 Notice. New forum software under development. It's going to miss a few functions and look a bit ugly for a while, but I'm working on it full time now as the old forum was too unstable. Couple days, all good. If you notice any issues, please contact me. |
Forum Index : Microcontroller and PC projects : Arduino goes CLOSED-SOURCE PLC
| Author | Message | ||||
| Tinine Guru  Joined: 30/03/2016 Location: United KingdomPosts: 1646 |
Last time I checked, their industrial Portenta product had zero galvanic isolation. What a joke: Painful to watch Craig |
||||
| JohnS Guru Joined: 18/11/2011 Location: United KingdomPosts: 4147 |
The Portenta H7 is somewhat like the CMM2. I wish I knew how like it / what changes MMBasic would require... hmm... well, if they can actually get the CPUs to make the boards! John Edited 2022-12-24 20:58 by JohnS |
||||
| Tinine Guru Joined: 30/03/2016 Location: United KingdomPosts: 1646 |
RP2040 (or multiples) + MMBasic is a dream combo for this. I would also incorporate one or more Click sockets. Job done. I would use the VGA for diagnostics....LED indicators are a PITA when you need to climb a ladder or plop down on the floor (reality) just to check I/O status  We have everything  |
||||
Mixtel90 Guru Joined: 05/10/2019 Location: United KingdomPosts: 8304 |
Sheesh.... I managed to sit through 12 mins. How can people program like that? Mick Zilog Inside! nascom.info for Nascom & Gemini Preliminary MMBasic docs & my PCB designs |
||||
| Tinine Guru Joined: 30/03/2016 Location: United KingdomPosts: 1646 |
Geoff and Pete need to form a commercial spin-off. Name it MM-Logic and take over the industry |
||||
| scruss Regular Member  Joined: 20/09/2021 Location: CanadaPosts: 95 |
Some of the STM32H7 and F7 chips are slowly coming back on the market. Not the STM32H743IIT6 used in the CMM2, though. OpenMV have finally managed to restock their amazing computer vision boards that use STM32 high-end chips. They so nearly disappeared from the market, and they were great programmers with a good product. I'm enjoying no longer being an official Arduino reseller. Around 2017, we could sell a couple of hundred Arduino Starter Kits, about 500 Arduino Unos and maybe 50 Arduino Megas a year. By 2022, we were only shifting about 50 Starter kits, 50 Unos and 3-5 Megas per year. Other suppliers has completely taken the market, selling better products for less - for instance, Cytron's Maker UNO we could not keep on the shelves. Contrast that with the more esoteric Arduino boards like the 101: we sold two, but one got returned and sat in our warehouse for years. While the Arduino starter kit is somewhat overpriced, it's still a really well-put together intro to physical computing. The component selection, the packaging, the chunky but very readable paper guidebook: nobody else comes close. Arduino's markup for dealers was ridiculous. We typically had to buy a year's supply to get any discount at all: we'd be passing them on at cost otherwise. I don't know what future they have. I mean, they started by knocking off their board from a grad student ... |
||||
| darthvader Regular Member Joined: 31/01/2020 Location: FrancePosts: 95 |
It will be fun to see what append ;) But seriously , go to Daimler , Porsche , Audi or other big industry and try to tell them you go for install the arduino PLC :) Lol , i think it take less than 2 min to be fired and black listed from the company :) Just use Siemens and allen bradley <-(in Germany) or telemecanique <-(in France) and you are OK ;) Theory is when we know everything but nothing work ... Practice is when everything work but no one know why ;) |
||||
| Mixtel90 Guru Joined: 05/10/2019 Location: United KingdomPosts: 8304 |
I've pointed out something similar previously. Industry wants a "name" that they know and can see the glossy brochures for. Beancounters don't care what's inside the box providing it looks the business and the cost can be covered in the contract. That doesn't mean it has to be cheap, far from it, in fact if they think it's too cheap they will tell the engineers to look for a bigger "name". Try to sell a PicoMite PLC in it's own enclosure and under it's own name and with a great version of MMBasic and you might sell five (to big companies, that is) if you are lucky. If Telemecanique or Siemens used a RP2040 in one of their PLCs they'd sell thousands, even with a vile programming language. Small companies are more forgiving but use less PLCs and are all looking for the cheapest that are carried in stock (or are on short delivery) so the competition (especially from the Chinese) is intense. The quality of the software development system is relatively unimportant. At the moment engineers are trained up on one or more PLCs, usually from the same manufacturer so there is a "house" similarity between the different models. Throwing another system into the mix - one that isn't used by anything else and is perceived as using a "dead" programming language - isn't going to be popular. You'll probably find far more engineers that are happier in C or C++ than in MMBasic. IMHO the Arduino, Raspberry Pi and PicoMite are dead in the water when it comes to "proper" industrial PLCs. Not unless one of the "names" adopts one and sells it as part of their range anyway. It doesn't matter how good the hardware and software are, when you can just include the cost of a stupidly priced PLC in the job price and still get it then that's how things will continue. After all, manufacturers, distributors etc all work on percentage mark-up. If the item is cheap and relatively slow moving then there's not enough mark-up to make it worthwhile. End of Mick's Boxing Day rant. :) Mick Zilog Inside! nascom.info for Nascom & Gemini Preliminary MMBasic docs & my PCB designs |
||||
| Tinine Guru Joined: 30/03/2016 Location: United KingdomPosts: 1646 |
