Low Voltage on Parallel Pin

Hi!
It's my first time using LinuxCNC after some months of GRBL. However everything went surprisingly smooth: I installed it on my old PC with an Intel Q6600 and an Asrock P5B-DE MB. 4 GB of RAM and onboard Parallel Port, all other Hardware removed. As a motor driver I'am using this infamous 4xTB6560 China driver but mine has another pinout .
It has an enable pin for every axis which I am trying to trigger with

in my custom.hal and for X and Z that works just fine! Pin 17 on the other hand just puts out 1.55 V when High.






That seems to benough to be sensed by an Input Pin (Connecting 17 to an Input, parallel port tester behaves just as expected) but not for the driver board. The Y Axis just wont move.
Of course there are plenty of possible workarounds like rewiring or using the 4th axis driver as Y but for elegance's sake I was hoping some configuration issue to be the cause.

Here is my bios and parallel port Settings, in case it matters:

Good morning,

Pins 1 14 16 and 17 are control pins and are made differently inside the parallel port "hardware". I suggest you use a breakup board, that already have the electronics built in to take the power needed from all the pins. This is in my point of view the most easy (and safe) way to use the parallel port.

theirs two ways to achieve , however it includes a bit of hacking ...

either fit a pullup resistor of say 10k from pin 17 to 5v on the pcb .. or
cut and jumper the track that goes to pin 17 across to the x enable pin 4 , you dont need to use a seperate enable for each stepper , they can all be the same pin .
yes these chinese boards can be a headache !

Hi

As emcPT says, you would be best off with a BOB, preferably a buffered one which will output 5v.
3.3v parports can cause many strange problems, simply because of the lower voltages not making it on longer cable runs etc.
It is never really a good idea to wire direct to a MB parport

Your 1.55v is undefined in terms of TTL, because it is above 0.8v and below 2v, all active HIGH signals should be 2.4v + and LOW 0.8v or less.

regards

Thanks a lot for your ideas!
I have soldered a bridge between all 3 enable pins and that seems to do the trick for now. But now the X Step Pin 16 stopped working (Yesterday it worked fine)

That breakout board you suggested: What would that be? My china-board already has all inputs optocoupled (Though the designer decided to use common ground for logic and power side ... who needs such shenanigans like galvanic isolation anyway)

I have found an MCS9865 PCI Parallel port. Could that thing maybe produce proper output voltages on those control pins?

purged wrote:

Thanks a lot for your ideas!
I have soldered a bridge between all 3 enable pins and that seems to do the trick for now. But now the X Step Pin 16 stopped working (Yesterday it worked fine)

That breakout board you suggested: What would that be? My china-board already has all inputs optocoupled (Though the designer decided to use common ground for logic and power side ... who needs such shenanigans like galvanic isolation anyway)

I have found an MCS9865 PCI Parallel port. Could that thing maybe produce proper output voltages on those control pins?

It does rather sound as if your on board port might be faulty, so trying a PCI board could be a good way forward.

The BOB I was meaning, is the type which triggers a 5v signal from the weaker 2.4v one from the parport, whilst also isolating.
Functionally the pull-up cncbasher was describing.

Try the PCI card before thinking about that. Just about all computers built in at least the last 12 or so years have had 3.3v TTL and most of them perform OK.

regards

So the same problem of low voltage occurred when using the pci card. How strange? I couldn't believe it, so I went "low level" and used a C Program with parapin to test it again. Error remains ... you get an idea? I finally did (after 2 days of fighting this trivial problem ...)

I measured the voltage driectly at the connector of the PC and guess what: Turns out it was a broken cable all along
If I used that salvaged one from my "old cable box" I might have tried that earlier but the defect one was brand new. Bummer!

It happens, at least you have found the problem

When I first started I used a proprietary parport cable that had come with a printer years before

I had all sorts of problems getting one axis to work, until I traced every single pin and found that two of them were swapped over internally in the cable, presumably to suit that manufacturers printer wiring.

At one point there were also 25 pin serial cables around, which looked for all the world like parallel port cables but half the pins were not connected.

Anyway, onwards and upwards

regards