Hacker News Reader

Comment by gw98 | Hacker News Reader

I don't get the market position of this.

A number of Chinese companies produce DIN rail mounted PLCs which already cover this market, cost virtually nothing and actually emulate Mitsu GX ones so work with existing off the shelf tools and have the same real time guarantees as the commercial units. If you're going to cheap out, these are fine.

But this isn't the problem. The expensive bit of a PLC is the dude perma-welded to the cabinet. Last thing people want to do is rock the boat by expanding the risk to a vendor like Arduino, which quite frankly doesn't have a great reputation or history of tested industrial or reliable hardware. It's better to pay 3x as much for a well established vendor.

View on HN

I'm thinking (but can't confirm) that the intended market for these is companies that are too small to care about things like "well established vendors", where budgets are measured on the magnitude of hundreds of dollars and not thousands, and things are built from scrap or misc parts. I'm guessing that in some countries other than EU/Americas/etc, where things are done by "I know a guy that can figure it out", and people wearing safety sandals, this device would be convenient. I have reason to believe this market segment is both invisible and massive.

For a small scale industrial process, even a really hacked together setup can work surprisingly well, and while you have a chance of running into some show stopping problem, there's an equal chance of the hacked together setup continuing to function for the required time. If labour to fix stuff and/or small enough scale to temporarily have a little downtime are effectively free, this can work out positively.

I can see that things like these would be a good step up for those who were originally not even using PLCs. I think it provides a good "entry point" from Arduinos to something at least DIN rail mounted, and it's a way to experiment with ladder logic and whatnot. I personally would consider using it just for the convenience of that option.

I've watched someone in my social circle building a company around pre- and post-workout snacks and your line

> even a really hacked together setup can work surprisingly well

Is exactly right. Now that they got those three machines working, efforts go mostly into marketing, not into engineering.

That market is already covered by the cheap dev boards, arduino clones and Chinese PLCs at considerably less cost than the officially branded parts.

You can't even buy an Arduino Uno for the price these guys are selling PLCs for: https://www.aliexpress.com/item/1005004887420307.html

> I can see that things like these would be a good step up for those who were originally not even using PLCs

I agree that that's what Arduino thinks the market is, but I disagree that their conceptual market will buy this.

Arduino-branded hardware is typically 5x the cost of the clones (remember, it's OSH) at no noticeable improvement in quality. DIN-rail mountable arduino clones and modules with I/O isolation that run on 24V power already exist at very reasonable price points, and from reputable dealers such as Automation Direct. I've used ruggedized modules such as Industruino (although their prices have definitely gone upmarket) and they're nice. IME, the only people who buy actual Arduino hardware are either beginners who don't realize that the much less expensive clones are every bit as good, or business buyers who don't really care, because it's not their money.

But maybe I'm wrong and there's enough of a price-insensitive market to make this a success. I certainly hope so. I'm a freelancer, so the bigger the market, the better for me :-)

I have designed & built an arduino clone-based product that mounts on DIN rail because the hardware it communicates with also does. Multiple installations running large machine control applications without a hitch. The clones are fine.

I definitely see where you're coming from and I think these are valid points. In my personal experience though, there definitely does exist a market of "price insensitive, as long as it's under x". I think Arduino probably fits in there.

I feel like at least based on the quality of some boards vs clones that I have, there are definitely cases where I would have been "fired for not buying Arduino [IBM]" due to stuff like weird USB serial chips requiring driver hackery etc.

And I've definitely been in the situation where you tell someone it requires some part that costs "several hundreds of dollars" and they're like "oh no problem that's not an issue at all", because they were looking at the magnitude and not penny pinching.

I suppose that would be the market for this kind of thing. There's the other thing that you mention, where if you're using Arduino for this purpose, you might not be experienced enough to trust yourself to choose the right clone. Arduino provides sane defaults and generally the installations I've seen in the wild tended to use fairly close to "defaults".

You may have run into a counterfeit FTDI chip, then. The normal $5 arduino clones from reputable manufacturers often have genuine chips. The $2-3 ones are trouble.

Most of the clones have stopped even using the FTDI knockoffs nowadays. There are alternative USB-to-serial chips from other manufacturers, like the CH340 series, that work very well.

Yeah Automation Direct is already in this space with their Arduino PLC[1] which has compatibility with their existing modules/etc.

I'm only adjacent to the space and have use their stuff for some home automation but from what I've seen it's more important that you can get replacement parts and reliability.

[1] https://www.automationdirect.com/open-source/home

$43 - $66

https://www.automationdirect.com/adc/shopping/catalog/progra...

> For a small scale industrial process, even a really hacked together setup can work surprisingly well, and while you have a chance of running into some show stopping problem, there's an equal chance of the hacked together setup continuing to function for the required time. If labour to fix stuff and/or small enough scale to temporarily have a little downtime are effectively free, this can work out positively.

Can confirm that this is happening to some extent in biotech startups. Most of them don’t have the resources to hire grad students en masse to do the tedious labwork so they’re investing into lab automation. At the seed/series A level there usually isn’t enough money to splurge on brand name lab equipment beyond a few core pieces of equipment and with experiments running, there’s enough tech savvy downtime to experiment with hacky but cheap solutions. I’ve even seen HomeAssistant driven lab automation setups driven by AppDaemon apps with ESP32s used to control a mix of pumps and sensors (with CI/CD, though the whole thing was backed by a proper LIMS). It felt incredibly hacky but uptime was stellar once the SD card was replaced with an SSD.

