【convoy】H4 and B35AM options are updated

I’ve noticed that Convoy’s Biscotti driver are locked (to prevent flashing), and some people already mentioned to make a cut with a knife on some pin number. I’ve tried doing the cut on a Biscotti driver, but still couldn’t flash successfully — it could be my “cut” wasn’t deep enough though.

I was wondering, if these bare Biscotti drivers can be fixed with a reflash?

May I ask which Biscotti .hex file to use for re-flashing affected Biscotti drivers (since I now have a few Convoys with Biscotti drivers) and fix the fuse issues or timing issues, if that’s possible?

Thanks for the info.

Though that’s a disappointing one. Please let me explain…
Most of my friends who actually use flashlights are backpackers. They tend to pick headlamps. Always plastic ones; Black Diamond, Petzl and such. Sometimes I try to convince them to get something brigher, more efficient, with better tint, tougher, with more battery….and I always fail because even 92 g of Skilhunt H03 with 18650 cell is a lot for them. Similarly skinny light with 18350 would have a chance. And one that allows to pick tube size - even more so. I see that H1 with 18350 will weigh roughly as much as Skilhunt H03 with 18650. I see little reason to use 18350 then…

Also - I’m a backpacker myself. And a weight weenie on top of that. I have quite a few uses for H1 even as it is now (most importantly: a worklight, after putting a TIR in it) but it won’t become my ultimate headlamp; there will still be significant uses where it’s just too heavy.

Anyway I fully expect that H1 will be on my list of lights worth recommending - but not to everyone. Significantly - not to most of my friends.

Yes. Just cut pin 5 and reflash it as usual.

I just bite the chip leg off the board

The one from Toykeeper’s Repository. (She’s the author :slight_smile: )
Edited: Corrected by TK in the next post.

I’m not sure the copy in trunk is calibrated correctly for Convoy’s drivers. Each flavor of driver has slightly different calibration values. The one I sent Simon is in the Convoy branch instead. However, it’s otherwise generally recommended to use trunk whenever possible, since it’s kept more up to date than individual product branches. It just might not have the calibration values set right for individual products. Those values are still only approximations though, and it’s ideal to measure each driver individually with the tools in the battcheck/ directory if you want things to be exact.

TL;DR: It should work, but the LVP activation point might not be quite right and the battcheck mode might be a little off.

Lucky i messed up the URL then. :person_facepalming: Cheers for the correction and info :slight_smile:

I did this a while back with some free diffuser samples. Works well with the L2 because it has a fairly thick lip on the reflector to hold it in place.

In my opinion Biscotti is what elevates the Convoy C8 and S2+ from great hosts to great lights that I happily recommend to non flashoholics. I’ve used a few Convoy Biscotti purchased from Banggood drivers myself and I haven’t had a bad one yet.

I’d rally like to thank you ToyKeeper for making your firmware freely available. The majority of my lights are running Biscotti, Bistro, or Anduril and I don’t think I would have got into the hobby if it wasn’t for the BLF drivers and your firmware.

The axes of the CIE 1960 colorspace are labeled “u” and “v”. Duv is “delta uv”, or the distance between two coordinates. As used in maukka’s reports, it’s the distance from the blackbody line at the correlated color temperature of the light to the actual color coordinate of the light, or more simply, how red or green the light is.

More than +/- 0.005 is a very large Duv - very obviously red or green, easily noticed even without comparison to another light source. A positive Duv is green, and a negative Duv is red. +0.0075 as seen with this C8/219C sample is extremely green such that the color of everything illuminated with it will appear distorted despite the high CRI.

Almost nobody here likes light that is very green. Some people like light that is noticeably red (and seek out rare LEDs like the Nichia 219B sw45k to get it), but almost everyone will be happy with light that is purely white (Duv less than +/- 0.001).

I hate this UI. I find it difficult to understand why anybody would design a UI like this.

The biggest problem is that changing modes too quickly results in strobe. Accidental strobe is no fun at all, especially if you’re trying to switch from medium to low, so you click the button again quickly after you get high. The combination of a single click to change modes and a double-click as a shortcut to anything is bad, but especially bad if it’s strobe.

Long-press for off also tends to confuse people. If I had a light to someone without giving instructions, they will expect a single press of the button to turn it on, and another press to turn it off. There’s a reason Thrunite, Olight, Acebeam, Emisar, Haikelite, and the popular open source firmwares for e-switches (Narsil and Anduril) all have roughly the same behavior:

  • Single-click: on/off
  • Long-press from off: lowest mode
  • Long-press from on: change brightness
  • Double-click: highest mode

Simon, I imagine you’re buying drivers with this firmware already on them, but do you have any way to get a different user interface on them? This one is very bad.

Do i understand it right. The sideswitch on the H1 has no led behind it?
Why not?
To make the reflector cavity compatible to use TIR optics sound like a very good idea!
This would allow people to change it from throw to flood like they prefer.
And the use of a TIR would allow me to mod the H1 with a TIR backlight like i did with my H03.

the manufacturer is trying the 85V now ,and when it’s updated ,
i will ask he to print V3.0 on the PCB board

Thanks for your explanation.
i will launch H2 H3 H4… in the future,
You made me understand the needs of the users. In terms of weight, I will design a headlight that is more suitable for the user.

Clear, thanks for your explanation. :+1:

Thanks for your suggestion.

I’ve told this to driver manufacturer.
The firmware will be changed later.

The DUV is super high, which means this emitter will look very green :confounded: . Hopefully Simon can source some hi cri emitters with low DUV like SST-20 with FA1, FA4, FD1, FD4 tint bins or Oslon Squares with below the BBL tints.

You can buy the headlamp strap for $1-$2 sometimes they call it “headband”.

yes, i understand,in short , i should order redish LED

Thank you !!!
I’m so happy to read this that i’m going to cry.

Probably, most people will be happy with emitters spot on black body line (duv +0.001 to –0.001). I think, reddishness (or rather rosiness) to duv –0.003 is acceptable (and helpful with recognizing blue and red details), below that I consider it a bit of special use.

In practice, while SST-20 LEDs are promised to be on black body line with tints FA2, FA3, FD2, FD3, they are a bit more yellowish, when used on lower currents. This is why SKV89 recommended to source the tints above (FA1, FA4…), because they can hit the black body line better (especially, when not just bare led is measured, but optics as well).

In practice, I found the FB4 tint of SST-20 very appealing for outside use, if it is driven around 2A, because the tint changes, so it is noticeably rosier, than if multiple same LEDs were driven at a much lower current each.

It is very difficult to source Nichia 219C with proper duv values, and this is more visible for emitters in the 2700K-4000K CCT range. But for me, even a slightly greenish (duv up to +0.003) 219C is fine, if they are in the 5000K-5700K CCT range.

To cover the ‘cool white’ range of CCTs with high CRI emitters, do you consider to get Samsung LH351D emitters (they have 90 CRI for both 5000K and 6000K CCT)?
Those emitters are a bit less throwy, than 219C, but overall, they have somewhat more output (also, it is safe to drive them far past 4A with good heatsinking) - and maybe a bit less green.