**** BEGIN LOGGING AT Fri Apr 17 02:59:57 2020 Apr 17 03:01:43 I looked. Apr 17 03:01:44 Nice. Apr 17 03:02:36 I might test it soon. I know nothing about basic as of now. Apr 17 03:03:41 Okay... Apr 17 03:10:00 You know tar? That's all you need to know. Apr 17 03:10:56 I may be asking the wrong question here, if so, please feel free to point me in the right direction and I'm happy to dig into docs/etc. Apr 17 03:10:56 I've got a BBB which I'm looking to use with a TI DLP dev board. I'd like to use the parallel RGB data on the GPIO pins with a framebuffer that I can write to directly from userland, but I'd like this to be independent of the HDMI port / controller on the BBB. Apr 17 03:11:06 Is this reasonable? Apr 17 03:11:08 Download tar, extract it, cd to bwbasic-3.2a, sh compile, that's it. Apr 17 03:12:29 KenUnix: I might try it later. I have the page marked. Apr 17 03:12:30 set_ sounds like your kind of question... Apr 17 03:12:52 No, that is definitely not my question. I have used the DLP board yet. Apr 17 03:13:00 IanWizard: ehh what Apr 17 03:13:17 From what I've read so far, it sounds like the RGB888 output on the pins is only possible when it's driven by the HDMI controller, but I've been looking at a lot of docs and datasheets for a lot of different projects lately, so I may just be lost in it all. A last resort would be simply bitbanging GPIOs with the data. Apr 17 03:14:09 zmatt: https://elinux.org/Beagleboard:Cape_Expansion_Headers#Black_eMMC_and_HDMI_Pins Apr 17 03:14:11 IanWizard: no, the AM335x drives the parallel RGB data, either to some external consumer (e.g. an LCD cape or TI's DLP cape) or to the HDMI framer Apr 17 03:14:43 the HDMI framer (when used) receives the parallel video data and converts it to hdmi Apr 17 03:15:55 MattB0ne: I have a Pololu driver. I can rehash some fury I took out on it years ago w/ an update if necessary. I will look it over and get back to you. Apr 17 03:16:11 My question is if it's possible / reasonable to decouple the RGB pins from the HDMI, so that both may potentially be used? Or would that require manually driving pins with the data. Apr 17 03:17:10 IanWizard: it is physically impossible. what you're saying contradicts the schematic Apr 17 03:17:19 zmatt: It sounds like I'd have to handle it manually. At which point I'm probably looking at using a PRU. Apr 17 03:17:26 Yeah. Apr 17 03:17:40 TY zmatt :) Apr 17 03:17:40 there's no point in using the PRU for video instead of using the display controller Apr 17 03:18:13 you wouldn't be able to use HDMI independently of your DLP projector in either case Apr 17 03:18:21 If I want a separate display pushed out over the pins than the HDMI. Apr 17 03:18:34 the am335x only has 1 video output Apr 17 03:18:46 (and a mediocre one at that) Apr 17 03:19:56 set_ Gotta run. Leave ma a message on GitHum under Issues. Apr 17 03:19:56 True dat. (mediocre) Apr 17 03:19:56 I'm referring to using a separate framebuffer, which would be pushed out over different pins, to drive the DLP board. Apr 17 03:20:34 At which point, I doubt it could keep up without being handled by a PRU (which I've yet to play with), and outputting over separate pins. Apr 17 03:20:40 set_ Gotta run. Leave me a message on GitHub under Issues. 