In preparation for my next MOSFET tutorial, I'm using the DCOI driver with an added bootstrap capacitor to drive the high side N-channel MOSFET
By Julian
Youtuber, shed dweller, solar charge controller aficionado
Good morning all…
Youtuber, shed dweller, solar charge controller aficionado
This site uses Akismet to reduce spam. Learn how your comment data is processed.
Full schematic please
Hello! These MOSFET videos helped me tremendously. I have a question, and if you can help me, I would REALLY appreciate it. What I'm trying to do is make a circuit that will auto-range a panel meter. Using a resistor divider of 900k, 90k, and 10k, I can give the meter 3 ranges, and move the decimal point with some logic. It seems the easiest way to switch the meter to different resistors would be with some reed relays, but I want to use something more elegant than that — something silicon. (note, the meter doesn't like the low side being switched, it wants that to be ground. So I'm forced to switch the high side of the meter's input). I was able to make it work with an N-channel MOSFET, but I had to pull the gate voltage high enough with a 9v battery, like you showed in these videos. (Thanks for the tip!) The meter was accurate (Including the least significant digit!) with the MOSFET both on (low range), and off (High range). But for a final design, I don't want 9V batteries on every MOSFET, and I'm not sure all the circuitry involved with the opto-coupled bootstrapped capacitor would be considered an 'elegant' way to do this. One idea I had was to simply use FET optocouplers to move the high side of the meter between resistors. Do you think they'll be able to replace the functionality of reed relays? Or do you have any other ideas? Thanks for reading, I hope I can find some help. 🙂
Nice one. I am now playing around with that some time and there are 2 things which I am currently wrapping my head around. When powering the circuit without the the arduino signal, the gate of the mosfet seems to be supplied with 12V, slowly dropping. I assume this comes from the cap. So the mosfet puts some power on that point, which can be a problem. Also, when turning on the mosfet, the 24V from the bootstrap start to drop slowly, depending on the capacitor size that i used. In my test with a 2.2uF it takes somewhat around a minute. The problem when increasing the cap is that when going too high (tested with a 2200uF) the cap does not supply 24V, but only around 15V. When the switching time is not too long, it slowly ramps up to 24V. Any ideas on how to work around that?
I believe the capacitor does rise to 24V. This is what I believe is happening. There are two stages: when the circuit is first activated (the bootstrap) and after bootstrap (normal operation). When both optos are off the capacitor poles are at +12V and -12V respectively (assuming the capacitor has discharged from 24V that is might have been charged with previously) and gate and source are at 0V. When opto1 first comes on during bootstrap gate is connected to the positive pole of the power supply and rises to +12V and source drops to -12V. The circuit is balanced. When opto1 goes off and opto2 comes on gate is now connected to the negative pole of the power supply which creates an imbalance in the circuit: the positive side is +12V in the capacitor, the negative side is -12V in the capacitor, +12V on gate and -12V on source. The negative side of the circuit are now at 0V which is not stable (it should be at -12V) and has to be rectified. The most obvious way would be for the +12V of the gate and the -12V of the capacitor to neutralize one another which would constitute as a discharge of the capacitor. This is however made impossible by the diode on the positive side: the only way the positive side of the capacitor can discharge is for negative charge to be attracted from the positive pole of the power supply. The positive pole does have negative charge, it just has more positive charge than negative charge. However, the diode makes this attraction impossible which means that gate and the negative side of the capacitor also cannot attract one another. The imbalance on the negative side and the stability of the capacitor charge break the bond between gate and source. -12V is attracted from the negative pole of the power supply but this movement of charge also happens on the positive pole. This makes the capacitor rise to +24V and -24V respectively. But on the negative side -24V is needed into total: -12V to drop the capacitor to -24V and -12V to bring gate to 0V. The second -12V comes from the source which brings both gate and source to 0V. I haven't tried this circuit myself so I'm not sure if this is correct but based on my understanding the mosfet does not switch on during the first cycle (during the bootstrap that is). Can you confirm this? Anyways, this is the end of the bootstrap cycle. When the normal cycle starts the capacitor is at +24V on the positive side. When opto1 comes on again gate rises from 0V to +24V and switches on the mosfet. When opto2 comes on gate is connected to the negative pole and drops to 0V which turns off the mosfet. If you measure the voltage across the capacitor you should be reading that at first the capacitor sits at 12V with both optos off and when opto1 and opto2 cycle the capacitor rises to 24V and stays there. Can you confirm this?
Awesome job on your videos Julian! I have enjoyed them immensely.
Any idea on the upper limit of the switching frequency on the DCOI? This obviously IS the Holy Grail of Mosfet switching in say a H-bridge circuit or something similar.
I like lights, but I would really love to see something more in the area of motor control. 😉
Dave
LOL at the reason for using tants 🙂 Nice 1 Julian
John