Hello, all! I’m having an issue with a new machine my company is designing. I’m trying to figure out if certain solutions are do-able, and from there which ones might be worth pursuing. I'm OK with doing a lot of learning as I go as long as it's towards something that's possible to achieve.
Background:
We need to make a precision measurement with a probe. The probe currently connects to a ~$900 handheld instrument that makes the measurement, does some math, and displays the measurement on the screen. We have several of these on-hand already. The handheld instrument can be hooked via USB (it has a micro USB port) up to a PC -- using Hyperterminal, according to a brochure from the manufacturer. It can send the measurements and receive commands via ASCII strings over this USB port.
The other option is to spend $5500 for a replacement to the handheld terminal for one “designed to talk to for automation." Then either ~$500 on an RS232 I/O module or an unquoted upgrade to make the terminal talk via profinet plus the cost of a profinet gateway since I am using an Allen Bradley PLC. In short, around $6000.
Whatever process we go with, we will likely be duplicating on future machines, since the company is marketing itself as more of a precision manufacturer. This ~$5000 in equipment savings will be duplicated (several times per year, for several years), so it’s definitely worth the time to lower these costs. I got my bachelor’s in MechE this last May, so I’m making entry-level MechE pay. I can justify spending some time on this.
Solutions:
I have a few ideas. Solution 1 is to just spend $6k on every machine going forward on the expensive terminal and whatever gateway I’d need.
Solution 2 is an ethernet to USB adapter, then USB to micro USB adapter, and run an ethernet cable to one of the ethernet switches connected to the AB PLC. Links for example adapters are at the bottom. Does anyone have any experience doing something like this? My background isn’t in IT stuff, and I just learned on the last machine I worked on that ethernet cables use more than 1 type of communication protocol. Quite the learning experience. Would it be likely that I could make a generic .eds file for the handheld terminal, and send/receive strings straight from the PLC?
Solution 3 is similar to 2, but I found a gateway that does basically the same thing as the adapter. The link for an example gateway is at the bottom. Is the gateway the same thing as the adapter but at ~10x the price?
Solution 4 would be to buy a raspberry pi or low-end PC and stick it in the machine’s electrical cabinet. Then fumble my way through making a python program that receives commands from the PLC via ethernet, and sends/receives data with the terminal via USB. Does anyone have any experience with that? Does that sound do-able, specifically the PLC to PC/Pi connection? I wrote a couple Python programs in college for school work, so I feel like the PC/Pi to handheld terminal communication is in the realm of “known unknowns” in that I know where to start googling to make this happen if it I need to. This seems like a potentially attractive solution in the long run, because this could also potentially be expanded later down the road (on this or future machines) to include saving the measurements and e-mailing them to production management/customers.
Solution 5: We have a lot of in-house vision system talent. My boss and I discussed mounting the $900 handheld terminal inside the machine. The robot that's already going to be inside can press buttons on it. We could put a camera with character recognition in there pointing at the handheld terminal. My company has some decent discounts with Keyence because we buy a lot of stuff from them. I installed a few vision systems in existing equipment that had decent shape recognition software built-in, and the setup was $1600 for all the equipment/software. I imagine a version with character recognition wouldn't be too much more. It needs to come in under $5000 (after discounts) to save money. I haven't talked with Keyence yet because I hate the idea of pointing a camera at a piece of equipment that has a communication port on the side.
To wrap up: Which (if any) of these solutions sound possible? Which would you recommend pursuing?
Thanks!
