My first build! -- RoomSweeper "clone"

Candle Power Forums

Help Support Candle Power:

ejot

Enlightened
Joined
May 27, 2009
Messages
372
City & State/Province
East Fishkill, NY
Recently I’ve been wanting to delve a bit deeper into the hobby and start doing some mods. Finally decided to start with a modest project – direct-driven P7 into an E-series head – a la the “Room Sweeper”. Not nearly as exquisite as a Milky version, but bright just the same.

I had absolutely no real clue what I was doing and messed up just about every step at least once. :) But I am mostly pleased with the final result, for a first try. My photography isn't the best either – both my equipment and skills are lacking in this dep't.

First up was hacking out most of the reflector on the E1e head. From the back, I first trimmed the pedestal flush off, and then opened up the size hole I needed through everything that was left. Once I got started cutting, I didn’t want to stop – and I like bare aluminum – so I made a few other changes to the head and then polished it.

IMG_1595s.jpg




Teflon leads were soldered to the P7 DSWOJ. Not being very experienced, my solder blobs are fairly fat. I covered them with Plasti-Dip since they were very close to shorting on the material under the reflector. Hopefully it won’t melt. :thinking: It’s not pretty, but my plans called for this area to get covered.

IMG_1598s.jpg




Attention was then turned to the heatsink. Starting with 1” tellurium copper rod stock, I ended up with this after polishing:

IMG_1605s.jpg




Since I didn’t know how to center the emitter, I made a centering and mounting jig, which came in handy many times throughout the rest of the build. It is designed to press on the outer plastic ring of the P7 and leave flat clamping surface, so the parts can be clamped in a vice while gluing the LED:

IMG_1609s.jpg


IMG_1613s.jpg




For my first go at attaching the LED, I used AA compound, not the glue. I was pretty unhappy as the centering didn’t turn out quite right. Worse, when I put the LE in the head and did a test run, I got a disgusting donut hole right in the middle of my beam. :sick: I thought the emitter wasn’t deep enough, so I started shimming it (Incidentally, these shims were a real *%$#^ to make!). This didn’t help so then I figured the emitter was too deep. Began cutting away some of the thickness behind the reflector in the head, and retrying. Finally the donut began to disapper! :grin2:. A few more iterations of trimming and trying, and I was happy with the beam.

Of course, now my heatsink was no good. It was too shallow. :sigh: Since it was both easier and more effective to just make a new sink than the required adapter parts … out came the copper bar stock again. To better center the emitter, I cut a 0.020” deep centering area (taking care to adjust my other dimensions to take this into account):

IMG_1750s.jpg

(Yes – the lathe is from 1920. :p )

IMG_1755s.jpg

(Sadly, this left a much rougher mounting surface for the LED, that I could not polish out. :duh2: )



Once again, I used my centering jig to mount the emitter. Even though the centering function was ineffective and now redundant, the jig was still useful to tightly clamp the emitter down to minimize thickness of the thermal adhesive. This time I used the epoxy, along with two strands of hair under the plug to keep electrical isolation. Clamped into a bench vise, and then checked that there was no continuity where continuity was not welcomed – thankfully all was good:

IMG_1767s.jpg

You can see some thermal compound residue on the plastic part of the emitter package. I actually mounted and dismounted the LED four times with thermal compound before I was confident enough to do it with epoxy.



The negative connection was an easy blob of solder in the hole, then filed flat. For the positive side, I first made an isolating sleeve from Delrin and a positive contact from brass that pressed tightly inside the sleeve. (Yes, I made this out of aluminum first. Yes, I now know that you can’t solder to aluminum.) A small hole was drilled into the brass contact right below the join between the two parts, and a bare wire soldered to the contact. Then I soldered the bare lead to my positive lead coming from the LED:

IMG_1774s.jpg




The final job was to pot the whole area around the contacts. I used MarineTex grey epoxy. Super great stuff for many jobs – but don’t use it for this!! (More about this in a moment).

IMG_1777s.jpg




After drying overnight…

IMG_1782s.jpg




Starting the cleanup …

IMG_1785s.jpg

Notice anything funny about the epoxy? Right … it’s METAL FILLED. :banghead: I checked and it doesn’t seem to conduct at all, so I just went with it. :shrug: But needless to say, I will look for another product.



