On Wed, Jan 29, 2020 at 3:34 AM Les Newell <[email protected]> wrote:
> > > Why are they not using > > LinuxCNC to make transmission parts at the Toyota factory? ... It must > be something about > > LinuxCNC that makes it unsuitable. > > Large companies want a known brand high end machine with a comprehensive > support contract. Big CNC manufacturers use either their own control or > a big name control manufacturer such as Siemens. If they have their own > control they aren't going to switch to anything else because they have a > huge amount of time and money invested in their current systems. If they > use someone else's control they want a complete hardware + software > package with support. Either way they aren't going to even consider > LinuxCNC and that is unlikely to change, no matter how good LinuxCNC is. > The mindset at the larger companies is not one of the 'time is money' rather than a cya attitude. The manager of the larger company answers to an entity above. The entity above doesn't make the decision to install LinuxCNC or not. The manager of the company makes the decision. The manager will decide based upon whether the decision is justifiable to keep his job. The paradigm shift lever is VERY difficult to move from expensive but justifiable to cheaper but very possibly not justifiable. The known entity of expensive set it and forget it is a hard nut to crack. For a smaller company a customer will evaluate the equipment list to determine the capability and reliability of the vendor's equipment. The old adage of "buy the same equipment as your customer" plays very well in the vendor evaluation process. A secondary consideration in the reliability evaluation is "how many of one machine does the vendor have?". In my world, for production parts, a customer would like to see three machines available for two machines if work. That assures machine capability at all times. A machine can go down and the vendor will still have the ability to deliver parts. We are talking about the 'get in the door' time frame. As trust is developed a customer will allow a vendor to violate the ratio but it takes a long time to develop that trust. Another factor in the trust arena is many companies move the buyers around to break the relationship between the individual buyer and vendor. Trust leads to abuse. In those cases the vendor is always facing the trust issues. There remains a trusted history between the customer and vendor but the individual buyer will 'cya' until the individual buyer begins to trust the vendor. Rinse - Repeat Remember - 1000 ataboys can be erased by one 'ah shit'. Another story In 1997 I purchased a 5 axis license from OpenCNC (MDSI2). OpenCNC (closed source) ran on QNX RTOS (closed source) using hardware from a list of . The Open meant OpenCNC didn't have hardware to sell but they supported a limited list of approved vendors (all closed source). The hardware vendors were very proud of their offerings. I ended up with three 5 axis machines running OpenCNC. I had to have a license to run each machine. Software and hardware was something just south of USD20,000.00 per machine. One of the machines was running OpenCNC (QNX) until 2017. Open had a slightly different meaning than we think of open today. :) I will say I liked their product. Their total package including documentation was complete enough I was able to install, configure and run it with minimal contact with OpenCNC staff. They were very helpful and knowledgeable. A very polished organization. I purchased their API and would probably still be running their software today but in the next release they left QNX and settled exclusively on Windows and Venturcom RTOS. They would no longer support QNX. I almost cried. I called my contact and told him they just lost me as I would not purposely install Windows on my machine. For me this was preLinux. I came from (I am no VMS guru) VMS and Windows. I know the difference between a productive OS and a pretty OS. I am almost glad OpenCNC took the direction they took because if they had not left QNX I probably would not have found LinuxCNC. When I bought the software OpenCNC was already in one of the big three in Detroit. The requirement was OpenCNC has to look and respond EXACTLY like all the other controls in the shop. I was told it took the tech department in said shop one day to develop the human interface screen and MDSI configured OpenCNC to run just like all the other controls in the shop. I don't know if that project is ongoing. A 300 man shop South of Wichita started installing OpenCNC after I had completed my installs. I knew a couple of the techs installing OpenCNC there and a couple operators running it. All reports were the techs and the operators loved it and lobbied for it to be installed on every machine. I don't know if that project is ongoing as my contacts have either died or retired. A four or five man group from Cessna visited my shop to evaluate OpenCNC. I believe it was an operator, a maintenance tech, a software developer and a finance guy. They installed it on at least one flat bed router. Reports are they loved it. Modern Machine Shop did a feature on the Cessna install. I don't know if that project is ongoing. Bruce Norse (of Compact II by MDSI and OpenCNC by MDSI2) started what became OpenCNC to develop a factory automation application. This was to collect information from any and all controls in the factory and present it to management. Somewhere around 2000 Briggs and Stratton bought MDSI with the plan to utilize the factory management software world wide. I am not sure how that worked out but in relation to OpenCNC it seemed to stop development. All of this was during the 1997 through the 2001 time frame. I don't know why OpenCNC seemed to go into hibernation then as it was used and loved by small shops and very large shops. Maybe Briggs stopped it on purpose or with Bruce no longer in ultimate charge there was no overriding dream to develop it further. The "OPEN" world had an opportunity to grow and flourish at that time and seemed to just fizzle. > > Smaller CNC manufacturers are a different story. There are a number of > smaller plasma and CNC router manufacturers using LinuxCNC with > considerable success. Tormach is another example of a smaller > manufacturer using LinuxCNC to run a range of CNC machines. > > > and is basically just a kt of > > parts and not a product and it does not scale well to large factory floor > > sized systems. > > LinuxCNC is a small open source project. It could scale well for large > machines but it doesn't have the comprehensive support you would get > from a closed source control manufacturer. For end users time is money > so support is a major deciding factor. > > Les > > EMC (Enhanced Machine Control) seems to have fizzled out. LinuxCNC is largely the result of 4 of 5 guys giving their time and intelligence to the project. In no particular order I thank Matt, Jeff, Chris, John and Stephen. Their contributions were done without profit motive. Another group of men with contributions far in excess of their profit motive is Jon, Peter and Steve. What a group of fine men. This list is not necessarily complete. If anyone would like to add to the list then I encourage the addition. I appreciate each and every one on this list. I like the latest discussions and hope a way forward is found to continue the development and proliferation of LinuxCNC. Regards Stuart > > _______________________________________________ > Emc-users mailing list > [email protected] > https://lists.sourceforge.net/lists/listinfo/emc-users > -- Addressee is the intended audience. If you are not the addressee then my consent is not given for you to read this email furthermore it is my wish you would close this without saving or reading, and cease and desist from saving or opening my private correspondence. Thank you for honoring my wish. _______________________________________________ Emc-users mailing list [email protected] https://lists.sourceforge.net/lists/listinfo/emc-users