8) 8) Apr 17 03:20:40 parallel video requires a lot of pins though Apr 17 03:21:20 True, but for the resolution being used, I could probably get away with three shift registers. Apr 17 03:21:40 Though I'd probably just do parallel for simplicity. Apr 17 03:21:53 really? how slow is the pixel clock? Apr 17 03:22:27 I'm speculating, haven't applied any actual numbers to it. Apr 17 03:22:55 And TBH, I don't really need a separate output, I just wanted to know how reasonably that would be doable. Apr 17 03:23:08 The answer being "not very". Apr 17 03:23:51 no. the pru cores only have 17 output pins each, and they're not even all available on the BBB, especially not if you don't want to disable hdmi Apr 17 03:30:56 MattB0ne: I have a win desktop to goof w/ right now. If you want, I can try to configure the older 18v7 simple motor controller from Pololu to test things. I even had a book on it at one point. I am sure I dredge it up somehow. Apr 17 03:31:48 (I'm not still trying to make this work, just thinking publicly. Also, I haven't worked with the PRUs at all, though it's been on my todo list for some time.) Apr 17 03:31:48 At 12 pins each I could presumably split it into the upper and lower 4bits, with the framebuffer shared between the PRUs. Or, since it's pretty data agnostic, just two sets of 12bits each. Apr 17 03:32:08 and for example for 640x360p60 with reduced blanking, the CVT pixel clock rate is 17.75 MHz. that would leave only 11 PRU cycles per pixel clock cycle Apr 17 03:32:29 Really, quite far beyond the point of being worth while for something trivial such as this. Apr 17 03:32:57 also PRU cannot fetch data from ddr3 ahead of time like the display controller can, they synchronously wait Apr 17 03:33:04 p30. Doesn't need to run at the full 60 Apr 17 03:33:26 Yeah, it's becoming unreasonable. Apr 17 03:33:28 and a useful framebuffer doesn't fit in PRU memory Apr 17 03:33:31 I'll stop thinking aloud. Apr 17 03:35:24 My brain is still waking up after a long hiatus; so these thought exercises are helpful to flex it a bit. Reintroduce all the crap I used to know more readily. Apr 17 03:35:37 I mean, maybe with a DLP controller that has an embedded framebuffer (like the DLPC2607) there is no requirement for a specific pixel clock? Apr 17 03:35:49 as in, if it's configured to run asynchronously Apr 17 03:37:21 in "free-run" mode I mean Apr 17 03:37:41 in that case you could even use normal gpios if you don't care even a little bit about performance Apr 17 03:37:57 (assuming there are enough gpios that aren't occupied by hdmi) Apr 17 03:39:11 and BT.656 interface mode only uses 9 pins (8 data pins + pixel clock) Apr 17 03:39:47 oh it actually says minimum pixel clock is 1 MHz Apr 17 03:40:03 What is the maximum allowed amperage that the barrel jack can handle on a BBB? Apr 17 03:40:44 set_: uhh, you mean how much amps the BBB will (at most) draw from the barrel jack? Apr 17 03:40:52 Sure. Apr 17 03:41:09 How man amps will the BBB draw from the barrel jack. Apr 17 03:41:09 ? Apr 17 03:41:17 2A max (limited by the pmic), although that's hard to achieve in practice Apr 17 03:41:26 Okay. Apr 17 03:44:07 So, the pmic on the BBB allows for only so many amps to be drawn in via the barrel jack. Okay. Apr 17 03:44:52 Where you two plugged in via tty or usb earlier when things went awkward? Apr 17 03:45:54 IanWizard: you can also write images to the DLPC2607 via i2c (intended for splash screens, if not loaded from spi flash)... but performance may not be adequate ;-) Apr 17 03:51:44 IanWizard: but on a final more serious note... I think if the one of the modes that only require 8 data bits can be used (1 pixel per 3 clocks) then it may be possible (even at 60fps), but it would not be a simple project Apr 17 03:52:52 it would also conflict with hdmi audio, since if both video and hdmi audio are used you have almost no pru pins available Apr 17 03:55:28 Yeah, I agree that it's not worth the hassle. More than that, I don't want to spend as much time on it as it would require. Especially since I really don't need it that badly. The project is thus far just me playing around