Machine data collection networks – Wired and Wireless
Machine data collection requires machines to be connected in a variety of ways to the computer that has the software for analyzing and taking action on the data.
Industry 4.0 involves these:
1. Capture of data from the machine.
2. Transfer of the data to a computer that has software for analyzing the raw data.
3. Reporting the data to people, or sending it to other software.
The computer can be on-premise, meaning within the physical boundaries of the organization, or on the Cloud. Connection can be via a mobile phone network or via wired or WiFi LAN and broadband. These are the options available for machine data collection.
The sensors on machines send data to a server within your organization. The connection between sensors and the server can be wired or WiFi LAN. Users can see reports if they are on the same LAN, from within the shop floor.
Cloud based system
The sensors on machines send data to a Cloud server outside your organization. The connection between sensors on the machine can be direct IOT, via a mobile phone network, by wired LAN or WiFi LAN. Users can see reports from anywhere on earth, on the Web. In the case of the direct IOT via the mobile phone network, there is no IT infrastructure required on the shop floor for data transfer. In the wired or WiFi LAN systems, you need LAN cabling, switches, access points, repeaters, etc. In the on-premise system, you additionally need a server within the organization.
Here are the layouts for various systems of data transfer from the sensor to the server.
Cloud based systems
IOT hardware to mobile phone network to Cloud
WiFi LAN to Broadband internet to Cloud
Wired LAN to broadband internet to Cloud
LEANworx Cloud is a Cloud based system that supports all the 3 connection options above – IOT direct to cloud, WiFi LAN and Wired LAN.
On-premise server systems
WiFi LAN to on-premise server
Wired LAN to on-premise server
Requirements for various machine data collection systems
Yes means Required, No means Not required.
|IOT sensors direct to Cloud server||WiFi LAN to Cloud server||Wired LAN to Cloud server||WiFi LAN to on-premise server||Wired LAN to on-premise server|
|Server in your organization with mirror server, RAID||No||No||No||Yes||Yes|
|Infra for server:|
room with physical protection, air-conditioning, 24/7 power
|IT personnel for server maintenance||No||No||No||Yes||Yes|
|IT personnel for LAN maintenance||No||Yes||Yes||Yes||Yes|
|LAN cabling in shop floor||No||Yes||Yes||Yes||Yes|
|Wired or WiFi LAN infra on shop floor:|
switches, WiFi access points
CNC machines – machine data collection
Data from CNC machines can be collected in various ways based on the features available on the machine.
Method 1: Send out data through RS232 serial port using Macro calls
E.g., On Fanuc or Haas. Data is sent out through CNC machine’s serial port using the DPRNT command whenever there is a cycle start/end, spindle on/off, etc. Machine’s serial port is connected to a server via serial-to-LAN converters, over wired or WiFi network. Data is sent to the server, reports are visible on PCs on the local network.
- Limited to CNC machines that have Fanuc Macro B option or equivalent.
- Macro B (or equivalent) option costs money.
- Requires IT infrastructure – cabling, serial or LAN hardware, server, IT personnel.
- Long installation time because of LAN and RS232 port connections.
- High capital cost and maintenance cost of network.
- Requires macro calls to be inserted in every program. Does not work if people forget to insert these, or deliberately do not insert them.
- High downtime because of variety of equipment involved – cabling, LAN hardware, server. If any of these fails, the system fails. It is also susceptible to sabotage because everything is in the open.
- Mobility of machines is poor – to move machines within your shop, you’ll also have to rewire the serial connection from the machine and the LAN cabling from the machine to the server. Machine monitoring is down for the duration of this rewiring, which is typically a few days.
Method 2: Send out data through the machine’s ethernet port
E.g., with Fanuc Focas, OPC-UA on Siemens, MTConnect, Haas ethernet, Mitsubishi ethernet. Data is sent out through machine’s ethernet port. Machine’s ethernet port is connected to a server via serial-LAN converters, over wired or WiFi network on shop floor. Data is sent to the server, reports are visible on PCs on the local network, OR Data is sent to the Cloud, reports are visible on internet on PC, tablet or mobile phone.
- Large variety of data can be acquired – spindle power, FRO status, spindle temperature, etc.
- Requires machine to have ethernet communcation capability.
- Long installation time for cabling and LAN infrastructure.
- High capital cost and maintenance cost of network.
- Limited to newer machines, typically after 2015.
