In terms of quality, this is the 2nd best host out of the 3. Most of the light is good quality, except the pill. The actual pill is not whats shown in the picture on the site. I’m not sure what material the pill is made off, looks almost like a combination of aluminum and brass. The threads on the pill is quite bad and very sticky. The build was fitted with a 22mm DD from MTN and 20mm MT-G2 on copper with Arctic Silver 5. No need to enlarge the LED opening in this reflector. The light does around 11-12 amps with purple 26650 Efests, but I’m running the light with 5200mah Keepowers with which it does around 7-8 amps. The light runs extemely well for the first 15 seconds where after the brightness just takes a massive dive. The ceiling bounce starts around 300 lux, after around 1 minute its on 150 lux! The light does get hot quickly, but still, why the HUGE loss in output in such short time.
This is a very good quality host, by far the best out of the 3. It has an aluminum pill with smooth threading. Screws in tightly with the light body. Its a chunky, heavy light which should make heat management good. The build was fitted with a zener modded q-lite driver in combination with this adaper and 20mm MT-G2 on copper with Arctic Silver 5. Had to enlarge the LED opening in the reflector. The light does around 6 amps with purple 18650 Efests. Once again, the light runs extemely well for the first 15 seconds where after the brightness just takes a massive dive. The ceiling bounce for this one starts around 245 lux, after around 1 minute its on 130 lux! The light does NOT get hot quickly, much better than host 1. Still, why the HUGE loss in output.
Quality wise, this host is close to host 1. The anodizing is not good and the lens is kind of hazy. Heat management on the other side is very good. The light is nice and chunky. The build was fitted with a 22mm DD from MTN and 20mm Dedomed MT-G2 on copper with Arctic Silver 5. Had to enlarge the LED opening in the reflector. The light does around 10-11 amps with purple 18650 Efests, but I’m running the light with 5200mah Keepowers with which it does around 7-8 amps. So, this light’s figures are much different. The ceiling bounce for this one starts around 225 lux and holds it very well thereafter. After +–2 minutes its around 200 lux. The light does get hot, but less than host 1. This light is thus the exception. The output does NOT drop as in the case of host 1 & 2.
So, why do I have the issues with host 1 & 2. Host 2 is not even pushed that hard? Is it heat or might it be the drivers? If it would have been the drivers, why is host 3 doing good, its the same driver used in host 1. All lights run 22 gauge wire.
Did you ask for turbo timer when ordering these drivers? A 60 sec timer would explain a sudden dip.
.
After the dip, quickly turn off, then back on the light and see if goes back up to a high output.
Hi shrick. Hard to tell without seeing the lights themselves, but it sounds like the first host has a thermal bottleneck or 2 close to the emitter. It sounds like some heat is getting to the host as you say it heats up quickly, but it's a small host. So it could get hot even with a bottleneck.
The second host sounds like the AMC7135's are overheating. 7135's have an built-in thermal protection that reduces current to like 250mA. Do you have the driver thermally connected to the host or is the driver suspended by the wires to the adapter and led?
We can't tell you for sure. Not enough information, to tell. In order to tell for sure, you would need to monitor the voltage and the amperage continuously, so that you would have data to help support what is going on. Also, you would need to read temperatures of the mcpcb and pill/head.
The problem is that there can be many reasons. If someone says, "this is the problem". It's only a guess. Not even an educated guess, since the only way to make an educated guess would be to have data backing up the guess.
My uneducated SWAG would be, 1.) Host 1 and 2 are not meant for using a led like the MT-G2. At best they would handle an XP-G2 and at worst, maybe an XM-L2 at 3 amps. 2.) Host #3 would be better than host 1 and 2, due to the finning on the head/pill area. 3.) You need a host that has A-more mass, B-a solid led shelf made right in the head, C-more mass. 4.) the driver should be attached to it's own heat sink, to remove heat from the driver itself. Whether it's a 7135 driver, FET driver, etc., heat lessens the ability of the driver to keep amperage at first turn on level, just as heat adversely affects the led. 5.) batteries also loose voltage rather rapidly when they are hit with high amperage. Some are better than others, but at amperages over 3-4 amps. batteries will drop rapidly.
So, without doing a lot of measurements with expensive equipment, find a much bigger host, with a solid led shelf and lots of mass. Then heat sink the driver and use the best of the best for cells --- and expect only a few minutes run time, no matter what. If you want your light to stay bright for over a couple of minutes, (bright meaning no drop in output), then keep the amperage under 3 amps, as a general rule.
i have purchased few mtg2 on noctigon stars that were very poorly reflowed, i could see half of the led’s heatpad hang in the air, could even insert a peice of paper between the led,and the star. so despite DT copper star, i would overheat very soon.
