Am I wrong or does this topic perfectly matches with my thread about the MR30 and my dispute with ROFIS / Banggood's CS team?
https://budgetlightforum.com/t/-/53709
FWIW, Imalent's DN70 boost driver can draw 11.5 amps from the battery and maintains turbo until cell's voltage is down 3.1V, i.e. until the LVP is being triggered. Comparing this to ROFIS' sh***y boost driver that's quite a difference.
Alright JasonWW. I’m gonna play around a bit with copper braid to see if I can somehow reduce the resistance in the tail. It’s still a bit disappointing that I can barely get 5 proper turbo blasts…
Do you know how the Lumintop ODF30/C, Olight R50 Pro Seeker, or other single 26650 XHP70/70.2 compares to the MR70?
Nope. XHP70.2 and boost driver combo can’t be compared to the MR30 with XHP35 and boost driver. XHP70.2 draws a lot of current in turbo. As I tested already, the cell makes a difference, and cleaning the contacts. The MR70 just runs into voltage sag faster. Your MR30 doesn’t seem to be defective. Your cell might be though. 5 full Turbo runs of 3500 lumens on the stock cell ain’t that bad I guess. Other cells just perform a bit better.
Oh man, you make me read your long thread plus a long TLF thread. :person_facepalming: Okay, 40 minutes later…
LOL.
Your MR30 seems different. Since it does not respond to cleaning, etc… I think you have a driver issue or maybe bad contact of the battery tube due to glue.
I wonder if they would send a new driver?
As a quick test, use wire to bypass the tail cap and see if it holds turbo longer. I found the ODF30 tail pad thing did not flex much. When it had soft contact, turbo runtime was greatly reduced. I noticed that turbo lasted much longer when I was measuring amperage at the tail. Bending that metal tab up fixed it.
Here is my ODF30 review with amp draws at different voltages.
Here is a nice video from Mad Max showing how the ODF30 boost driver draws more current as the voltage drops. When the current gets too high the driver steps down to a lower level. So you see how a little extra resistance can lower voltage slightly and this causes the light to draw extra current then step down prematurely. This video shows the basic functioning of all boost drivers. (minus the specific voltages, etc… That of course varies from one boost driver design to another)
The lowest amperage I could get with my ODF30 was 8.35A. Max got 7.6A. This is because of the voltage sag in my battery. I was not at 4.2 volt exactly. It was probably at 3.9 or 4.0 while under load.
Thanks! I’m definitely gonna play around with it tonight. Busy charging at the moment. If I can get it to do about 7-8 full Turbo runs I’ll be happy. One tends to forget just how much light it produces from a single 26650…
You get much longer total turbo run times using the unprotected Shockli 5500 battery, right? Or the same as the protected Rofis battery? 3.78v and 3.92v are very far apart. My lights seem to stay around 3.8v for a very long time. Once it gets below 3.8v I know the battery is mostly dead.
With the Shockli I get 7-8 turbo runs that don’t dip. After that I get slight dips when activating turbo… It’s like I get the full 3500 lumens and after a few seconds it drops to about 2700 to 3000 lumens. Then after the 1min turbo timer kicks in it drops to the 1500 lumens high level.
This turbo mode with drops goes on for about another 3 - 4 times before it just drops straight to high when I try and activate turbo.
With the Rofis cell the same happens, but less of it. Indicating to me that the light experiences more voltage sag with the Rofis cell.
I have a Keeppower 6000mah unprotected high drain cell that’s still quite new and it performs the worst out of all 3 cells. 3 - 4 full Turbo runs then it starts dropping the output in turbo.
I can understand why the Rofis cell does that. the protection circuitry (all of them) add some resistance which lowers voltage.
The KP 6000 is odd, though. It should perform just like the Shockli 5500 since they are supposed to be the same cells. Are they the same length? Are the battery ends clean?
This is where it’s nice to have a clamp style ammeter so you can see how many amps you get at the tail cap.
I have a proper Fluke clamp meter but it’s at work and I’m on study leave for a few days. Will get it this weekend and play with it.