with some hardware I got a couple years ago and never used at the time. Apr 17 03:56:40 I appreciate the input and guidance, zmatt. I'll undoubtedly be popping up in here more in the future. :) Apr 17 03:57:14 Hopefully providing at least as much help to others, as I end up seeking myself. Apr 17 03:57:58 I've also got a couple of PocketBeagles that I need to get set up for a separate project. Apr 17 03:58:20 For now I'm going to get some sleep though. TY again :) Apr 17 03:59:16 back Apr 17 04:01:50 thanks for taking a good look for me set_ Apr 17 04:02:05 i pour a little drink on the floor for the poor BBB that met its maker today Apr 17 04:02:13 poured* Apr 17 04:02:23 still a little sour on the whole mess Apr 17 04:02:36 I am going to find my green and see if I can diagnose what is going on Apr 17 04:02:48 I have a windows machine for the config tool. So I have been on that Apr 17 04:03:11 There is a mode where that can see if I am getting through to the board but I am still having issues talking withit Apr 17 04:03:27 I am going to try again with it properly powered. Per zmatt's point Apr 17 04:03:42 I was only giving it 5V when it needs 6V+ Apr 17 04:03:48 Oh. Apr 17 04:03:53 perhaps also see if you can confirm with measurement that the level shifter works as it should... considering it looked like it was the cheapest no-brand no-spec level shifters on the internet Apr 17 04:04:03 hopefully that fixes it Apr 17 04:04:04 that too Apr 17 04:04:12 MattB0ne: that might have caused things to not work, it cannot possibly have caused damage to the bbb though Apr 17 04:04:41 yeah BBB taking a dirt nap was probably me. I moved a bit and I heard it disconnect Apr 17 04:04:49 fried right there Apr 17 04:04:58 still odd though Apr 17 04:05:05 do you have a recommended brand for the level shifter Apr 17 04:05:06 oh, unless ground disconnected Apr 17 04:05:24 since I bought a BBB might as well get a better shifter Apr 17 04:05:29 dimension engineering! Apr 17 04:05:40 I found their stuff to be sturdy and expensive. Apr 17 04:05:42 MattB0ne: not really, there are innumerable many of them Apr 17 04:06:00 just something better than what I landed on Apr 17 04:06:01 MattB0ne: but one that actually has a part number and a datasheet would be a good start Apr 17 04:06:02 lol Apr 17 04:06:11 and unidirectional Apr 17 04:06:39 unidirectional is a plus, but keep in mind you'll need (at last) one lane in each direction Apr 17 04:07:09 I am still missing the connection that was made. Is there a photo or a schematic? Apr 17 04:07:12 bidirectional level shifters are useful for more applications though, even if they're a bit gross Apr 17 04:07:33 ok Apr 17 04:07:46 connection between the BBB and jrk? Apr 17 04:07:50 Yes. Apr 17 04:08:00 i was using UART4 Apr 17 04:08:05 Okay. Apr 17 04:08:06 so pins 11 and 13 on P9 Apr 17 04:08:18 I had that going to level shifter Apr 17 04:08:33 Okay. Hmm. Where you in TTY or USB mode from the config. tool? Apr 17 04:08:39 I had the Vin and GND coming from P9 as well Apr 17 04:08:44 to the jrk Apr 17 04:08:48 so that is how it was powered Apr 17 04:08:51 Oh. Apr 17 04:09:01 MattB0ne: when you said "you heard it disconnect", you mean a wire got loose or what? Apr 17 04:09:07 I thought these two things were powered separately. Apr 17 04:09:27 my computer makes a noise when you pull something from the USB Apr 17 04:09:31 ah Apr 17 04:09:50 You are supposed to but I hated lugging the PSU Apr 17 04:10:10 so when I was just working on the QT part I would have the 5V from P9 go to Vin Apr 17 04:10:29 on the Low side I had 3.3V and GND going to the logic shifter Apr 17 04:10:34 and that is pretty much it Apr 17 04:10:43 usb