5 ms·

Thank you for posting this, it reaffirms what I've been thinking for so long. There are many opportunities to retrofit old systems of all types with modern low-

Thank you for posting this, it reaffirms what I've been thinking for so long. There are many opportunities to retrofit old systems of all types with modern low-cost embedded technologies. Around 2019 or so I was approached by an engineer who had a small business retrofitting very old machine tools with modern motion controls. Think very large lathes and planers. The problem they had was that in order to get these systems working with newer controls, they had to make time-consuming modifications to the old machines, in some cases modifying the axis motors and that cost added up quickly. The engineer realized that it was theoretically possible to take the old analog position signals and convert them into something a modern motion controller could read. That converter box would make the retrofit pretty much plug & play, but he didn't have the programming expertise to make it happen. We probably built the first iteration of the converter for under $50 in parts and less than 50 hours of development time. That had me searching for other similar opportunities. I have found a few, but they tend to be one-offs that aren't worth the time unless you're already building something similar. Either that or my ability to see opportunity sucks!

lostlogin 3mo ago
> retrofitting very old machine tools with modern motion controls. I have a friend who has automated an old food production line. Eg peanut butter making, Muslie roasting etc. With no background in programming he got machines hooked up measuring various things - eg peanut butter levels with ultrasound, roast temps, rotation rates. Zero version control and eyebrow raising methods but massive improvement in safety and throughput. Really impressive.
- eru 3mo ago
  I suspect this is where our AI agents will really come to shine: helping people who might have a technical bend but are not programmers do cool things.
  - fragmede 3mo ago
    Yeah. Anyone who hated writing code but had to, like biologists and chemists, now don't have to.
    - EPWN3D 3mo ago
      The less physicist code exists in the world the better.
      - torginus 3mo ago
        The problem with super smart people is that they often succeed without having to or bothering to understand even the most basic common sense techinques. I had a friend who told me a story, that when he was a kid, he found a copy of the Borland Pascal DOS compiler on his dad's machine. It came with a demo demonstrating the graphics library, but that was the extent of his introduction to Pascal, most of the syntax he figured out by trial and error. He went ahead and build a quite elaborate tank game, with things like enemy AI, multiple weapons and destructible terrain. The code was a fever dream, but the game was pretty impressive, and enjoyable. One of the things he didn't intuit, was that you could not only use built in functions, but define your own. So when he needed a code block a second time, he just copied it, so you can imagine how pleasant that code looked.
    - eru 3mo ago
      Yes. And in addition, Claude and Kimi and friends will happily do version control and write tests etc. They'll even volunteer to do so and recommend.
    - silisili 3mo ago
      I'm mixed on this. They may not(or may) write elegant code, but they have clear intent about its parameters and purpose. I've seen LLMs mess up numbers by order of magnitude on multiple occasions, which could range anywhere from annoying to fatal, depending on the field. Like all things, helpful as long as it's checked very carefully.
    - baranul 3mo ago
      This is arguably a dangerous fallacy, particularly for anything critical. When the person doesn't know programming, it becomes harder to guide agents and do reviews, leading to overlooking issues and accepting mistakes. This gets amplified for any creative solutions or fixing future problems, where it is something the LLM was not trained on or has limited data about the subject. Plus add how much money spent on tokens. Maybe a way for non-programmers to better understanding what they are saying, is to think about translation. It's kind of like saying a person doesn't need to know anything about Spanish to now sell LLM translated books from English to Spanish. Yes, the LLM can do translations for you, and quickly. But, as you don't know Spanish, you won't know how good the translation is. You are less likely to catch all kinds of mistakes, from bad grammar to wrongfully translated idiomatic expressions. The less Spanish that you know, the more likely you are to be impressed or fooled into a false sense of security. Then when native and fluent Spanish speakers read your books, they may rate it as mostly gibberish and nonsense.
BobbyTables2 3mo ago
