There was a comment on reddit where Simon said he was out of stock on hosts. Simon mentioned on WhatsApp that he’d use shorter screws.
To fully fix the LHP531 version it needs a redesign of the mcpcb and maybe the driver? Then dual springs most likely need to be added by us still. Without dual springs they overheat with high drain cells.
I think he intends the single springs to be the main current limiter
There are multiple posts in this thread which are still there and they claim the driver was built incorrectly. When, clearly that is wrong. So, I would suggest that you do not go off making posts across all platforms about the driver being in error until you discuss with Simon and Convoy.
My outside the box thinking, the guy probably has to deal with a lot of garbage. Let’s not make it worse for him. In addition, you don’t want to put incorrect information out there when someone new may come along and see that.
I only asked a question about the driver. Simon said he’d redesign the mcpcb, so I was questioning whether changes on the driver were needed to accommodate, that is all.
“I will design the seven LEDs in a fully parallel configuration, rather than dividing them into three circuits.”
While there’s nothing wrong with the driver using higher Vf emitters, it’s clearly not suitable for use with the LHP531 without changes to mcpcb?
Currently, high drain cells are desoldering wires on the driver using high current cells, in their factory state, no mods.
I find it very difficult to believe that wires are de-soldering because of the batteries. But, I’m willing to test it. When I have an opportunity I’ll test my 3x21c which has SFT42r emitters. I’ll use the stock single springs, and 50Pl batteries. By your statements, it basically will fall apart right?
No, you need low Vf emitters like the LHP531.
Here’s an image by a fellow reviewer who had this happen. He switched out his 50PL to P45B cells and the problem went away.
What am I missing here, because the Koef tests show LHP531 rated voltage is between 2.6V to 3.2 V; SFT42R data sheet says 2.5V to 3.1V.
Edit: could that photo you posted have been an isolated improper solder? I’m not following how a solder blob would 1) reach a temperature high enough to de-solder itself while in use (yet none of the other wires de-soldered) and then 2) somehow re-solder itself while using a different battery?
You keep posting these things then claim that Convoy has a defective product. I’m sorry but I don’t see it.
They certainly do the job, but at the cost of them overheating and collapsing.
I’d recommend lower current cells, something like 25 CDR, and adding a note to the description stating 30 sec max bursts on FET Turbo.
Have you not being following the whole 3X21C discussion? It was concluded by others that the emitter mcpcb was most likely to be the cause of failure. One wire set is powering 3x, while the other two wire sets are only powering 2x. That desoldered wire in the photo is the negative from the set to the 3x emitters.
No, he resoldered it. Tried again with 50PL’s which half desoldered again, then resoldered and switched to P45B’s.
You still haven’t answered my inquiry about the fact that what you’re saying is that the wire is de-soldering itself with a high drain battery but fixes itself with a non-high drain battery?
All I can say is none of this makes sense but I’m happy to test. Unfortunately I bent the factory springs on mine while adapting it for a 46950 battery. More springs will be in my next order. I’ll happily try to burn up my own light to find out if this makes any sense.
It makes a lot of sense. You have one set/pair of wires powering 3x emitters, they’ll be pushing higher current than the other pairs only powering 2x emitters, thus overheating with higher current cells.
I guess without altering anything, you could use lower gauge wire on that pair.
Not sure what to tell you. Still doesn’t make sense. All things considered about the light, it should be fine. A good solder joint with a conductor of this gauge doesn’t just “desolder” itself.
Which, I won’t claim to be an expert by any means. But the solder in that photo you posted looks like a cold joint to me. Ask me how I know.
Happy to agree to disagree and move on. I really didn’t want to flood this thread again about the 3X21C when it’s all been discussed. Simon agreed to redesign the mcpcb because there is an issue using only the LHP531.
Click on my linked comment earlier showing four other people confirming this.
Happy to discuss further via PM.
I have two kinds of anduril driver for S21E,
one for S21E SST40 ,max 8.5A
another for S21E 519A ,max 6A
The logistics company will change the label.
Anduril 2 would fix the temperature issues with 3x21c… just saying 
What temperature issues?
Anduril wouldn’t help anything when the issue only occurs on momentary fet turbo. We believe the wire desoldering is caused by the LED mcpcb configuration where one pair of wires are powering three emitters. Simon will sort this out in time.
Some unrelated good news: the new black 5A driver is a substantial upgrade as an NiMH driver, compared to the old red 1.5A version. Moonlight went down from 20mA to 15mA (measured at battery), and turbo is now 50% brighter.
Right now I’m running it with SFT40 3000K and comfortably pumping out 200-250 lumens (estimated), which means the drive current at the emitter is close to 1A.
The new slim T3 tube feels much better than the old thick one, which made the light feel slippery given how stiff the switch is. The new tube stays in the hand much better and does not slip when the switch is pressed.
I get the feeling you really don’t like Anduril 2 
Indeed. Wearable devices have to be light weight. Here about H1. I guess Convoys 14500 model will weight like 18650. So what that it is cheap, like half price of H04 . But it so annoying . It looks similar but have too many flaws .
