On 08.02.16 13:16, andy pugh wrote: > I need to connect to 8P8C connector on the VFD which has pins > 1 A' > 2 B' > 3 A' > 4 R > 5 D > 6 B' > 7 DP5V > 8 SG(GND) > > Is that ever going to work? The YSB dongle seems deficient in > terminals by at least 1, possibly 4...
The above looks a lot like full duplex, with A'B' being one differential line, and A'B' being the other. (Seems to be too many ' there.) (We need to know which is A, or the data is inverted.) D,R are commonly the labels for Tx,Rx on the inboard side of an RS485 driver chip. What can they be doing on the external connector? (Unless they're alternative single-ended inputs? Weird.) Or is R the (100 ohm, ISTR) resistor which may be connected between ends, in lieu of binding the GNDs, when there is up to the permissible (7v ISTR) voltage offset? But then what's D? The two-terminal dongle would be used for half-duplex, with separate grounds, relying on less than 7v float. On 08.02.16 22:42, Bertho Stultiens wrote: > On 02/08/2016 10:12 PM, andy pugh wrote: > > (And, back to the original question, how do I wire the 8 pins on the > > VFD to the 2 pins on the adaptor?) > > Ok, that looks reasonably "easy" and looks like rs422. It has two AA and > BB connections, which are most likely a full duplex differential lines. > If either AA or BB are opposite polarity, then you can be quite sure > that it is differential (use a pull-up when measuring). And that will identify whether it's AB which make a differential pair. > Then you should be able to see on an oscilloscope if any line sends a > peep on power-up. If so you have identified the rx/tx lines, the > polarity of the tx line and the default baudrate. > > Alternatively, the tx/rx lines should be identifiable using injection > with a series resistor. The tx lines will stay put, whereas the rx lines > will follow the injection. Yes, if it's RS422, with the driver permanently enabled. But if it's being tri-stated when not sending, it'll look just like an input. On 08.02.16 15:19, Jerry Scharf wrote: > This isn't the worst thing I've seen. Rs485 is a two wire differential half > duplex signalling system. The doc says that it can run up to 38.4kbps. In reverse gear. ;-) RS485 driver datasheets have for decades been quoting 10 Mb/s for shortish lengths, and 1 Mb/s for half a km line lengths, AFAIR. The wikipedia page talks of up to 35 Mbit/s up to 10 m and 100 kbit/s at 1200 m. There are slowed-down low-interference driver chips which will only do 2.5 Mb/s. ISTR it was the physical layer of Appletalk, the network used by Apple Macintosh, back when my beard wasn't grey. For an undocumented VFD, untangling the Modbus trickery sounds like a non-trivial project. Sending 1 Mb/s overs RS485 to an Arduino with D/A sounds simpler. (PWM into a LPF might respond too slowly, once ripple is filtered adequately?) What about 20 mA current loop? Sending an analogue signal as a current, rather than a voltage is reputed to be significantly more noise immune than sending it as a voltage. Erik ------------------------------------------------------------------------------ Site24x7 APM Insight: Get Deep Visibility into Application Performance APM + Mobile APM + RUM: Monitor 3 App instances at just $35/Month Monitor end-to-end web transactions and take corrective actions now Troubleshoot faster and improve end-user experience. Signup Now! http://pubads.g.doubleclick.net/gampad/clk?id=272487151&iu=/4140 _______________________________________________ Emc-users mailing list Emc-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/emc-users
