OK, so I intend to build a Li Ion battery inside an original battery pack that had either Ni Cad or Ni MH cells installed, I have both available.
Initial measurements lead me to believe that I can get a 7S6P array of 18650 cells in a casing with some room left over for a BMS, wiring and packing. Perhaps a bit ambitious but hey, what are targets for?
Then we come to deciding what cells to use. My thought at the moment is to use Samsung 30Q 3000mAh cells with a peak output of 20A. In a six parallel array this should give a possible 120A before too much voltage sag. My guess is that the peak instantaneous demand from the bike would be somewhere in this region but this is still to be determined. I can probably get higher capacity cells but 18aH should be more than enough and I suspect the higher output cells will be more robust both in terms of load performance and longevity.
But, what is the load demand? Just how much power does the motor need? I have been unable to get hold of a service manual and even if it gives a specification of the motor, I know from experience that the figure quoted is "nominal". It has to be really. So, what I intend to do is this. I have made up a simple SLA battery pack with a capacity of 5Ah and I have fitted is with some instrumentation wiring connected to a battery powered data logger. With this I can record the current a voltage output of the battery whilst riding under a variety of conditions. If I am able I will try to post some useful graphs when I've done it.
Now, many will say why bother? The bike is known to have a 250w motor and a 24V battery so the current must be no more than about 10A. It happens that during my career, I have been involved in battery powered motor system design for people and products that are well known and successful. The success was not accidental, it was design and technology that worked - ask Elon Musk.
I will obviously need to see the real time measurements recorded on the bike battery motor but from experience I would not be surprised to see peak currents in the range of five to ten times the quoted demand. It will very much depend on the controller sophistication. OK, these current peaks may be very short term and could probably be ignored without seriously compromising the performance of the bike, but, they will have an impact in response times and efficiency, certainly noticeable and possibly significant.
I'll post some more when I get a bit more detail on this but for now I have a question. I find this sort of thing fascinating and love the research involved but know that many find it unutterably boring. So, readers, are you interested? Should I post more detail and discuss the findings? I won't be offended if the general opinion is that I'm a BOF, it would never do for us all to be the same.
Initial measurements lead me to believe that I can get a 7S6P array of 18650 cells in a casing with some room left over for a BMS, wiring and packing. Perhaps a bit ambitious but hey, what are targets for?
Then we come to deciding what cells to use. My thought at the moment is to use Samsung 30Q 3000mAh cells with a peak output of 20A. In a six parallel array this should give a possible 120A before too much voltage sag. My guess is that the peak instantaneous demand from the bike would be somewhere in this region but this is still to be determined. I can probably get higher capacity cells but 18aH should be more than enough and I suspect the higher output cells will be more robust both in terms of load performance and longevity.
But, what is the load demand? Just how much power does the motor need? I have been unable to get hold of a service manual and even if it gives a specification of the motor, I know from experience that the figure quoted is "nominal". It has to be really. So, what I intend to do is this. I have made up a simple SLA battery pack with a capacity of 5Ah and I have fitted is with some instrumentation wiring connected to a battery powered data logger. With this I can record the current a voltage output of the battery whilst riding under a variety of conditions. If I am able I will try to post some useful graphs when I've done it.
Now, many will say why bother? The bike is known to have a 250w motor and a 24V battery so the current must be no more than about 10A. It happens that during my career, I have been involved in battery powered motor system design for people and products that are well known and successful. The success was not accidental, it was design and technology that worked - ask Elon Musk.
I will obviously need to see the real time measurements recorded on the bike battery motor but from experience I would not be surprised to see peak currents in the range of five to ten times the quoted demand. It will very much depend on the controller sophistication. OK, these current peaks may be very short term and could probably be ignored without seriously compromising the performance of the bike, but, they will have an impact in response times and efficiency, certainly noticeable and possibly significant.
I'll post some more when I get a bit more detail on this but for now I have a question. I find this sort of thing fascinating and love the research involved but know that many find it unutterably boring. So, readers, are you interested? Should I post more detail and discuss the findings? I won't be offended if the general opinion is that I'm a BOF, it would never do for us all to be the same.