into the BBB for ssh Apr 17 04:10:58 Okay. Apr 17 04:11:09 it was working but I was getting that suspicious tty error permission denied Apr 17 04:11:16 from Qt Creator Apr 17 04:11:18 yeah that *should* be fine (in the sense of not frying things), if the level shifter works okay and all connections are solid Apr 17 04:11:48 MattB0ne: wasn't that determined to be an irrelevant error because it tries to open tty0 or tty1 or something? Apr 17 04:12:19 zmatt: there was the issue of the input that you solved Apr 17 04:12:29 i was running under sudo Apr 17 04:12:30 yeah that was unrelated Apr 17 04:12:34 so I stopped doing that Apr 17 04:12:36 which you shouldn't Apr 17 04:12:37 yeah Apr 17 04:13:13 i dunno will give it another go next week when the board comes in otherwise i look for another controller Apr 17 04:13:23 or buy the updated one Apr 17 04:13:26 they are offering Apr 17 04:13:33 the jrk G2 Apr 17 04:13:50 just make sure you have proper connections, especially when part of the circuit is 5V Apr 17 04:14:14 depending on how the level shifter works, I can imagine if the 3.3V wire to the level shifter got loose you might have trouble Apr 17 04:14:43 Icstation TXS0108E 8 Channel Logic Level Converter Bi-Directional High Speed Full Duplex Shifter 3.3V 5V for Arduino Raspberry Pi Apr 17 04:14:51 i had sort of stubby jumpers too Apr 17 04:14:58 some come longer than others Apr 17 04:15:13 https://www.pololu.com/product/2595 is something I found. Apr 17 04:15:14 having good jumpers with pins that fit properly in the BBB is a good idea Apr 17 04:15:16 this one had long wires which was nice but the connection part was stubbier than I would like Apr 17 04:15:42 the TXS0108E is pretty nice I think Apr 17 04:16:23 (for a bidirectional level shifter) Apr 17 04:17:06 nice buying it Apr 17 04:17:18 I have been reserved w/ my BBB w/ testing specific circuitry. Apr 17 04:17:43 I know that logic-level shifters are a thing to keep around and have handy if one embarks on that journey. Apr 17 04:17:55 5v to 3.3v and so on... Apr 17 04:18:21 MattB0ne: note however that it's entirely possible the level shifter you're using right now works fine... I just said it may be worth *verifying* that it works properly :P Apr 17 04:18:33 Right! Apr 17 04:18:33 right Apr 17 04:18:51 And get a good circuit going for usage. Apr 17 04:19:00 how can you test w/ mulit-meter? Apr 17 04:19:06 Not just the logic-level shifter. Apr 17 04:19:27 Test it. You have three leads. Apr 17 04:19:31 Right? Apr 17 04:20:13 MattB0ne: hook it up with 3.3V on the low side, 5V on the high side and on one the high-side IOs, verify with multimeter that the low-side IOs don't exceed 3.3V Apr 17 04:20:54 I have one but it has a trimpot on it. Apr 17 04:21:24 MattB0ne: actually the TXS0108E might not be as great as I thought Apr 17 04:21:34 I have to get a tiny screwdriver to manipulate the thing. Apr 17 04:21:36 doh Apr 17 04:21:45 why Apr 17 04:22:17 or dunno Apr 17 04:22:35 you know what, I hereby retract any opnion on the matter :P Apr 17 04:23:02 Hey MattB0ne: Do you have a pot on your logic-level shifter? Apr 17 04:24:04 Test the Vin on the level shifter w/ volts on the DMM && GND w/ COM. Apr 17 04:24:28 Then, if you can, turn your trimpot. Apr 17 04:24:35 set_: what on earth are you talking about Apr 17 04:24:35 Aw! Apr 17 04:24:55 no Apr 17 04:24:57 You can see if the voltage is changing on your logic-level shifter. Apr 17 04:25:01 set_: no I didnt Apr 17 04:25:11 Oh. Apr 17 04:25:12 Okay. Apr 17 04:25:13 didn't think I needed one Apr 17 04:25:23 you know I'm not even going to try to understand what you're saying Apr 17 04:25:27 afk again Apr 17 04:25:30 It is for fine tuning if you