Hi Pete, Based on the photo in your link a google search brings up this: https://uk.rs-online.com/web/p/motion-control-sensors/2914305/ and in turn this although it's not clear that the data sheet is exactly for yours but the other sheet only states push pull without any clear voltage levels. https://docs.rs-online.com/450e/0900766b8002f9dc.pdf
Anyway the output is listed as low is <= 0.5V and high is >= 2.5V but doesn't say if it goes up to the 12V encoder supply voltage. I'm guessing it does. Depending on whether you have the K or D version the drive current could be as high as 30mA but let's assume a max of 10mA. With your 1K resistor in series and assuming no drop on the input circuit 2.5V/1000 Ohms = 2.5mA. Next a look at the buffer data sheet. https://www.nxp.com/docs/en/data-sheet/74HC4050.pdf With a 4.5V supply (nominal 5V) the high voltage threshold is 4.4V minimum. That it's working at all suggests you are seeing at least that out of the encoder so likely is close to the 12V supply voltage. The question then is now much current is flowing in the wires. The buffer has a very high input resistance with a leakage in the 5uA range. What that means is any kind of random noise ends up impressed on the signal wires and can result in spurious pulses. With the encoder capable of driving 10mA I'd look at something like the HCPL2631 opto isolator. Input current is between 6.3mA and 15mA. It has delays in the 100nS (10MHz) so the max 300kHz from the encoder won't be bothered by it. Assuming your encoder does produce 12V from the Push side of the push/pull outputs then 12V/0.015A=800 Ohms. I'd stay with the 1K resistors and then even with just 6.3mA the encoder output has to be at least 6.3V. And for induced electrical noise to create 6.3mA is very unlikely. The only catch-22 might be if the encoder does produce 12V for the high level and is also just rated at 10mA you will have to change to a 1.2K to 1.5K resistor instead but then with that higher voltage you will still be above the minimum 6.3mA required by the HCPL2631 John Dammeyer > -----Original Message----- > From: Peter Hodgson [mailto:[email protected]] > Sent: October-08-21 12:22 PM > To: [email protected] > Subject: [Emc-users] Fwd: Rogue Index Pulses > > Hello All, > > I'm hoping someone with more electrical control experience than me may > be able to offer advice on why I am seeing occasional rogue index pulses > which are messing up my G76 cycles on larger threads. > > The Halscope below shows an example of the occasional ghost index pulse > which can create a new 'thread start' and trash my parts as photo above. > > A schematic of the encoder setup which is a quadrature 500ppr > incremental encoder can be downloaded here: > > > https://www.purbrookengineering.com/index.php?option=com_content&view=article&id=11 > > > The signal goes to a Pico Systems USC board via a voltage level shifter. > We have a 0.75uF capacitor to filter the 5v supply to the voltage > shifter and 1kohm resistors in series on the input signals from the > encoder. The encoder cable is shielded, grounded only at the encoder > case end and is not close/parallel to any power cables. The 12v supply > to the encoder is from a voltage regulator and the 5v supply to the > level shifter is from the Pico control board. > > Jon @ Pico provided me with excellent support to get me this far with > the setup which now cuts smaller threads quite reliably but with larger > threads on my small lathe I need many passes and I haven't been able to > get rid of these occasional ghost pulses which eventually cause a bad pass. > > I think these pulses are 'noise' but I cannot find what might be causing > them. > > Any suggestions please? > > Pete > > > > _______________________________________________ > Emc-users mailing list > [email protected] > https://lists.sourceforge.net/lists/listinfo/emc-users _______________________________________________ Emc-users mailing list [email protected] https://lists.sourceforge.net/lists/listinfo/emc-users
