In my last post I announced to have a look at CNC automation in more detail. I also mentioned that just putting a robot in front of your machine tool won’t do the trick and I mentioned that in the first step you will need to make your production fit for automation.
But where and how can you start? Today I am concentrating on the following four aspects which are the building blocks of your automated production:
- Process
- Environment
- Machine Tool
- Workforce
I am not saying that this list is scientifically proofed to be fully comprehensive, but I promise at least if you can tick it successfully off, you should be good to start planning a robot purchase in more detail.
It all starts with a proper process
I know – most of us get the shivering if they think ISO9001 – calm down! If you are not certified yet, you won’t have to prior you proceed reading (although in the long rung it will be worth it, but that’s another story). For now, just pick one of your machine tools and stand next to it in the very moment when the worker is starting to set-up a new part. Can he/she do the change completely without requesting any support? Does he/she have everything he/she needs to do the change in time? How is the quality in line secured? You should spend some time and have a look at different machines and – if applicable – during different shifts before you are ready to sign it off!
Environment
If you are just considering building a new production hall, well then you are lucky. For the rest of us it’s time to think a moment of your strategy and analyze if our current set-up on the shop floor really fits to our strategy.
Besides following a low-cost strategy, I mentioned basically these two strategies, on which I am going to concentrate here:
- High-end parts and other specializations
- Highest efficiency
You may ask me now: what does my strategy have to do with the environment? Bear with me for the moment. I am coming back to that in a minute.
Environment can be everything: layout of your shop floor, cooling you are using, chip disposal, your intralogistics strategy, programming environment, and so on. In our example I want to concentrate on one fundamental factor: temperature.
We put a gauge block of 100 mm in height beneath a micrometer. Only through my breath hitting the gauge block for a moment did its length change by several microns! Now think about the size of your machine tools!
Temperature changes are the biggest enemy of accuracy. Contrary to physics, however, production halls that are not air-conditioned are found frequently, when you are visiting medium-sized companies.
Please don’t tell me you are producing parts with a tolerance in length of +/-.001 mm if you have a non-isolated hall and the machine is standing next to a door so that you can expect sudden and frequent temperature changes of 20 °C and more as the door next to it leads to the loading ramp. I once worked for a company with such a set-up so unfortunately this believe is not unrealistic. And no: only if your screen says X 2,175.004 mm, it doesn’t mean that your tool tip is standing there.
Now take away your experienced worker from this machine who – after 5 years of experience – knows exactly how he must compensate the lack of stability and you have absolutely no chance to reach your quality goals. Remember: your plan is to automate this machine!
If we have a look at this simple example and coming back to our two strategies, I do believe you agree that your machine tool should not stand next to any loading door.
I am not saying you need 21 °C +/- .5 °C the whole year in order to produce the desired quality. But you want to have at least a stable condition which is withstanding severe temperature changes.
What does high efficiency mean if we break it down to a more practical approach? If I learned something during my time in the industry, I could say that there is no typical production setting. It’s always very specific and this diversity is great as it opens enough room for exactly your niche.
But at the extremes, high efficiency means a low mix of parts with a very high volume (cycle time 30 s and less). Best examples: cell phone casings or parts for the drive train as e.g. cylinder blocks. As you have – literally – millions of parts, you are working with statistical process control and have the chance to adjust machining parameters on the fly. If you are controlling temperature or at least temperature changes, however, the hysteresis will be a lot smaller compared to the one in an uncontrolled environment.
If high-end-parts production is your cup of tea, I’ll really encourage you to invest in proper temperature control. Why is that you may ask? Because now we are coming to the other extreme: high mix with low volume, at the extreme: batch size one. This is typically the case in die and mould industry. If you don’t meet the desired quality in one shot, you will have to start reworking the parts, which is killing your margin and the promised delivery time. Believe me – I witnessed this unfortunately quite often in my live: good hope doesn’t make good parts!
Machine Tool
I could write a series of posts about machine tools alone, but as we are talking about robotics here and as this was meant as a primer for starting with machine tending operations, I will just touch them briefly.
As I already mentioned earlier, as a kid or young adult I saw manual machine tools on the shop floor and still today they are typically being used for educational purposes – although I believe the skilled worker of today should rather be educated in robotics from the beginning and in working with CAD/CAM systems than turning the wheels manually, but that’s again a different story.
Later – when I started my career in the machine tool industry – transfer lines were still common in light duty engine production. When I started in engineering, I met the last silver necks using still their drawing boards. One of them called transfer lines “piff paff” machines – an analogy for a single axis modules moving just back and forth, boring some holes.
During the next decade, the requirements in high-volume production changed completely. I started in machine-tool industry in an automation group – believing I would plan and commission robot cells only to find out that my main job was to plan and implement (with suppliers) line automation. So, my tools at that time mainly were gantry loaders and conveyors.
But car builders diversified. By pushing more and more different car models to the market, they needed to become more flexible – also on the shop floor. Transfer lines started to fade away. Today, if you have a high volume of prismatic parts, horizontal 4+ axis machining centers are your weapon of choice. You can configure them to the operations so that sometimes you have a B-Axis, B- with integrated A-Axis, A-Axis bridge… or you are even using several spindles so that you can do the operation in parallel on 2 or more parts in one shot in one machine.
High volume typically means splitting the workload between spindles in several operations and balancing it out so that you meet the desired cycle time with each spindle. You can imagine that it is not the target to change parts on such a line several times a day. For high-volume production, the machines are typically adjusted to your process. You purchase them for a single project (as an extreme).
On the other side there is low-volume production with a high-mix of parts. Here you want to buy flexibility because often you don’t know this week what you are going to machine next week. So, the machine needs to be as flexible as possible. If you want to get highest flexibility, go for 5-axes machining centers.
Today there are 5-axis machine tools available which are even designed for milling mould inserts – talking about accuracy.
One word about your machining process. As you want to run your machine tools autonomously with a robot in front, you want to make sure that they behave predictable. That means you want to have machine tools which have a thermal stability.
What goes without saying is that the accuracy of the machine tool needs to fit to your process requirements. And unfortunately, I must mention that – if you just look after these two, thermal stability and accuracy – you can say the higher your requirements, the higher the purchasing price.
And no: you cannot reach the same results with a low-budget machine which gets some mathematical compensation. The only way to find out if the desired machine tool is a perfect match for your requirements is the good old test cut.
If you want me to deep dive into this topic, let me know in the comments below and I might consider doing some feature about machine tools in the future.
Workforce
I briefly started talking about education above, and unfortunately it reflects reality. Skilled workers with experience are hard to get for the shop floor. And if you want skilled workers who can set-up your NC process and the robot, it’s like you won the lottery.
Okay, maybe I’m exaggerating a bit, but you get the point: if you want to automate, you need to invest in education. There are certainly suppliers who promise you that you can set up your automation process intuitively and without any training, but the question is surely what you want to achieve with your automation and whether a minimal result is enough for you. Also, this might be the moment considering to working with external partners if you don’t have the expertise or capacity in house.
As this post already became rather long, I will stop it here and next time we will have a closer look at different possibilities to finally start automating your machine tools.
Let me know if you took a similar approach automating your CNC machines on your shop floor or how you started using robots in your production. Also, I am always curious to getting to know your lessons learned.
If you liked reading this post, I’ll appreciate if you share it to your network or if you leave a comment below which might help me identify other topics of interest for you.