Industry wants to make products....period. I am currently quoting a customer for my RP2040 + Prop-2 system because he needs working machines since several months ago. I have hundreds of machines throughout Ford, GM, Chrysler and all of their tier-1 suppliers. The RFQ always stipulates AB/Siemens/Beckhoff. Nope. You get a generic PC motherboard....Never was a deal-breaker. Lander Automotive (Birmingham, UK): In 2015, I provided them with my Android tablet based 6-axis CNC. Their inside controls engineers recently contacted me to say that this has inspired them to start switching away from traditional suppliers in favour of Node-Red, Android tablets and something like the RPi. To paraphrase Eben Upton: "in spite of the shortages, we've managed to support our industrial controls manufacturers". Craig |
||||
| Mixtel90 Guru Joined: 05/10/2019 Location: United KingdomPosts: 8304 |
Your multi-axis motion control field is specialised though, Craig. You couldn't substitute one of their control systems with a "middle market" PLC unless it ran slowly. Mick Zilog Inside! nascom.info for Nascom & Gemini Preliminary MMBasic docs & my PCB designs |
||||
| Tinine Guru Joined: 30/03/2016 Location: United KingdomPosts: 1646 |
Give me an example of what the RP2040 can't handle. As you know, PLCs replaced mechanical relays. Using a state machine approach in MMBasic, not much one can't handle and coming from one who lives with this stuff, VGA output is a huge benefit. Scenario: Machine stops unexpectedly. Company is too small to justify a technician. They call their local contractor who can't be there for a couple of days (whatever). Technician shows up and finds that it's nothing more than a loose sensor. It's common practice that if they are charging by the hour, they pretend that they are stumped to bump-up the billing. Now imagine a VGA display that clearly explains why the machine has stopped. Loads up step-by-step diagnostics text file, etc. Every service tech can relate to having to hop on a plane to some far out place at great expense only to find a trivial issue. String a few dirt cheap RP2040s together for multi-processing I/O expansion Edited 2022-12-27 00:06 by Tinine |
||||
| Mixtel90 Guru Joined: 05/10/2019 Location: United KingdomPosts: 8304 |
It doesn't matter what it can or can't handle. My point is what does someone looking for a PLC *perceive* to be the right one for the job? I've been through this at work, arguing that most bottom end PLC jobs could be done just as effectively and cheaper using a PIC chip, but at the end of the day whos name is on the box? That's what will make it acceptable to the majority of low-middle end PLCs. I appreciate your point of view, but in my experience the customer is happy to see a name like Telemecanique (or Merlin Gerin), Siemens or SquareD and will pay for it. They are nowhere near as happy with an un-named PCB in a box or on a DIN rail mount even if it's saving them money. They are even a bit shy of Mitsubishi and Omron sometimes, although both have been in the business for many, many years. Mick Zilog Inside! nascom.info for Nascom & Gemini Preliminary MMBasic docs & my PCB designs |
||||
| Tinine Guru Joined: 30/03/2016 Location: United KingdomPosts: 1646 |
No, you are correct but this is usually dictated by some know-nothing jobsworth who doesn't want to get in trouble. If your entire plant employs one system then great but this is not reality. The company buys each piece of equipment based on its process capability/suitability. Poor maintenance guy finds himself hopping from Telemec to MG to AB to Siemens, yada yada. And then we have the ridiculous variety of fieldbus systems to deal with. They have to plug-in a laptop and try to remember how to use the overcomplicated diagnostic software. And then they find that this laptop doesn't have the dongle, etc. The real world is a total freaking mess and it's all because each one of the traditional suppliers wants to tie you in to their particular eco system. And then you have the in-house guys who upgrade older equipment from relays to PLC. Typically, one guy gets into it while the others don't want to deal with "this computer stuff". I had one of my systems on the front end of a production line which would randomly freeze up due to another piece of equipment. They'd call their "computer wizard" (guy who learned to program the PLC) to save the day. I got sick of this and so one time I told them to let me look at this PLC program before they called him. He'd programmed a deliberate latch-up based on time and number of cycles. He never stepped foot in the plant again. In this day and age, there is no reason to not have a running display of what's supposed to be happening at all times. It's all marketing BS and brainwashing. EtherCAT makes life more difficult for people like me. Who in their right minds would be in favour of throwing hundreds of megabits/sec in a sea of unavoidable EMI. There is absolutely zero benefit to the end user. That OMRON system above, you see the price? It's not even their flagship. Their flagship boasts 5 axes in 50uS "the world's fastest" 1) This number is meaningless because no mechanical device can respond this quickly. You would be doing well to achieve 100Hz (prove me wrong) 2) My Propeller laughs at this but I won't embarrass myself by using BS to promote a product (I have 6 axes in < 800 nanoseconds). Craig |
||||
| JohnS Guru Joined: 18/11/2011 Location: United KingdomPosts: 4147 |
Reminds me of "no-one got fired for buying IBM" (mainframe days) More recently, Microsoft (Windows etc). However, those sentiments (fortunately) are not universal - there are people who care more about things like reliability & cost. Those may keep them in business or allow them to compete successfully. John |
||||
| Mixtel90 Guru Joined: 05/10/2019 Location: United KingdomPosts: 8304 |
I completely agree, John! I did find just a few people that would accept alternative ideas - and those ideas almost always worked out very well indeed for all parties. Unfortunately none of them resulted in large repeat orders - which wasn't our field anyway, it was all bespoke singles or small runs. Saved thousands using Russian vacuum contactors on a couple of jobs. :) Mick Zilog Inside! nascom.info for Nascom & Gemini Preliminary MMBasic docs & my PCB designs |
||||
 Print this page |
 |
The Back Shed's forum code is written, and hosted, in Australia. | © JAQ Software 2025 |
