Hi folks,

I'm trying to design a circuit where I can send a single PWM output from an Arduino (Teensy 4.1, 3.3V logic) to multiple MOSFET gate drivers in a looping sequence. The gate drivers are 3.3V logic compatible. I've tested it with a single gate driver wired up to the PWM output.

For example, say I have 10 gate drivers and the PWM pin is outputting a 10% duty cycle, 1 kHz signal. I'd like the first PWM pulse sent to gate driver #1, second pulse sent to gate driver #2, all through the numbers up to tenth PWM pulse sent to gate driver #10, then the sequence would repeat and the 11th pulse would be sent to gate driver #1 and so on.

Could anyone suggest a way for me to achieve this, without adding much latency to each pulse? I've had a look at the 74HC4017 Johnson decade counter with 10 decoded outputs and it seems like it might be a good fit, but I'm unsure if it will allow the PWM waveform to be replicated to the outputs.

Many thanks,

Lee

 

Hi folks,

I'm trying to design a circuit where I can send a single PWM output from an Arduino (Teensy 4.1, 3.3V logic) to multiple MOSFET gate drivers in a looping sequence. The gate drivers are 3.3V logic compatible. I've tested it with a single gate driver wired up to the PWM output.

For example, say I have 10 gate drivers and the PWM pin is outputting a 10% duty cycle, 1 kHz signal. I'd like the first PWM pulse sent to gate driver #1, second pulse sent to gate driver #2, all through the numbers up to tenth PWM pulse sent to gate driver #10, then the sequence would repeat and the 11th pulse would be sent to gate driver #1 and so on.

Could anyone suggest a way for me to achieve this, without adding much latency to each pulse? I've had a look at the 74HC4017 Johnson decade counter with 10 decoded outputs and it seems like it might be a good fit, but I'm unsure if it will allow the PWM waveform to be replicated to the outputs.

Many thanks,

Lee

4017 is a good start, then you need an AND gate on each output.
There are gate drivers with an AND gate on the input.

 

4017 is a good start, then you need an AND gate on each output.
There are gate drivers with an AND gate on the input.

Great, that sounds like a good solution. The gate driver that I'm using (Infineon 1ED3124MU12H) sadly doesn't have an AND gate on the input. It has non-inverting and inverting inputs.





Luckily, I had the foresight to order some logic gates at the same time as the 4017 so I'll look at combining the PWM signal with the output from the 4017 and directing that to the gate driver non-inverting input pin.

 

Great, that sounds like a good solution. The gate driver that I'm using (Infineon 1ED3124MU12H) sadly doesn't have an AND gate on the input.

Actually it does. It comes after the inverter on the inverting input.
If you invert your PWM signal and connect to all the inverting inputs, you’ve done it.

 

Of course, each of the outputs will be limited to a 10% duty cycle. Is that what you want?

 

Well that certainly sounds simpler than involving some AND gates. Excuse my ignorance as I'm a self taught electronics noob, but how exactly would I connect the output from the 4017 with the inverted PWM output and into the two gate driver logic inputs?

 

Yes, that's what I want - just duplicating the exact same PWM signal but on different gates sequentially. I don't need to vary the PWM signal on an individual basis, just overall. So, for example, if I change the duty cycle to 20%, the 20% PWM signal should be applied to each of the gate drivers in sequence.

 

Yes, that's what I want - just duplicating the exact same PWM signal but on different gates sequentially. I don't need to vary the PWM signal on an individual basis, just overall. So, for example, if I change the duty cycle to 20%, the 20% PWM signal should be applied to each of the gate drivers in sequence.

The “solution” won’t do that. Each output will have 1/10 the frequency and 1/10 the duty cycle of the input. Your 20% duty cycle becomes 2%.

What you are asking for is actually trivial. Just connect the original signal to all 10 outputs through a buffer that can drive them all.

 

Actually, thinking about it you're right. Each gate driver will have 1/(gate driver count) the frequency and duty cycle of the main PWM signal when you look at it from that point of view.

Here's a timing diagram to explain what I want to achieve. It might help provide some colour to the words. This is a simplified example where there are 3 gate drivers in total.

 

Hi If,
I would suggest you drop the PWM description, to a saying you want the same pulse On time to be sent in sequence to 10 Gates
Do you agree?
E

Update: is this what you require, only first 4 pulses shown for clarity.

 

Yes that's a fair point if it makes it more understandable. The source of the pulse isn't really relevant. The fact it is a PWM signal doesn't have much bearing I suppose.

 

hi If,
This shows stepping the ON period from 5m to 55m in 10m steps
E

 

I don’t think it is a PWM signal at all. If it was, the timing relative to others (the in sequence part) would be irrelevant.

It is time to tell us what you are trying to accomplish by sending pulses in sequence, because that makes no sense if it is just a PWM signal.

 

hi If,
This image shows the last 4017 pin Q9
E