The Keeppower is slightly shorter but a bit fatter than the Shockli. The positive ends also looks different. And this is the Shockli from MTN. I had a pair of Liitokala 5000 mah cells that outperformed the Keeppower on voltage sag but not capacity (as expected) but sold them both in flashlight builds.
Isn’t rule of thumb the higher the capacity, the less current a cell can generally supply?
Kind of. You need to have batteries made around the same time for it to be true.
For example, the Panny B first made in 2012 was 3400mah and could do 4.8A continuous. The Panny GA made in 2015 is 3500mah (about the same capacity), but can do 10A continuous.
Once you start comparing cells made around the same time, within a year or two, then you can compare them more directly. Manufacturers can alter their chemical mix to give desired results. You can trade capacity for higher amps and vice versa.
I think the cyan blue Liitokala 5000mah (20A continuous) came out around mid 2016?. The Shockli 5500mah came out about late 2017 and is closer to 5700mah, but still does the same 20A continuous. The Golisi 4300mah is less capacity, but can do 30A continuous. So newer batteries, in general, tend to be better. Newer cells seem to be extending the capacity while not really losing much current.
A quote from Baditude on the e-cigarette forum. (CDR = continous discharge rate)
Very interesting… So not all cells of the same brand and type are equal. From what I can see with my 3 different 26650 cells is that the Shockli has lowest voltage sag under heavy load.
Gonna check the tail current this weekend and I’ll upload a video of possible. Look, I know there’s probably a reason why this light is limited with turbo runs. I appreciate the fact that it can sustain 1500 lumens for a very long time (even though you have to bump it up from 1000 lumens after 6min) and that is very handy for such a compact light.
The downside is that after using it on high mode for an hour or so starting with a fresh cell, you won’t get 1 single turbo run from it. I’m looking at the Acebeam EC50 GEN III now… Hope to find some goos info on how that performs, especially as the cell depletes.
I still think that a lower capacity cell with better voltage sag characteristics will perform better in a single cell power hungry XHP70.2 light.
Correct, each manufacturer (at least the big ones) tries to produce several models of 18650, 21700, etc… Occasionally you will see similar cells like the Panny B 3400 and Panny GA 3500, but this is a case of a new model superceding on older one. Also the Samsung 48G being replaced by the 50E. Over time the chemical mixtures get refined and you get better overall cells. Or like in the case of the VTC5D, that was a cell requested by the Dyson Vacuum company for their battery powered vacuums and they had very specific requirements for it.
This is normal. Take for example a FET based driver. You can run it at an hour on a middle level. Even though you can still activate turbo, it’s a reduced output turbo no where near what it was when the cell was fully charged.
Look at a Buck driver. When the cell is fully charged it can hold its turbo for a while, but once the voltage gets to a certain point it falls out of regulation and you get reduced turbo output.
Like other driver designs, boost drivers have their advantages and disadvantages. They can maintain low to med levels much longer than other driver designs. They can also be more efficient than most other designs.
I think the reason why the MR70 is limited with turbo runs is poor design or a defective driver. Another possible reason is they are trying to get too much output from the emitter. Thsee boost lights trying to get 3500lm, 3800lm, 4000lm, just ask for more battery current and that limits their turbo run time.
If you want longer turbo run time, lower the requirements. Look for lights outputting 2500-3000lm. They should be able to do that at lower voltages. Also make sure to use the best battery. No protected cells. Look for lights with factory double springs or bypassed springs.
Personally I’m waiting for Lexels or Richards boost drivers to became available. They are adjustable for turbo output and should run nice ramping UI like Narsil or RampingIOS. I’ve already got the host light ready to go.
Oh yeah, Sofirn just released a new version of the SP33 light. It’s a 26650 boost driver design, 5 levels, one click on/off, press and hold to cycle the levels. It uses an xhp50.2 rated at 2500lm turbo and has no built in charging, but it’s only $23. Seems pretty sweet. Mine is on the way to me. 