the datasheet from the bus transceiver 74HC245 shows a minimum voltage of 3.15V when driven at 4.5V
John-100 schematics of the TX14175 show the voltage at the transceiver to be at +5V
Is the 3.35V delivered by the parallel port sufficient to drive the output high? I'm worried that might be a little tight and cause sporadic missed steps. Can anybody confirm what kind of margin is usually considered safe?
Any recommendation to boost the voltage of the parallel port to 5V?
I do not have a PC with a 3V3 printer port I can test but
I expect if you terminate the outputs with this :-
the 1K pull up resistors will help to pull any output at a logic 1 up some thing close to 5V
I have searched for data on the super I O IC's found on new mother boards
but I have not been able to find out exactly what the IC's internal circuit is thats connected to the I/O pin
I have found in an on line copy of The Art of Electronics (2nd edition) this partial circuit of the original centronics port
it does not show the address decoding in detail but give you the basic idea
John
PS the centronics port diagram is not in the printed copy of 3rd edition I have !!!!
Last edited by john-100; 06-08-2017 at 06:52 PM.
Reason: correct typo
That's interesting. If I get that correctly, I could pull the line up to 5V without risks of damaging the 3.3V PCI card.
I'm not sure I understand the need of a 1K pull up resistance at both end of the wire though...Is that design for very long wire where there could be significant impedance?
I'm in the process of acquiring a 5V parallel card, sounds like the easy way to test that hypothesis... As long as I indeed get a 5V card, there seems to be some confusion on how to find those. I'm banking on the fact that if the PCI has only the 5V notch on it, it should output 5V. Misplaced hopefulness maybe.
Not knowing how to troubleshoot new instruments just drives me crazy, busted a stepper driver out of pure stupidity and exhaustion trying to figure out that missed steps problem, the 6560 IC exploded! Now I'm trying to understand before I do more stupid stuff...
the IEEE1284 recommended circuit shows the two resistors are only needed for bi directional tranceiver circuits
for signals only going in one direction just need one pull up resistor
at the frequencies involved
the 39 ohm resistor in series with the signal wire is to match the impedance of the cable as they act like 50 ohm transmission lines !!!
a while ago I traced the circuit of a C10 breakout board from pictures off the web
on the C10 they have used 1200 ohm pull up resistors connected to +5V
many the TB6560 stepper drivers have known issues that causes them so self destruct
if the 5V supply to the TB6560 is not up first then destructive currents flow when the motor supply is applied
the other problem is the motor supply voltage
on some boards they quote Absolute Maximum Ratings of 40V !!!!! - that guarantees a short life
on others they correctly give the Operating Range 4.5 ⎯ 34V and in the kits supply a 24V power supply to under run the IC
either a wire breaking or you disconnecting a stepper motor with the power still on will also destroy the TB650 or any other stepper driver
According to the schematic the board already has 10k pullup resistors to the +5 volt supply so there should be no problem with 3.3 volt input as default will be 5 volts and port will pull it low to change state..
I expect the 10K pull up resistors on the TX14175 will pull the I/O IC's up to a valid high for the 74HC245 input
the only issue I can think of is the possible limit on the step frequency using 10 K resistors instead of 1k
I'm assuming the I/O IC output can only provide a weak pull up current of its own and the extra current from a 1K pull up resistor will
enable it to charge the cables capacitance faster than when using 10 K resistor
Just a quick update to say that using a 5V output parallel card fixed my issue. Ended up getting a Rosewill RC-302 from ebay and voltage was spot on to 5V.
For those that might be hitting that post with no idea on how to find a proper 5V PCI parallel card, I would propose to look for cards that lack the 3.3V notch (see attached picture). I assume that universal cards 5V/3.3V will output at the lower voltage of 3.3V (this is actually what I observed). I would also like to assume that manufacturers won't bother lowering voltage to 3.3V if the input is 5V only. If your PC doesn't have a 5V or 3.3/5V PCI port I think you're out of luck
To complete, I would say clearly that the TX14175 does not seam to live well with 3.34V signal from a onboard parallel card (Dell Optiplex 380), at least in my test settings. This voltage was also slightly higher than what I is expected for a 3.3V card. Other cards I tested were dead on 3.30V and mine at 3.34V. I did not test the machine with that last card but it might account for more sporadic and hard to track missed steps issues.