No problem :) Let me know if you need any help!
I don’t think the PWM on the arduino is slow for your application. Motors are actually great for filtering. Even if the current is not filtered, mechanically it is so slow that you can go as low as 100Hz and still drive the motor acceptably well.
I also worry about the LM7818. It drops about 6V, so at 330mA it burns 2W, which seems like the maximum for the package that I quickly pulled up. If you have a very efficient buck converter that gives you at most about 1A to drive the motor. Not a lot of overhead.
Also as I said already, you could just skip the whole thing and make everything work from the 24V using PWM. Since you want to drive it with an arduino that is not a difficult thing.
If I were you, I would not step down the 24V, but use that to drive the motor with pwm. It requires a bit different H bridge but overall it would be simpler. But if you already have a converter module that is good enough this works too.
In that case I would omit the extra transistors and use only the optocouplers.
What sort of application is this? Having 18V somewhere and a 5V supply too that can output several amps is quite unusual.
Obviously. No idea what I was thinking.
Is it okay that there is no stop state? It can only drive the motor in either direction, bit there is no OFF state with your inputs.
Please don’t use non perpendicular lines and draw in a clearer software if possible. Its quite hard to read.
What is the point of the push-pull BJTs? With the optocoupler you could drive the nmos gates directly. If you need more current to drive the gates (unlikely for a small motor), I suggest you buy dedicated gate drivers. They are tested and not expensive. If you have the 18V available already, using the optocouplers directly would be a lot simpler.
I also vote for a writeup. This sounds awesome!
Not in this wave, but right after covid “ended” they wanted us to return to the office. There was some back and forth and since then the agreement is hybrid, at most 60% home office.
Im quite happy with that. I work in-person 2 days a week, but I do embedded so it is often needed to do more. When it makes sense I have 0 problem with doing more days.
Ugh, and the code is so bad. I love it!
I think the real solution is repairable gadgets. A microusb port costs pennies and if the phones were repairable at all, it would be a 5 minute solder job. The same is true for USB C as well. It might be more durable, but it can still break and it feels so stupid to replace a decive because a single cheap part broke.
I liked microusb. I don’t know what everyone is doing with their connectors, but I never had any contact issues.
Then it still doesn’t take away anything from you. How would an additional embedded web based control panel impact you negatively? If you don’t use it just don’t use it. But it ensures longevity and makes the device entirely self-contained.
Its still stupid. No reason there is no embedded web interface. That is such an easy thing to do. Like routers have been doing for ages.
Yes, really. If my move tool is set to layer move, dont change it just because I used the select tool for something completely unrelated. That is the typical dumbed down big colorful button approach that I hate in modern corporate software.
I do embedded. Its all C. You can’t replace it.
But… but… Glass is not single use. That is the whole point. I don’t like this article.