Bot freezing under v3.5.3

[geneb]

I ran into a pile of trouble last night with the bot freezing up on me. I actually had to power the computer off because the software wouldn't respond to end-task commands. This has happened a few times in the past, but hasn't been that big of a deal because I was setting my Z height at the top of the board. I could just restart and all was good.

Well for my latest project, I've been zeroing to the table surface instead. 2 freezes later and I've ruined a full sheet of material.


Does the new 3.5.9(8?) build resolve the freezing issue?

tnx all.

g.

[Brady Watson]

It's your computer. It is either something running in the background that you are not aware of, competing with the communication stream to the Bot or it is a USB conflict. A new 2.0 USB card with a 2.0 hub plugged into it should cure the hardware issue. Running MSConfig & unselecting non-essential programs, plus running the TaskManager to find out what other non-essential programs are running should take care of the rest. Upgrading to the latest control software AND updating the control box firmware is always a good idea.

-B

[geneb]

So essentially you're advocating a hardware fix for a software problem. Astonishing.

So I should ignore the perfectly good USB 2.0 ports on the motherboard and scramble to find a short-riser USB 2.0 card because the _software_ is broken?

I don't mean to be rude, but that's like advocating you replace the transmission in your car when you run out of wiper fluid.

g.

[Brady Watson]

I thought you were looking for a solution to your problem...I didn't know you needed to like the answer. My tools ran *perfectly* on v3.5.3 & still do on v3.5.9 - both Alpha & Standard. One desktop, one laptop. The desktop didn't about a year ago until I installed a cheap ($40) USB/PCI card & all hiccups went away. Good luck getting it all sorted out.

-B

[andyb]

Gene,
Brady is correct. I had the same problem that you are having. I even went back to running an older version of the control software to fix the problem. I was wanting to run the newer version of the control software, so I upgraded my computer to a newer used Dell with a 2.0 USB on the motherboard. I still had the problem. In the SpeedTest utility the Effective Transmission Rate (ETR) was 73. At Brady's and other suggestion, I went to Wal-mart and purchased a $20 2-port USB v2.0 hub. The ETR jumped to 82 and I haven't had the problem since. Try the $20 hub first. If it doesn't work you can always take it back.

Andy B.

[geneb]

I guess the way the answer was given rubbed me really hard the wrong way. It's _NOT_ my computer.

A more accurate description would go along the lines of, "There's a bug in the software that is known to react badly with some USB controllers. Until that bug is fixed, you might want to try installing an auxillary USB 2.0 controller and see if that helps."

Now considering that the controller is using a link that's not USB end to end, but is USB to serial, is it possible to just use a regular 9 pin serial cable to connect between the host and controller box?

I don't mean to be so abrasive about this, but I write software for a living and anytime I hear "the hardware is the problem", my cop-out alarms go off REALLY loudly.


g.

[geneb]

Andy I just ran the comms test that came with the 3.5.9 distribution and it reported the transmission efficency(sp) at 89.2. The system already uses a USB 2.0 hub between the box and the computer. I needed to get that to resolve another problem I had with the 'bot going stupid on me. I managed to find a short usb card that I'll try, but I just wish they'd fix the problem and be done with it instead of blaming the hardware.

[andyb]

Gene,
Luckily, I didn't have to add another USB card. The hub fixed it for me and you're right, 22 years in IT myself, I think it is the software also. But hey, it's working great for me now so I can't complain.

Andy B.

[geneb]

Just a quick update - I installed the new USB card and my efficency(sp!) number went down to 89.1. Go figure.


We'll see if I end up ruining another sheet of material.

[andyb]

Two words, air cut. :-)

Andy B.

[geneb]

I've tried tricking Murphy like that before. He's wise to that little maneuver(sp).


The process finished correctly. Unfortunately, I don't know if 3.5.9 fixed it or the PCI board did. Fortunately, I don't really care as long as it didn't quit on me. *laughs*

BTW, I did that "8" setting on the driver controllers and was amazed at how much more quietly the motors ran!

g.