Gen 2 Elite JM has come (batch 2)

This is just to share some experience of moving on with the machine.

Generally everything went smooth - putting together QCW with folding legs and wheels, cutting and attaching the waste board, assembling axis and wiring. Upon the first start there was a problem with homing the x axis. Onefinity technical support was very responsive and on Monday I had a troubleshooting video session. The problem was that the end switch pin on the x axis was too short. Adding 2 layers of masking tape resolved the problem. Tuesday morning the same problem appeared again. Another two layers of masking tape helped and I hope this is for good. Onefinity said that the manufacturing team was informed about the issue (I was not the first) and they are working on a better solution than a masking tape.

One more hiccup was an incorrect setting of the motor stepper resolution for x axis. I had it set to 200 while it should be 125 steps/mm.

I am not too sure how the machine passed QC with this problems.

There are still a couple of obstacles. One is minor but annoying: the wifi network does not get back after restarting HMI. I have to press “Test” button to get it back. This is counter intuitive as well, there are two buttons “connect” and “test”. The “connect” one does not help but the “test” one does almost always. The second issue is with the joypad. I received it today and still cannot get it working. HMI detected it but there is nor reaction on pressing the buttons. I am going to look at it tonight.

Otherwise the machine looks impressive and promising!

2 Likes

Regarding the joypad, I’ve found mine is often unresponsive though recognized as well.

The solution so far has been to unplug and replug the USB dongle on the side of the HMI usually no less than twice but no more than four times and it becomes responsive.

This isn’t perfect, and there’s probably some work that needs to be done in the software, but in the short term, since it only has happened to me at startup it’s not a show stopper.

1 Like

Thank you for the tip.

I tried unplugging and re-plugging and it helped. I observe the same behavior, at startup I need to do that and then it works fine. My concern is that the HMI USB connector will eventually stop working after some re-plugging cycles…

2 Likes

I dont have a gen 2 but I have a feeling if the motor resolution issue was a setting on the controller its probably that the machines get tested with a bench controller and its just your hardware.

In another post I recommended buying a right angle connector: Redline HMI and Joypad Dongle

Though I’m somewhat confident that Redline will improve the performance in the future, though I’d guess it’s a lower priority issue as they’re still rolling out features, you probably have thousands of cycles on a typical port before it becomes and issue.

That being said, maybe buying the adapter dongle and plugging and unplugging in that would at least relieve some of your concern about wearing the port out in your HMI. That doesn’t mitigate possible wear on the dongle itself of course, but that’s a less expensive part to maintain if for some reason you needed too.

1 Like

The joypad and wifi are two known issues and are high priority after the windows/mac software and atc releases.

3 Likes

I recommended buying a right angle connector

Good idea, ordered! I just probably saw too many broken connectors in laptops and phone while I worked for Motorola.

I am a software engineer and my initial impression is that the HMI software has rather, to put it lightly, some room to be improved. There are imperfections in the UI (which are typically very easy to fix at the development stage and not let it go to production), counter intuitive workflow (like the one with wifi test button) and even bugs. Yesterday I spotted a bug (a minor one) and asked onefinity support how to file a report to redline (no answer so far). However I also hope that Redline will polish the software in the future releases.

2 Likes

I’ve found support at Redline to be very responsive and helpful when I contacted them regarding the RealTimeCNC software image for a Pi build as well as their documentation. Hope your experience with them is as positive as mine was. Contact: support@realtimecnc.com

2 Likes

Feature requests, ideas, and software bugs can be posted here:

The Redline/Realtime controller is just getting started and we take feedback very seriously.

3 Likes

Here is some more progress.

Last weekend I finished with accommodating a dust collection, grounding the machine, spindle breaking-in, updating the HMI software, flattening a wasteboard, tramming the spindle and making a grid on the wasteboard.

I use wi-fi connection for the HMI and I bought the recommended external wi-fi adapter. The HMI documentation has a link to amazon. Nevertheless updating the HMI software took 3 tries, two of which hanged the HMI to the point that it required a hard reset. The updated software v.1.4.6 looks a bit refined in terms of the user interface however the problems with a lost network connection and a necessity to unplug/plug a joypad dongle are still there. It becomes more and more annoying for me that each time I switch the machine on I have to “jump” around to make all the equipment working. In particular I found that if right after I switch the machine I also “restart” it then the wi-fi network appears functioning and I can transfer my files to HMI.

The wires for grounding the machine are not provided by redline. The recommendations are to ground both the dust collection hose and the HMI unit. I can understand why the hose grounding wires are not provided - this part differs for everyone but the wire for the HMI unit is the same for everyone. Redline says that static can cause malfunctioning so the grounding is essential but the wire to ground the HMI unit is not provided. It’s a bit strange to me. The good part is that grounding worked very well for me. Cutting MDF did not produce any static.

