jellochaos

I was searching for a way to design a simple synchronous rectifier with N-ch mosfets, to avoid the voltage drop of a diode bridge. I found some interesting ideas, a little expensive though. Now I found this "ideal diode" from Linear. http://www.linear.com/pc/productDetail.jsp?navId=H0,C1,C1079,P85544 It is similar to the way a synchronous rectifier works, it has a circuit that controls a mosfet's gate. It also seems like the simplest way of making an active rectifier! But I also wonder why they don't mention that in the datasheet. Would it be possible to use 4 of those ideal diodes as a bridge rectifier with low losses?

I haven't tested it with a lower resistance but I have tested it with a higher valued one. The voltage varies with the load. The higher the resistance, the higher the voltage. But with a higher resistance there was less current flowing. So I can connect the LED diode directly to the rectifier and the voltage of the dynamo will become equal to the forward voltage of the diode? Which is 3V in my case. Is there a way to calculate the current that will be fed to the LED? But even if it's possible (it seems possible), I will keep the LED driver because I can precisely set the LED current and use its PWM dimming option. Also, I have 3 of them. :p I will remove the 5V voltage regulator which I intended to put before the LED driver. It seems useless and I don't know why I put it. If I operate the LM3405A, the led driver, with a DC voltage which is not constant (13V-18V), will it be able to maintain a constant LED current? I don't see why not but an other opinion won't hurt me. Thanks

Here is a table I made when I tested the dynamo with a 30Ω load. It's just basic measurements to aid me in the initial design. It seems that in high speeds the output doesn't change significantly. Maybe it has to do with the coil's impendance increasing as the frequency of the AC current increases with speed. But I'm just guessing, I don't know much about generators. Anyway, I am looking for a spice model for the LM2907 to test it. I think it's the best choice because frequency monitoring seems to me the most reliable way to determine when to make the switch.

I just found that National Semiconductors makes the LM2907 which is a frequency to voltage converter. I didn't read its datasheet extensively but I think it can be used as a switch that provides energy to a load (a relay in my case) when the input frequency is above a set level. I don't know if it's the most efficient way but it sounds like my best bet.

But wouldn't that overcharge the battery? When it's fully charged I will have to manually disconnect it. Also, a zener doesn't sound that efficient... I am planning on using the MAX712 to charge my NiMH batteries when the circuit is on the "charge" state. When the circuit is on the "Light from dynamo" state, I will feed the rectified output to a 5V switching voltage regulator (~90% efficiency), then to a current mode led driver (~85% efficiency) which will drive a CREE XP-G R5 white led at 0.7A probably. When it's on the "Light from the batteries" state, the batteries will supply current to the LED driver. But I can't think of a low components, energy efficient way to make that switch from the dynamo to the batteries and vice versa at a certain speed.

Hullo, I am designing my first "serious" circuit. I will use it to manage the power from a bicycle dynamo (permanent magnet generator?). I want it to switch from the dynamo to a backup battery when the output of the dynamo is not sufficient to drive a high power led and some other things. First of all, I will use a full wave bridge to get a DC output. Then, I was thinking of using a relay followed by a comparator with a fixed voltage reference to make the switching from the dynamo to battery and vice versa. To help you help me, At 8km/h no load the output voltage is 15V. At 8km/h with load the output voltage is 8V. At 12km/h no load the output voltage is 17V. At 12km/h with load the output voltage is 10V. Let's say I set the threshold level of the comparator to 10V. Under 10V, the battery is employed. The problem: The bicycle is still (0km/h) [The comparator sees 0V from the dynamo and switches to the battery. No problem here.] The bicycle accelerates to e.g 8km/h and maintains this speed [The dynamo has no load and its voltage is 15V, the comparator+relay switches to the dynamo. Now the dynamo is under load and its voltage drops to 8V, the comparator+relay now switches to the battery. The dynamo has no load and its voltage rises to 15V, the comparator+relay switches to the dynamo. That is the problem I am facing. If my logic is correct, this oscillation will keep happening as long as my speed is below 12km/h.] The bicycle accelerates to e.g 15km/h and maintains this speed [The dynamo's output is above the threshold level, the comparator+relay switches to the dynamo. The dynamo's output WITH load is still above the threshold level. No problem here.] The bicycle decelerates to 8km/h and maintains this speed [Like before, the same problem with the oscillation occurs.] The bicycle decelerates to 0km/h and stays still [The comparator sees 0V from the dynamo and switches to the battery. No problem here.] In conclusion, the problem is when the bicycle's speed is 0km/h<U<12km/h. What can I do to overcome this? I hope that my question is clear because english is not my native language. If something is not clear or wrong please point it out.