| Electronics > Projects, Designs, and Technical Stuff |
| Circuit for MosFets in parallel for extra current capacity.(Solved) |
| << < (4/10) > >> |
| nick_d:
I agree with all that has been said here. One thing I am slightly confused by is the need for a gate resistor, it has been said that a design seen in the wild did not use gate resistors and got away with it, but why would this be bad? If you just connect all gates in parallel and drive them with one hefty gate driver chip, I would have thought that all the gate capacitances would appear in parallel and be charged in parallel, leading to all MOSFETs switching on at the same time (subject to slight differences in their Vgs as discussed). Anyhow. On the subject of gate drive generally. Because of the huge gate capacitance as mentioned, what you are doing (directly or indirectly) is using a smaller MOSFET, which has a smaller gate capacitance and hence is easy to turn on and off (perhaps by a digital output from your ESC), and then using that smaller MOSFET to drive the gate of the power MOSFET. And as already mentioned, the reason this is important is because the MOSFET becomes a resistor when partially turned on, so it can heat up during the switching. Hence the switching must occur as quickly as possible. The problem then is that to drive the gate, you need a separate smaller P-type MOSFET for charging and N-type MOSFET for discharging the gate. Not only that, but since both will be connected to the gate simultaneously and hence also to each other, dead time is desirable between turning one of the gate driver MOSFETs off and the other on. While you can build all this from discrete parts, the dedicated gate driver chip is highly recommended for simplicity. (And it has also been mentioned above that it has improved waveform, I didn't know this but maybe it compensates gate inductance). Note the ESC might have the gate driver built in. Check it. Now for my suggestion, the "poor man's gate driver" and that is the 74AC logic family. The thing is though, you would need to have logic-level power MOSFETs, which require a gate voltage of only 5V to turn on, not the 12V as calculated above. I know you already have the traditional (not logic-level) MOSFETs in stock, but if the choice is to order a gate driver or a new MOSFET then the logic-level approach may be attractive. You could get a 74AC541 octal buffer and put 2 of those buffers in parallel to drive each of your 4 gates. No gate resistors required, and each gate driven separately. cheers, Nick |
| spec:
+ Jim-0000 Thanks for your replies- very helpful. There is no need for any apologies- just ask any questions that you like. By the way it is me that is uninformed about modern model aircraft, although I have built and flown quite a few as a nipper- no electric motors in those days, just thermals, elastic bands, and diesel. And radio control was an electronic marvel well beyond my means or understanding. :) In general, I do not see any problems in any area that cannot be resolved, and will cover your points in replies #13 and #14 in further posts. But can I ask a few more questions about the application: Am I right is saying that the MOSFETs will be used to control the speed of an electric motor which spins the propeller in a model aircraft which flies. How big is the model aircraft? Will pulse width modulation be used to control the speed of the electric motor? Or will the motor just be turned on and off relative infrequently, say no more than twice a second? What is the construction of the model aircraft. If the model aircraft is not skinned in aluminum, would it be possible to skin a part of the aircraft in aluminum or are there any thermally conductive areas on the aircraft. The thinking here is to use the aircraft as the heatsink for the MOSFETs. A large area would not be required. What signal would be used to control the MOSFETs- do you have details of voltage etc? A picture of the model aircraft would be great. Once I get sufficient information about your application, I will do an outline circuit which we can discuss if you like. |
| spec:
--- Quote from: Jim-0000 on December 30, 2018, 09:19:12 am ---What is RDSS? --- End quote --- When an NMOSFET gate and source are at the same voltage the resistance between the NMOSFET drain and source is, essentially, an open circuit so that any voltage between the drain and source will not cause any current to flow from the drain to the source. But if you make the gate more positive than the drain (4V for the FQP50N06) the NMOSFET will conduct a small current. This voltage is the threshold voltage (Vth). But if you increase the G/S voltage the resistance drops to a very low value. In the case of the FQP50N06 the resistance with 10V G/S drops to 0.022 Ohms (22 mili Ohms), which is essentially a short circuit. This low resistance is RDSS (resistance drain/source): can't remember what the second S represents. And that is all there is to RDSS (it is a bit more complicated far a critical design, but that need not concern us at this initial stage). |
| spec:
--- Quote from: Jim-0000 on December 30, 2018, 09:19:12 am ---The onboard main power in the model aircraft this is planned for is 14.4 volts. So, that is no problem. I could just wire a supply off the main LiPo battery supply. --- End quote --- Excellent :) |
| spec:
--- Quote from: Jim-0000 on December 30, 2018, 09:19:12 am --- The size and configuration of the heatsink might be a limiting factor. --- End quote --- From what you have explained about the application I don't see any problems, so far, with the heat sinking that cannot be solved. When you mention heat sinks people think of big, expensive aluminum structures with fins all over the place, and in many high-power applications this is the case. But in other applications, you can use the unit structure as the heat sink. For example, I have built many 100W audio amplifiers, which had no separate heat sinks. Instead, the whole aluminum top cover was one big and efficient heatsink. |
| Navigation |
| Message Index |
| Next page |
| Previous page |