- Requires local IT infrastructure – cabling, LAN switches or WiFi access points, server/ internet router
- Requires a team of IT personnel to maintain the IT infrastructure.
- High downtime because of variety of equipment involved – cabling, LAN hardware, switches/access points, server, router.
- Mobility of the system is poor – to move machines within your shop, you’ll also have to rewire the LAN connections to the machine. This can be improved by using WiFi LAN.
Method 3: Monitor status of signals at machine’s relays
This is done through a sensor that tracks digital signal lines from the CNC machine’s PLC. Send the data direct to the Cloud. Hardware sensor reads status of relays to determine cycle start/end, spindle on/off, etc. Sensor directly transmits data to Cloud via mobile phone network. Reports are visible on internet on PC, tablet or mobile phone.
- Short installation time – 30 minutes typical.
- No IT infrastructure required, and hence no IT maintenance.
- No capital and recurring cost of IT infrastructure and maintenance.
- Works only with machines that have signals on 24V digital outputs.
- No downtime caused by IT equipment failure.
- Cannot be sabotaged, because the sensor hardware sits inside the machine control panel with nothing coming outside.
- Mobility of machines is very high. The sensor hardware moves with the machine, and starts tracking as soon as the machine is powered on in its new location.
- Can only get on/off type of data – cycle start/ end, spindle start/stop, machine fault. Cannot get program number, feed rate override status, spindle temperature, etc.
Comparison between various methods of machine data collection on CNC machines
Method 1 is the least preferred, for reasons of cost, installation time, and poor reliability. It is obsolete.
Method 2 gives the most data, but is limited to newer machines and involves LAN connections.
Method 3 involves the lowest cost and installation time and has the highest reliability, but cannot track machine internal parameters.
LEANworx Cloud Industry 4.0 supports method 2 and method 3 on CNC machines.
Machine data collection from machines with no electronics
You will notice that there is nothing here that indicates the level of automation of the machine. Machine data collection means tracking its status, getting data from it – whether it is running or idle, parameters like speed, temperature, pressure, vibrations, etc. What you need to capture depends on the type of machine it is, and what its purpose is. If you can connect a sensor to a machine, you can track it, irrespective of its level of automation. The type of sensor depends on what data you want to acquire, and today’s sensors can track a huge variety of parameters. The data that is acquired is analyzed to report to humans, or control machines. With a manually operated machine, the software cannot control the machine automatically. It can however report to the human who is operating the machine, to take necessary action.
With the appropriate sensor to track machine activity and parameters, data can be collected from any machine at all, and the machine made part of your Industry 4.0 system.
If the machine does not have appropriate electronic signals, just retrofit the signals with sensors. ‘Sensors’ sound scary and expensive, but are actually very affordable. For most applications you will just need on/off type sensors – spring loaded mechanical switch, inductive or optical proximity sensor. They cost as little as Rs. 1000.
What is Proof and ABV on alco(hic!)hol bottles ?
Alcohol by Volume, or ABV, is the percentage of ethyl alcohol in the total volume. You’ll find ABV Alcohol by Volume also written as ABV, V/V, or alc/vol. All these mean the same thing – 40 ml of alcohol in every 100 ml of contents (the other 60 ml being mainly water): 40 % V/V, 40 % ABV, Alc. 40% by vol, 40 % vol, 40% alc/vol.
Proof is a more complicated and confusing term. The term ‘Proof’ originated in England 500 years ago. Rum was taxed at different rates depending on its alcohol content. The rum was tested by soaking a pellet of gunpowder in it. If the gunpowder could still burn after the soaking, the rum was rated as ‘above proof’ and taxed at a higher rate. Gunpowder would not burn after soaking in rum that had less than 57.15% ABV. So rum that contained this percentage of alcohol was said to be “100° (one hundred degrees) proof”.
To convert Proof to ABV, just multiply by 0.5715. See the label below, and note the year of manufacture. It’s 86 % ABV (that’s a crazy amount – the average whisky is about 40 %), which means 86 % alcohol and 14 % water. The label actually says “flammable”.
In the US, proof merely meant 2 x ABV. So if a bottle said 100 % Proof, it meant 50 % ABV.
Very confusing, this Proof business. Happily for us tipplers of today, Proof is obsolete.
ABV, which is easy to understand even with faculties dulled by alcohol, is the standard used the world over. Here’s a page that has ABV for various types of liquor.