Posted October 22, 20186 yr Hi. I´ve recently bought a VP 3000-24 whith a 25.5 V 200 Ah lithium battery. Solar panels are 5, 60 cells 265 w,, and the charger is a 50 A PWM. I could see dinamic response is very, very slow. Does anybody know if there is a firmware to change it? PWM is 300 Hz, but the Duty Cicle increases/decreases very slowly. That is a problem when de 18650 bank is charged to 28 Volts and a big load is disconected,
October 23, 20186 yr Welcome Javi I am not sure how quick the Axperts are suppose to respond, but @Chris Hobsonknows the Axperts well. Maybe he can answer this question. I guess the PWM might be the culprit as a PWm is slower than a MPPT to my knowledge.
October 23, 20186 yr I am afraid I know little of the PWM units. From what I understand the duty cycle would determine the charging voltage so it should not be changing rapidly. The charging current would be determined the resistance or a limitation imposed by BMS or the SCC. So if you had a large load that disconnects I can understand that the SCC would be slow to back off. One could lower the float voltage to @weber / @Coulomb recommendation to give one a little headroom.
October 23, 20186 yr 10 minutes ago, Chris Hobson said: duty cycle would determine the charging voltage How it is supposed to work, is the difference between the target voltage and the actual voltage represents an error factor and depending on whether that error is positive or negative (over- or undervoltage) you adjust the duty cycle (closed control loop as it is called). Technically a larger error should also lead you to adjust faster, though firmware designers might opt not to do that and simply adjust linearly (evil!). For various reasons, even if you do adjust according to the size of the error, you may want to dampen the adjustment, usually to avoid oscillating: where you back off too much, then overshoot again, repeatedly until you finally settle back at the required level some cycles later. The typical way to do this is using a low-pass filter, where you feed back only a part of the error, for example you may decide to feed back 60% of the error. If you had a 10A overshoot on the current, that would mean that you go back to 4A, then 1.6A then 600mA, and then finally get there. So for a short bit of time, especially with very fully charged lithiums, you will overshoot. And it all comes down to how the firmware in the charge controller does it, and if you cannot fix the firmware, your best bet will be to get an external MPPT.
October 23, 20186 yr Author 4 hours ago, Chris Hobson said: I am afraid I know little of the PWM units. From what I understand the duty cycle would determine the charging voltage so it should not be changing rapidly. The charging current would be determined the resistance or a limitation imposed by BMS or the SCC. So if you had a large load that disconnects I can understand that the SCC would be slow to back off. One could lower the float voltage to @weber / @Coulomb recommendation to give one a little headroom. PWM charger conects and disconects solar panel from battery. For example, if solar panels are able to give 50 A at the battery voltage, and the duty cycle is 20%, the charge will be 50x0.2=10 A. It does 300 times per second.
October 23, 20186 yr Author 4 hours ago, plonkster said: How it is supposed to work, is the difference between the target voltage and the actual voltage represents an error factor and depending on whether that error is positive or negative (over- or undervoltage) you adjust the duty cycle (closed control loop as it is called). Technically a larger error should also lead you to adjust faster, though firmware designers might opt not to do that and simply adjust linearly (evil!). For various reasons, even if you do adjust according to the size of the error, you may want to dampen the adjustment, usually to avoid oscillating: where you back off too much, then overshoot again, repeatedly until you finally settle back at the required level some cycles later. The typical way to do this is using a low-pass filter, where you feed back only a part of the error, for example you may decide to feed back 60% of the error. If you had a 10A overshoot on the current, that would mean that you go back to 4A, then 1.6A then 600mA, and then finally get there. So for a short bit of time, especially with very fully charged lithiums, you will overshoot. And it all comes down to how the firmware in the charge controller does it, and if you cannot fix the firmware, your best bet will be to get an external MPPT. Yes, at the end i had tu use a Xantrex C60, other PWM charger, which is faster than the Axpert.
October 23, 20186 yr Author 4 hours ago, plonkster said: I'd love to put a scope on that thing though, to see how it really adjusts... I see you use MPPT chargers, so you wouldnt see that in a scope...
October 23, 20186 yr 5 hours ago, Javi Martínez said: I see you use MPPT chargers, so you wouldnt see that in a scope... You scope the gate connection of the FET (the upper one if it is a synchronous model). Can do the same with a PWM model. Then you can see the ratio adjust visually :-)
Join the conversation
You can post now and register later. If you have an account, sign in now to post with your account.