SFT70 vs SFT25, Measurement Test on Convoy M2

I modded my Convoy M2 from 3V8A SFT-25R 6500K to 6V5A SFT-70X Gen2 5000K. I prefer throwers, so I measured lux with phone at 5m to roughly get the candela, 3V8A SFT25 SMO reflector got 100kcd, 6V5A SFT70.2 SMO/OP reflector got 65kcd. Considering the SMO and OP reflectors show almost no difference in candela for this use case, I ended up choosing the OP one to get better beam pattern. SFT70 gen 2 doesn’t improve much over the gen1. I’ll continue to call it SFT70 for short.

Back to the topic, generally if you are considering SFT70 vs SFT25 and their matched driver features in a single cell light, you need to weigh these trade-offs.

6V boost driver SFT70: Larger hotspot but shorter throw, better regulation (maintains high output even at low battery)

3V buck driver SFT25: Longer throw but smaller hotspot, relatively poor regulation (output gradually decreases at low battery)

However, these are just rough summaries. While a boost driver can theoretically maintain regulation, this isn’t always the case in reality. The SFT25 output will decrease as the battery voltage decreases, but it may still be very throwy. So I did some measurements to get more specific info.


In case too many curves to see clearly

The test was in water to prevent thermal control kicking in. However, the Convoy 6V5A SFT70 still gradually dropped the turbo to a relatively low level within 30s, and around 15mins, the output decreased to a lower level. I tried re-turning on the light near the 15mins, and it regained full power turbo, but still inevitably dropped afterward. Testing with batteries at different voltages also showed that when below 3.6V (abt 30% capacity, according to this), the Convoy 6V5A cannot maintain full power turbo whatsoever. I didn’t re-turn on the 3V8A SFT25 in the test because I tried it before and it would recover a little, but it basically followed the curve.

Overall, Convoy 6V5A SFT70 maintains full power turbo only when the battery voltage exceeds 3.6V about 3.7V. After that, it gradually decreases, though the drop is not significant, lumens from 100% to 56%, ANSI throw from 500m to 400m. Convoy 3V8A SFT25 will gradually decrease output from the start, with lumens from 100% to 36%, ANSI throw from 630m to 370m. However, throughout the test, it thrown farther than SFT70 for most of the time.

I didn’t take beamshots, but if you want a more realistic reference I would recommend check out the @Weerapat_K test( SFT70 vs SFT40 | Pioneman K35). Not exactly SFT70 vs SFT25, but close.

Note: My light uses uncoated sapphire lens with lower transmittance instead of the default AR-coated glass. If you use Convoy’s standard setup, you can actually achieve higher output, approximately 3V8A SFT25: 110-120kcd; 6V5A SFT70: 70-75kcd. But the battery I used is Ampace JP30, it’s a tabless battery with ultra-low IR and ultra-high CDR. If you use any non-tabless battery, your output curve may decrease prematurely.

9 Thanks

beamshots, I would say within 300m, SFT70 gives me a better feeling. But it does struggle to reach very far distances, and the massive spill illuminates the air, somewhat blocking visibility.

4 Thanks

Where did you find sft70 gen 2 5000K. I can only find 6500K. Is there 3000k version, or there is not big difference to old 3000k?

I bought it on Taobao in China.
SFT70 gen 2 also has 3000K option, the full name is SFT-70X-WGH, refer to this datasheet, But please note that other CCT are unavailable for now, and may be for a long time to come.

Thanks for the quick reply, and data sheet, missed it. I’ll stay looking for 4000 or 3000K, or go with the gen 1.