On Thursday 04 May 2017 02:32:41 Erik Christiansen wrote: > On 03.05.17 22:40, Gene Heskett wrote: > > On Wednesday 03 May 2017 22:29:43 Chris Albertson wrote: > > > RS422 is balanced and opt isolated at the Mesa card end so you > > > likely have better noise resistance then with SPI. > > > > rs485 is the single ended cousin. Could a atom board keep up with > > that with a pci-e rs422 card in it? That does have a bit of an > > attraction, particularly from the noise safety dept. Silly Q: can > > the pi do rs422 over its gpio? > > Hook a RS485/RS422 transceiver to a Pi's serial TXD & RXD, and > terminate each end of the twisted pair cable in 120 ohms, then it's > done. All the chip does is TTL/CMOS to differential Tx & Rx. > > I bought some surplus DS3695A RS485/RS422 half-duplex transceivers a > few days ago for US30c each, but they draw over 27 mA, idling, - a > MAX485 gets by on 300 uA, though it'll only do 2.5 Mb/s. The > SN65HVD485 will do 10 Mb/s on long lines, and draw 1 to 2 mA. (They > may be the one Jon likes?) > > Just watch pinout, driver enable logic, and whether it's full/half > duplex. The former has separate Tx & Rx cables (4 wires) vs 2 wires. > > Incidentally, it's the RS485 which is multipoint - just in case you > want that, and go to put plastic down. > > Erik > The ones I bought to play with are usb on one end, and 3 terminal on the other. Bought them at the same time I bought that clone vfd, paying about $3 ea at the time, only to find when I took it apart, that the footprint for the seriel connection was totally absent from any of its boards. So I'm running it directly with a spinx1, which is receiving a PDM signal from the 7i90. Control works great until nearly 200hz, but the motor inductance is so high its drawing under an amp/phase at 200 hz, and the torque is of course vanishingly small. I'd guess under cutting loads, 120-140hz is about maximum. I'm sure I don't have it well tuned yet, the lf cutoff at 15 hz seems too high as the start is pretty brutal. With the clamps on the chuck pulled up tight so I can't unscrew the chuck, I can have it doing 100 spindle revs using the lowest belt position, type m4 to reverse, and its doing that same 100 revs in reverse less than a second later, wash, rinse, repeat going back to m3.
My target is minimum overshoot for rigid tapping. I have some hal stuff I can watch with a halmeter, tells me how many encoder counts it overshoots so I can set the g33.1 params to not hit the bottom of the blind hole & break the tap. Getting more familiar with pyvcp, I'll probably add a number box to the gui at some point. I'd like it to display the distance, but the k value fed to the g33.1 hasn't been found so I can multiply it and convert the encoder counts into actual overshoot distance traveled. Best I can display might be degrees turned fwd since the reversal was given. I freeze that peak reading until the next peck if using a peck cycle. Sounds like if I buy some more rs422 stuffs, they ought to be based on the SN65HVD485 chip. I'll be on the lookout. I am in favor of any isolation from the noise I can get even if my current "lashup" is working well. But on ebay its opto and slow and expen$ive, or usb and a 5 dollar bill. And those for sure aren't isolated. Or am I looking with the wrong search terms? URL for something like you are writing about plz. Doing an experiment this day, might take a while. I managed to find the root of the card on the pi, and with a terrabyte hd plugged into the pi, cloned the one bootable card I have left to a 32Gb file on the hd. I just brought a dead card, one of my readers, and that drive in here and plugged it into my usb tree, and dd is now happily writing that 32Gb image to a card that won't boot. Since the hd had an un-expanded filesystem install of raspian jessie on it, I first had to find a way, using the tools available on the pi, to expand the / filesystem from 3.2Gb to 850Gb, and do it without losing any data. Between parted and resize2sf, I did appear to get it done. If it writes at the usual 4.5 megabytes/second, it will be done about midnight. :( Cheers Erik, Gene Heskett -- "There are four boxes to be used in defense of liberty: soap, ballot, jury, and ammo. Please use in that order." -Ed Howdershelt (Author) Genes Web page <http://geneslinuxbox.net:6309/gene> ------------------------------------------------------------------------------ Check out the vibrant tech community on one of the world's most engaging tech sites, Slashdot.org! http://sdm.link/slashdot _______________________________________________ Emc-users mailing list Emc-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/emc-users
