I've forgotten...what is good for a P-60 drop-in?

[Tom_E]

Sorry - first I've seen this thread. So the details are 4.49A, de-domed XM-L2, OP P60 reflector, 852 lumens. Something is wrong if it's a U2 1A. What's the bin and tint? Is the LED on a copper star? I see it's a SinkPAD in the pic...

I've gotten better #'s in P60's at lower amps. Figure 20% off for de-doming, so what you have is about 1,070 lumens domed.

My #'s: XM-L2 U2 1A/Noctigon, 3.80A QLite (2x 7135's), SMO reflector, 502B host, AR lens: 1,217 lumens at 30 secs

Same pill, 501B, unwrapped: 1,173 lumens at 30 secs

Same pill, same 501B wrapped, 1,234 lumens at 30 secs

Same pill, L2T host (back bezel), AR lens: 1,108 at 30 secs

Numbers can be all over the place in P60's. P60 hosts with black fairly large bezels can easily knock down 100 lumens. The L2T stock is a bad host for output, for example.

OP reflectors will always knock down throw, but will get equal or better lumens output. Extensive P60 tests showed that, published here somewhere.

[Rufusbduck]

@DBCstm

Scott, it happened all right!

No, I didn’t go all the way out for the 97 yd shot of the red drum, it’s NASTY out there!

That first shot at the red drum is about 65 yds. The zoomed in shot didn’t turn out, snow messed up focus.

Looks like perfect weather to stay in and make your wife’s day…

…modding of course.

P60’ would be great if not for the inherent design flaws of a brass pill and poor fit both of which are surmountable but limit the performance of unenhanced models. It wouldn’t be a “drop in” or fit so many different makes if it were threaded or a press fit and even a Malkoff still needs to fit well to work well. Even though it works I don’t classify aluminum foil crammed in under the heading of “fits well”. There are levels of performance that are suited to the size of a host and the way it’s put together.

God love you Dale but you just love to ignore the first and max out the second. I think if you need hand warmers you should wear gloves instead.

[Tom_E]

I've found SolarForce's seem to block output at the head more so than cheapo P60 hosts - this was also highlighted in the extensive P60 tests done -- can't find the thread anymore, but showed the loss's from blocking bezel designs. Also, as always and as you know, tailcap springs need to be copper braided.

[DB_Custom (Dale)]

This drop-in is being built for a friend. I only put it in the L2P as that’s what I had to try it out in. No small brass pill here, solid copper shell with silver plating. The tubular style aluminum reflector fits down inside quite snugly, should also give some mass for more heat sinking.

I’m using a Dr. Jones driver with 4 added chips, I’ve carefully set the Hi mode to the 16th level. So I don’t know why it’s so low. (comparatively speaking) I also don’t know exactly which emitter it is, I had it already on a SinkPAD and haven’t a clue when I did it or what I was using it for and why it’s not in a light already. The memory thing. I might have had it in my Raysoon F13 and changed it out for de-domed U2 1A. I just dont’ remember. It’s probably a T6, although it does have a decent tint even de-domed.

I thought it should have higher OTF lumens, which is why I started this thread…needed input on what it should be doing so I would know if I should troubleshoot. I’ll get a couple of beamshots tonight shooting the red oil drum at 97 yds, see how it acts and then decide if I want to try for more power to put it where I was expecting.

Thanks for the help! Appreciate all the feedback, even the God loves me approach! lol I don’t always do this for me ya know…occasionally I even build one not powered to the max, just because.

[Yellowhorse]

@DBCstm

3 Nichia 219’s making 1600 lumens from 4.65A? Not gonna happen.
I have the high powered 219 that supposedly makes 476 lumens at 1.5A, from the data sheet. So in theory, 4.5A would yield an emitter lumens total from 3 lights of 1418. None of the emitters I know of will make 100% efficiency, or even close.
And I also have a Triple 219 Hi CRI light that makes 493.35 lumens (30 sec) at 4.16A. Somebody is Over-rambunctious with their numbers. This particular light is in a full copper silver plated shell, with the high end electronics and bumped to the higher amps by me. It’s not a cheap light, it’s been measured in a lightbox for OTF lumens, and I really really like it. But it’s no beast.

Thanks for those links, some very nicely made high end P-60’s!

Try re-reading the link to Oveready on the triple led drop in. You missed a few details. There are several different options for this. Here they are:

Triple/60 H3c 1-2 cells 1600L (1200L) H-M-L Cool XPG2

Triple/60 H3n 1-2 cells 1500L (1125L) H-M-L Neutral XPG2 (4750K)

Triple/60 H3n 1-2 cells 1100L (825L) H-M-L Neutral 219 (4500K)

Make sense now? Only 1100 lumens using the 219’s. I have the H3c using Cool XPG2’s (1600L) and the H3cn using the 219’s (1100L). HUGE difference in output between the two engines. Personally, I don’t care much for the 219 version. I thought I might be impressed with the better CRI numbers on it but I’m not. To many lumens lost to make up for the better color recognition.

[DB_Custom (Dale)]

Geesh! I’m gonna have to start drinking coffee again!

Sorry bout that.

[Tom_E]

Overall it's closer than I thought at first, so the blocking P60 host, possibility of it being a T6, etc., are all possible causes of it getting what it's doing.

[DB_Custom (Dale)]

It’s funny, we get to chasing the big numbers and get addicted to it! I think it’s doing pretty doggone well right where it’s at. I’ll still get some pictures when it gets dark and see how it’s throwing, but primarily for reason of interest in this reflector, as I haven’t used this combination before.

Interestingly enough, this is right about the output I was looking for in a P-60 I built for my wife to replace the el-cheapo I had put in her P1D, that one didn’t turn out so well, with a thermal control that shuts it down too quickly.

