I/O Interface Board Recommendation For Lathe

andypugh wrote:
It's all a hobby for me, though, so whether it is writing drivers or....

I hope to be making parts that I can SELL using my new CNC lathe. Writing drivers may be a hobby to you, but it could substantially help put food on my table. No pressure. I should send you a check for the 5i25 driver!

I do hope to pay some of this forward, and be a contributing member of the EMC2 community. I don't want to be a leech.

With the Mesa I/O hardware on order and about ten other orders inbound, it does feel like I'm over the big hump, but it's still a long way to the finish line. I probably won't post a Look At All The Stuff I Did Wrong project thread, but I will post some details of the finished project with an eye toward making it easier for the next EMC2 traveler. Realistically, I'll probably reach a point (or two!) once I dig into EMC2 where I need a little more help beyond what I can learn from reading the voluminous EMC2 documentation and scouring the info on this site. If nothing else, the newness of the 5i25 hardware will probably throw a curve or two at me, but I'll try not to wear out my welcome here. My long term goal is to give more than I take.

On the plus side.... Once I have it all figured out for the lathe, the CNC conversion of my G1006 Grizzly bench mill should be a lot easier, with a lot of cut and paste. I'm already ordering extras of many small parts like connectors, with the milling machine conversion in mind. I'd sell my old Grizzly and buy a well used Bridgeport class mill, but I can't tolerate the idea of moving the Grizzly out of the basement and a Bridgeport into the basement. I don't think I have that in me, so I'll CNC the import mill that I have.

Thanks for your posts as you go through your retrofit. I'm going through the retrofit of an Emco 120P lathe, and have been wrestling with the same sort of I/O issues. Sounds like the 5i25 and 7i76 combo is going to be a good one - it seems expensive at first, but given that it will replace a breakout board with optoisolation, a relay driver board, and provide enough I/O to replace the secondary control board I was considering having to build, all while being reliable, and providing smoother step control, it starts to sound pretty appealing. In my case, the Emco has a gad-zillion inputs and outputs, 5 inputs and 2 outputs alone to operate the 8 position tool turret, not to mention lubricant pumps, indexers on the ball screws, limit switches, etc. 32 inputs sounded liked a lot at first, but as I built my list of required I/O, I think I'll need it.

I'll definitely be following along with your thread, and I'm interested to hear about how it goes, and the process for writing the EMC interface stuff to interact with all that I/O.

-Sam

That sounds useful! Where can I find this sample config?

-Sam

BTW anyone using a 7I76 should download the latest manual (V1.4)
earlier versions have many pin-out errors...

Sam (phidauex) - That Emco sounds like a nice lathe. I looked at some Emco lathes of the "CNC shop class" variety before finding the Clausing fairly close to home. The tool changer sounds handy. I'm hoping for a simpler version of that capability using 8" of gang tooling. My parts are small, less than an inch in diameter, so I should be able to index 1.5" in X, and have a new tool. If I'm clever, I might get six tools in my gang tooling. So far, my I/O count is small, but I wanted room to grow. I know I'll want to add stuff later and interface it to EMC2. When I spoke to Peter at Mesa, he told me that the digital inputs are actually analog inputs, and the FPGA is programmed with the ON and OFF voltage levels to support whatever voltage devices you have, and as much hysteresis (noise rejecting dead band between ON and OFF) as you'd like. That's a very flexible solution. I think Peter said that it ships with the FPGA programmed for a 5V spindle encoder and the rest of the inputs are 24V, which should work well for me. I'm trying to keep all my field devices as 24V devices, but it seems likely that I'll need some other voltage on an input and reprogramming the FPGA may be easier than building a transistor circuit to convert the voltage.



Andy - Thank you yet again for your work getting the new Mesa hardware working with EMC2. I'll upgrade from EMC2 version 2.4.3 to 2.5 when my Mesa hardware arrives. I ordered USB wifi dongles in anticipation of the need to upgrade Ubuntu and EMC2, but the Linux friendly wifi doesn't work with the stripped down realtime Ubuntu. It probably doesn't have a lot of extraneous drivers for stuff like wifi. That puts me in a catch 22. To upgrade Ubuntu I need the internet, so every time I upgrade, I'll lose the wifi that I need to apt-get the stuff I need to get wifi working again. I can either carry the PC/monitor/keyboard upstairs to plug it in to an Ethernet connection, or run Ethernet to the basement shop. I think I'll do the latter. I envision carrying the PC back to the basement thinking I'm good to go, and then realizing something didn't work and I need to start over, or I forgot some file or driver. That can get frustrating. I don't want to spend a lot of time and effort supporting wifi when wired networking is free, easy and reliable. Add "Ethernet To Shop" to the many unexpected additions to my lathe project To Do list. I'll be supporting EMC2 on a milling machine after this, and I've even been looking at EMC2 controlled desktop lasers, so that To Do item will be time saved on future projects as well.



John - Big thanks to you as well for the 5i25/7i76 config file. I'll be needing that! As soon as I get internet to the EMC2 PC and my Mesa hardware arrives, I'll be downloading it.



Peter - The one week old 7i76 version 1.2 manual is in the recycling bin. I downloaded and printed the V1.4 manual. I placed my Mesa order a week ago and you advised there would be a delay of a week or two. It sounded like Andy's EMC2 integration might be the holdup, but that's apparently done, so I guess there are final tweaks in the hardware or FPGA code. Another week isn't a problem for me, and unfortunately, at the rate I'm going, shipping two weeks from now still won't impact my schedule. I'm looking forward to getting the Mesa hardware so I can start wiring up the lathe, but I have a lot of other stuff to do first.



Project Update - I finally got the lathe off the dollies, and leveled. I'm starting to wire up the spindle. Then I'll use the lathe as a manual lathe before starting the serious part of the CNC conversion. Once it was off the dollies, I finally took a look inside the headstock. I was unsure how I'd connect the CUI encoder to the hollow spindle as those generally mount on the centerline of a rotating shaft. There will be a 5' length of steel bar stock on the centerline of that spindle. There is the motor shaft and an intermittent pulley shaft, and while those run at some ratio of the spindle speed, I wouldn't be able to generate the once-per-revolution index pulse I'll need for threading operations, and to stop the chuck at an indexed position so I can integrate an automated chuck closer later. Thanks to Jon Elson's post at CNCZone.com, and his page describing how he monitors the spindle RPM on his Bridgeport ( pico-systems.com/bridge_spindle.html ). I've decided to follow his example and generate an absolute quadrature encoder signal by using automotive magnet & Hall effect sensors to detect the 33 gear teeth on the spindle gear and a nearby divot that was already drilled in the OD of the spindle.





I ordered four of the Allegro ATS667LSG Hall effect gear tooth sensors from Digi-Key yesterday.

Digi-Key - 620-1328-1-ND (Manufacturer - ATS667LSGTN-T)
search.digikey.com/us/en/products/ATS667...20-1328-1-ND/2042758

One sensor is for the encoder index pulse, once per revolution. Two sensors are for the A & B quadrature encoder signals, in my case, 33 pulses for each of the two sensors for a 132 count quadrature encoder. The fourth sensor is a spare. Clausing made that job fairly easy for me. I'm seldom that lucky, so it's a noteworthy event. I hope this is a good omen for the lathe CNC retrofit to come, but I suspect The Fates are lulling me into a false sense of security.

Now if life and work would stop intervening, I could get some work done on my CNC lathe conversion. Or, from another perspective, if my paying work would let up a bit, I could do the work that I hope will lead to more paying work in the future!