USB Cable Length - PMDX 422

[rae_gordon]

I have a new 422 that I am utilizing via the USB cable.   It is using the USB for power and it appears that there is an issue with the USB Cable length.  When I run it utilizing a 3' usb cable it seems to work fine once I get the cable orientated.  I purchased an 8ft cord to relocate my PC and it will not communicate at all with the 422.  Is this a known problem,  It makes sense that this would be the case.  If it is, will powering the 422 via a 5v external source fix this?

I am not utilizing the 422 as a power source for any sensors or 5v devices.

Windows 7 ultimate
PMDX 0.31.147
Bootloader 1.7.63
Firmware 0.40.130

Mach4 Hobby - 4.2.0.2703

[Bob at PMDX]

I have not heard or seen this being a problem.  I routinely run my SmartBOBs through a standard USB cable plus one of our USB A-A extender cables.  I don't know off the top of my head if that total length is 8 ft or not, though it *should* be since the extension cable is 6 ft and the original USB cable should be at least 3 ft.  But I"ll check that when I get into the office later.

Here are a bunch of questions, some of them pretty basic, but please bear with me.

Can you give me any more details about the symptoms you are seeing.  For example, do you see the SmartBOB turn on all of its LEDs briefly when you plug it into the longer USB cable?  Does the green LED continue to blink about once per second?  What does Mach4 show you to tell you it isn't working?  Do you get a "I couldn't find a valid PMDx-SmartBOB device" kind of pop-up window during startup?

Have you tried this longer cable with any other USB device?

Try disconnecting everything from the SmartBOB (PMDX-134 and anything connected to the screw terminals.  Does this make any difference?

When I run it utilizing a 3' usb cable it seems to work fine once I get the cable orientated.

What exactly do you mean "once I get the cable oriented"?

Without changing anything that is connected to the PMDX-422, can you plug the 3' USB cable and have it work and the, not changing anything else including position of the PMDX-422, plug the 8' cable in?  Does it work in this case with the 8' cable?

And finally, please enable the debug log file as described here:
http://www.pmdx.com/PMDX-Forums/index.php?topic=97.0
and send the log file.  You can either upload it to the forums here (just the log file, not the entire Mach4 package), or email it directly to me at bob at this domain.

Bob

[rae_gordon]

3' cable
Started Debug, Reference all home, Jogged around, referenced all home, All worked fine. I did not attach the file since it worked fine.

8' cable
restarted, Error Port Autoscan did not find PMDX-Smartbob-usb device
Removed all input and outputs from card and restarted Mach.  Would not reconnect, so removed usb an reinserted.  Once I did this, it was able to connect.  Started reconnecting devices and trying to isolate the problem. Restarted everytime I reconnected a cable to include unplugging the usb.

I am able to communicate with all outputs connected, xyz, but no inputs in a low state.
USB power meters at 4.94v with the longer cable 5v with the shorter, board output voltage is at 4.44, with EStop voltage at 4.52v open.  I was able to do a limit overide and jog around for a small bit, then it no longer took my inputs.

FYI i am utilizing optical inputs for the x,y, and z inputs.  Standard Estop buttons.  If noting is connected and the Estop is activated, shorted, then I cannot connect, or at least most of the time I cannot.  It appears that any draw down on the inputs causes the board to lose communications.

Since I was unable to communicate do you still want me to run the debug

[pedio]

Don't know if this will help - I put a USB hub by the PMDX board and connected the hub to the computer via a long USB cable. I mainly did this so I could use a ShuttlePro on the machine as a pendant; however, it seemed to correct a few problems. Has been working for several months.

This was early in my testing of Mach 4 so I don't remember what the problems were. I also don't really know if they were a Mach 4 or communications error. Too many components were changing at that time. It may be worth trying.

[Bob at PMDX]

Thanks for the testing info.  The voltages you mention are about what I would expect - a little bit of a voltage drop in the longer cable.  And sorry that I thought you had a PMDX-134 - I was getting my support threads  mixed up.

(1) One unanswered question from my previous post: You said in you original post that "When I run it utilizing a 3' usb cable it seems to work fine once I get the cable orientated." What exactly do you mean "once I get the cable oriented"?

(2) When you close the EStop switch or ground the EStop input and then lose communications with the PC, is the green LED next to the USB connector still flashing about once per second?  Is the red LED just above it on or off?

(3) When in this state (EStop grounded and no communication with the PC) please measure the USB bus voltage and/or the voltage on any of the "+5V" screw terminals.

(4) We ship the PMDX-422 with a spade lug installed shorting the EStop input to GND on connector J10.  Do you get the same behavior if you install this (or a short piece of wire) instead of your EStop switch?

(5) Which terminals on the PMDX-422 do you have your EStop switch wired to?

(6) What about with the 3 ft USB cable?  Does closing the EStop switch make the SmartBOB stop communicating with the PC?

(7) Do you have anything connected to the 26-pin ribbon header on the PMDX-422?  If so, look at jumper JP1 along the edge of the board near the ribbon header and J10.  Is this set to "On" or "Off".

Bob

[rae_gordon]

I want to note that when I loose communication, a restart of just mach will not restore communication, I must unplug the usb in addition.

Quesions:
  1) when I had the usb stetched tight to allow the computer to reach the monitor and Control panel I would get an intermediate loss of com.  "where it will stop communicating until I restart mach4.  I dont remember if I had to unplug the 422
2) still flashing once per second.  Red light is solid
3)USB voltage is 4.94, power terminal is 4.4v
4)I have 2 e-stops.  one on the machine wired to terminal 10, the other on the control box wired to the Estop adjacent ABC terminals.  both wired normaly open.  They were originally Normaly closed when I started this, but I have since changed them.  So now I am testing in the normally open which allows me to communicate unit I hook up the limit switches.  Also if I run it for a duration with the Limits overidden in this state it will work for a short while then stop. with the Red light
5)see above
6) no the 3ft cable seems to work fine
7) Nothing is connected to the Ribbon connector, JP2 is on A and JP1 is off.