[Tom_E]

You are too generous! I gave my wife one of those $2 AA zoomies from WallBuys. Of course it was mod'ed with a XM-L2 U2 1A on copper...

[DB_Custom (Dale)]

I am fairly impressed with this combination, not sure about the de-domed XM-L2 tint though. It’s not bad, but is headed towards green. In normal use it’s hard to distinguish, but compared to most of my other lights it’s obvious.
I used the Canon G1X, ISO 1600, f/5.6 for the 28mm, f/5.8 for the 112mm. Red oil drum is 97 yds.

[comfychair]

Will the choice of host really block light with the bezel, when the front opening of the reflector is only 20MM?

[DB_Custom (Dale)]

Comfy, I think that’s with the 26.5mm standard reflector in a P-60. Pretty sure this 20mm isn’t getting any light blocked by the host.

[Tom_E]

I can't find the thread, wonder if it was deleted, but a guy had all these measurements on a spreadsheet, showing results of P60 tests. With my simple collection of P60 hosts, I noticed out of the 501Bs, 502B, S-R5, and a SF L2T, the SF L2T performed the worse and does have a big black bezel on it.

Here's some test results I just did:

- P60 drop-in: Nanjg at 3.85A, XM-L2 U2 1A/Noctigon, 22 gauge wires, copper braided springs, SMO reflector, copper braided spring, alum foil wrapped

- SolarForce L2T host, copper braided tail spring

- fresh Panasonic PF cell

Results, in sequence:

- SolarForce flat SS bezel: 1,234 lumens at 15 secs

- L2T stock black bezel: 1,156 lumens at 15 secs

- SolarForce flat SS bezel: 1,204 lumens at 15 secs

Unfortunately, you can see some effect of the accumulated heat and/or cell droppage. I gave it a couple of minutes between tests, but probably not enough. I would say the effect of the black bezel is about 60 lumens loss from this test, or roughly 5%, compared to the low profile SS bezel.

5% is fairly significant for just a bezel difference.

[DB_Custom (Dale)]

But is it significant enough to actually be able to see? If you took a light out on a given night that had 60 more lumens than you’re used to, you quite likely wouldn’t even notice. Especially over 300 lumens.

JMPaul put an AR lens in the C8 he bought from me and took it from 77Kcd to 85Kcd. That should be a visible difference in throw.

So I suppose if you add up the little things, you’d get enough gain to be able to actually see it.

Like in my just completed M8, Sanyo UR18650FJ cells will make 5.25A while Samsung 20R’s net 6.77A! Sometimes the best mod is a battery swap.

[Tom_E]

5%'s do add up to 100%, eventually . This is what we do, little by little, with heavier wires, copper braiding springs, better cells, better thermal treatments, copper MCPCB's, higher bins, AR lens, centering/focusing emitters, etc.

It's all about efficiency - getting the most out of the parts. There's trade-off's of course of cost/looks/functionality vs efficiency.

If I had a really nice light but with a big black bezel, I'd think about grinding it down, removing the anodizing, polishing up the aluminum. Of course with a SolarForce, you always have the option of buying their low profile SS bezel.

Edit: Finally found it! Here: https://budgetlightforum.com/t/-/19458

I really have no reason to doubt these results - they seems very thorough, and lots of repeated tests. Basically he shows about a 4% improvement between the stock L2T bezel and the B6 SS bezel (here) - the two I tested.

[gitira6]

@comfychair

Ha! I recognize that little reflector… they do work well. I think I have about a half-gallon ziplock bag full of them by now…

I can relate to that, I have a ziplock filled with parts and reflectors too!

[Old-Lumens]

@comfychair

'Better hot spot' - you're gonna have to define 'better', lol. Could mean too many different things to different people. It may seem weird to pair a de-domed LED with an OP reflector, but it gives a really nice, usable output, doesn't it?

It works good with a de-domed XPG2 too, maybe better than with XML2. Depends on what you want from it though I guess.

Here's a teaser of a discarded prototype for a C8, I'll keep trying until I can make one with the plating still intact. Polishing isn't much fun. :(

Send it to me, I will polish it… and I will even send it back to ya.

[DB_Custom (Dale)]

Ok, rebuilt the copper shell P-60. I put a stock Qlite in it at 3.04A and I’m seeing 3.00A at the tail. I’m using an XM-L2 U2 1A, domed, and getting a start value of 969 OTF lumens. 969 at 3A as compared to what was it 863 at 4.5A? This is much better, more efficient, will run longer on a single 18650. But now it’s blue instead of green. sigh

It’s not bad though, and is definitely an intense P60.

[Tom_E]

The 20% loss rule for de-domed, and the de-dome and OP combo is quite odd - usually you de-dome for throw, but the OP knocks down throw. 20% still doesn't account for the full difference though, so maybe that de-domed XM-L2 was a T6?

Is that epoxy on the stacked 7135's and the wire connections? I've never seen that before, but could help I suppose - it's very neatly done work.

[DB_Custom (Dale)]

Of course, you’re right, but the shell design demanded pretty much that particular reflector and I didn’t find it in mirror finish. So you work with what is available.

The white stuff on the chip legs and contacts is Arctic Alumina Thermal Adhesive, put there as an insulator to avoid a short. I’ve taken to doing that when space constraints make me stack on the spring side and I want more power, I sometimes stack 3 high on the spring side and if using a spring I don’t want it to be able to compress into the legs. I put the AATA on the connections in an effort to prevent the reflector from shorting out, not having dealt with this one before. Mini potting, if you will. Usually, if stacking 3 high I remove the spring altogether and put a brass post there so it cannot short. Also gives better conductivity.

Wasn’t really necessary I suppose with that cylindrical spring that comes stock on a Qlite. Meant to put a different spring on it.