I wasn’t too happy with all the air bubbles that were left either, but after some more work got a reasonable finish and also added a small chamfer. Here is the completed LE:

IMG_1807s.jpg

IMG_1808s.jpg




And the completed head with glo-ring (remember I said the LED contacts would be covered?):

IMG_1822s.jpg




The original plan was to de-anodized and polish a Vital Gear FB2 body for use with AW protected 17670. However, I just picked up a bare McLux 2x123 body that I’m hoping to use instead. Hopefully the battery fits, if not I’ll bore it or use for something else. Also picked up a Balrog 1x123 bare alu body for IMR pocket rocket status. :D For now, it’s on the still-anodized FB2:

IMG_1816s.jpg




Well, this thing sure does light up a room! Wow! I took some comparison beamshots at work, but now I realize I can’t post them as we’re not allowed to publish any photos with our lab in it! Here is what I can offer for now until I get a chance to reshoot at another location:

IMG_1829s.jpg

That’s not the tint :green: ! (It’s a blue wall.) :duck: Against white, it is a very bright, neutral white with a slight yellow tinge. This was taken in an otherwise pitch black room, 3’4” from the wall with the camera and emitter @ the same distance.



Glow-ring! Ran out of batteries before I could get a better shot:

IMG_1831s.jpg




It’s been a lot of fun building this light and I have plans for many more. I welcome any constructive criticism that will improve my builds as I am still learning just about everything! lovecpf
 
Wow. For re-inventing the wheel you did a remarkable job.
I was concerned the +'ve was exposed to battery bumps and glad to keep reading to your solution to that. Some people are fastidious about the mounting surfaces for the emitter slugs, polishing it from 200 to 400 to 600 grit with sand-paper that was spray-glue mounted onto a pain of glass. Who can say? It is a nice heat-sink you made. Very nice. Keep at this and you will go far. Then I can start ordering parts (heatsinks) from you.

There is a person, wquiles, who documents the custom mods he does with many pictures and explanations. If you have time to read through a few of the pictorial threads every so often, I wonder if you'd pick up some process ideas that would smooth out the design elements. If you had a larger plan and had to think less, it may allow you to focus more time on perfecting the implementation of your ideas. During his mods he often ends up re-building many portions of the light and you covered a lot of similar ground in your first build documented above.

When do you start taking orders? Can't wait to see build # 2.
Best,
Linger
 
Last edited:
Wow. For re-inventing the wheel you did a great job

LOL - Indeed I didn't have any idea how this would "normally" be done, and just made it up as I went.

Am humbled by the compliments – :o – thank you.

Will try taking some measurement of current @ tailcap on a fresh battery tomorrow, and will work on some beamshots.
 
Nice build ejot.

Is your lathe really from the 1920's?

Let me guess-a SouthBend or a Cincinnati.

Now all we need is some outdoor beamshots and maybe some idea on how hot the body gets after say 5 minutes.
 
As an Amazon Associate we earn from qualifying purchases. Product prices and availability are accurate as of the date/time indicated and are subject to change.
Linger --
Will's write-ups were a big inspiration for me to try building something. Always carefully planned, perfectly executed with huge attention to detail - really sets the bar high as a level of work to strive for.
I really appreciate the feedback and agree on developing a broader plan and procedure first. Now I know some more of the things to keep in consideration.


ANDREAS--
The lathe is actually a Sheldon, which I believe was in Chicago. I am going to research the serial number for actual year, but it seems older than the 40's literature I've seen so far.
There is also most of a South Bend at work, in pieces. I hope to get that together someday because it is a far better lathe with much more capability. :hitit:
As for heat after five minutes ... if I hold the head the whole time, I can hold all five minutes without letting go. If I just hold the body, the head is too hot to continuously hold after 5 min, but the body is still very comfortable.

I'll try taking some temperature data sometime, along with my previous promises. Work has me very busy right now (13 hrs today) :tired:
 
Someone posited that your hand will dissipate 30watts, beyond that the heat will accumulate.
 
SWEET!

I realy like the way you machined the epoxy and added the ring of GID. :grin2:

Very clean build and thank you for sharing it. :thumbsup:
 
Thanks!

Your bump reminded me that I'd promised current and temp measurements and never did them. It's back on the flashlight to-do list (along with 77 other items :whistle:).

I use this light all the time, it's got a few scratches and scars now to prove the workhorse status. Very useful when you need a lot of light up close in a confined area. Running it with a McE2S adds even more versatility.
 
Was going to paint a little area at the base of the head so I could just hit it with a Fluke 65. This number by itself probably not very interesting, but if I measure a few other production-type lights as a comparison ....

If that doesn't work out I'm sure I could dig around a bit and piece together a thermocouple setup.
 
A thin gauge self-adhesive K-type and temp read out would do the trick. In fact I have few self-adhesives sitting around so let me know if you could use one and I'll send it over. :)

Come to think of it I need to start doing some temp measurements myself along with the 50 other things on my to-do-list. :shakehead

Again, thank you for the very well documented tutorial.:twothumbs
 
Last edited:
Back
Top