[Bob at PMDX]

I ran a PMDX-422 here in the office through 2 ea 6' USB A-to-A extension cables, then though a normal 3' USB A-B cable.  I was unable to create any issues with the PMDX-422 communicating with  the PC.

I *was* able to reproduce your symptoms somewhat by plugging the USB cable only part way into the USB jack on the PMDX-422.  The USB connectors are designed so that the ground terminal makes contact first, then the +5V terminal, and then the two data lines.  With the USB cable plugged only part way into the connector on the PMDX-422, there was power to the PMDX-422, and the data lines would initially make contact so the PMDX-422 would be able to talk to the PC.  If I wiggled the cable or connector, or pulled it just ever so slightly out of the PMDX-422's USB jack the data lines would disconnect (even if only for a split second) and cause the USB port to reset which then makes the plug-in unable to talk to the device.

The symptoms you describe, combined with your statement that the 3' cable seems to work fine and my experiments above, make me think there is an issue with the longer cable.  It may be in the cable itself, or it may be that one of the connectors is not seating completely or is making intermittent contact.  If so, you should be able to replicate this problem by wiggling the cable and/or connectors.

Can you try that same long cable with another USB device?  Wiggle the cable and connectors and see if there are any problems.

Hmmmm...  Are you plugging the USB cable directly into the USB jack on the PMDX-422? Or are you using a panel mount USB jack that plugs onto the row of pins just behind the USB jack?

Bob

[rae_gordon]

I thought it was the connector also initially.  and have tried wiggling it and checking how far it seated, but that did not seem to be the issue.  I am using the onboard connector.  I purchased what I thought was the panel mount connector with the pin connectors to remote another usb, but when it came I was just a usb extension that panel mounted, but plugged into the front of the unit, it was a mistake on my part when ordering.

I have not purchased another usb cable, but I guess for process of elimination, I will go ahead and order one.

Also, keep in mind that every change that I made, I had to unplug and re plug the cable.  I have a 5v power supply driving the opto sensors and I will try to connect it to the board to see if that helps.

When you tested with the longer cables, did you notice a voltage drop at your board due to the additional length?  I was seeing a .6v drop with my longer cable, if you are not seeing any drop that may be the issue.

If I do run 5v power to the board, will I have to power down the whole unit, or press the reset button if it locks up?

[rae_gordon]

FYI I just ordered the Header to USB mount to give it a try in case the connector is bad.  Plus I needed one anyway.

[Bob at PMDX]

I thought it was the connector also initially.  and have tried wiggling it and checking how far it seated, but that did not seem to be the issue.  I am using the onboard connector.  I purchased what I thought was the panel mount connector with the pin connectors to remote another usb, but when it came I was just a usb extension that panel mounted, but plugged into the front of the unit, it was a mistake on my part when ordering.

That cable should be OK.  I see that you ordered a USB "B" Panel to header cable.  We also threw in a 6 ft USB A-to-B extension cable for you to try as well.

Also, keep in mind that every change that I made, I had to unplug and re plug the cable.  I have a 5v power supply driving the opto sensors and I will try to connect it to the board to see if that helps.

You can try the external 5V supply, and I will be interested to hear the results.  My gut feeling is that this is not a power supply issue.  If it was a supply issue, I would expect the PMDX-422 to reboot or mis-behave (in some way other than not communicating over the USB).  Since you said that when the comms stopped working you did not see the power-on LED test and that the green LED continued to do its slow blink, that suggests to me that the firmware was still running.

When you tested with the longer cables, did you notice a voltage drop at your board due to the additional length?  I was seeing a .6v drop with my longer cable, if you are not seeing any drop that may be the issue.

When you said that you measured the USB supply voltage at 4.94V, I presumed you meant you measured the +5V pin on the USB jack itself (or on the 5-pin header next to the USB jack).  I also thought that the 4.4V that you measured was at one of the "+5V" screw terminals on the PMDX-422.  If my two presumptions are correct, then you have approx. 0.06V drop through the USB cable.  The 4.4V you measured is consistent with the voltage drop *ON THE PMDX-422* due to protection circuitry that protects the USB interface from overvoltage on the +5V terminals.

But to answer your question, I saw negligible drop across the 15 ft of USB cable (less than 0.1V - I'm at home now and don't have access to my test notes to get the exact value).

If I do run 5v power to the board, will I have to power down the whole unit, or press the reset button if it locks up?

There is no reset button on the PMDX-422.  The push-button is (currently) only used for forcing the boot loader to run in the case of corrupted firmware.

While you are waiting for the new USB cables, I would like you to enable the debug log and try to capture the communications failure (see my first message in this thread for details).

Bob

[rae_gordon]

Sorry for the delay on this, it appears that the problem I was having was due to the USB connector on the 422, I plugged in the header to USB and using that connection, everything seems to be working.  I am using my longer cable that I was never able to use.  I tried the cable that you sent, and I still had the problems through the front connector.  Is there any circuitry between the Board USB and the Header pins.  If there is not, then the only thing I can imagine being wrong with the unit is possibly a bad solder joint.

[Bob at PMDX]

Glad to hear you got it working.  Nope, there nothing but traces (copper) between the USB jack and the header.  It is possible there is a marginal solder joint.  Even though we visually inspect every board, we are human and occasionally miss something.  If you are happy with your current setup, good.  If not, and if you want to send the board back to us for repair, we'll take a look and see what we can see.

Bob