I once took a really crappy RS232 cable to India as part of some equipment to train our remote developers. The cable barely worked in our lab in the States. I told our hardware engineer that it wasn’t going to work in India, and I was right. So in India I ended up having to wrap the entire wire bundle in a wire that I soldered to ground on both sides. Soldered it together with a plumbing soldering iron. I am a software engineer, but I have an electrical engineering degree. The VP that I was traveling with couldn’t believe that the crap I made worked. Realistically, I couldn’t either.
This makes sense actually. The issue with the bundle was regarding some part of it that was subject to RF interference and you shielding and grounding stopped all of that. Packets hate noise. This applies double to coaxial cable, in my experience, especially spanning up.
Okay, don’t get me wrong I’m impressed and I also enjoy macgyvering things like that… But if it’s for a work thing, surely it can’t be that hard to go out and buy a new cable from any old shop nearby? I would think the cable is common enough to still be in stock in a lot of places, even if it’s ancient.
I built an RS232 cable from parts from RadioShack 25 years ago, with no soldering, just electrical tape. It’s surprisingly easy if you don’t need speed. Mine capped out at 1200 or 2400 baud. Was it good? No. Did it work? Absolutely.
Yeah, the protocol itself is pretty robust. The cable I had didn’t have enough noise immunity for the dirty power the building had in India (afternoon brown outs when the voltage dipped when the air conditioners ran). The Faraday cage that I made around the cable helped with the noise and also (and I believe more crucially though I had no scope to confirm) gave the two boards a common ground. I had a little trouble with before I left, but it didn’t work at all in India until I modded it. Made the hardware engineer buy me a beer when I got back.
I once took a really crappy RS232 cable to India as part of some equipment to train our remote developers. The cable barely worked in our lab in the States. I told our hardware engineer that it wasn’t going to work in India, and I was right. So in India I ended up having to wrap the entire wire bundle in a wire that I soldered to ground on both sides. Soldered it together with a plumbing soldering iron. I am a software engineer, but I have an electrical engineering degree. The VP that I was traveling with couldn’t believe that the crap I made worked. Realistically, I couldn’t either.
India loves this barely working lifehacks known as “Juugad”
This makes sense actually. The issue with the bundle was regarding some part of it that was subject to RF interference and you shielding and grounding stopped all of that. Packets hate noise. This applies double to coaxial cable, in my experience, especially spanning up.
Okay, don’t get me wrong I’m impressed and I also enjoy macgyvering things like that… But if it’s for a work thing, surely it can’t be that hard to go out and buy a new cable from any old shop nearby? I would think the cable is common enough to still be in stock in a lot of places, even if it’s ancient.
This was a proprietary cable specific to our board design. Believe me, I wish we could have used a standard cable.
I built an RS232 cable from parts from RadioShack 25 years ago, with no soldering, just electrical tape. It’s surprisingly easy if you don’t need speed. Mine capped out at 1200 or 2400 baud. Was it good? No. Did it work? Absolutely.
Yeah, the protocol itself is pretty robust. The cable I had didn’t have enough noise immunity for the dirty power the building had in India (afternoon brown outs when the voltage dipped when the air conditioners ran). The Faraday cage that I made around the cable helped with the noise and also (and I believe more crucially though I had no scope to confirm) gave the two boards a common ground. I had a little trouble with before I left, but it didn’t work at all in India until I modded it. Made the hardware engineer buy me a beer when I got back.