I think you’re absolutely correct. I suspect it is practical to do a small one-time run of maybe a 1000 copies of a purely hardware device and sell them. Of course, that is just kicking the can down the road a bit farther… At least dealing with obsolescence with purely analog parts is simpler… As soon as software appears, things become less ideal. Now the HW and SW become highly specialized to a single vendor’s single family of devices. Plus, bugs and internet connectivity almost ensure a potentially unbounded number of updates or forever broken functionality. Even toolchains fade into obsolescence… Sure, one can redesign for a new microcontroller from a new vendor. Maybe with new RTOS too. Of course, that is almost as much work as the initial design and probably done by someone else at a later date. Yuck! (And I saw this as a software engineer!) Definitely something to be said of the old habit of a simple analog design with schematics fully provided to the end user. This is something probably still needed today, but perhaps commercially challenging, especially with countries like China in play…
ryukoposting 3mo ago
I think I see some parallels with the aftermarket for old cars. Holley's entire catalog is "here's a thing that'll modernize your 1960s engine without replacing the whole damn engine" This kind of thing is a low-volume, low-margin business by nature. It's those kinds of businesses that the megalomaniacs of our field tend to ignore. There are people with needs, and money they are willing to exchange for a solution to those needs. But there's no path to world domination by selling carburetors (or bowling alley control systems) so those problems never get solved by anyone with the means to do so. It's a shame, really.
- red369 3mo ago
  I once expected an aftermarket industry to pop-up to hybridize old cars. Old rear-wheel drive manual cars sound easiest. Generator attached to the flywheel instead of the starter motor, and then find some way to squeeze a drive motor by the rear diff (probably more complicated, but surely not impossible). I thought something might pop up once companies like Bosch started making generic e-axles for manufacturers to use. With a small Li-ion battery (and all the usual inverters, etc.), it could work like a manually operated version of the dual-motor Honda hybrid system. Put it into neutral, and whatever energy is in the battery could drive the motor on the rear axle. The engine could automatically kick-in and drive the generator if there's not enough in the battery. It would never be as economical as a purpose-built hybrid, but old manual cars are often reasonably economical on the open road, and terrible in city traffic. Targeting only the urban fuel consumption would go a long way towards keeping them useable, with the added bonus of making manual shifting optional in stop-go traffic. Small motors would be enough for city speeds, and with a small battery, it sounds cheap to me at least. I doubt it could ever be a simple kit, but I thought people would have a go.
  - yobbo 3mo ago
    Stuff like this exists for bicycles. For mopeds and motorcycles there are electrification kits. It should be possible to just replace the flywheel altogether with a generator/motor that both brakes and powers during each cycle, cancelling vibrations. Then more generation during deceleration, and more power during acceleration.
madaxe_again 3mo ago
Yup. I live on a river. We flooded catastrophically in 2019, which made me suddenly care about the government’s network of river level meters. Turns out, they abandoned most of them in 2008, as it was the local council who were meant to maintain them, but they couldn’t afford it - a chat with the president revealed they’d been paying €15k a year for each station to maintain it and they’d cost about €100k each. I went and looked at the battered remains. Sounding tube. Ultrasound transducer. Box of 90’s electronics. Anyway, long story short I banged out six of them for them for about €80 a piece - the expensive bits were the batteries and the dinky solar panel. ESP32, ultrasonic transducer, GSM and LoRa (as two are outside of the cell network), haven’t had to touch them in five years and we now know when it’s time to evacuate, rather than wading out in the night while trees sail past us at 60kph. So much stuff like this.
- fragmede 3mo ago
  They didn't offer any of the €15k a year to you, did they?
  - madaxe_again 3mo ago
    They hadn’t been paying since 2008, hence it was all derelict trash - they don’t even have the budget to maintain the roads or keep the school open, never mind monitor river levels. My interest was not drowning, rather than pecuniary.
    - cechmaster 3mo ago
      so what are they doing?
      - madaxe_again 3mo ago
        Care and education, and whatever they can manage with whatever’s left over. Same as pretty much every council everywhere.
- japhyr 3mo ago
runtime_lens 3mo ago
[flagged]
berkes 3mo ago