Ethernet to USB could be something like this: https://www.alliedelec.com/product/...riYkrBY_5FUfVsxymsRoCIRgQAvD_BwE&gclsrc=aw.ds
USB to micro USB could be something like this: https://www.alliedelec.com/product/...4Y2ZebT3Ia1ksNxF-JBoCIykQAvD_BwE&gclsrc=aw.ds
Ethernet to USB gateway is something like this: https://www.express-inc.com/Product...P9dFvfgfttyeHkW5eXMOktgo4mUWfs8hoCHMgQAvD_BwE
Background:
We need to make a precision measurement with a probe. The probe currently connects to a ~$900 handheld instrument that makes the measurement, does some math, and displays the measurement on the screen. We have several of these on-hand already. The handheld instrument can be hooked via USB (it has a micro USB port) up to a PC -- using Hyperterminal, according to a brochure from the manufacturer. It can send the measurements and receive commands via ASCII strings over this USB port.
The other option is to spend $5500 for a replacement to the handheld terminal for one “designed to talk to for automation." Then either ~$500 on an RS232 I/O module or an unquoted upgrade to make the terminal talk via profinet plus the cost of a profinet gateway since I am using an Allen Bradley PLC. In short, around $6000.
Whatever process we go with, we will likely be duplicating on future machines, since the company is marketing itself as more of a precision manufacturer. This ~$5000 in equipment savings will be duplicated (several times per year, for several years), so it’s definitely worth the time to lower these costs. I got my bachelor’s in MechE this last May, so I’m making entry-level MechE pay. I can justify spending some time on this.
Solutions:
I have a few ideas. Solution 1 is to just spend $6k on every machine going forward on the expensive terminal and whatever gateway I’d need.
Solution 2 is an ethernet to USB adapter, then USB to micro USB adapter, and run an ethernet cable to one of the ethernet switches connected to the AB PLC. Links for example adapters are at the bottom. Does anyone have any experience doing something like this? My background isn’t in IT stuff, and I just learned on the last machine I worked on that ethernet cables use more than 1 type of communication protocol. Quite the learning experience. Would it be likely that I could make a generic .eds file for the handheld terminal, and send/receive strings straight from the PLC?
Solution 3 is similar to 2, but I found a gateway that does basically the same thing as the adapter. The link for an example gateway is at the bottom. Is the gateway the same thing as the adapter but at ~10x the price?
Solution 4 would be to buy a raspberry pi or low-end PC and stick it in the machine’s electrical cabinet. Then fumble my way through making a python program that receives commands from the PLC via ethernet, and sends/receives data with the terminal via USB. Does anyone have any experience with that? Does that sound do-able, specifically the PLC to PC/Pi connection? I wrote a couple Python programs in college for school work, so I feel like the PC/Pi to handheld terminal communication is in the realm of “known unknowns” in that I know where to start googling to make this happen if it I need to. This seems like a potentially attractive solution in the long run, because this could also potentially be expanded later down the road (on this or future machines) to include saving the measurements and e-mailing them to production management/customers.
Solution 5: We have a lot of in-house vision system talent. My boss and I discussed mounting the $900 handheld terminal inside the machine. The robot that's already going to be inside can press buttons on it. We could put a camera with character recognition in there pointing at the handheld terminal. My company has some decent discounts with Keyence because we buy a lot of stuff from them. I installed a few vision systems in existing equipment that had decent shape recognition software built-in, and the setup was $1600 for all the equipment/software. I imagine a version with character recognition wouldn't be too much more. It needs to come in under $5000 (after discounts) to save money. I haven't talked with Keyence yet because I hate the idea of pointing a camera at a piece of equipment that has a communication port on the side.
To wrap up: Which (if any) of these solutions sound possible? Which would you recommend pursuing?
Thanks!
Ethernet to USB could be something like this: https://www.alliedelec.com/product/...riYkrBY_5FUfVsxymsRoCIRgQAvD_BwE&gclsrc=aw.ds
USB to micro USB could be something like this: https://www.alliedelec.com/product/...4Y2ZebT3Ia1ksNxF-JBoCIykQAvD_BwE&gclsrc=aw.ds
Ethernet to USB gateway is something like this: https://www.express-inc.com/Product...P9dFvfgfttyeHkW5eXMOktgo4mUWfs8hoCHMgQAvD_BwE