have higher voltage too. Apr 17 04:26:38 Some of the logic-level shifters have a pot on them. This way, you can turn to exactly which logic you need from a higher voltage. Apr 17 04:30:23 https://www.dimensionengineering.com/products/de-sw033 is another one I found online while looking earlier in the year. Apr 17 04:30:26 Enjoy. Apr 17 04:32:10 1A continuous output! Apr 17 05:04:30 So, I guess one can use, in theory, a LDO reg. w/ a DC-DC converter to better equip the BBB w/ not being zapped into orbit. Apr 17 05:19:21 Just for reference, I have this "enormous" book on circuits. If you want a good circuit, listen to the book. Books! Apr 17 06:06:38 set_: that's usually a bad idea. you should use the supply of the hardware you're interfacing with, not use a random other supply and generate a lower voltage from it Apr 17 06:07:13 oh you didn't actually link to a level shifter Apr 17 06:07:19 but to a voltage regulator Apr 17 06:07:35 so it was just something random and irrelevant, ok Apr 17 12:25:44 I mostly analyse the power supply based on what is being done. An LDO is not necessarily the correct solution. Just like you can solve all problems with C but it isn't necessarily the correct programming language for what you are doing. Apr 17 15:25:40 . Apr 17 15:28:36 Per zmat suggestion, I have tried to check the real-time responsiveness on Beaglebone. Basically, I see the following: Apr 17 15:29:13 nanosleep resolution is very high. Better than 80us Apr 17 15:29:39 But the nanosleep() error is up to 6ms' Apr 17 15:30:44 I couldn't make pthread_setschedparam() working. It returnsno error, but does not set the params Apr 17 19:27:27 dreamhiker71 zmatt not zmat Apr 17 23:28:35 set_ something to laugh about. https://embeddedemulation.blogspot.com/2020/04/april-powers.html Apr 17 23:35:43 GenTooMan: YOu got it down, man! Apr 17 23:36:30 Except for what was mentioned as incorrect, that thing looks nice. Apr 17 23:36:50 I have no clue as to how you got it but you got it. Apr 17 23:38:51 I use KiCAD to do the schematic PCB and 3d views. I had to add the LCD and yellow connector (I choose yellow so it would be easy to find). Apr 17 23:39:43 Nice! Apr 17 23:39:56 Well I made models for them then added them to the schematic and PCB. The LCD is actually a pad with a 3d model attached. Apr 17 23:40:25 Oh. Apr 17 23:40:47 KiCAD does more than I figured. Apr 17 23:41:18 you can see it in the first image by looking at the letters J2 (the pad is right below it on the edge of the board). Apr 17 23:42:37 it's definitely not as convenient as Altium. It does look to be able to get the job done. So now I can say "it works". LOL Apr 17 23:43:48 anyhow that's a BBB cape Apr 17 23:45:58 It works! Apr 17 23:46:01 Nice. Apr 17 23:46:20 Are you going to produce it into production? Apr 17 23:46:34 Or is it a prototype? Apr 17 23:47:33 preprototype Apr 17 23:50:19 Nice...I just read over your comments on the page. I saw the connector, too. Apr 17 23:51:14 It is not easy. That is for sure. Now, when other people are making things (boards, Capes, etc), using is way easier compared to actual engineering of ideas to fruition. Apr 17 23:52:12 This is a learned trait I am trying to break w/ knowledge. It is not easy. Apr 17 23:53:58 Now, w/ the last two or three sentences, I would like to state this idea. I can read just about anything, comprehend, and act but it takes me a long time to gain my aim in the subject matter. Apr 17 23:54:36 For instance, things come up. I put together a couple of circuits and they did not work. Apr 17 23:55:47 Why? I am not sure. Could I figure it out. Yes. How long would it take me? Sometimes, I just move on. So, if I am going to build