GT mini~reflash the latest firmware~

Like BSM said, with a FET light like this, a high drain battery will get you higher output than a high capacity battery.

You basically trade some of that high output for some extra runtime when the voltage gets real low.

Here is a comparison with the Sanyo GA 3500, it’s very similar to the 35E.

As you can see, they are both pretty good. There are just some small differences.

okay great gonna use the 30q then :slight_smile:

I’ve tried measuring the current when the 18650 is housed in a similar kind of battery holder. I think the battery holder’s springs plus the thin wires are causing a massive drop…

UPDATE TO MANUAL

After trying mode 4 and mode 10 it appears that the groups are reversed.

Modes 1-6 use full turbo as 100% (5A)

Modes 7-12 use the top of the ramp as 100% (2.5A)

We definitely need an updated manual to reflect this.

Edit: And here it is.
GT Mini User Manaual
.

The GT mini manual was printed before the “official” manual was sent to lumintop IIRC. So this manual is one they made without really understanding the details.

The manual we sent them has the correct modes.

GT Mini User Manaual

Cool, now we just need to get rid of the old manual links and update them. Thanks. :+1:

Regarding NarsilM versions.

The stock BLF Q8 (and it seems even the currently-produced models) come with NarsilM v1.0
(can be updated to NarsilM v1.2 — I wonder why they didn’t do so from the factory, for this bestselling light;
which fixes the strobes mode to be cyclable)

The BLF GT comes with NarsilM v1.2

The Lumintop GT Mini now comes with NarsilM v1.3

What is changed between v1.3 and v1.2 of NarsilM? Is there a changelog somewhere?

(I notice the Fireflies ROT66 also just comes with NarsilM v1.2 ; any other factory-produced lights that come pre-installed with NarsilM?)

I guess you could say Thorfire is lazy and/or doesn’t want to risk the newer version having a flaw or issue. So they just leave it as it is. It’s works fine as is, it just lacks a few features. It’s also possible that they need someone to tailor M v1.2 to work reliabily on the Q8. I’m not sure.

I don’t know if there is a changelog between M v1.2 and M v1.3. I don’t think any features changed, it was mostly cleaning up code, tweaks, etc… that the user will not see. M v1.0 to M v1.2 is where we had all the changes that the user notices.

Here is a link about NarsilM in general, but it’s not up to date, I think. NarsilM - configurable e-switch firmware for Multi channels, v1.3

No other lights come with NarsilM, although Lexel and Texas_Ace do make drivers that use it if you want to convert an existing light to using it.

I am not entirely sure what changed in 1.3 to be honest. Some back end bug fixes I think.

Somehow I ended up with the NarsilM 1.3 when working on a light and have just been using that ever since.

excellent
thanx

awesome photos

I had today also a GT mini open

as I noticed now also here in the pictures the MCU is not original Atmel, looks like a chineese copy

I had in first Prototype board of fireflies also this MCU and programming it failed

Lexel,

thanks for sharing those pics…

so that means it may not be possible to reprogram the “Chinese clone” of the Atmel ATTiny85 MCU chips in the GT Mini?

I think people have already reflashed them. Maybe those that have done so can tell us which mcu their driver has.

Lumintop contacted me and assured me that they do not buy fakes to save money and said “it is impossible to buy fakes of these we think”.

And to be honest, I don’t disagree with them. It is a strange thing to fake that is for sure, I find it hard to believe there would be a big enough market to make faking this worth it. Particularly when the savings would be so small.

They are looking into what is going on with them, I know they had trouble finding the tiny85 is china at all (along with every other manufacture that has tried to find it pretty much). Them simply being old stock from 10+ years ago is both very possible and probable.

I am not sure what is going on with the MCU myself but I do trust them when they say they did not know about it and they sure did not try to buy fakes. They simply have never complained about the price of doing something the right way when I insist on it, why would they do that with something so important and also so cheap?

I re-flashed the mcu on the prototype they sent me and it worked outside of some connection issues at first. After some alcohol cleaning of both the chip and connector it worked. The flux / solder they use does seem to have some kind of film that makes getting a connection difficult without cleaning.

Doubt that it’s Lumintop’s problem.

It’s probably the supplier which is having fakes mixed in. It’s a major problem in electronics manufacturing.

It is very easy to fake chips by a manufactory with the proper gear

years back big manufacturies added some sort of hard particles above the silicon chip layers so you can not polish each layer open without destroying the first layers
the question is if the copies reach the original chips specs or not

It is quite simple why you can buy those only in China and Taiwan, they get copied in Taiwan and sold to chineese suppliers

I doubt why an original Atmel factory should do spelling error of their size code that is there for ages and all batch markings removed on old stock and then export tens of thousands to a Taiwan chip manufactory and chineese suppliers

copying chips is not that hard, you simply polish each layer of them use a very good mictoscope to make of each layer a template

I never said it could not be done, I said it doesn’t make a lot of sense to copy this particular chip as it does not seem to be very popular due to how hard they are to find (generally popular items will have large stockpiles in china and be easy to get).

Regardless, it does the job and has not had any issues that I know of. That is all that matters. They are looking into the issue for future drivers now.

making this chip in Taiwan costs the factory like 10 cents or less selling those at 60 cents in 1000 pieces is quite a very good profit, as Atmel chips are pretty expensive compared to other chips

Its quite simple for Lumintop to find out if those are original, send pictures of them to Microchip and ask if those are original old stock or not

Why do you think they are made in Taiwan?

Do you have a trusted source?

It is not that simple, they do not even see the chips before the final drivers are delivered, the production house takes care of most of that.

And once they are delivered the drivers are already tested and working, so why go looking for a problem when they already work fine?

It is completely understandable that they would not even think there was a problem. I didn’t even think to check on the prototype.

You didn’t even notice the problem until you had an issue, because that is just how things are, if there is no problem, why go looking for one?

I am not sure what your agenda with this entire discussion is? They work, they are looking into it for future drivers, there is nothing they need to do about the drivers already made at this point.