xie cheng driver technical

whatever

100 kW
Joined
Jun 3, 2010
Messages
1,297
Trying to nut out the nitty gritty details of the xie cheng drivers and how they work in fine detail.
Below is a pic showing the xie cheng drivers and how they have changed over time. Bottom left section shows very basic xie driver ( circuit was posted on esforum long time back), top section shows xie driver from 2008, bottom right shows xie driver with 116 chip. Because I'm not an electronics buff I have to break down these circuits into their basic bits in order to understand them. Help needed with how these things work. The bottom left pic being very basic is much easier to understand so I'll start with that one, try get a good understanding of it first.
driver1.jpg
 
looking at just the basic driver diagram
this pic shows all three phases and drivers, note the circuit has a pwm chip also, although the 846 and 116 have built in pwm ( I think!), the pwm not relevant to this thread, as just concentrating on the driver circuits.
 

Attachments

  • driver3.jpg
    driver3.jpg
    59.5 KB · Views: 1,369
next taking just one phase, there are two drivers, one for high side and one for low side, anyone can tell me which one is high and which one is low? I"ve just guessed on the diagram, both drivers are almost identical, one has a large diode.....right at the top.......is it associated with the high or low side driver?
I think j13 is connect to pwm chip ( could be wrong!), not relevant in this thread
 
taking one driver ( high side) its broken up into two sections,
5volt section: one section runs from 5volt source and joins to the chip ( which turns transistor on/off in sinc with hall sensor signals)
15volt or 12volt section: this section turns the mosfet on/off, mosfet needs about 10volts to turn on fully, has to turn on fully if not it will get hot. This section has two transistors involved.
driver5.jpg
 
now just taking the 5volt section, hopefully getting into the nitty gritty now, can try to work out how this section is working in detail.
pic below shows the 5volt section, its not a complex circuit, my question is how is the transistor being turned on and off.
There is a signal coming from the chip which is ultimately driven by the hall sensor firing, I think there is some sort of processing of the hall sensor signal via some 'algorithmn" which then sends a signal to this 5volt section which turns the transistor on/off. Just how this transistor is turned on/off I have no idea!!!
There is a 5volt source via a resistor which then connects to the chip trace, what is the chip doing? is it changing its pin voltage? is it allowing a tiny current to flow? any one have an answer to this?
How does this configuration allow the transistor to be turned on/off by the chip? all a mystery to me!
What I'm trying to do is work out exactly how this thing works, the way its connected and the role of resistor and chip ( later drivers also have a capacitor involved in this section.......not relevant at this level )
This 5v section transistor is connected to one of the 12v transistors, which is next step in understanding the driver.
This 5v section is identical on low and high side drivers
driver2.jpg
 
I think there's an error in the schematic for the "basic" driver. It doesn't make sense the way it's drawn, the high-side driver would never work very well at all. Here's how I think it should look:

View attachment Mod Basic XieChang.JPG

It's still a pretty mediocre driver, but it at least works in this configuration. The capacitor I added is for the bootstrap circuit that provides the drive voltage for the high-side FET. If the FET is just driven from the battery voltage, it won't be able to fully turn on.
 
thanks for that, a bit beyond my current understanding though. That basic circuit someone posted it ages ago in relation to the xie cheng controllers, I thought it might be helpful for getting the basics sorted out.
Would you agree in the xie controllers there are two sections in the drivers:
1. a 5volt section associated with hall sensor firing, sending a signal to a transistor in driver circuit
2. the above transistor once on, then allows a higher voltage signal ( 12-15v) to turn the mosfet on/off via two other transistors ( role of driver caps/diodes etc still a bit down the track for my understanding)
really trying to nut out the basics on this
 
It's all one circuit, dividing it into 5V and 15V sections doesn't make very much sense because neither will function without the other. I'll explain in very high-level terms how the circuit works.

The bootstrap is critical because the FET wants to have a gate voltage ~10V higher than its source voltage to fully turn on. For the high-side FET, this requires a voltage above the battery voltage. When the low-side FET is on, the center of the half-bridge (where the motor phase is connected and where I drew the capacitor to) is pulled to near ground. During this time, the capacitor can charge through the diode D9 from the +15V rail. When the commutation switches and the high-side FET turns on, that point is now near battery voltage, but the capacitor still has ~14V across it. That provides a voltage of ~12V above battery voltage with which to drive the high-side FET. It's called bootstrapping and its extremely common in these circuits.

The drive circuit is configured as an inverter. That is, when the input signal from the MCU is driven high (+5V), the FET will turn off, and when the input goes low the transistor turns on. The first resistor R15 is simply a pull-up which ensures the FET is off unless told otherwise. When the input goes high it turns on the first NPN transistor Q3 (numbers are fuzzy). This pulls the node immediately above it down, which is connected to the base of the PNP transistor Q56. When this transistor turns on, it pulls the gate of the FET down to within ~2V of the source, turning it off.

When the drive signal goes low, Q3 turns off and R16 pulls that node to the bootstrap voltage. This turns off Q56 and allows Q7 to turn on. Current can then flow through Q7 and R52 to charge the gate and turn the FET on.

The turn-on circuit does not work very well, though. It only passes a current of a few 10's of mA, which turns the FET on in 4-5 us. That is very slow. I put together a quick SPICE simulation to verify this.
 
thanks so much for that detailed explanation, its very detailed and I can follow it!!!
Its cleared up how this simple driver works ( even though the bootstrap was missing), it is for me quite complex to follow it.
I've heard bootstrap mentioned in the forum quite a few times, and it now makes some sense, much appreciate the time you spent on this. resistor r15 had made absolutely no sense to me. For my level of understanding, I can understand for example you saying that r15 has a function of keeping fet off, but my understanding of how it does it, is beyond me. I'll go over it a few times and see if I can understand a few missing links in my understanding.
 
Back
Top