Your gates are ringing. This causes heating of the MOSFETs and the IC, and also discharges the boost cap(s) rapidly, preventing the high-side MOSFETs from turning on fully. Additionally, it looks like the AHO is going high before BLO; BLO needs to go high first. Remember the charging of the boost cap thing I was talking about in the 3rd to the last paragraph in my last reply?


You have not indicated what MOSFETs you are using, nor what your motor's run current and stall current is.

Yes, I remember what you said about the charging of the boost cap. As I said in my reply post I was counting on HDEL and LDEL to make the high side go first, do I need to turn BLO high first manually?

Do those 22Ω resistors need to be 1/4W, or will 1/8W be sufficient?

The problem is that the motor was ordered from Dayton, and has very little documentation whatsoever. We asked them for a datasheet and they sent us the "user's manual." From I measured, at 24v run current with no load is about 1A, moderate load is about 4A, and stall current is slightly more than 15A

I'm probably going to order some motor drivers soon-- this H-bridge is proving to be more of a science experiment.

 

Yes, I remember what you said about the charging of the boost cap. As I said in my reply post I was counting on HDEL and LDEL to make the high side go first, do I need to turn BLO high first manually?

It looks to me by the waveforms displayed on your 'scope that the high side MOSFET is trying to turn on before the low side.

Take a look in the Applications Note, page 6, figure 7, and study the bootstrap capacitor charging path.
Note that in order for the AH boost cap to charge, the AL MOSFET must be turned ON, as that is the current path to charge the cap. If you depend upon the BL MOSFET turning on to charge the cap, you have a large inductive load between the low side of the cap and GND, and it will take a relatively long time to charge the cap. I'm not making this stuff up.

Do those 22Ω resistors need to be 1/4W, or will 1/8W be sufficient?

Try the 1/8 Watt.

The problem is that the motor was ordered from Dayton, and has very little documentation whatsoever. We asked them for a datasheet and they sent us the "user's manual." From I measured, at 24v run current with no load is about 1A, moderate load is about 4A, and stall current is slightly more than 15A

OK, so what N-ch MOSFETs are you using?

I'm probably going to order some motor drivers soon-- this H-bridge is proving to be more of a science experiment.

Well, you'll need a fairly hefty motor driver. There aren't many integrated solutions that I know of that will take a 15A load.

 

Note that in order for the AH boost cap to charge, the AL MOSFET must be turned ON, as that is the current path to charge the cap. If you depend upon the BL MOSFET turning on to charge the cap, you have a large inductive load between the low side of the cap and GND, and it will take a relatively long time to charge the cap. I'm not making this stuff up.

No need to get condescending. The problem with holding AL on to charge the cap is that I need to turn AL off or risk a shoot-through in my current circuit. If I went out and bought an inverter, I could probably make it work at the cost of more complex input algorithms to avoid a shoot-through while letting the high half charge sufficiently.

If anyone is reading this and attempting to build an H-bridge, I would suggest giving a few clock cycles for the BS cap to charge, and then only switching the bottom FET while the FET on the high side is held on. Switching both FETs simultaneously above 10kHz just seems to give the IC's charge pump too much of a hassle while not giving the BS capacitor enough time to charge back up.

I'm going to be using this: http://www.dimensionengineering.com/Sabertooth2X12.htm

The FETs I was using are :http://www.nxp.com/documents/data_sheet/BUK7510-55AL.pdf

 

No need to get condescending.

I'm not being condescending. I've been trying to be very patient with you.

If you look back at post #7 in this thread, you'll see that I wrote early on to be certain to read the Application Note, including a link to the AN. I really did mean for you to not only read it, but to understand it as well. If it's your first time dealing with a component like this, you need to spend extra time going over the Application Note, perhaps re-reading the entire AN several times over the course of a few days until it "sinks in".

Since reply #7, it seems that I've spent much of my time in this thread locating/quoting snippets of information in that same Application Note which answer the questions that you have. That Application Note is really quite good. Probably the most difficult to understand part for a beginner would be the board layout tips.

The problem with holding AL on to charge the cap is that I need to turn AL off or risk a shoot-through in my current circuit.

You don't have to hold AL on very long; just long enough to initially charge the boost cap - either that, or wait long enough for the internal charge pump to charge it before starting the switching on and off. If the gate ringing is brought under control using the series gate resistors, the charge pump should be able to keep up with the demand.

The MOSFETs you chose have a rather large gate charge of 125nC. You would have been better off to choose MOSFETs with a lower Vdss, the same or somewhat higher Rds(on) and a much lower gate charge. All other things being equal, the lower the gate charge, the faster the MOSFET can turn off, which is where much of the power dissipation in the MOSFET is likely to occur.

If I went out and bought an inverter, I could probably make it work at the cost of more complex input algorithms to avoid a shoot-through while letting the high half charge sufficiently.

You just have to turn AL on for a moment, and then the charge pump should take care of things after that.

If anyone is reading this and attempting to build an H-bridge, I would suggest giving a few clock cycles for the BS cap to charge, and then only switching the bottom FET while the FET on the high side is held on. Switching both FETs simultaneously above 10kHz just seems to give the IC's charge pump too much of a hassle while not giving the BS capacitor enough time to charge back up.

I really don't know why you were switching both the high and low MOSFETs together to begin with. If you were using a driver like an IR2110, you'd use switching on the high-side MOSFET to keep the boost cap charged.

I'm going to be using this: http://www.dimensionengineering.com/Sabertooth2X12.htm

The FETs I was using are :http://www.nxp.com/documents/data_sheet/BUK7510-55AL.pdf

Like I mentioned above, a lower gate charge (specified as Qg) requirement would help the power dissipation issue, along with the ringing on the gate.

 

I really don't know why you were switching both the high and low MOSFETs together to begin with. If you were using a driver like an IR2110, you'd use switching on the high-side MOSFET to keep the boost cap charged.

I just wanted to build a prototype to make sure the IC did what I wanted period. I didn't really care if the layout introduced noise in the circuit-- I just wanted to make sure that it turned on the transistors as desired, etc. and see at what switching frequency the noise overcame the signal. Because of this I had a hard time filtering out what were tips to make the circuit work *better*, and tips that make the circuit work period.

At the time it seemed logical to manipulate the inputs like I did. If I want the motor to go forward, I want the top left and the bottom right transistors to be on. If I want it to go backward I want the top right and the bottom left transistors to be on. Since I'm really only concerned with going forward and backward, it made sense to only tie up two of the PWM pins on the processor to interface with the bridge, control the two pairs I was concerned with, and call it a day.

The part that I didn't understand (obviously now, less so when you were trying to explain it ) was how essential the bottom transistors were in the bootstrap's ability to pump up the gate voltage, and how it takes time to do so. I was so busy troubleshooting other problems, that it was only until about 3/4 of the posts in that I started to get to the really essential parts of making the circuit work.

Thank you for your help! Hopefully this thread will help someone out there with more time on their hands to build an H-bridge. It was *almost* a cost effective, power efficient solution, I was just running out of time.