Texas Avenger driver is the FET driver that Texas_Ace uses.
Yeah. That last one would be doable though: since we now have access to an 8 cell battery tube(hopefully), we could do this.
We would only need 20A to achieve it, although, and a massive light.
Either they are using the 18-24V input the X9R, which costs a ton of money compared to even the retail MT09R, or it is limited to a max of 12 000 constant lumens, which using 6 XHP70.2, would be very efficient, and doable with only 80W, even counting inefficiencies.
Finally, since we do have dual and triple channels drivers already, would it be easily implemented a dual channel buck/FET driver, switching to the FET once it goes something above 10A for the buck driver?
Would be lovely.
Do you mean 4S2P (like the 32,000lm Imalent DX80)?
Yes, 4S2P.
The TA driver is a 2 channel N-FET (N channel FETās are about 10x lower resistance then P channel) driver.
It is a good general use driver that does everything well and at a reasonable price point but it not the best at any one particular thing.
I generally tend to go towards the jack of all trades option as a rule which is how I would describe the TA driver.
How do you figure only needing 20a? Are you thinking a 8S setup in the 8 cell tube? Yeah, that could work nicely, sadly it is only an option for a very few lights.
In theory it is possible to have a buck and FET driver on the same PCB, although voltage ringing and EMF issues could pose serious issues without a really good design.
Doable but not as simple as it sounds. Not to mention making an already expensive driver even more expensive. Also having room for the components could be hard as well.
Yesā¦ the firmware came with low voltage indicatorā¦ I remember one of my haikelite start blinking while the light is on ā¦ good featureā¦
Ahh, so thatās why you described the TA MT09R as FET+Resistor driven at lower levels.
The P Mosfet acts as a linear regulator which canāt do lower levels as regular linear regulators such as the 7135, right?
That is a very smart idea, but one that does not work well for very very low power levels.
But moonlight does not matter much in a light like the MT09R.
Edit: And yes, I was thinking of the 8 cell tube.
I do not understand the EMF issue though. Why would that cause problems on a 2 channel 10A buck/FET driver? I know that small boost/buck drivers have to deal with that issue very carefully, like the 17mm boost drivers Lexel and Schokli are developing, but why would that cause issues in a 46mm driver?
Well, actually both FETās are n-channel. 1 runs a resistor bank to provide low modes and the other is full power for the high modes.
By adjusting the resistor values and firmware settings I could make the LED just faintly glow if I wanted to, so not suer what you mean by it not being good for low modes? It has basically unlimited range for adjustment of he low modes. The hardest part is the transition between the channels. Took me dozens of hours of tweaking to get that to my liking.
Well, the FET causes massive EMF / voltage ringing. Took me 3 prototypes to get just an FET driver stable at these power levels. Add in a buck driver that needs to be stable on the same PCB and it would be even harder. Space would be a big issue as well.
The larger PCB is nice but it would still be a tight fit. It can be done but it would also take a very well designed pcb to be stable enough for a production light (which is why I try to make all my drivers ready for now days). Factor in production ready build practices and things get even harder. Not impossible but hard.
Oh ok then. I had not understood that you had full control of the low modes.
Thanks for the answers.
Also, related to the electronics, have you noticed that since a few months, high binned SMD capacitors have exploded in price? I have been trying to get my hands on some, but I could not find a source that would not charge an arm and a leg for it. Is there a reason for this?
Yep, both channels are fully controlled.
I have not purchased a lot of high capacity caps recently but I have noticed that the SMD caps have been harder and harder to find and the prices seem to be higher as well.
According to this article
electronic companies mergers in the last years.
They cut overlapping products and long lead times start to hit the market.
āpassive components (like capacitors) with 30 weeks lead timeā
Components donāt fall from the sky, it takes often astonishingly long times to produce them.
The articel is a good reminder of that.
And a newer article
assumes it will last till 2019
Joe
So I got all three haikelite modded by TAā¦ I am very impressed with the driver, specs, and high quality ledsā¦ I decided to test the lumen with the Texas ace lumen tubeā¦ first number is turn on and 2nd is about 20 seconds.
Mt09r triple xhp70.2 5700k. 21000 lumen / 19500 lumen
Mt09r triple xhp35 5700k. 5400 lumen / 5000 lumen
Mt07s xhp70.2 5700k. 7600 lumen / 7000 lumen
Very nice. Iām surprised to see 7600 calibrated lumens for the single XHP70.2. I was expecting more like 6k lumens. How does the MT07S throw compared to the other two? Is it domed or dedomed?
Indeed, it lines up with the numbers I got before the calibration, it was a bit over 10,500 lumens on the old calibration. Normally I would of rated it at ~9000-9500 but could not pass up the chance to say 10k lol.
That P2 LED handles the power really well. The 4000k does really well but it does loose out when you really start pushing it hard in a setup like this. I will still personally take the 4000k 80cri with a bit less lumens but to each their own.
In other semi-related news. The XHP70.2 GTās that sold on ebay yesterday for $495 suddenly promoted lumintop to get in contact with me again and let me know they are officially moving forward with this light after over a year of trying to talk them into it. So thanks to whoever bought those for that!
Still got details to work out before I start a thread for it but it is in the works officially now. They are saying 2 months, which based on the GT GB, I figure 3 months is a good guess lol.
I am also putting my personal XHP70.2 GT up for sale to free up some cash flow. One thing I have learned is that Money goes into China like a greased pig but getting it back out again is like pulling teeth. I still have not been paid the other half from Haikelite, although they said the driver are supposed to be done this week, so hopefully that will happen soon.
I have not put my light together yet so if someone buys it I can build it to your specs and I will sell it for $425 on here since I donāt have to pay ebay fees and I prefer to cut my Family at BLF a deal.
I have 4000k 80cri, 4000k 70cri P2, 5000k N4, 5700k P2 LEDās as options and can dedomed them as well
Anyone interested can PM me.
Lastly lumintop ALSO suddenly showed interest in another project I have been pitching them for almost a year now as well. It would be the biggest and most expensive BLF light so far by a fair margin but boy, would it be a light like nothing else on the marketā¦ā¦
Next up, try to make a CFT90 version of the light 
I already pitched that idea but it got exactly zero traction once they saw the price tag on the LEDās. Canāt say I blame them, at almost $140 each for the LED alone it would cost as much as the light itself!
Now an optic version of the GT is something that thanks to a member on here I do have in the prototyping stage. Depending on what the results are it could turn into a production unit but I kind of doubt it since it needs a wavien collar to work properly.
Great news TA! Really looking forward to seeing what you and Lumintop cook up!
I would be interested if lumintop make a blf gt with cft90ā¦ I do have a blf gt cft 90 but it have ring around the hotspotā¦ maybe if lumintop do minor update on the reflector, and mounting, cft90 will sell good for around $450-$550.
That ring is interesting, I wounder what causes it? Maybe light hitting the edge of the reflector base since it does not have a centering ring?