> There are many opportunities to retrofit old systems of all types with modern low-cost embedded technologies. I'd leave out the "old" there. An example: Friends work(ed) as engineers in theaters and music-venues. The costs for (DMX) equipment is insane. Some lamps, rigs or controllers are so expensive that shows only rent them. While there's lots of cases where you don't want to cut costs¹, quite often a cheap variant cuts it just fine. So apparently there's a lot of Pie's, Arduinos, ESP32s and old androids used to handle stuff. Custom, hacked, duckt-taped solutions. All of them are certain a startup offering generic but cheap modules that engineers can customize, would be an instant hit. Especially if "service" can be bought with it. Or a friend who installs (climate) control systems in (food)factories, who cludges together arduino's and some hacky web-interfaces, who is certain that the industry would love some "off the shelve" solutions that installers can build their own hardware- and user-interfaces on is a very much wanted product. The industry (in EU) is according to him, one built on "lock ins", where a factory starts off with AcmeInc controllers and then slowly Acme will start charging almost "extortion-fee" alike figures just to fix stuff on a sunday night. Or, a friend who has a fruit farm with lot's of freezer-storage (running hundreds of KWs), whith whom I cludged together some Arduino's and home-automation to control the freezers to run when electricity is cheap or free. Stuff like this is available for consumers at reasonable prices, but the moment the Amperages go to industry-level, the costs explode. Where a simple remote controllable power-switch can cost €1000s. I'm certain that there's hundreds of thousands of small warehouses, supermarkets, traders, logistics that could save a few hundreds to thousands of €'s a year, by controlling their freezers, heaters, pumps, pressurizers, dryers and whatnots smarter with "turn key" modules built from generic low cost hardware. -- ¹ relevant: There's a lot of "Nobody Ever Got Fired for Buying IBM" behind it: if you go for the cheap €800 chinese knock-off and mid-show something starts failing, its your fault. But if that €10.000 controller starts failing, it's no-ones fault.
- kfse 3mo ago
  With respect to bowling alleys, for things like professional shows or food safety it's a bit more than "nobody got fired for buying IBM". The stuff's gotta be able to survive field conditions and not cost millions of dollars in refunded show tickets or spoiled foods. Manufacturers test and make themselves accountable for this. Cheap, customized modules work until they don't, and by then the engineer who put the custom system together in the first place isn't there to fix them. Who would you call when some quick-and-dirty integration code that was burned into these embedded modules without ever being put in a source versioning system fails in some weird edge condition? OP is an SRE, it's ok for him, but the average farmer or show operator doesn't have the skill.
TacticalCoder 3mo ago
> Thank you for posting this, it reaffirms what I've been thinking for so long. There are many opportunities to retrofit old systems of all types with modern low-cost embedded technologies. At my shooting range, where security is very important, we're replacing the old antiquated "fire stop" (when everybody on "shared" lanes has to stop shooting for there's at least one person going to change the targets) system with a modern system of electro-magnets and embedded devices. Mechanical safeguards where, by default, the 10 lanes (there two times 10 lanes, the 50 meters ones and the 100 meters ones) have to stop shooting and then an embedded system with a little controller for each lane that needs to be working to give the greenlight / "fire frei" ("free to shoot"). We mostly decided on the switch after a failure of the old system: the "free to shoot" system has recently been given even though there was still one person outside, changing the targets (nothing bad happened but this is a serious incident). We asked for quotes for a system that is always "door locked" / red light / "fire stop", unless everything works fine and every lane gives the greenlight (each shooter has now one badge). Same thing as in TFA: cost in hardware is minimal compared to the quoted prices. Like basically 1/50th of the quoted price.
MisterTea 3mo ago
I do retrofits and I'll just assume the equipment he was using didn't have an analog encoder option or was ignorant of such equipment (likely using some cheap hobby level CNC thing.) I've worked with a few motion control systems such as ACS and Aerotech and they provide inputs for just about every encoder type imaginable. I retrofitted a CNC with 1500W 90V servo motors with analog resolvers using an Aerotech system, zero mechanical upgrades.
- HeyLaughingBoy 3mo ago
  I'm going to forget some details, but he was not using hobby level CNC products. I still have a Kollmorgen AKD motion controller that he sent me as a test bench. Most of the old tools were using 0-10V sawtooth analog outputs for motor position and, IIRC, their preferred controllers didn't support that. The initial version of the box I made took that 0-10V and pretended to be a resolver so the AKD could control the motor. Like I said, I don't remember all his design decisions, but they wanted to retain the original position sensors.
  - MisterTea 3mo ago
    Ah, that sounds like something really weird then. Haven't seen 0-10V sawtooth signals for encoder output though Ive seen 0-10V for laser infarometers and other distance sensors.
bashkiddie 3mo ago
> but they tend to be one-offs that aren't worth the time Friend of mine used to retrofit CNC machines. He's left the business, because the machines that can be serviced moved to poor countries - where customers tend not to pay him at all. The modern influx of CNC have some "intellectual property" protection and cannot be serviced.

