[HN Gopher] Hardware Is Hard?
___________________________________________________________________
Hardware Is Hard?
Author : prdpx7
Score : 7 points
Date : 2026-04-18 19:34 UTC (3 hours ago)
(HTM) web link (prdpx7.github.io)
(TXT) w3m dump (prdpx7.github.io)
| cooldk wrote:
| Hardware is slow
| Neywiny wrote:
| Thoughts as a computer engineer:
|
| 1. The motor demo didn't work for me (Firefox on Android). Nice
| to say it was made by Claude.
|
| 2. Voltage isn't like gravity. It's literally called potential.
| Voltage is the water being high up on the mountain. Gravity is
| the gravity, which was kinda the point of the analogy, no?
|
| 3. A Tesla has thousands not hundreds of 18650s. But I guess
| that's a superset.
|
| 4. For motor speed, I would use P = I x V. It shows the way the
| two work better, and the resulting P = I^2 x R = V^2 / R.
| Resistance is constant.
|
| 5. PWM wasn't explained right here. You can most certainly vary
| the voltage without PWM, and it's not just any old modulation you
| want. You want a binary output so your transistors don't go
| linear, and you can control the average power. PWM is easy for
| linear scaling. But you could presumably use other modulations.
|
| 6. It's important to understand what the H bridge is doing, why
| it matters, and what it offers. Shoot through and dead time are
| important. Understanding that the microcontroller outputs X
| current at Y voltage, and the motor needs J current at K voltage.
| You could drive a small enough motor directly.
|
| 7. Could maybe do a 2-cell charger with balance leads as a
| usability upgrade.
|
| I think this is a good start and always kudos to those who learn,
| even more to this trying to teach others. Hopefully there's a
| part 2 going into more very simple theory. Especially a
| comparison vs the L298N.
___________________________________________________________________
(page generated 2026-04-18 23:01 UTC)