as you have, I would take time and not just cram and jam it out. No offense. Apr 17 23:56:11 Just something to think about for now. Apr 18 00:11:32 Hey, does anyone need for people to make more content for the BBB and other related boards? Apr 18 00:11:59 persistence to overcome is what is rewarded I think are the correct words. Apr 18 00:12:38 For example, I noticed on other pages online, people are putting in many articles related to specific hardware/software. Apr 18 00:13:20 I mean... Apr 18 00:14:03 Forget it. I must not understand yet. Apr 18 00:14:34 dreamhiker71: actually the modern linux calls for getting/setting the scheduling policy of a process/thread are sched_getattr and sched_setattr Apr 18 00:15:27 dreamhiker71: see also "man sched" for details about the various calls available to interact with the linux scheduler, the scheduling policies available, and how their use is constrained by privileges and/or resource limits Apr 18 00:18:19 the short summary on the latter is: if a process has the CAP_SYS_NICE capability (e.g. if it's root) then it can configure any scheduling policy it wants, otherwise you can only configure a realtime priority up to the RLIMIT_RTPRIO soft limit Apr 18 00:20:08 note that if your program runs as systemd service (typical for things that need to be started automatically during boot) then you can give it realtime priority and/or the ability to set realtime priority on itself (nonzero RLIMIT_RTPRIO) by setting appropriate directives in the service file Apr 18 00:36:51 for small tests (e.g. latency tests) this should allow the debian user to use rt scheduling: https://pastebin.com/yiaFJFn6 Apr 18 00:40:09 zmatt: if I wanted to send two bytes via pyserial Apr 18 00:40:58 I do so with ser,write(b'\xAA',b'\xaa') Apr 18 00:41:03 ser.* Apr 18 00:41:09 b'\xAA\xAA' Apr 18 00:41:46 https://pythonhosted.org/pyserial/pyserial_api.html#serial.Serial.write Apr 18 00:42:39 you can also create a bytes object from an array of integers (in range 0-255) if you find that more convenient, e.g. bytes([ 0xAA, 0xAA ]) Apr 18 00:43:31 the struct library (https://docs.python.org/3/library/struct.html) may also be of interest, especially if you need to encode/decode 16-bit integers Apr 18 00:44:07 ok Apr 18 00:47:31 i am on my back up board Apr 18 00:47:37 the bb green Apr 18 00:47:44 and I have a real power supply hooked up Apr 18 00:48:00 and I am able to send commands to set the speed Apr 18 00:48:00 i am still having problems reading Apr 18 00:48:18 that spitting back what you sent seems to be normal behavior Apr 18 00:48:21 that is sort of shitty Apr 18 00:48:40 you figure there would be acknowledged or some sort of code Apr 18 00:49:21 it's not normal. their example code in their documentation would break if it received echoes of commands back Apr 18 00:50:17 I cannot get reads off this thing Apr 18 00:50:25 but I can write to it Apr 18 00:50:27 e.g. they send 0xA5 (get feedback) or 0xA3 (get target) and expect to read 2 bytes in response Apr 18 00:50:39 i try that and via pyserial Apr 18 00:50:50 can you share your transcript? Apr 18 00:50:58 and all I get is b'\xA3 Apr 18 00:51:01 ok Apr 18 00:51:45 have you tried their example code? (pages 49-50 of https://www.pololu.com/docs/pdf/0J38/jrk_motor_controller.pdf ) Apr 18 00:51:52 ok I lied Apr 18 00:52:07 lol Apr 18 00:52:29 https://pastebin.com/JFkwCNWN Apr 18 00:53:00 I did use there sample code but I think something is wrong with how I process the response Apr 18 00:53:04 maybe not parsing correctly Apr 18 00:53:09 this still makes no sense Apr 18 00:54:09 your right I thought it was giving me something different but it was just an old command Apr 18 00:54:25 