lostlogin 3mo ago

> retrofitting very old machine tools with modern motion controls. I have a friend who has automated an old food production line. Eg peanut butter making, Muslie roasting etc. With no background in programming he got machines hooked up measuring various things - eg peanut butter levels with ultrasound, roast temps, rotation rates. Zero version control and eyebrow raising methods but massive improvement in safety and throughput. Really impressive.

eru 3mo ago
I suspect this is where our AI agents will really come to shine: helping people who might have a technical bend but are not programmers do cool things.
- fragmede 3mo ago
  Yeah. Anyone who hated writing code but had to, like biologists and chemists, now don't have to.
  - EPWN3D 3mo ago
    The less physicist code exists in the world the better.
    - torginus 3mo ago
      The problem with super smart people is that they often succeed without having to or bothering to understand even the most basic common sense techinques. I had a friend who told me a story, that when he was a kid, he found a copy of the Borland Pascal DOS compiler on his dad's machine. It came with a demo demonstrating the graphics library, but that was the extent of his introduction to Pascal, most of the syntax he figured out by trial and error. He went ahead and build a quite elaborate tank game, with things like enemy AI, multiple weapons and destructible terrain. The code was a fever dream, but the game was pretty impressive, and enjoyable. One of the things he didn't intuit, was that you could not only use built in functions, but define your own. So when he needed a code block a second time, he just copied it, so you can imagine how pleasant that code looked.
  - eru 3mo ago
    Yes. And in addition, Claude and Kimi and friends will happily do version control and write tests etc. They'll even volunteer to do so and recommend.
  - silisili 3mo ago
    I'm mixed on this. They may not(or may) write elegant code, but they have clear intent about its parameters and purpose. I've seen LLMs mess up numbers by order of magnitude on multiple occasions, which could range anywhere from annoying to fatal, depending on the field. Like all things, helpful as long as it's checked very carefully.
  - baranul 3mo ago
    This is arguably a dangerous fallacy, particularly for anything critical. When the person doesn't know programming, it becomes harder to guide agents and do reviews, leading to overlooking issues and accepting mistakes. This gets amplified for any creative solutions or fixing future problems, where it is something the LLM was not trained on or has limited data about the subject. Plus add how much money spent on tokens. Maybe a way for non-programmers to better understanding what they are saying, is to think about translation. It's kind of like saying a person doesn't need to know anything about Spanish to now sell LLM translated books from English to Spanish. Yes, the LLM can do translations for you, and quickly. But, as you don't know Spanish, you won't know how good the translation is. You are less likely to catch all kinds of mistakes, from bad grammar to wrongfully translated idiomatic expressions. The less Spanish that you know, the more likely you are to be impressed or fooled into a false sense of security. Then when native and fluent Spanish speakers read your books, they may rate it as mostly gibberish and nonsense.

BobbyTables2 3mo ago

I think you’re absolutely correct. I suspect it is practical to do a small one-time run of maybe a 1000 copies of a purely hardware device and sell them. Of course, that is just kicking the can down the road a bit farther… At least dealing with obsolescence with purely analog parts is simpler… As soon as software appears, things become less ideal. Now the HW and SW become highly specialized to a single vendor’s single family of devices. Plus, bugs and internet connectivity almost ensure a potentially unbounded number of updates or forever broken functionality. Even toolchains fade into obsolescence… Sure, one can redesign for a new microcontroller from a new vendor. Maybe with new RTOS too. Of course, that is almost as much work as the initial design and probably done by someone else at a later date. Yuck! (And I saw this as a software engineer!) Definitely something to be said of the old habit of a simple analog design with schematics fully provided to the end user. This is something probably still needed today, but perhaps commercially challenging, especially with countries like China in play…

ryukoposting 3mo ago

I think I see some parallels with the aftermarket for old cars. Holley's entire catalog is "here's a thing that'll modernize your 1960s engine without replacing the whole damn engine" This kind of thing is a low-volume, low-margin business by nature. It's those kinds of businesses that the megalomaniacs of our field tend to ignore. There are people with needs, and money they are willing to exchange for a solution to those needs. But there's no path to world domination by selling carburetors (or bowling alley control systems) so those problems never get solved by anyone with the means to do so. It's a shame, really.

