My expertise in controls is primarily focused on Rockwell Automation PLCs. However, I believe they may not be suitable for this project due to sensor sampling speed limitations. I am uncertain about other options to explore in this regard. With Rockwell Automation hardware, I have mainly worked on simple assembly benches and CIP servo motion projects. I am currently assisting a small company in gaining more experience in this area, as well as enhancing my own skills. The project at hand involves spinning a part at 150rpm using a servo and encoder, and counting teeth on the part using a 5v crankshaft position sensor. The sensor detects 58 teeth (60 with 2 missing for a register position) by capturing rising and falling edges, resulting in a square waveform on an oscilloscope. The goal is to graphically plot or log voltage spikes of the teeth against servo rotational position, analyze the data to determine peak to peak measurements and extrapolate degrees of rotation between teeth. The recommended servo/encoder for this project is a Kollmorgen absolute hiperface encoder. It is essential to accurately capture the register position of missing teeth and log voltage spikes against rotational position. Quality assessments will be made based on the spacing of the teeth, with bent teeth resulting in part rejection. The data acquisition process should be completed within one rotation of the part (0.4 seconds at 150rpm), including the identification of missing teeth during random loading. I have looked into National Instruments (NI) as an alternative, but integrating the hiperface encoder appears to be complex. Consideration is being given to using a simple incremental encoder instead. Options from a company called Promess, such as the UltraPro product, could meet the project requirements but are expensive. There is also interest in Kistler KiBox and Beckhoff solutions, although the latter may require additional knowledge in programming and software requirements. Beckhoff's compatibility with EtherCat, a communication protocol used by Kollmorgen, makes it a potential choice, although further research is needed. Any guidance or recommendations on suitable control systems for this project would be greatly valued.
Although it may seem like a vision application, it may not align with your cost objectives. When considering a PLC-based solution for a system operating at 150 RPM with 60 teeth, it's crucial to have a setup that can handle transitions between states (off > on or on > off) every 3.33 milliseconds. One way to achieve this is by using high speed counters to sample the signal. Connect the signal from the wheel under test to one high speed counter, and connect an encoder with a higher number of pulses per revolution to another high speed counter. The challenge then becomes transferring the accumulated count from the high speed counters to the PLC in under 3.33 milliseconds. For example, if the second encoder has 3600 pulses per revolution (equivalent to 10 pulses per degree) and you set the starting point at zero when the test piece is in place, it should go off at 30 counts, on again at 60 counts, off at 90 counts, and so on. Any deviation from this pattern could indicate a missing or damaged tooth in the test piece. Ensuring the success of this method relies on the PLC's ability to retrieve the accumulated count from the high speed counters within the specified timeframe. If achievable, the next step would be to define what constitutes a "significant deviation" in this scenario.
Hello Steve, thank you for responding. I appreciate your innovative thinking and agree that using vision technology could expedite the process at a lower cost. However, my client's customer insists on using the specific crankshaft sensor currently in production. I have concerns about whether a Rockwell Automation PLC can process these signals quickly enough. Perhaps if I could access information on scan times and utilize a dedicated PLC solely for this portion of the machine, it might be feasible. Kollmorgen's specifications for an AKD drive with EtherCAT indicate the ability to capture both the sensor and encoder data at a rate of 14kHz, or 4kHz if the data is sent to an external EtherCAT master. However, this method, similar to trending in Rockwell, may not be suitable for our needs. If it were viable, I would require an Allen Bradley PLC communicating via Ethernet/IP through a gateway to EtherCAT for controlling the Kollmorgen drive. After data logging, speed becomes less critical. In theory, I could transfer the logged data from the EtherCAT master to Ethernet/IP through a gateway for the CompactLogix to make pass/fail decisions. Alternatively, a simpler approach could involve a dedicated EtherCAT PLC for data acquisition, logging, decision-making, and basic digital I/O communication with the CompactLogix.
The 5069-HSC module boasts a rapid 0.2 ms RPI throughout the entire backplane, ensuring efficient and high-speed performance.
In document 5069-TD001O-EN-P, the specific specifications you mentioned are not visible. It is unclear what you mean by "across the backplane" and how it compares to other factors. However, it seems like the speed would be sufficient. Can this be achieved on multiple channels simultaneously? Let's say I am using a 5069-L310ERM with 80% memory usage as an example for program size and PLC scan load. Would I need to create complex schedules to manage the scan time? Currently, the rotation is at 1 revolution every 0.4 seconds, with data logging starting after detecting two missing teeth and continuing for 360 degrees of rotation 3.34ms later. I am unsure of the scan time for my previous program. Could a 4096 pulse encoder provide high enough resolution? Your thoughts and assistance are appreciated. I hope this does not come off as asking someone to do the work for me.
