It's important to note that Bosch motors (and Shimano/Brose/etc) can deliver much more than 250w,
It is more important to establish the difference between peak and nominal power ratings. The definition of nominal is the power that can be sustained indefinitely, peak power is the maximum power that can be reached but for only a limited period of time. To keep a high speed one needs a high sustained power which is given as 250W. Brose, bosch etc can all peak around 600W, but this is for short periods , they don't sustain these values for a long time. For mountain biking where it is about small bursts of power to go over obstacles etc. they work very well, for high speeds where one needs to have this power continuously they don't.
A top model performance line Bosch produces 85Nm of torque, and it's been measured and tested by several as pretty accurate. A 1000w Bafang is claimed to produce 160Nm of torque, but it is widely accepted that it is exaggerated.
First of all that torque value at the crank does not mean much by itself and gives no information about power without the rpm information. I wish people were more careful when using these notions, I am seeing this mistake of erroneously using them interchangeably .
A bmw motorcycle having only 140nm torque can produce 100KW. The so called 85nm gen4 motor is not even a toy next to that motor. Even at its peak it barely reaches %0.6 percent of BMW's power output while the torque at the crank is %60 of BMW's. A 90kg human being putting all its weight on one pedal can generate 150nm of torque and it says nothing about the power this person can generate.
Given how these motors work, the max torque is achievable at only low cadences, when cadence gets higher you will hit the power limit (power=torque x rpm, at low rpms torque limit will dominate at high rpms power limit will be the limiting factor). For mountain biking where there are lots of start/stops and spending time at lower cadences, torque values can make a big difference. For high speed riding, where it is about keeping a continuous high output power for a long time at higher cadences these torque numbers don't make any difference whatsoever.
In terms of Bafang Ultra I have seen a max torque/rpm graph with the stock controller and 48v source (I think it was in a Frey's doc). It seems to achieve 160nm only at "0 rpm" but in the usable cadence range seems to be more in the 120nm range. This is another reason why these so called "max" torque values are meaningless. According to bosch's own graph however the controller is tuned so that it is pretty much flat in the 0-65 rpm.
That being said It seems like bafang can sustain 750W+ stock indefinitely hence for high speed it is still much more powerful than bosch gen 4, on the flip side it is still a toy compared to the BMW motor that we have mentioned previously.
it's been measured and tested by several as pretty accurate.
While I don't doubt the 85nm figure by bosch, please point me out a third party test which measured power output/torque vs cadence of gen 4 bosch motor. I couldn't find it on the net. These tests are pretty easy to make but no one seems to bother(or maybe they want to be in good terms with companies).
If we could truly derestrict a Gen 4 Bosch, you could most certainly make a 40+mph, 1000+w fire breather that keeps up with anything on the market, and was probably built on a much better base bike to start with.
This tech is mature and I don't buy this "hidden potential" thing. While it is a quality products, I am sorry but there is nothing miraculous going on in here.
Certainly? I would say unlikely. Continuous 1000W+ with the given controller and the 36v battery from that motor(for any light motor) is a lot. I would suspect that It will be a fire breather in the literal sense, since you still need to dissipate at least 300+Watts of heat in the best case. That motor seems to be designed to be a light motor for the given peak/nominal power output, to deal with less dissipated heat and it is very good at that, however that does not mean that it would also be great at sustaining much higher power outputs for very long periods. With custom controller, some mods and 48-52v batteries things may be different.
Many bike hackers have done it with the older Gen 2's.
Can you please point me to such projects where people modded a gen 2 motor for higher continuous power? I searched but couldn't find any.
Actually I thought gen 2 may be a better candidate for a high power mod since it is larger and may handle heat better if controller and battery is also updated. In the future, I would love to invest in a higher end controller and a larger 48v battery to turn it into a 500W nominal motor.