Quote:
|
Have you identified what kind of transistors Q7, and Q5 are? I am tring to understand how these are used, but since I have no idea what kind of transistor, or even type (NPN vs PNP), I am at a loss.
|
I'm glad you asked that. Q7 and Q5 are common emitter inverting amplifiers(NPN BJT transistors.....like a 2N 3904) which will 'invert' a high to a low, and a low to a high. The collector of this transistor arrangement is connected to the gate of its corresponding FET. Without them, you will not have the correct logic levels at the gates of your FET's(h-bridge) which will indeed cause your h-bridge to be 'firing' all wrong. You will be directly shorting out your battery through your FET's if/when you apply power to them IF you don't have those particular signals inverted.
If you are applying your own home-made h-bridge, take your signal from the pad which the stock FET's would be soldered to....NOT where IC1 provides the signal to the transistors/FET's(at those resistors you mentioned). If you bypass the transistors and take the signal from IC1 directly(from the resistors), you will have to implement your own inverting method to this signal(common emitter inverting amplifier). If you take the signals from the pads that the FET's would normally be soldered to, you can just build the H-bridge configuration with your own choice FET's and the signals will already have the appropriate logic levels.
This is something that Robert overlooked in his design of his V1 power management board. If you look at his installation instructions, he tells the customer to remove those transistors(Q5, Q7) from the xmod board. He instructs people to do this because the very same type of transistor is implemented in his board design to do the very same thing for you. So he has you remove something that he turns around and sells to you, which is unnecessary. Then he instructs you to jumper the signal from IC1's 4 signal traces directly and connect to his board.
Quote:
|
I have seen 2 diffrent pictures of your H bridges - one had what looked like a 16 pin DIP IC, and the other used what I think is on the stock controller using 4 resistors and 2 transistors.
|
Yes. The later one you described was actually my first attempt at building an h-bridge(that's actually driven properly) for xmods. The one with the 16 pin IC was my latest. There's really not much to say, other than it was cool to do. The IC was there to isolate the signal lines from the higher voltage circuit. It is an opto-isolator with NPN transistor outputs. Neither had given me the performance that I had desired, but the first one was a more or less a failure. It had heating issues like you described and was originally designed to be installed in the roof of the body. This made for very hairy wiring, added weight, and if the signal lines came disconnected, the FET's were toast because the biasing must be maintained at all times.
I have another idea up my sleeve for an even better h-bridge design that SHOULD have minimal power losses due to signal propogation delays and gate driving. Of course, if I told you I'd have to kill you. So I won't tell you. I'm going to wait for the next generation of xmod'ing to come around to even talk about it. I'm sort of out of xmods at the moment, you could say. To be continued till Fall.......