It used to be that I had to go to Staubli or similar manufacturer and spec out a $50k robotic arm just to move plates from a liquid handler to the fridge. Now the same can be had for a few grand or less and there’s even open source options.

I see this PLC along those lines, where a lab might need to control a bunch of pumps and syringes with some synchronized via RS485 for realtime control and others on WiFi for less time sensitive stuff. A low cost PLC in that sweet spot between enterprise PLC and bare dev board could do really well.

These PLCs would be the perfect complement to ESP32s and RaspberryPi if they’re cheap enough to stockpile like RPis were preshortage. It’s unfortunate they haven’t published pricing yet, which tells me that it’ll be too expensive for hobbyists.

What are some good keywords to search for if I'm looking for some of these knockoff PLCs?

You can just go put "PLC" into AliExpress and you will get a bunch of FX01 clones, or you could put "GX Developer" since all of these seem to be compatible.

Unfortunately GX Developer doesn't appear to be free, but apparently Delta makes PLCs too (with actual UL certs too) which sell for about 130 USD on AliExpress and the software appears to be free (unsure). They also have reasonably (~80$) priced HMI LCD touch screen displays.

I think this is an attempt to break monopoly of those traditional PLCs. There are definitely going to be good use cases for ladder logic and "tried and true" programming model which was hasn't changed since 1990's. And in this case you are correct, the new Arduino makes little sense.

But sometimes you want something more modern, that you can program in the same language your main product is in, and apply the same expertise and tooling you already have.

I am not sure Arduino is the best option, given that their ecosystem is often very low quality, but there is definitely space for regular MCU in rugged case with protected I/O.

Arduino has a very good reputation for reliable hardware. Obviously this applies to official products. Shitty chinese clones have bad reliability issues...

Even some of those chinese clone PLCs are not spectacularly reliable... This is a professionally made device, developed in partnership with Finder which is a company with 70year of experience in making relays and other mission critical industrial equipment..

No the current Arduino ecosystem does not have a good reputation for reliable hardware at all. It does in your product echo chamber but quite frankly it's laughed out of the building in many professional settings and for good reason. The Chinese PLCs have somewhat more credibility there. The current hardware has little to no protection, poor power management, the software ecosystem is quite frankly dangerous in a lot of cases and the IDE only recently gained feature parity with tools we had 30 years ago.

I'm quite willing to entertain that you are trying to get into the market and that these issues may be resolved with this product but you have a large reputation mountain to climb first.

You have to build reputation not try and add it afterwards.

To note I use Arduino projects usually for one off simple bench automation but nothing further than that. I tend to use them as a host for flat avr-libc / avr-gcc projects though, mostly because the old IDE and toolchain was absolutely dire.

True or not, "reliability bordering on childproof" can be a compelling sales pitch. And reputation is bound by recall, of whuch Arduino has plenty: brand X might have the best reputation imaginable amongst those in the know, but if you aren't in the know and you haven't ever heard of X all that reputation won't affect you. For all you know, X could be terrible. If there's another brand Y you have heard of, and only half of what you heard was negative, brand Y has at least some reputation (the non-negative half) positively affecting you.

That's the basic mechanism of why so much advertisement is just about brand awareness, not talking about quality at all: "I've heard a lot about them, and not everything has been bad" is less bad than "never heard about them, even if they were really terrible I wouldn't know"

I don't think you understand the industry. Reliability is qualified through testing, certification and support. It isn't a function of marketing.

Marketing gets you to the testing table. But better not have any skeletons in the cupboard.

Edit: I suspect you're confusing engineering for the niche sector of makers, the latter of which are driven by fads and marketing (sorry if I piss anyone off there).

“The industry" isn't a single uniform group. Sure, people who spent their career procuring industrial controllers won't look at Arduino even once when they introduce a non-maker form factor, and chances are they wouldn't be impressed if they did. But I can easily imagine that there's an underserved space between the extremes of the maker juggling dev-boards not really intended for him and the career specialist who goes to all the right trade shows. Some company for example that has barely enough controller needs to set aside full time staff for that, but does not want to outsource either.

It's possible that Arduino fantasizes about taking over large fractions of the industrial PLC market and if that's the case I'll chuckle agreeingly with everybody ridiculing them, but I believe that there could far more reasonable and realistic opportunities for them to stretch a little in that direction.

Arduino is a marketing company that makes amateurish overpriced dev kits and sells them to amateurs. They have that market largely thanks to marketing moves.

Arduino products rarely sell to people who do any form of QC or comparison shopping - they sell to people who buy the first thing on their mind.

Most of the Chinese embedded companies are engineering groups that understand they are filling the low end of the professional cost-quality curve, but they are still on that curve. For what you pay, they make a great product.

I see multiple comments recommending random Chinese PLC device from Aliexpress over this one. Where is the reputation and brand recognition of Arduino positively affecting anything here?

It also seems to me that your definition of the word "reliability" isn't agreeing well with others - to me it seems that these industry guys are using the word along "semi believable certification documents and soft established likelihood of not catching fire" whereas yours are more along "blind idolization for moral qualities seen in individual corporate employees". That might need a fix.

A professional level device would have ESD protection, power filtering, and EMI controls. Something you'll never find on an Arduino product.

Atmel yes, idk about Arduino.

Arduino has a terrible reputation for hardware and software from a professional, reliable point of view.

It's fine from a easy, hacked together hobby project point of view, but that's an entirely different thing.

It's like Python vs Rust. I wouldn't introduce a beginner to programming with Rust, but I also wouldn't program my manufacturing line (or whatever) with Python.

Reply on news.ycombinator.com