Tramming the spindle was not exactly a very pleasant task. Aligning along the Y axis required for me about 0.4 mm shims of multiple layers of an aluminum foil at the bottom of the z assembly. Each time it means I had to unscrew quite a heavy assembly and screw it back. It would be so much easier if 1F added small setup screws at the corners of the z assembly! In this case there would be no need to take the assembly on and off each time. Also 0.4 mm of shims looked too much for such a precise machine but 1F support assured that it is perfectly normal.
Tramming along the X axis was quite easy, I needed only a couple of light mallet taps on one of the sides of the Z assembly.

Flattening of the wasteboard went not that smooth for me. I had a couple of passes using amana rc-2255 bit (2-1/2 diameter) on my 1.5kw spindle. Each time after finishing whole job successfully the controller reported error 6 at the moment the spindle moved to the home position. The 1F and redline support was extremely quick with replying to my question. They say that the bit is too big for my spindle:

The spindle itself helps to decelerate the bit rotation and due to the mass/size, it creates an over-voltage at deceleration.

and

The bit is much too big/mass for the 1.5kw spindle. A 2.2kw most likely could handle it, but we’ve found a 1.75 to max out the 1.5. Sometimes a 2 will work but not recommended. The error is the spindle trying to slow the mass and doesn’t have enough umph to slow it down, so it errors.

So both of them 1F and redline know about the limitation of the spindle. Yet, the 1F page where I ordered my 1.5kw 80mm spindle has not a single word that some bits will not work because of their size. This flattening amana rc-2255 bit is by far the most expensive bit I bought, around $190. 1F and redline knew that it is not going to work and did not share that information with customers. You can imagine what I feel.

Making a grid using 60 degree v-bit went smooth with no problems at all.

One hiccup was with the HMI software. After finishing the first pass of flattening the UI lost the cursor position on the touch screen. When I tried to press a button displayed on the screen the UI registered a press 6 inches away. Likely after a hard reboot the error cleared.

Also, I stepped on one more minor issue. The spindle connector is covered by a rubberized cup. When a cable is connected the cup is hanging around. In my case I installed the spindle so that the cup was hanging towards the z assembly. Once it was trapped between the moving parts which caused the z motor error.
The spindle has no key so the cup can be oriented in any directions. I was lucky with mine. If I were 1F I would note that in their installation video and instructions. Or the cup could be made detachable.

1 Like

Other than the error at the end it sounds like the 1.5KW handled the flattening bit just fine? I have a 2.5” Slab Slayer by R.I.P. Precision Tools. I do plan on using it and I believe it is lighter than the Amana RC2255. I would think they could help you reprogram the VFD for a slower ramp down. One of the points of getting the spindle for me was running that flatting bit, so hopefully they can help us find a work around. I realize that 1.5KW is only about 2HP, but it should be able to handle a 2.5” bit. When I bought that bit my spindle at the time was 3KW (4hp) so it was no issue. Obviously we have to use a grain of common sense when selecting bit sizes, but personally I don’t think a 2.5” flatting bit when used with reasonable feeds and speeds should be off limits. Especially when considering it spun up fine, ran your file fine and only had an issue at spin down. @OnefinityCNC can we have a slower spin down cycle to avoid this error with larger bits?

1 Like

Correct. I did not notice any problems neither while flattening was in progress nor in the produced result. The error appeared only when everything was finished and the spindle moved to the home position. As far as I understand at this moment the spindle started to slow down.
Error 6 is described as “Over-voltage at deceleration”. The doc has also error 5: “Over-voltage at acceleration”. I did not observe error 5 with rc-2255.

1 Like

Page 52 looks like decel is set to 6 seconds from factory. You could change to 10 seconds and see if that gets rid of your issue. If it does then you could try less time until you get your error again. Then back it off by a second. Here is a link to videos on how to reprogram your VFD.

1 Like

Thank you for the tip!
Last night I setup a test toolpath, just like 6 inches without any cutting. I guess what important is to spin the bit and then to slow it down. The error cleared for me with 9 seconds setting.
I am not sure what redline has saved with the default 6 seconds. The difference is only 3 seconds and the tool changing does not happen very often. The vast majority of the time is consumed by cutting, including the production sites.

2 Likes

One thing to be aware of is if you are using ATC that you may need to increase the time before it does auto tool change for increased spin down time. Glad it worked for you. I did talk to the manufacturer of the Slab Slayer but and he said generally they suggest a minimum of a 2.2kW for his 2.5” bit. He said it is doable with a 1.5kW but to be careful and know with heavy use it maybe decrease the life span of the spindle. I understand these limitations and the “risk” but I feel like with conservative feeds and speeds all should be well and good. Just my opinion and I am definitely not an expert :winking_face_with_tongue:

2 Likes

Can you give a little more detail around the masking tape fix? My Y axis will not home every time, and I would like to know what you mean by “end switch pin”. Thanks!!

Hey bill,
Email official tech support so we can get you fixed up! Support@onefinitycnc.com

Good point.
It’s not applicable to me at the moment though. I do not have an ATC and have no plans for it.

2 Likes

:face_with_bags_under_eyes: Had to ask support about something else and asked for a status update on my batch 2….they said it should go out in the next two weeks :sob: lol

2 Likes

We’re working on a production update to be released later today.

1 Like