I am delighted to assist. As a proud owner of one of these cards, I can offer additional demonstration information if required. The card's data is transmitted to the PLC via the backplane at a specific RPI. Additionally, it has configurable outputs that can be triggered locally without relying on the backplane RPI. According to the HSC manual (page 20), the Requested Packet Interval (RPI) is a crucial parameter that determines the data exchange rate between the controller and the module. The RPI value is set during the initial module configuration and can be adjusted as needed during operation. The valid RPI values range from 0.2 to 750 ms. Considering the project perspective, there is still 20% of memory remaining on an L310ERM to accommodate the addition of the card. The period of a periodic task can be set as low as 0.1 ms, which is exactly half of the minimum 5069 RPI value.
It sounds like you're tackling quite an intricate project—kudos to you for reaching out for guidance! From your description, it seems that high-speed precision and data capture are crucial. Given that, I'd personally recommend exploring Beckhoff's solutions further. While it's true that it has a steep learning curve, the benefit is the high degree of customization and control it offers. It's a potent choice considering its compatibility with EtherCAT. Learning any new system might seem daunting initially, but the long-term gains for both you and your company could be significant. Nonetheless, if budget constraints are a major concern, looking into lower-cost alternatives or used equipment might be a wise move. Good luck with your project!
While it might be a bit challenging at first, I'd definitely suggest looking further into Beckhoff. Given its compatibility with EtherCat, you've got a clear advantage there. Moreover, the TwinCAT software platform it offers could be ideal for your requirements as it allows logging of a high amount of data at high speeds. Although it does require some programming knowledge, the learning curve could be well-worth it given your project needs. You may also want to look at Mitsubishi PLCs; while less common, they have impressive processing capabilities and might just fit the bill for your high-speed data acquisition needs. Lastly, don't disregard NI completely. Yes, the integration may be complex, but their support forums and customer service are excellent. They might be able to guide you through the setup process without too much trouble.
It sounds like you’re navigating some interesting challenges with your project! Since you're already considering various options, I'd suggest looking into Delta Tau's Turbo PMAC controllers as they have a reputation for handling high-speed applications with precise timing, which could solve your sampling speed concerns. For data acquisition, their ability to interface smoothly with different encoder types might ease the integration headaches you're worried about with the Kollmorgen setup. If you're leaning towards using a simple incremental encoder, just make sure to account for the potential drift in position if you can't maintain absolute referencing. It might also be worth reaching out directly to vendors like Promess or Beckhoff; sometimes they can provide demos or trial hardware that could greatly aid your decision-making process. Good luck with your project!
It sounds like you're diving into an interesting project! Given your focus on speed and precision, the suggestion of switching to a simple incremental encoder might be a good starting point if it can provide the necessary resolution without the complexity of integrating an absolute encoder like the Kollmorgen. If you haven't already, consider looking into the Arduino or Raspberry Pi platforms as alternatives for data acquisition; they can often handle high-speed signal processing fairly well and may offer a more straightforward programming experience. Additionally, checking out existing libraries for capturing encoder signals could save you a lot of time. Good luck, and I’m sure you'll find a solution that meets the project's demands!
✅ Work Order Management
✅ Asset Tracking
✅ Preventive Maintenance
✅ Inspection Report
We have received your information. We will share Schedule Demo details on your Mail Id.
Answer: - Rockwell Automation PLCs may not be suitable due to sensor sampling speed limitations, especially for capturing the required data from the spinning part accurately.
Answer: - The project involves a servo, encoder, and a crankshaft position sensor to spin the part, count teeth, and analyze data for quality assessment. These components are crucial for achieving accurate results and identifying any defects in the part.
Answer: - The Kollmorgen absolute hiperface encoder is recommended for accurately capturing the register position of missing teeth and logging voltage spikes against rotational position. Its precision and capabilities align well with the project requirements.
Answer: - Alternatives such as National Instruments (NI), Promess UltraPro, Kistler KiBox, and Beckhoff solutions have been explored. Each option comes with its own advantages and complexities, requiring careful consideration based on the project's specific needs.
Answer: - The need to complete the data acquisition process within one rotation of the part (0.4 seconds at 150rpm) highlights the importance
Join hundreds of satisfied customers who have transformed their maintenance processes.
Sign up today and start optimizing your workflow.