Problem with Oshpark board

[wight]

@GulfCoast

I turned it on high and immediately went to pin eight with the open end of the + switch wire and it still ramps down.

I really appreciate everyone’s help so far.

I’m not that familiar with lumodrv. Are you certain that it is not the low voltage protection kicking in? If R1 is missing and R2 is present then I’m sure LVP will kick in immediately.

(Also FYI - the values you used for R1 and R2 in your build will prevent LVP from functioning, if 19.1k and 4.7k are properly soldered in place it will simply never kick in.)

[GulfCoast]

It will ramp down after I set level. The ramp down is slower and more deliberate speed than normal ramping.

[Old-Lumens]

@wight

@GulfCoast

I turned it on high and immediately went to pin eight with the open end of the + switch wire and it still ramps down. I really appreciate everyone’s help so far.

I’m not that familiar with lumodrv. Are you certain that it is not the low voltage protection kicking in? If R1 is missing and R2 is present then I’m sure LVP will kick in immediately. (Also FYI - the values you used for R1 and R2 in your build will prevent LVP from functioning, if 19.1k and 4.7k are properly soldered in place it will simply never kick in.)

19.1k and 4.7k are what is on the stock NANJG and Qlite drivers. That’s why they are set for this one too. Basically, it’s just a big NANJG linear driver.

[GulfCoast]

@Old-Lumens

When the light ramps up or down on it’s own, you have a short which acts as the switch being held down, (acts as a complete circuit). You probably have solder bridging, (bridging across two pins, or bridging to ground), somewhere in the circuit, either on one of the the MCU pins, or the switch. I have been this route and it’s why I went to hot air reflow. I also went to using just a tiny bit of solder on every joint, so you could hardly tell it was there at all.

Do you think a sledge hammer would fix it? What kind of hot air reflow set up do you use?

[wight]

@Old-Lumens

@wight

@GulfCoast

I turned it on high and immediately went to pin eight with the open end of the + switch wire and it still ramps down. I really appreciate everyone’s help so far.

I’m not that familiar with lumodrv. Are you certain that it is not the low voltage protection kicking in? If R1 is missing and R2 is present then I’m sure LVP will kick in immediately. (Also FYI - the values you used for R1 and R2 in your build will prevent LVP from functioning, if 19.1k and 4.7k are properly soldered in place it will simply never kick in.)

19.1k and 4.7k are what is on the stock NANJG and Qlite drivers. That’s why they are set for this one too. Basically, it’s just a big NANJG linear driver.

I definitely know what the driver is and how it works. Maybe you mistook my statement about lumodrv; lumodrv is a firmware who’s behavior I am not intimately familiar with. The use of those resistors, properly soldered into place, on a Zener-modded driver of this type will prevent LVP from ever kicking in. [I am unequivocally telling you that.]

The best advice I can currently give is to check the voltage between Pin7 and GND. If it’s less than 1v (like 0v…) then something is wrong. If it’s between 1v and 2v then your voltage divider is operating correctly for the given values and the problem is somewhere else (and LVP will never kick in).

[GulfCoast]

While I’m working on the suggestions in this thread I thought I post a video of what’s going on.

- YouTube

[wight]

@GulfCoast

While I’m working on the suggestions in this thread I thought I post a video of what’s going on.

http://youtu.be/VqWZ02FH0vk

Sure looks like LVP kicking in to me. (but I’ve been wrong before ;))

[Old-Lumens]

@wight

@Old-Lumens

@wight

@GulfCoast

I turned it on high and immediately went to pin eight with the open end of the + switch wire and it still ramps down. I really appreciate everyone’s help so far.

I’m not that familiar with lumodrv. Are you certain that it is not the low voltage protection kicking in? If R1 is missing and R2 is present then I’m sure LVP will kick in immediately. (Also FYI - the values you used for R1 and R2 in your build will prevent LVP from functioning, if 19.1k and 4.7k are properly soldered in place it will simply never kick in.)

19.1k and 4.7k are what is on the stock NANJG and Qlite drivers. That’s why they are set for this one too. Basically, it’s just a big NANJG linear driver.

I definitely know what the driver is and how it works. Maybe you mistook my statement about lumodrv; lumodrv is a firmware who’s behavior I am not intimately familiar with. The use of those resistors, properly soldered into place, on a Zener-modded driver of this type will prevent LVP from ever kicking in. [I am unequivocally telling you that.] The best advice I can currently give is to check the voltage between Pin7 and GND. If it’s less than 1v (like 0v…) then something is wrong. If it’s between 1v and 2v then your voltage divider is operating correctly for the given values and the problem is somewhere else (and LVP will never kick in).

Sorry, I will butt out. I shouldn’t have posted when I really don’t know anything about it.

[wight]

@Old-Lumens

Sorry, I will butt out. I shouldn’t have posted when I really don’t know anything about it.

I don’t think there’s anything to apologize for, or a need to butt out. I’m just letting you know that I know what’s going on there. Adjusting those resistor values is necessary for LVP to function.

I actually mis-read and thought that GulfCoast posted those words. You know just as well as I do that LVP will not function with those resistors in place…

[Sirius9]

Something similar happened to me today, I was upgrading regular NANJG 105C driver with additional AMC chips, added 4 of them, all went good, I tested briefly to see if it puts out light, assembled light and measured ~2.36A o.O wait what!!! I used Samsung unprotected icr18650-26f cell and ncr18650pf (both can certainly deliver more than 3A).
Then I took another NANJG 105C, measured about 2.5xA stock, added one AMC and measured about 2.8A/2.9A so its working, added one more AMC, measured 2.8xA It should be over 3.1A O_O just to see what will happen I added one more AMC (that is 11 chips in total) and again, measured about 2.8-2.9A… :_(
I used short and thick wires, measured amps at LED not frm battery, used charged cell/s!!! Not sure what to do next…