Show HN: A high-speed laser camera trigger for under $2

It not specific on hardware, it's general about all things not about programing and new products. Also it's not about new posts, at the new section I keep finding them. It's that they hit the front page less frequently and when they make it the fall to the second or third page in a mater of one hour, compared to the framework or product posts that stay very high with relatively less points. Maybe it's not the case, maybe it's just the habituation to HN contents after three years of daily reading. If it happens I don't know if it's due to the new curation team (for example now I find that conversations are not growing so much into long rants about small details as before) or to more people flaging non "programmer" stuff more than before.

I like that! I'm going to start giving estimates like that

"I'll do this for free, not including my hourly or the cost of materials"

Yeah, the LED in the opto isn't even connected correctly. I doubt it would ever light up. I left a post on the website explaining how it should have been setup.

Ah, I see the confusion. The power supply should be connected such that the positive terminal is on the left. If you look at the breadboard, you can see that this is, in fact, what's built. My description of the circuit's operation above is accurate when the power supply is properly connected.

See my comments below. The only error is that the OP reversed the symbol for the battery. Look at the breadboard; he/she built it with the power supply correctly installed.

You're probably right about the datasheet for the opto, but that neither clashes with my statement ("most optocouplers") nor invalidates the broader point (Vcesat is ~300mV; the diode is most definitely off).

Aside: I spent several years as the lead engineer on a line of digital isolators (modern replacements for optocouplers: CMOS integrated circuits with much higher speed, better immunity to common-mode transients, and no need to use expensive and generally boutique processes supporting LEDs). Most of the time, opto manufacturers oversell the capabilities of their optos at low voltage, and generally speaking in industrial applications optocouplers are driven much harder than the currents at which forward drop numbers are quoted in datasheets.

Aside #2: bipolar transistors absolutely do operate upside-down! Reverse beta is usually far worse than forward beta, because the collector region tends to be much larger than the emitter and as a result there is substantial carrier recombination.

In fact, when you want to use a bipolar transistor as a low drop switch, and if you don't care how much base current you have to use to do it, running the transistor upside-down tends to result in lower drop and faster operation. There are a few reasons for this, but it boils down to this: the speed of the bipolar transistor as a switch is limited by charge storage in the base when the device is saturated, or more precisely, the amount of time it takes to remove this charge when trying to turn the device off. The base-collector diode behaves differently than the base-emitter diode because of the difference in emitter and collector doping. As a result, when upside-down, the base-collector junction stores much less charge, the lower beta actually means that the charge is removed faster, and the higher forward drop of the base-emitter diode means that reverse Vcesat is lower than the forward Vcesat.

The downside, to which I alluded above, is 10x or more base current to actually drive the transistor. Depending on the quantum efficiency of this phototransistor and its reverse beta, were the OP to use the phototransistor upside-down in this circuit, it might not be able to shunt sufficient current to keep the optoisolator off.