red369 3mo ago
I once expected an aftermarket industry to pop-up to hybridize old cars. Old rear-wheel drive manual cars sound easiest. Generator attached to the flywheel instead of the starter motor, and then find some way to squeeze a drive motor by the rear diff (probably more complicated, but surely not impossible). I thought something might pop up once companies like Bosch started making generic e-axles for manufacturers to use. With a small Li-ion battery (and all the usual inverters, etc.), it could work like a manually operated version of the dual-motor Honda hybrid system. Put it into neutral, and whatever energy is in the battery could drive the motor on the rear axle. The engine could automatically kick-in and drive the generator if there's not enough in the battery. It would never be as economical as a purpose-built hybrid, but old manual cars are often reasonably economical on the open road, and terrible in city traffic. Targeting only the urban fuel consumption would go a long way towards keeping them useable, with the added bonus of making manual shifting optional in stop-go traffic. Small motors would be enough for city speeds, and with a small battery, it sounds cheap to me at least. I doubt it could ever be a simple kit, but I thought people would have a go.
- yobbo 3mo ago
  Stuff like this exists for bicycles. For mopeds and motorcycles there are electrification kits. It should be possible to just replace the flywheel altogether with a generator/motor that both brakes and powers during each cycle, cancelling vibrations. Then more generation during deceleration, and more power during acceleration.

madaxe_again 3mo ago

Yup. I live on a river. We flooded catastrophically in 2019, which made me suddenly care about the government’s network of river level meters. Turns out, they abandoned most of them in 2008, as it was the local council who were meant to maintain them, but they couldn’t afford it - a chat with the president revealed they’d been paying €15k a year for each station to maintain it and they’d cost about €100k each. I went and looked at the battered remains. Sounding tube. Ultrasound transducer. Box of 90’s electronics. Anyway, long story short I banged out six of them for them for about €80 a piece - the expensive bits were the batteries and the dinky solar panel. ESP32, ultrasonic transducer, GSM and LoRa (as two are outside of the cell network), haven’t had to touch them in five years and we now know when it’s time to evacuate, rather than wading out in the night while trees sail past us at 60kph. So much stuff like this.

fragmede 3mo ago
They didn't offer any of the €15k a year to you, did they?
- madaxe_again 3mo ago
  They hadn’t been paying since 2008, hence it was all derelict trash - they don’t even have the budget to maintain the roads or keep the school open, never mind monitor river levels. My interest was not drowning, rather than pecuniary.
  - cechmaster 3mo ago
    so what are they doing?
    - madaxe_again 3mo ago
      Care and education, and whatever they can manage with whatever’s left over. Same as pretty much every council everywhere.
japhyr 3mo ago

runtime_lens 3mo ago

[flagged]

