Testing newer batteries in the PhD-M6 ...

wquiles

Flashaholic
Joined
Jan 10, 2005
Messages
8,507
Location
Texas, USA, Earth
Back in 2008/2009, Eric, myself, and many forum collaborators tested "many" incandescent bulbs while developing the PhD-M6. At the time, we settled on the battery technology we had back then, namely the protected 17670 LiOn cells. Nearly all of the experiments and the fine-tuning of the soft-start, and set voltages were set based on those cells.

I recently sold my personal black edition PhD-M6 to another forum member and upon receiving it, and trying it with new cells, he had an insta-flash with an old/used MN-20. I had (many times) tested that M6, MN20, and 17670 cells until I was sure it was working great, so of course I was disappointed to hear about the insta-flash.

Upon asking a few questions, I learned he was using very-low resistance, high-amperage 16650 batteries, which were not the protected 17670 cells we/I used during development & testing. Although all of my PhD-M6 programming setup is long gone, I still have my true-RMS (AC+DC) Fluke and my old but still trusty Tek scope.

In order for me to know/learn what might be happening, and to help him out on what cells to try different (if any), I purchased 3 different sets of 16650 protected cells, plus a set of non-protected 17670 cells. Based on the testing I have done today:
- Only the protected 16650 cells behave along the original cells the PhD-M6 was developed around
- The high-amperage 16650 cells behave like the unprotected cells - both subjecting the bulb to a higher voltage, and sooner, thus making it more "likely" to insta-flash.

Here is my basic setup and batteries to try out (not in picture are the UltraFire unprotected cells):
20241025_124411.jpg


KeepPower 16650 2100mAh protected cells:
20241025_131912.jpg


Soft-start in action:
20241025_131939.jpg



Orbtronic 16650 protected cells:
20241025_132044.jpg


Soft-start:
20241025_132107.jpg
 
And here is the very high current 16650 cells:
20241025_132209.jpg


Soft-start:
20241025_132230.jpg



and the un-protected 17670 cells:
20241025_134835.jpg


Soft-start:
20241025_134934.jpg




The pack is setup for the MN-20, at 7.6vrms.

Everything in the scope is exactly the same in all cases. And for background, the actual PWM frequency is in that 245Hz range (I don't remember "exactly" what is in the software).

Given the vertical/voltage scale, both the high-current and the unprotected cells hit the bulb with a loaded voltage of about 8volts, which when grows quicker than with the other two bulbs, so the bulb/filament has less time to warm-up before steady state is reached.

I am not saying with 100% certainty that this is why that MN-20 bulb instaflash, but I can be certain that this high-current and the unprotected cells "do" hit the bulb harder and earlier, than the other more "normal" protected 16650 cells. So my recommendation at this point is to NOT use unprotected cells with the PhD-M6, and do not use high-current cells either.

At this point, based on this (limited) testing, the Orbtronic cells would be the ones I would advice using with the PhD-M6 pack.
 
Last edited:
Thanks, Will! Nice writeup and interesting results. Will stick to my protected 16650s. I also wanted to add with some of these aging SF lamp assemblies, the risk of gas escaping is always a possible cause of insta-flash.
 
Thanks, Will! Nice writeup and interesting results. Will stick to my protected 16650s. I also wanted to add with some of these aging SF lamp assemblies, the risk of gas escaping is always a possible cause of insta-flash.
Thank you. Good to know!
 
Hello Will, can you make a specific recommendation as to which cells to buy and where ? My pack is stashed away but I still use my M6´s with the MN15 and MN20 and SF123s. I´d like to use the pack again.
 
Hello Will, can you make a specific recommendation as to which cells to buy and where ? My pack is stashed away but I still use my M6´s with the MN15 and MN20 and SF123s. I´d like to use the pack again.

From his post:
At this point, based on this (limited) testing, the Orbtronic cells would be the ones I would advice using with the PhD-M6 pack.
 