maybe? Apr 18 00:54:28 i dunno Apr 18 00:54:38 all I can confirm is that my writes are succesful Apr 18 00:54:48 and match what I would get via the windows tool Apr 18 00:55:18 so, just to confirm: what happens if you disconnect the wire between the jrk's "TX" pin and the level shifter? Apr 18 00:55:50 and run a command? Apr 18 00:56:12 yeah just do some testing again Apr 18 00:56:37 does it still show your commands being echoed back or does it now read nothing anymore (just b'') Apr 18 00:56:40 ? Apr 18 00:56:56 https://pastebin.com/UqN3zGGn Apr 18 00:57:22 so I unplugged TX coming from the jrk so I could still send a command Apr 18 00:57:26 okay the problem has nothing to do with the jrk. that narrows it down to the level shifter or software weirdness Apr 18 00:57:46 so this is pyserial Apr 18 00:57:50 do you think that could be it Apr 18 00:57:57 disconnect wire from level shifter going to the uart rxd pin of the beaglebone Apr 18 00:58:00 and test again Apr 18 00:58:42 you should still be able to send commands as before. again, does it still echo? Apr 18 00:58:56 https://pastebin.com/j5ypJ28y Apr 18 00:59:24 oh crap you were doing ser.read(2) instead of ser.read(256), I overlooked that Apr 18 00:59:30 that makes the tests useless. redo again Apr 18 00:59:49 you're just reading old data that's still in the input buffer Apr 18 00:59:49 ok Apr 18 00:59:59 so we don't know how old that data is Apr 18 01:01:37 https://pastebin.com/c7hRuZ0d Apr 18 01:01:42 the two cases Apr 18 01:01:50 with me manually stopping the motor Apr 18 01:02:23 looks like that first command still had a ton of garbage in input buffer Apr 18 01:02:46 (because you hadn't done a ser.read(256) previously) Apr 18 01:03:11 that second write/read combo is with no wire to the beaglebone rxd right? Apr 18 01:04:26 you'll need to redo the first (i.e. the one where the beaglebone rxd is hooked up, but the other side of the level shifter isn't hooked up to the jrk TX) Apr 18 01:07:31 to avoid getting bitten by old data still sitting in the input buffer, you could also do ser.reset_input_buffer() right before a write. while testing you can combine that and the write into a one-line by separating them by a semicolon, e.g.: ser.reset_input_buffer(); ser.write(b'\xe1\x20') Apr 18 01:07:47 *into a one-liner Apr 18 01:08:00 ok Apr 18 01:10:52 you're still retesting the first case? Apr 18 01:11:42 https://pastebin.com/MUb1Lp9y Apr 18 01:11:51 that is the first case Apr 18 01:12:21 huh, does semicolon not work? maybe I'm confused Apr 18 01:12:38 i dunno i just split it Apr 18 01:13:01 https://pastebin.com/BegT1KnM Apr 18 01:13:03 2nd case Apr 18 01:13:16 oh you just had a typo in that line Apr 18 01:13:23 \xel instead of \xe1 Apr 18 01:13:38 doh Apr 18 01:13:53 so when I break Tx back Apr 18 01:13:55 I get nothing Apr 18 01:13:58 okay so disconnecting the TX from jrk makes the weird echoing go away Apr 18 01:14:14 just as sanity check... if you reattach TX the echo'ing is back? Apr 18 01:14:22 let me check Apr 18 01:16:01 let me check https://pastebin.com/76PRLLyr Apr 18 01:16:17 so echo is gone but I cannot read the level back though Apr 18 01:16:43 you sure you didn't mess anything up while wiring things up? Apr 18 01:18:12 there's another option to test serial communcation: if you configure the jrk into "USB dual port" then the BBB should be able to communicate via the jrk with your host pc, see image on page 33 of https://www.pololu.com/docs/pdf/0J38/jrk_motor_controller.pdf Apr 18 01:18:41 then you can just send something on one end and see what you get at the other end Apr 18 01:19:44 i tried that Apr 18 01:19:55 motor responds but the software doesnt Apr 18 01:20:04 ? Apr 18 01:20:08 pretty confident the wiring is correct but I will check again Apr 18 01:20:24 those write commands I am sending are for the motor to go slow Apr 18 01:20:57 so I see the motor moving but I do not see gui update Apr 18 01:21:03 well you're getting different results now than you were earlier, even though you're supposedly in exactly the same configuration as earlier Apr 18 01:21:04 but let me check over everything again Apr 18 01:21:10 ehh what? Apr 18 01:21:13 gui update? Apr 18 01:21:16 what are you talking about? Apr 18 01:21:33 the windows config tools Apr 18 01:21:42 does not update when I send commands Apr 18 01:21:50 why would you expect it to? Apr 18 01:22:06 that mode you are talking about Apr 18 01:22:14 that's not what I'm talking about Apr 18 01:22:35 the mode I'm talking about isn't about sending motor control commands by anything at all Apr 18 01:23:50 let me try that now Apr 18 01:26:38 regardless though, you're now supposedly back in the same configuration as in the first transcript you shared today, yet you're getting different results Apr 18 01:26:48 after removing wires and putting them back Apr 18 01:30:58 doubled check and I put the last Tx wire in the wrong spot Apr 18 01:31:02 here is what I get Apr 18 01:31:07 which is sort of echo Apr 18 01:31:08 https://pastebin.com/mzpyhEpU Apr 18 01:32:05 basically it is echoing again Apr 18 01:32:16 if I unplug it goes away Apr 18 01:32:50 i am going to switch to that other mode USB Dual port Apr 18 01:33:19 yeah that's an echo Apr 18 01:33:22 \x20 is a space Apr 18 01:34:02 what do you mean Apr 18 01:34:19 just type b'\x20' in python Apr 18 01:35:02 a bytes object is treated kind of like a string Apr 18 01:35:09 I though it would be 2*16+ 0*1 Apr 18 01:35:17 so 32 Apr 18 01:35:26 yep, and ascii 32 is a space Apr 18 01:35:34 hence b'\x20' is b' ' Apr 18 01:35:57 bytes objects are not a convenient way to view these responses ;P Apr 18 01:36:07 but they'll do for debugging Apr 18 01:36:23 lol Apr 18 01:36:33 so in the example max speed is 0x7f Apr 18 01:36:42 i can switch to that and see if it behaves different Apr 18 01:37:48 well in USB dual the motor is not seeing anything Apr 18 01:37:52 but I get the echo back Apr 18 01:38:08 could it be how my UART is configured maybe Apr 18 01:38:43 the motor isn't supposed to see anything, *you* are supposed to see if you receive stuff on the other side (if you open a terminal on what the jrk calls the "TTL Port") Apr 18 01:39:16 and conversely if you type there it should be received by the beaglebone Apr 18 01:40:07 oooohhhhh so it has nothing to do with the tool Apr 18 01:40:15 TTL or serial Apr 18 01:41:02 i see i see Apr 18 01:41:23 i gotta find my TTL cord Apr 18 01:41:23 so it is like a debug mode Apr 18 01:45:39 my problem will be actually hooking up the TTL Apr 18 01:45:51 i only have one Tx and Rx pin on the jrk Apr 18 01:46:12 which I have the beagle hooked up too Apr 18 01:50:16 tried unplugging the usb which I had in there the whole time Apr 18 01:50:20 still getting the echo Apr 18 01:58:54 MattB0ne: "TTL cord" what? no Apr 18 01:59:39 the whole point of it is that it would allow you to confirm that serial communication works in the hardware setup you have Apr 18 01:59:48 changing the hardware setup would make that test useless Apr 18 02:00:20 anyway, I'm kinda done for today Apr 18 02:00:37 afk Apr 18 02:03:08 back Apr 18 02:17:41 effw Apr 18 02:55:46 Hello! Apr 18 02:55:46 Apr 18 02:55:55 ? Apr 18 02:55:58 hmm Apr 18 02:56:07 I think my computer has gone goofy. Apr 18 02:56:27 I guess it is reboot time. Apr 18 02:56:59 its only been 3 months for mine **** ENDING LOGGING AT Sat Apr 18 02:59:57 2020