact with the inner walls of the host for excellent electrical and heat transfer.
Next I got one of Moh's drivers out to set up and sink to the pill.
I started by adding leads to it.
Then I got some two part thermal adhesive ready to sink the driver to the copper plate.
Next I set it and used a pair of helping hands to keep pressure on it as it hardened. Give about 15-20 minutes to be sure it is fully hardened.
Now I needed to set up the pills battery contact. So I ran the wire through the pill and striped the end.
Ran it though the contact board.
Then I just soldered it in place.
Next a little more of the thermal adhesive to secure the contact board so it would not come out of place and again used the helping hands to keep it where I wanted it while it hardened.
Next I made a little heatsink for the resistors. It is shaped like this so I didn't bypass the resistors by making contact with them on both sides with the aluminum.
And again I used the helping hands to keep pressure on the heatsink while it hardened.
Now that the pill is all set up I screwed it in to the host. I used a small Allen wrench with a drilled hole that is there to help screw the pill in tight.
Next I prepared a diode in a module.
Added heatshrink and soldered the leads to the diode.
Next I slide the module up through the heatsink and the heatsink into the host body. I could tell how well the tolerances are with this host as you put the heatsink in it shoots the module out like a geyser from the air pressure in the host.
Then I set the module to the height I wanted and tightened down the set screws.
Screwed the head on and thought I was done.
Oh wait. Now I tried to but the batteries in. Here is how tow 16340's fit. Not even close for getting this tail cap on and something I should have checked before getting started.:banghead:
Here is the extra room you get with an unprotected 18650 and it is a nice tight fit. Which is good. This is not a host you have to worry about not holding the battery tight enough.
BUILD FAIL!!!
:cryyy::cryyy::cryyy::cryyy::cryyy:Alright now I have to go back to scratch. I just wasted an hour and a driver as in my frustration I pulled it out with a pair of pliers destroying it.:evil:
So I regrouped and came up with another game plan. I did not have any boost drivers and did not at first know how I was going to proceed. Then I remembered I had an old microboost that I had thrown in my parts bin because I figured it was toast.
When setting this driver up I accidentally pulled one of the resistors off and lost it in the copper mesh that I clean solder off my iron's tip with. So I proceeded to try the driver anyway and it was giving 1.8A and heating up very quickly. I figured this driver no longer had the ablility to regulate and thew it to in my parts box and moved on as I did not have time to mess with it at that time.
This is the resistor that came off.
This is how the remaining resistors are bridged.
EDIT after another test I believe this is the correct way to bridge the resistors. EDIT
Ok so I decided that I would pull it back out and see if it could maintain 1.8A with good heatsinking. As Morgan discovered in his dual microboost skyray build the flat package here is the one that produces the most heat and is where you should heatsink this driver.
So I followed that and made a nice little aluminum mounting sink and sinked the microboost to the pill. I used the contact wire that I had set up in this pill earlier so I did not have to reapeat that step.
Next I proceeded to test the driver on my test load and it held relatively constant considering how over driven the microboost was being run. The pill did get pretty warm so I stopped about 2:30 into it but in a host the heat can escape with great efficiency into the massive body of this unit. Although the pill was getting hot I had done a run with my finger on the driver to feel for any hot spots and It seems that all the heat was being pushed into the pill which was a very good sign.
Next I decided to get a little perspective on how it would react as the battery drains. This was odd to me because I already noted the unit was pulling 3A with 4.2V but it seems as I lowered the voltage from the power supply the current draw stayed constant and the current output dropped. Down to 3.6V the unit was still producing over 1.5A. Which I though was pretty awesome considering this driver is only rated for 1.1A. It seems this mod kicks this driver into more of a constant voltage mode like the NJG-18 driver.
I felt pretty comfortable at this point so I paired the unit with a decaned M140 diode that I would not cry to bad if I lost and the results were very good.
Just to note I used my Ophir head which gives virtually instantaneous readings. All units have the initial peak and drop in the first few seconds of about 200mW that you would not even detect on a slower response LPM like my Laserbee.
On my laserbee II 3.2W it's peak reading would probably be about 2090mW. I always use the Laserbee II when quoting people what a laser peaks at as saying this is a 2.4W laser would not be accurate but a 2W would be.:beer:
Anyway here is a three plus minute run with the unit driven by an 18650.
Ok so I have been playing with this unit for almost a week. I have run it between 3-5 minutes at least 8 times and 30second to 2 minuet runs many many times to see if it would run into any issues. None yet.
So I do want to say this if you try this mod you do at your own risk as this was one driver being tested. This needs to be tested more to see if it really is safe and reproducible.
I don't have any more drivers at the moment so when I do get some more I will be trying this again and will report my findings.
Anyway if this turns out to work consistently this could be a good middle of the road option between a NJG driver and a dual flex setup.
More data will be added as it comes and now no review is complete withought some glorified beam shots.
Last edited:
