Determining whether those images showcase an AS-I top can be quite challenging. The Thinktops are compatible with AS-I, just like the GEA Tuchenhagen TVIS and Sudmo models. My recommendation is the AC1422 master; it's a dual master unit that eliminates the need for multiple devices. Additionally, it's Ethernet/IP compatible, allowing for installation in a control panel located near the manifold rather than relying solely on the ControlLogix panel. It's crucial to mount it as close as possible due to distance considerations affecting performance. Keep in mind that you'll require a specific power supply, likely the AC1258 if opting for dual masters. Further, flat cables connected to M12 connectors and short M12 cables leading from the flat yellow trunk to the valve are necessary. A quick tip: if you're looking for flexibility in your trunk layout, you can bypass the small M12 cable by utilizing a trunk-to-flat connection followed by a flat to M12 connector leading directly to the valve top. Lastly, don't forget to invest in a handheld addressing tool. I typically purchase one for every project and leave it with the maintenance team after commissioning, ensuring smooth operation moving forward.
Paully's proposed 5.0 AS-I networking solution appears to be a promising option that I've commonly observed in practical applications. It presents a cost-effective choice, although newcomers may find the initial learning curve somewhat challenging. Personally, I would prefer the Allen-Bradley (AB) approach.
If you have several 1769-I/O32 modules (comprised of 15 input and 8 output), along with a few CompactLogix power supplies and a 1769-AENTR available, you can connect them to a 1756-L73 controller using a 1756-EN2TR EtherNet/IP bridge. This setup requires you to configure two banks: one bank with 11 modules plus the 1769-AENTR and power supply, and another with 12 slots plus a power supply (expandable to a maximum of 16).
However, have you considered utilizing PointIO entirely? Since you're implementing a 1756-L73 controller—which has no known issues with PointIO—you can support up to 64 modules for each PointIO bank (using a 1734-AENTR and a power supply). This setup allows for a total of 512 I/O points across 8 modules (8 x 64 = 512 I/O points) for each bank. You also have the option to add another 1734-AENTR for an additional bank of 64 modules or can use a bus extension cable (1734-EXT1(3)) with another power supply to expand the existing EtherNet/IP setup.
Additionally, it's worth noting that the physical footprint of four adjacent PointIO modules is smaller than that of a 32-point 1769 module, making it a more efficient solution with significantly lower implementation costs.
dmargineau commented: "I would opt for the AB route." Click to expand… I’m curious about your approach to the physical installation. The manifold is quite large, and each valve will necessitate several cables—specifically, four-conductor cables for solenoids and eight-conductor cables for feedback systems. These cables should be installed using tray cables, so it’s essential that they meet the necessary ratings for both the application and wash-down requirements. That's a significant amount of cabling, and running everything back to a single panel may create congestion. To manage the cabling efficiently, you would likely need to divide the I/O racks among multiple panels. Even then, the costs for both the cabling and labor for the installation seem quite high.
Excessive Questions Surrounding System Design
Why does every solution seem to require home runs? Why not consider a compact I/O panel for every two or four tanks? Would the combined expense of a dedicated panel plus an AENT be greater or lesser than the cost associated with cabling and installing trays or conduit? Without understanding the physical layout, it's challenging to assess this.
One key advantage of a distributed system is the ability to implement a Lockout/Tagout (LOTO) procedure on one section while maintaining operations in others. How often is this capability really required? That remains uncertain. Additionally, how many spare parts should be kept on hand? There's also the possibility that rather than having five valves per tank, a future need might arise for a sixth. What type of valves are being used, and what’s the reasoning behind having three discrete outputs (DO) per valve—are they meant for Open, Close, and Lift functions?
If solenoid valves are involved, perhaps a component like the SMC EX260 can effectively manage the discrete output load and the required discrete inputs as well. There are numerous strategies to tackle this challenge, and they can all be effective. While there may be a most cost-effective option available, one must ask: how much time and resources will be dedicated to pursuing that solution? When considering the broader aspects of design, installation, and maintenance, the potential savings of a few hundred dollars on hardware may not be as significant as it seems.
Aardwizz raised an interesting question: Why is it always about making home runs? Why not consider implementing a single small I/O panel for every 2 or 4 tanks instead? Would the combined cost of a panel plus the AENT be lower than the expenses associated with cables, trays, and conduit? We currently lack information regarding the physical layout of the setup. One significant benefit of a distributed system is the ability to perform lockout/tagout (LOTO) on one section while continuing to operate the remaining parts. But how often is this necessary? It's tough to say.
The primary challenge lies in the substantial 80-valve cluster, which represents a considerable amount of I/O connections in a compact area. Aardwizz also inquired about the number of spare valves required. What are the chances that rather than the standard five valves per tank, there may be a demand for a sixth valve in the future?
Additionally, Aardwizz expressed support for exploring a networked option. He questioned the types of valves in use and the reasoning behind having three digital outputs (DO) per valve—namely for Open/Close/Lift functions. If these valves are solenoid-operated, perhaps a device like the SMC EX260 could efficiently manage the digital output and even handle the diagnostic input (DI).
It’s likely that these valves are probably mix-proof or comply with the Pasteurized Milk Ordinance (PMO) regulations in the U.S. These valves would incorporate solenoids and feedback devices all housed within their tops, accounting for the high I/O counts. The three digital outputs serve specific purposes: the main valve actuation, the upper seat lift, and the lower seat lift, with the DI representing the relevant feedback.
Aardwizz noted that there are numerous approaches to resolving this issue, all of which could be effective. While a "most cost-effective" solution may exist, one must consider how much time and money would be expended in pursuit of it. Relative to the overall costs of design, installation, and maintenance, how significant is it to save a few hundred dollars on hardware?
Ultimately, the original poster must determine the exact type of valves they possess since they were acquired "used." This identification will narrow down their options. If the valves are equipped with DeviceNet or AS-I control modules, that would indicate the direction to take. Otherwise, discrete I/O wiring will be the necessary route.
Paully's5.0 expressed curiosity about managing the physical installation process. Considering the extensive amount of cabling involved, running everything back to a single panel location may result in a congestion issue. To elaborate, a Point I/O 'Rack' can accommodate as many as 99 modules (totaling 792 I/O). This configuration allows for the splitting of modules into up to five banks through the use of Expansion Power Modules (1734-EP24DC) and 3 or 9-foot Expansion Cables (1734-EXT1-3). Utilizing multiple adjacent enclosures can effectively resolve any potential challenges related to cabling and wiring installations.
In my previous experience with dry-hopping, I learned the importance of avoiding the networking of valves. Even a single cup of water can potentially disrupt the entire system. Typically, mixproof valves are equipped with just two solenoids for each head: one for valve diversion and another for seat lifting. These valves are often interconnected with other components in the circuit via airlines. It’s worth noting that feedback primarily indicates whether the valve has been diverted, as the seat lifting action occurs only for a brief moment.
I am currently facing some challenging decisions regarding the setup of remote I/O in a large production facility, which will involve approximately 200 inputs and 80 outputs, with around 10% being analog. I plan to use a CompactLogix L3 processor, but I’m uncertain about which remote I/O system to choose. My instinct tells me to stick with DeviceNet and Flex I/O, as I have extensive experience in this area and have been utilizing this technology for many years. However, I’ve been considering whether there might be newer, more advanced options available on the market.
Unfortunately, my local sales representative isn't very knowledgeable—they seem to only understand PLCs from articles in magazines. Most of the new products I'm encountering come with connectors, which confuses me since I've never grasped their necessity in the field. It often seems like the available cables are never the right length for my needs. Additionally, I must consider that my installation is situated in a highly corrosive environment filled with water, salt, and various chemicals. Should I stick with the tried-and-true systems I know, or is it time to explore modern alternatives?
Utilize Ethernet IP Remote I/O for optimal performance in your projects. While I personally use Omron, renowned brands like Allen-Bradley (AB), Schneider Electric, and others also offer this technology. It has proven to be highly reliable, and I've successfully implemented it across approximately 40 projects with no problems encountered. This solution is ideal for seamless connectivity and efficient data transmission within industrial automation systems.
According to dmargineau, a Point I/O 'Rack' has the capacity to hold as many as 99 modules, totaling 792 I/O points. This setup can be divided into as many as 5 banks by utilizing Expansion Power Modules (1734-EP24DC) along with 3 or 9-foot Expansion Cables (1734-EXT1-3). Furthermore, positioning multiple adjacent enclosures can help address potential cabling and wiring installation challenges. However, I was under the impression that the maximum number was 63 digital I/O modules, and the limit would be lower if you incorporate analog modules.
You're absolutely right; the limit is a maximum of 63 modules of any kind. Although the AB Architecture Builder permits the addition of up to 99 PIO sockets, it restricts the actual number of modules that can be integrated to 63. It’s important to note that adding analog modules does not impact the maximum allowable count of modules.
It's important to note that if multiple devices require communication with the rack, especially when using analog and specialized modules, there are some considerations to keep in mind. George has a detailed explanation that can provide further insights on this topic. You can check it out here: [PLC Talk Forum](http://www.plctalk.net/qanda/showpost.php?p=618700&postcount=6).
In summary, my personal record stands at an impressive 53 Point I/O modules housed in a Remote Rack, consisting of 45 digital modules and 8 analog modules. This setup has been operating seamlessly 24/7 for nearly five years, powered by an L72 controller and an EN2T EtherNet/IP bridge. Additionally, I’ve encountered configurations with up to 63 modules in PIO racks, including a mix of fully digital, hybrid, and high-speed counter (HSC) modules. When it comes to maximizing I/O density in a limited space, Point I/O truly excels. While there were hardware issues when the 1734 series was initially launched, I haven't come across any recent concerns. Moreover, a 1734-AENT module is priced under $400, making it a cost-effective choice for expanding your network. You can easily add another EtherNet/IP node without significant expenses.
I appreciate all the valuable suggestions and advice! After conducting additional tests, I've confirmed that the valves are not configured for AS-I compatibility, which has led me to consider a discrete I/O solution instead. Depending on the functionality of the existing electronics for each valve, we may need to replace the valve heads and controls. If so, I’ll reassess the feasibility of an AS-I setup, as I believe it would provide the cleanest implementation with minimal wiring.
Thankfully, management recognizes the installation's complexity, granting me some extra time for project planning. Currently, I am inclined to choose POINT I/O for discrete I/O solutions based on the following factors, as highlighted in earlier discussions:
- Compatibility with the 1756 series
- My personal experience with the system
- Consistency with our facility’s existing infrastructure
- Smaller footprint compared to compact I/O
- Cost-efficiency
For the tanks, I plan to utilize a total of 90 POINT I/O modules, excluding spares. I have some questions regarding the extension cables, as I haven't used them before and couldn’t find detailed information beyond page 49 of the selection guide:
- Does using the cable mean that the module limit for a single AENTR remains at 63 modules?
- Is the cable primarily for organization, without altering functionality? For example, would 40 modules on one AENTR equate to 20 modules on another AENTR with an additional 20 modules on the extension cable?
- Are there different power requirements when using the extension cable, or am I merely overthinking this?
Today, I will calculate the necessary rack lengths. We have a 60" by 36" enclosure available, but fitting all the wiring into one space may pose a challenge. As others have suggested, we might need to use multiple panels to effectively manage the setup.
Dryhops inquired: "Does the use of a cable mean that the module limit for a single AENTR is still capped at 63 modules?" Upon expansion, it is confirmed that the Point I/O racks indeed support a maximum of 63 modules. Dryhops further asked: "Is the cable merely for organizational purposes and does it not affect functionality? For example, would having 40 modules on one AENTR be equivalent to 20 modules on that AENTR and another 20 on an extension cable?" The answer is yes; you can initiate a new Point I/O row or extend into another enclosure without needing to alter any configurations or functionality.
Dryhops also expressed concern: "Will the power requirements differ due to the cable run, or am I overthinking this?" The response is no, your concerns are valid; the power requirements for both the bus and the connected devices are critical. Different modules may require varying supply currents and voltages.
For anyone involved in designing Logix systems, I highly recommend the RA Integrate Architecture Builder (IAB) utility tool. This tool allows you to create your Point I/O setup with up to five banks of PIO sockets and then seamlessly add I/O modules. The IAB utility automatically integrates the necessary 1734-EP24V DC power expansion modules based on the quantity and types of I/O modules selected, as well as the presence and lengths of any 1734-EXT1(3) extension cables. For more information, check out the official site: [Rockwell Automation Product Selection and Configuration](https://www.rockwellautomation.com/global/support/product-selection-configuration/overview.page) and view this informative video: [YouTube Link](https://www.youtube.com/watch?v=-VWWYH0S0HY).
I understand that you've opted for a particular point, which is a solid choice. However, I'd like to add a quick suggestion based on my current experience. I'm utilizing the SMC EX600 series for pneumatics in my ongoing project. In this setup, the SI (Ethernet/IP) module is centrally positioned, with pneumatic valves organized on the right side. On the left, I have multiple 8x M8 3-pin digital inputs and 2x M12 analog inputs. This configuration significantly reduces wiring complexity, as I can rely on pre-assembled sensor cables. Additionally, my setup requires just one power cable (4 wires: 2 for bus power and 2 for safety voltage reduction) and a single Ethernet cable, streamlining the overall installation process.