Does it happen at low speeds? (ie: 5 IPM)
Printable View
Does it happen at low speeds? (ie: 5 IPM)
Yes, it is independent of speed.
Do you have access to an oscilloscope?
The next thing that I would check is to see if the PC is actually sending step pulses to the board during the time when you are not jogging it, or to see whether you can detect noise or some other signal going to the step input of the TB6560 (that may not be coming from the PC).
Unfortunately no, I don't have an oscilloscope. Open to suggestions on other ways to detect this - obviously just sound doesn't work as the motors are spinning when this is happening. I might be able to rig an LED across the relevant input...
Unfortunately an LED alone is not likely to give you any useful information.
The step pulses (or noise) would be extremely short in duration, on the order of at most a few microseconds, so you would not see that on an LED.
An oscilloscope is really the right tool for performing this test. Do you have a friend who is an electronics enthusiast who might be able to bring over a portable oscilloscope to do the test?
As an alternative (not as useful, but might work) maybe you could find someone nearby who has a logic probe that you could use...
If you are handy at soldering and have a nearby electronics parts shop, I suppose that you could buy parts or a kit to put together your own logic probe (or find such a thing on eBay).
In my experience, although you can't see individual pulses, an LED across a parallel port will glow if the pin is active - I'm not sure whether it will work with the tb6560 board attached though. In fact, now that I think about it the board does have some LEDs on it - I need to go back and look at the circuit digram because what I've suggested might already be there...
It's a useful suggestion, however my goal at this point would be "problem source isolation", that is, determining whether the LPT port is actually putting out pulses or whether noise or pulses from some other source may be getting into the TB6560 chip, or whether something entirely different and unexpected is going on. Once such a determination has been made, you can then focus your attention on getting to the root of the problem and addressing it.
That task will require some probing around at various parts of the circuit, cable, and connectors, with and without the board hooked up.
So, while the onboard LED(s) might conceivably provide some diagnostic information, using a movable probe would be more flexible and likely to give a more complete picture of what is going on.
With a plain LED as a test probe (possibly with a series resistor), you would also run the risk of loading the circuit which could affect and confuse your test results.
Hi Malcolm ,
you could use a logic probe like this one from maplin -
Logic Probe : Basic Test Equipment : Maplin Electronics
to test the printer port pins used as step signal outputs
then see if the computer continues to send step pulses after you release
the jog button
you will need an adaptor cable to ,either to take power from a usb port
or ps2 keyboard/ mouse connector for example
if the board is the one I'm thinking of
the step and direction signals will be available at the 15 pin D-type connector
normaly used to connect the blue manual control thats available for the
tb6560 stepper boards
if you need to test the board
a 5V supply is available at the 15pin Dtype connector - be carefull not to shortcircuit it! or you will destroy the TA6560 ic's
John
It sounds like your parallel port might not be compatible. Use a voltmeter to measure the high and low voltage levels with the drive connected. The lows should be <.8v. The highs should be >2v.
Are you able to try another computer?
Which of the marvelous TB6560 based boards is this by the way. Many sour flavors of this chip/board.
The mouvement AFTER the application JOG button is released could be just about anything and may not be related to the changes done to the board at all or simply coincidence.. could be keyboard button bounce, could be buffer overrun, could be timming was thrown off with the modifications to the board and the added speed you gained.
You have another computer/software you can test with ?
MP*10
Thanks for all the help - looks like I'll run up to Maplins and get a logic probe.
I get the impression that no-one else has had this issue when they put in the 100pf caps then. My problem is trying to understand how this could affect the apparent behavior of the software, and I'm hampered by not really understanding the relationship between the TB6560 clock and the timings in LinuxCNC and the parallel port pulses. Is anyone able to provide a good explanation of this relationship?
Aside from the numerous issues with these controllers, it's not uncommon for some modern motherboards to put out varying logic levels via the parallel port, that sometimes requires a logic shifter or buffer to clean up before being fed to the input stage of the controller. This may possibly be an issue with yours, have you also verified the BIOS has the appropriate parallel port type set for the controller? Enhanced, ECP, etc?
cheers,
Ian