Any kind soul out there willing to help?
Auto-switchover for E-Bike controller
- Thread starter deepak4you
- Start date
geekoftheweek
- Joined Oct 6, 2013
- 1,221
How about eliminating the mosfets and use the relay to switch the power directly? You could use another relay to disconnect the batteries from the circuit. If you want to avoid the extra power used keeping a second relay turned "on" maybe a mechanical latching relay would do the trick.
I don't know transistors too well, but I'm wondering if it has something to do with the circuit still being powered instead of going dead which is what the MOSFET would normally expect to happen.
I don't know transistors too well, but I'm wondering if it has something to do with the circuit still being powered instead of going dead which is what the MOSFET would normally expect to happen.
The rating current of D2 is two times or more of its original current, and the rating voltage of D2 is 100V or more, D2 was designed to avoid when the PFET1 is turn off and the current flows through the body diode to the battery (or maybe you don't need it), the PFET1 and PFET2 are the same type and the rating current at least 3 times of its original current, the rating voltage at least 54V*1.5 times, the Rds(set) is depends on the current, the better is to choose the Rds(set)<20mΩ.
Actually I am not the best advicer on this topic as I have bought a batteries and motor but still not converter. However I have explored the topic enough, to realize the best would stand SEPIC or Čuk converter, and then You need no any relay, all is done automatically. https://en.wikipedia.org/wiki/Single-ended_primary-inductor_converter
Thus, for my case I am sure the first free time I shall elaborate one diy, however its not so simple as seems. One excellent tablet for this aim is LT3757, however seems the very new LTC3871 is much better suited http://www.analog.com/en/about-adi/...ost-dc-dc-controller-increases-available.html
But if any choices had been made and its too late to change the most expensive detail, then problems I can foresee in Your version are: 1) the inductive high amperage load cannot be switched under that load, otherhow plasma on contacture. 2) max amperage is going over 90% of all immagineable relays, and those rest of 10% are unpardonable large. Thus, let study the datasheet of proposed relay very carefully.
P.S. Just looked Your circuit: You are switching not by a relay but by mosfets. That change a things about relay, however it brings into circuit a kind of permanent losses. At for example 1kW=48V x 20A at R(ds)=100mOhm about 2V x20A=50W thermal flux. For my understanding its rather much.
Thus, for my case I am sure the first free time I shall elaborate one diy, however its not so simple as seems. One excellent tablet for this aim is LT3757, however seems the very new LTC3871 is much better suited http://www.analog.com/en/about-adi/...ost-dc-dc-controller-increases-available.html
But if any choices had been made and its too late to change the most expensive detail, then problems I can foresee in Your version are: 1) the inductive high amperage load cannot be switched under that load, otherhow plasma on contacture. 2) max amperage is going over 90% of all immagineable relays, and those rest of 10% are unpardonable large. Thus, let study the datasheet of proposed relay very carefully.
P.S. Just looked Your circuit: You are switching not by a relay but by mosfets. That change a things about relay, however it brings into circuit a kind of permanent losses. At for example 1kW=48V x 20A at R(ds)=100mOhm about 2V x20A=50W thermal flux. For my understanding its rather much.
Last edited:
