Vonada's Engineering Maxims

[Syonyk]

Seen floating around the internet, and replicated on Wikipedia.

1. There is no such thing as ground.
2. Digital circuits are made from analog parts.
3. Prototype designs always work.
4. Asserted timing conditions are designed first; un-asserted timing conditions are found later.
5. When all but one wire in a group of wires switch, that one will switch also.
6. When all but one gate in a module switches, that one will switch also.
7. Every little pico farad has a nano henry all its own.
8. Capacitors convert voltage glitches to current glitches (conservation of energy).
9. Interconnecting wires are probably transmission lines.
10. Synchronizing circuits may take forever to make a decision.
11. Worse-case tolerances never add - but when they do, they are found in the best customer’s machine.
12. Diagnostics are highly efficient in finding solved problems.
13. Processing systems are only partially tested since it is impractical to simulate all possible machine states.
14. Murphy’s Laws apply 95 percent of the time. The other 5 percent of the time is a coffee break.

One I’ll add that I’ve heard before, in the context of new circuit board designers and cross sectional area of high current traces: “Copper isn’t a superconductor.”

Since several of us here play in these spaces or near them…

The worst I’ve found is actually the wiring conduit on my Ford 9N tractor. There’s a metal tube that runs from the front of the engine to the back, with openings on the side for spark plug wires.

The tube contains:
4x spark plug wires
1x generator wire, coming to charge the battery
1x ignition wire, coming forward from the ignition switch to power the points

I’ve literally never seen a digital voltmeter glitch and fail like it does trying to measure battery voltage on this tractor when running. One would expect some nasty spikes on the lines, running in parallel with the spark plug wires. One would not expect a digital meter, on the 6V battery, to glitch, reboot, blank out, and generally freak out entirely trying to measure battery voltage.

This is such a well known effect on these tractors that one of the “50 Tips” entries reads:

47. A digital multi-meter is a handy and usually inexpensive tool to have around the shop. But, most inexpensive digital multi-meters do not like the electrical “noise” produced by the N’s generator brushes. The test leads act as antennas and the meter gives some erratic readings as a result. Stick with the old analog meter for your old N.

I’m not so sure it’s generator noise as induced voltages from the wire conduit. In any case, I’m curious, but not curious enough to risk a good oscilloscope trying to figure out what’s going on…

[Canem]

These are good. I’ve run into several of them with both PCB design and machining.

Digital circuits are made from analog parts.

Wasn’t it Bob Wilder who said “Digital? Any idiot can count to one!”

Prototype designs always work.

This is either the most sarcastic thing I’ve read all week, or I’m a lot worse at this than I thought. Sure hope it’s the first one…

Worse-case tolerances never add - but when they do, they are found in the best customer’s machine.

GD&T is hard. Even Boeing gets bit by it from time to time. Stories about assemblies that worked fine for years, until that one aircraft where all the tolerances just stacked in the worst possible ways. Then one day the found the pressure dome was never going to fit that fuselage.

Diagnostics are highly efficient in finding solved problems.

Man tell me about it… I’ve got a couple projects now with intermittent glitches and failures I still can’t solve. They sit on the top shelf of my desk, taunting me.

Copper isn’t a superconductor.

A lesson I apparently can’t learn is how to put enough heatsink on power handling components. Currently I’m trying a method of “take how much heatsink you think you need, triple it, then add another fan.”

Then wonder too long if building some of my projects inside their own refrigerator would blow the power budget for the device.

---

Here’s my addition, sign seen outside a university lab:

“Theory is when you know everything and nothing works.
Practice is when everything works but no one knows why.
In this lab, theory and practice are combined:
nothing works and no one knows why.”

That’s an odd one for the 9N freaking out DMMs for sure. Got an old Simpson meter or similar around that might give some insight?

[Syonyk]

@Canem

That’s an odd one for the 9N freaking out DMMs for sure. Got an old Simpson meter or similar around that might give some insight?

I bought an old analog meter that works fine on it. It’s the “house” voltmeter, which works well for troubleshooting ride on toys, because the needle moves and it’s more interesting for the kids - I’m not caring about 0.01V on those, more “Wait, that switch failed…” sort of stuff.

The main use of the voltmeter on the tractor is to adjust generator charging rate, and I largely solved that problem with a solar battery tender - 6V/10W panel and a 6V charge controller (insert proper lead acid-isms here for voltage). That keeps the battery topped off so I can leave the generator brush set a bit lower and not worry about boiling the battery when I’m working hard.

[Vertiginous]

@Canem

Prototype designs always work.

This is either the most sarcastic thing I’ve read all week, or I’m a lot worse at this than I thought. Sure hope it’s the first one…

I read that as “it works as a prototype [but that’s no indication of how it’ll work in practice]”

@Canem

These are good. I’ve run into several of them with both PCB design and machining.

What sorts of things are you making (if you don’t mind me slightly derailing this conversation with my curiosity)? I haven’t gotten into PCB-level stuff yet, mostly focused on embedded software and signal processing, personally.

[Canem]

@Vertiginous

I read that as “it works as a prototype [but that’s no indication of how it’ll work in practice]”

Phew. I think you’re right, that makes more sense.

As for what I’m working on, I’ve been kinda going back and forth between little radio projects to better understand basic RF engineering (like working through Ward Silver’s book series “Hands-on Radio experiments”), and my more complete projects which are various Atmega/Arduino designs.

I’ve been on a kick lately of using those Semtech 1278 ‘LoRa’ radios for long range low-data-rate comms. Projects in a more shareable state are up on my Gitlab: Cj · GitLab

Projects like the Lora Relays and Gate Controller are 100% awesome and working great. APRS Repeater and Lora HHT I have the boards for but little code written. Lora Motion Sensor and Magneto Driveway Alarm are two in the category of ‘random glitch-fest I can’t for the life of me figure out’.

Comments/questions always welcome. If it looks like the code was written by some amateur self-taught hack, well, that’s because it is… My programming skills are still quite far below my circuit design skills.

[Syonyk]

And my firmware skills far exceed my board designs skills. I’ve been hoping to work on resolving that, but all my free time lately has been going to the solar build, poking through national electric code, etc.