Thanks guys. Is this just a drop in replacement for the 17670-s or are modifications required ? what kind of runtime can we expect with the MN20 ?
 
Yes, the Orbtronic protected 16650 cells are a drop-in replacement to the 17670 cells from 10 years ago.

Yes, longer runtimes as we went from 1600mAh to 2400-2500nAh, or roughly 1.5 times longer runtimes based on battery capacity.
 
Thank you Will. Hey guys wanted to say it has been very nice exchanging with the old guard again. I miss JS anyone know Jim´s whereabouts ? Hope all is fine.
 
Orbtronic protected 16650 cells

These are the cells I'm using in my 16650 application, but based on the info I currently have, don't expect to get the label rating for capacity. It will likely be somewhere South of that.

The Net / Bottom Line: When doing quick calculations on flashlight operating times, the default / ballpark number I will be using for capacity is: ~1750 mAh.

cc: @LED61
 
And here is the very high current 16650 cells:
View attachment 69221

Soft-start:
View attachment 69222


and the un-protected 17670 cells:
View attachment 69223

Soft-start:
View attachment 69224



The pack is setup for the MN-20, at 7.6vrms.

Everything in the scope is exactly the same in all cases. And for background, the actual PWM frequency is in that 245Hz range (I don't remember "exactly" what is in the software).

Given the vertical/voltage scale, both the high-current and the unprotected cells hit the bulb with a loaded voltage of about 8volts, which when grows quicker than with the other two bulbs, so the bulb/filament has less time to warm-up before steady state is reached.

I am not saying with 100% certainty that this is why that MN-20 bulb instaflash, but I can be certain that this high-current and the unprotected cells "do" hit the bulb harder and earlier, than the other more "normal" protected 16650 cells. So my recommendation at this point is to NOT use unprotected cells with the PhD-M6, and do not use high-current cells either.

At this point, based on this (limited) testing, the Orbtronic cells would be the ones I would advice using with the PhD-M6 pack.
Great work. Very useful! Thanks
 
Hello Will, I just noticed that you uped the voltage in your pack that you sold from 7.5V to 7.8V to drive the MN20. I just did a wall test of beam whiteness between the MN15 and MN20, both on new primaries in the M6, and the MN15 is definitely whiter and brighter. IIRC, the MN15 with the two parallel stacks is being driven right around 7.8V, is this what prompted you to change this voltage ? I have a C5 pack with 7.5, and you know I have used even the MN21 with your modified M6R at 7.5 volts, which was awesome but definitely cut into the bulb´s life and as hard or impossible as they are to find, I better not. Anyway there is no way you can modify these packs anymore right ? I would also prefer 7.8V for the MN20 and MN15, and would not mind at all a setting for 7.0V for the MN21.
 
Hello Will, I just noticed that you uped the voltage in your pack that you sold from 7.5V to 7.8V to drive the MN20. I just did a wall test of beam whiteness between the MN15 and MN20, both on new primaries in the M6, and the MN15 is definitely whiter and brighter. IIRC, the MN15 with the two parallel stacks is being driven right around 7.8V, is this what prompted you to change this voltage ? I have a C5 pack with 7.5, and you know I have used even the MN21 with your modified M6R at 7.5 volts, which was awesome but definitely cut into the bulb´s life and as hard or impossible as they are to find, I better not. Anyway there is no way you can modify these packs anymore right ? I would also prefer 7.8V for the MN20 and MN15, and would not mind at all a setting for 7.0V for the MN21.
Great question, but to clarify: I didn't update anything - I lost the ability to re-program packs many years ago now, so I can't reprogram nor make new PhD-M6 packs. Since I still have my true-rms voltmeter (Fluke 189) and Tek scope I can still test and measure the set points, but I can't change them.

That V2=7.8Vrms on the pack I sold recently happen to be the setting that particular pack was programmed years ago. The recommended set point was 7.5Vrms, but since 7.8Vrms was within the range (7.2v to 8.1v) I tested it with the MN20 for a while and found it to be great for the MN20.
 
Top