ELS and Solid State Controls for my PM-1340GT

[ptrotter]

I have been working on the Clough42 ELS on my PM-1340GT lathe and decided that if I am going to be using solid state electronics for the lead screw, than I might as well build solid state controls for the spindle. After all the VFD is a solid state device itself. Here is what I have so far:

I have completed the installation of the servo and rotary encoder. I used the same mounting method as @xr650rRider used on his. My servo actually clears the gear cover with no modification.

I have the ELS electronics in a test bed and everything works properly. You can see my test bed here:

For the solid state lathe controls I used a slightly modified version of the controls that were designed by @B2. More information can be found at https://www.hobby-machinist.com/thr...nents-pm1440gt-vfd-3-phase.95058/#post-883089.

I designed the circuit using Kicad and had the printed circuit board built and the surface mounted devices installed by JLCPCB. The PCB cost about $4.00 from JLCPCB. I installed connectors and a relay on the board bringing the cost up to $30.00.

Here are pictures of the raw board as received from JLCPCB and after I installed the connectors.

To test the lathe controls, I built a test bed that includes a front panel with all the switches as well as a panel that includes switches that are not on the front panel to simulate the brake switch, cover interlock switch and the proximity sensor, and the lathe forward/neutral/reverse switch. This way I could test all the functions of the board. To simulate the connection to the VFD, I built an LED board with an LED for each VFD input to verify that proper signals are output from the board for any control input. You can see that here.

Everything tested out perfectly so the next step is to build a new cabinet to hold my VFD as well as the ELS and control so that everything can be in one enclosure.

More to come in the future.

 

[markba633csi]

Beam me up, Scotty! LOL

 

[B2]

Hi Paul,
Nice posting!
I have also made a couple of PCBs for my PM1440GT to replace my hand wired version, but I have yet to test them out. I am amazed at the fast turn around time from JLCPCB. @bdstark has a couple of these boards and is testing one of them. It seems to be working fine. One of these days I will post about the pcb and a couple of minor improvements. The original posting is here: https://www.hobby-machinist.com/thr...tronic-components-pm1440gt-vfd-3-phase.95058/

 

[bdstark]

Very nice - I just ran through testing the PCB myself. The VFD simulator board is a great addition and very helpful. I have been hanging resistors and LEDs off the connectors and was thinking of making a similar board. I'm hoping to get the lathe spinning this weekend - it's been 9 weeks from delivery to getting it up and running. I made a quick video - https://photos.app.goo.gl/6t9FP1Svbp9zLvGD8 - while testing the latching and interlocks, but there is no audio so if you don't know what/why I am doing something it won't make a lot of sense. (the left red LED is bright when ready to go, less bright with certain interlocks where jog would work, and off with full interlock disconnect, the two columns of LEDs on the right include forward/reverse from the lathe switch (a switch on the bench in this case), and the red/blue indicating jog forward/jog reverse, and I have one of them reversed at the switch.)

The video shows a chuck key interlock I designed and printed - using a NO switch, the chuck key must be inserted or the lathe won't run. I find it odd that isn't protected against - using the chuck key is more common then opening up the side motor panel, which has an interlock. I don't want it to start running if I was changing gears (though rare with the TL or ELS). More frequently though, I don't want it to start if I forget the chuck key in the chuck. If anyone is interested in the chuck key interlock design they can be downloaded at https://drive.google.com/drive/folders/17pNm_YiBlkyD03g5XBWKMuYDfYjak_Iw?usp=sharing

The PCB from @B2 has been fantastic, and being able to bench test the functionality before attaching to a $900 VFD is comforting. The solid state design is pretty simple, certainly less expensive then the relays, while being more flexible with the limited space. Being able to iterate in design, without large wiring changes at the lathe/VFD is great and something that can't be easily replicated with relays, especially indoors at my desk. Relays are simple to reason about, and the solid state design takes extra mental effort, but the complexity is due to additional features that can be extended, something relays can't easily replicate.

I'm hoping to get DRO scale information in a future iteration, as a way of replacing the proximity sensor. I have a 1440TL, and the micrometer carriage stop has been backordered for a few months, so I have no place to mount the proximity sensors right now. I'm also hoping to get the button/switch/rpm/etc... statuses, using a Raspberry Pi, to make a display of more than just LEDs. Something like displaying all three versions of the tachometer - what the dial is set to, what the VFD says its at, and the true value from the hall effect sensor - would be helpful at times. More helpful would be the ability to perform calculations on those values, for example to display SFM instead of RPM, or the number of rotations. Another feature could use the Pi to actually adjust the VFD instead of a potentiometer. (If Yuri from TouchDRO happens to read this, I would love to see if there could be some possible integration points.)

Excellent work - I look forward to future updates.

 

[ptrotter]

@bdstark, Thank you for your response. As I had said above, my design is based on @B2's design with a few minor changes to meet my specific needs so I'm sure your board is working fine. I replaced the LEDs on the board with simple diodes since I don't expect to be able to see the board when it is installed. I found the process of designing the PCB and getting it manufactured to be interesting and fun. I first did a through hole version for testing in case I needed to change any components. Since it worked fine, I thought it might be fun to do the surface mounted version. It continues to amaze me that I can submit an order to JCLPCB and have it delivered from China in about a week at a cost that is amazing. JLCPCB has a minimum order of 5 boards and the cost for the SMDs for all 5 boards was $.75 (yes, 75 cents). The front panel in my test bed is also a PCB designed in Kicad and manufactured by JLCPCB. I don't know if it is rigid enough for production use, but it looks great and cost about $4.00. That is much better than the $90 it would cost for the same panel in aluminum from Front Panel Express.

I don't use the proximity sensor much, but since I already have it, I included the capability into the board. I designed mine for a normally closed switch, as opposed to the normally open one that B2 used, so it simplified my design a little. I used a shorting type phone plug on the panel for the sensor, so I can unplug it and put it out of the way.

Integrating with a DRO sounds interesting. I'm working on the Clough42 ELS and this got me to wonder if any integration with that would have any value. I expect that calculating SFM would be fairly easy, but I guess you would have to enter the diameter of the part manually some how. There are so many things that can be done with things like Raspberry Pi that the possibilities are endless. I seem to spend more time on machine upgrades than I do using the machine.

I like the idea of the chuck key interlock. Just a couple of weeks ago, I left my chuck key in the chuck for the first time. Fortunately no damage or injuries, so I think this is a good idea. I guess I'll fire up the 3D printer, I love my 3D printers.

I need to finish testing the ELS and make sure I have the config settings correct for my leadscrew pitch and gear ratios and then build an enclosure for the ELS and lathe controls. I'm trying to decide if I should remake my VFD enclosure, making it big enough for both, or using a separate enclosure for the new stuff.

 

[CliveFX]

Hey there, are you having any issues with torque from your motor? I am sizing my 1236t conversion, and the guy at Rockettronics wants to sell me a 20nm motor!