Contacting Rofis directly is a dead end. I also tried that. They don’t want to help because they are probably afraid to admit their products are defective. Banggood will sort them out! I’m selling mu other Rofis lights and will replace them with Olights. At least they have proper customer service and quality.
I totally agree with you Kevin. I hope that this subject here will draw enough attention to warn other potential customers before buying the MR30 and maybe to have Rofis reconsider their miserable attitude towards its customers. I will keep you guys updated if and how this subject will be resolved by Banggood's customer service.
Supplemental:
Not in direkt conjunction with my actual problem here but maybe worthwhile to know:
On TLF (German flashlight forum) I have now heard of 3 members having a "dead switch" with their MR30. Maybe more incidents are yet to arise.
I have now uploaded those two videos for further proof and informed Banggood's customer service to resolve my case.
1) My MR30 with the original protected Rofis 21700 battery at 3.86V resting voltage.
2) My MR30 with a standard size 21700 battery from Efest at 3.76V resting voltage.
I just tested mine with 2 different 18650’s.
Protected Sanyo NCR18650GA - Can run turbo, no problem.
Samsung 30Q INR - Too short to make contact with both springs.
If the cell voltage is too low to sustain turbo it should just go to high mode if you double click for turbo. It just bombs out completey.
Are you sure the unprotected 21700 makes contact with both springs? Maybe your springs are shorter than mine. I’m asking because the protected 21700 looks like a protection circuit that trips when the current increases due to the lower voltage.
FWIW, the unprotected Efest battery did work sometimes in the beginning but in most cases it did not. I cannot tell for sure how this is possible but your assumption sounds realistic, i.e. springs might be too short or too compressed to use unprotected cells. 18650s with adapter never worked no matter if they were protected or not. With the Rofis battery the turbo did work until the voltage was about 4.0V. It then stepped down right to the low level after about 3 seconds. I don't think the protection circuit was triggered when I doubleclicked the switch as the same happened with the unprotected Efest battery on those rare occasions when it had contact.
Okay, so check it out and let me know. I have seen in the past many differences from the same model light, like springs, retaining rings, MCPCB’s, glass lenses and even reflectors. My 1st MR30, for example, didn’t have a screw to hold the MCPCB in place. The new one has. And different glue was used on the threads.
Banggood has now asked me to check if the (unprotected?) battery can be charged inside the MR30 via USB charging. I am going to test it tonight.
Have you tried putting a spacer in with the unprotected 21700 cell to see if the light switches on?
Unfortunately, I don’t have an appropriate (big enough) spacer/magnet here and I don’t want to risk any shorts by fiddling with a small size magnet on the negative pole. Don’t get me wrong but I don’t think that using spacers is an acceptable permanent solution. If no one else is having these issues with unprotected 21700 batteries it must be a faulty item.
It’s not a solution, it’s fault finding. And in your case the only way to test the operation of the light, charging and discharging, is by using a spacer with that unprotected cell.
I just don’t see how an unprotected cell can cause the driver not to work, unless it doesn’t make contact (obviously).
The pcb on the Rofis 21700 should handle more current than what the driver can pull on the primary side, to supply the LED with enough current on the secondary side to emit 1600 lumens (without tripping). And it seems like yours trips the pcb indicating to a possible faulty cell protection circuit.
Lights that were designed to be used with a protected cell will always have more pressure on the cell, and in turn better contact, and have less total circuit resistance compared to when you use an unprotected cell with little to nou pressure on the cell. Some lights just have short springs with little compression but high tension.
Use coins as spacers on the negative side of the unprotected 21700.
Also bypass the spring if possible.
Boost drivers want as low of a resistance as possible.
With phosphor bronze springs/BeCu springs, it is not a problem at all at 5-6A.
But with steel springs, it can mean a 0,3V drop accross the spring, which is huge already, and as it heats up significantly, resistance will rise again, making the boost driver pull even more current.
TLDR: Bypass the spring and see if it helps with your problem.
@KevinZA1988: I will try your hint with the coin later on and see what happens. Thanks a lot for your support.
@BlueSwordM: Thanks for the advice. I guess that would require soldering skills /-equipment (that I don't have unfortunately) and it might void Rofis' (worthless) warranty as well as Banggood's promise to resolve the problem.
No you don’t have to. If you can let the one end of the copper braid/wire sit between the cell and the spacer, and try and wedge the other point where the tube meets the retaining ring of the spring.
Other thing you can try as to short out the battery negative and the tube end with the wire using one hand, and turn on the light using the other hand. It’s a bit fiddly with the unprotected cell end sitting deep in the tube, but it’s possible. That will be a direct bypass.
I guess there will always be issues with a production light here and there. I have 3 and they work on Turbo so well that I almost feel they get too hot before they step down. I bought them as gifts for the Holidays. I will look at the springs when I get home to see if there are any differences between them. If you guys have anything else you would like for me to compare between the 3 I have just let me know. I will try and post some pics as well. Never posted any pics here but I should be able to figure it out.
Thank you HONDARIDER!
If you could compare...
1) the voltage sag / turbo sustainability at different voltage levels (4.0V, 3.9V, 3.7V etc.)
2) if unprotected 21700 work without any issues
3) turn-on issues with the side switch (light will not start with click)
...that would be fantastic! :-)
Cheers, Thomas
So have you tried bypassing with a wire yet??
Not yet as I don’t have found an appropriate wire yet and I’m not back home yet. But I will try the advice with the coin asap. :-)
(just noticed too many „yets“...LOL :-D )
It will be a lot easier to just take a piece of wire and short the negative and the tube end, you will get your answer before you start trying to insert spacers and stuff…. Just saying 
Yeah, that makes sense indeed! ;-) I‘ll probably get ahold of some desoldering braid by tomorrow. Would this work to check if the MR30 can use the turbo below 3.9V without the tailcap?
Once again, I really appreciate your great help and your patience with me, guys! :-)
If you have a piece of electrical cord laying around, like a kettle or pc power cord, just strip it and remove one of the cores. A 1.5mm² wire can comfortably handle 16A. If you press it firmly against the negative terminal and the tube end you can run turbo. I’ve done it with my MR70 when I measured tailcap current in turbo. I measured 9A at startup with a fresh cell. The MR30, in theory, should draw about 6A at startup in turbo with a fresh cell.