So what can we conclude based on the data provided:
Assuming that all three emitters are equal bin and condition the fact remains that the emitter is more efficient at low temperatures (I know, no new news here). All three have the same wires (and we will assume spring modfications). Factory spec is max 3A. lots of folks have good results at 4,5,and 6 amps. The max output has been found to be at 11-12 amps and beyond that you just make heat up to 16amps. So it seems you are pushing these pretty hard so finding that the ability to shed heat is possibly the boundary you are up against is not suprising.
Assuming that 26650 of the same brand will drive a light harder then 18650. And that all batteries generally peform worse under increased load.
The basic observation that battery perforance degrades over time is again no new news.
“stronger” 26650 cells and “same” driver as #3 but worse performance. #3 is the heaviest and we know mass helps but that mass could be in a non crucial area, plus the difference is not that great.
18650 cells but still lower performance then #3
Our winner an alleged known good “heat sinking” light
So what are we left with, The big variables here are the materials and design and execution of each light. Well like the others have alluded to, it “probably” is the ability to shed heat. Maxing out the mass in copper custom pills is possibly an approach to run lights this hard (and / or maybe even go DD and forget the drivers if your spec is to drive lights as hard as possible for as long as possible.
Thanks for the feedback from all. I really want to know whats the culprit causing the trouble in hosts 1 & 2. I wish host 2 could have been the best performer as I really think its a great quality light. Anyway, I’m going to try and provide more detailed information regarding amps, volts & temperature readings, including photos of the builds. I’ll get back to you guys.
Try #2 with a DD driver and mid level cells. That would take care of overheating 7135 chips. MT-G2’s generate lots of heat so if two hosts of similar head sizes heat up at different rates then one has a thermal bottleneck and is containing the excess heat closer to the source. Thermal adhesives need to be applied so the bond layer is very thin but without any insulating air gaps.
Thanks for the recommendation on the DD driver for host 2.
The extreme drop in output is gone. I’m seeing the following results after fitting the DD driver… PS: I’ve added additional Arctic Silver 5 to all joining threads, it made a big different in keeping the output more stable. The lights does around 7-8 amps with 3200mah Panasonic BDs. Comparing with the values in post 1, the ceiling bounce starts off at around 290 lux and holds it very well. After around 2 minutes (I did not measure the time spot on, sorry) the lux was around 260. Thats a 10% drop, I think its not bad for how hard the light gets pushed. The heat transfer to the body is very well distributed from the neck to the head and tube. I still believe this is a great quality host. The only thing I believe lacks is a nice solid copper pill with much more weight. Would have also been great if one could get a 26650 version, but I’m happy with the light as is.
Regarding the Qlite driver thats performs so bad… I have a C8 running an MT-G2 with 18350s doing close to 6 amps. The C8 does not show the extreme drop in output? Why was the effect so extreme in this host?
Ok, so I’ve found this post from Richard, which answers 1 of my questions…
“I don’t think you’re missing too much! This is not the optimum solution, it is one that was designed to work with existing drivers and still be able to fit into lights with small 17mm driver compartments. With IMR 18350 cells I haven’t seen much any thermal throttling when driving an MT-G2 at around 4.5A when actually installed in a host. With two 20Rs sitting in the open air you see the 7135s begin to thermally regulate (cut current) at less than 30 seconds. The closer the input voltage is to output voltage (look to emitter vF curve) the less heat the 7135s will have to shed, this means that batteries with a little more voltage sag may actually be better for this setup.”
Old-Lumens, regarding my heat sinking question, I’ve found this post from you which answers that question…
“After doing many of these, I feel that the 7135 chips need heat sinking to keep them from getting too hot and dropping out. As to the best way to heat sink them? I do not know. Fujik isn’t enough. It needs to be something like direct to metal, so I use copper and it seems to help, as the light does not dim near as quickly. When I say quickly, I mean that with no heat sinking, pushing the MT-G2 to 9 amps, causes dimming in seconds, where with copper soldered to the outside tabs of the 7135 chips, it lasts for a couple minutes before dimming.”
So, which regulated driver will do well in driving a MT-G2 around 6 amps? On the other hand, its probably just better sticking with the DD driver when pushing high amps. The voltage sag on the batteries will be high and regulation will probably not be sustained for very long. Whats the opinion here?
Disclaimer: I haven't read anything in this thread except for the OP. I can tell you that your 7135s are overheating and derating the power. The adapter and high drain cells are the problem here. With that adapter you aren't getting hardly any thermal interface with the host. Get that driver in contact with your host and switch to some weaker cells and you'll be in good shape.
I have host 2 and 3… I put 22mm BLF driver from RMM in host 2 but never get it to work right.
also having problem with host 3 too, it was fine for awhile then work on and off… :~