berkes 3mo ago

> There are many opportunities to retrofit old systems of all types with modern low-cost embedded technologies. I'd leave out the "old" there. An example: Friends work(ed) as engineers in theaters and music-venues. The costs for (DMX) equipment is insane. Some lamps, rigs or controllers are so expensive that shows only rent them. While there's lots of cases where you don't want to cut costs¹, quite often a cheap variant cuts it just fine. So apparently there's a lot of Pie's, Arduinos, ESP32s and old androids used to handle stuff. Custom, hacked, duckt-taped solutions. All of them are certain a startup offering generic but cheap modules that engineers can customize, would be an instant hit. Especially if "service" can be bought with it. Or a friend who installs (climate) control systems in (food)factories, who cludges together arduino's and some hacky web-interfaces, who is certain that the industry would love some "off the shelve" solutions that installers can build their own hardware- and user-interfaces on is a very much wanted product. The industry (in EU) is according to him, one built on "lock ins", where a factory starts off with AcmeInc controllers and then slowly Acme will start charging almost "extortion-fee" alike figures just to fix stuff on a sunday night. Or, a friend who has a fruit farm with lot's of freezer-storage (running hundreds of KWs), whith whom I cludged together some Arduino's and home-automation to control the freezers to run when electricity is cheap or free. Stuff like this is available for consumers at reasonable prices, but the moment the Amperages go to industry-level, the costs explode. Where a simple remote controllable power-switch can cost €1000s. I'm certain that there's hundreds of thousands of small warehouses, supermarkets, traders, logistics that could save a few hundreds to thousands of €'s a year, by controlling their freezers, heaters, pumps, pressurizers, dryers and whatnots smarter with "turn key" modules built from generic low cost hardware. -- ¹ relevant: There's a lot of "Nobody Ever Got Fired for Buying IBM" behind it: if you go for the cheap €800 chinese knock-off and mid-show something starts failing, its your fault. But if that €10.000 controller starts failing, it's no-ones fault.

kfse 3mo ago
With respect to bowling alleys, for things like professional shows or food safety it's a bit more than "nobody got fired for buying IBM". The stuff's gotta be able to survive field conditions and not cost millions of dollars in refunded show tickets or spoiled foods. Manufacturers test and make themselves accountable for this. Cheap, customized modules work until they don't, and by then the engineer who put the custom system together in the first place isn't there to fix them. Who would you call when some quick-and-dirty integration code that was burned into these embedded modules without ever being put in a source versioning system fails in some weird edge condition? OP is an SRE, it's ok for him, but the average farmer or show operator doesn't have the skill.

TacticalCoder 3mo ago

> Thank you for posting this, it reaffirms what I've been thinking for so long. There are many opportunities to retrofit old systems of all types with modern low-cost embedded technologies. At my shooting range, where security is very important, we're replacing the old antiquated "fire stop" (when everybody on "shared" lanes has to stop shooting for there's at least one person going to change the targets) system with a modern system of electro-magnets and embedded devices. Mechanical safeguards where, by default, the 10 lanes (there two times 10 lanes, the 50 meters ones and the 100 meters ones) have to stop shooting and then an embedded system with a little controller for each lane that needs to be working to give the greenlight / "fire frei" ("free to shoot"). We mostly decided on the switch after a failure of the old system: the "free to shoot" system has recently been given even though there was still one person outside, changing the targets (nothing bad happened but this is a serious incident). We asked for quotes for a system that is always "door locked" / red light / "fire stop", unless everything works fine and every lane gives the greenlight (each shooter has now one badge). Same thing as in TFA: cost in hardware is minimal compared to the quoted prices. Like basically 1/50th of the quoted price.

MisterTea 3mo ago

I do retrofits and I'll just assume the equipment he was using didn't have an analog encoder option or was ignorant of such equipment (likely using some cheap hobby level CNC thing.) I've worked with a few motion control systems such as ACS and Aerotech and they provide inputs for just about every encoder type imaginable. I retrofitted a CNC with 1500W 90V servo motors with analog resolvers using an Aerotech system, zero mechanical upgrades.

HeyLaughingBoy 3mo ago
I'm going to forget some details, but he was not using hobby level CNC products. I still have a Kollmorgen AKD motion controller that he sent me as a test bench. Most of the old tools were using 0-10V sawtooth analog outputs for motor position and, IIRC, their preferred controllers didn't support that. The initial version of the box I made took that 0-10V and pretended to be a resolver so the AKD could control the motor. Like I said, I don't remember all his design decisions, but they wanted to retain the original position sensors.
- MisterTea 3mo ago
  Ah, that sounds like something really weird then. Haven't seen 0-10V sawtooth signals for encoder output though Ive seen 0-10V for laser infarometers and other distance sensors.

bashkiddie 3mo ago

> but they tend to be one-offs that aren't worth the time Friend of mine used to retrofit CNC machines. He's left the business, because the machines that can be serviced moved to poor countries - where customers tend not to pay him at all. The modern influx of CNC have some "intellectual property" protection and cannot be serviced.