**** BEGIN LOGGING AT Tue May 05 02:59:57 2020 May 05 03:00:08 ga night May 05 05:29:27 set_: it's possible it's an issue with my code, I never tested it after all May 05 05:30:40 set_: I've updated https://pastebin.com/RhKjeKC1 ... see if that helps May 05 07:09:37 @zmatt: Thank you. I will test it soon. May 05 08:40:46 anybody played with modules using remoteproc / rpmgs_send/recieve() ? May 05 12:08:12 m May 05 14:54:12 Hi Guys, Greetings!. any API is available for WiFi connectivity via user Interface? I want to develop a webpage to listout all the available wifi and connect to the available wifi in beaglebone black wireless. May 05 15:16:50 Hi Support, I need help with RMA on a board delivered apparently faulty. May 05 16:04:47 hi everyone May 05 16:05:05 how can i delete the old OS on BBB May 05 17:17:41 I'm having a bit of a hard time understanding: what's the difference in one of the BBB "timerx" outputs and a pwm output? May 05 17:17:53 or are the timers inputs on the bbb? May 05 17:54:21 rob_w: Yes, I've played. It was easier than expected. But still not as easy as UIO or the provided pru_rpmsg module. May 05 17:56:08 dinuxbg__, nice , can i have a little chat aobut that ? May 05 17:56:25 basicly i am on the pru_rpmsg implementation May 05 17:57:57 i ve increase the static rpmsg buffers yet i rather like them different May 05 18:02:52 rob_w: my understanding is that rpmsg buffers are meant for small message passing, and not large transfers. What changes are you planning? May 05 18:03:42 my scenrio holds some adc behind the pru , where it therefore samples and provides the values "with some speed" May 05 18:04:02 so i wrapped iio adc driver around the pru_rpmsg examples May 05 18:05:08 oh ..i am a am335x with PRU co-cpu May 05 18:06:14 so i love to get a small TX buffer to the pru and multiple RX buffers coming from the pru, so i can alternate and save speed May 05 18:07:37 so its is message passing but small in TX and bigger in RX and splitted ? May 05 18:09:08 but i guess i can emulate a 2xRx to some degree using the existing single in 2 blocks May 05 18:09:33 You should be able to reach an excess of 1MB/s transfer speed using the original pru_rpmsg module. TX and RX max buffer sizes are the same. May 05 18:12:56 how does this number come up ? May 05 18:14:39 I did bandwidth experiments, and if my memory serves right, I could reach 1MB/s. May 05 18:15:02 But at the end I decided to transfer large chunks of data via shared memory, and leave RPMSG only for control messages. May 05 18:15:25 so rpmsg doesnt use the shared areas May 05 18:16:01 so i ve to use uio stuff too .. dang it May 05 18:17:24 RPMSG uses shared areas, but constrains you to a "call -> response" synchronous flow. And shared memory buffers are memcpy-ed. I think this is the major reason to be slow. May 05 18:17:27 While there are a few people here (and I don't mean to pester) but what's the difference in the PWM outputs and the timer outputs on the bbb? I was thinking I could use the timers as inputs for a square wave May 05 18:19:00 biggi_45, i think the timers can be used like that May 05 18:19:33 rob_w: RPMSG and shared memory are not exclusive, so you can use both. I did that: https://github.com/dinuxbg/beaglemic/tree/master/driver . May 05 18:20:30 so the timers can be inputs and outputs? May 05 18:20:43 i didn't find any documentation stating one way or the other is why i was confused May 05 18:20:53 rob_w: Audio data is written by PRU in shared DDR-SDRAM buffer, allocated by ALSA. RPMSG is used to signal for available data in buffers. May 05 18:21:16 dinuxbg__, oh my god May 05 18:23:43 rob_w: sorry for the shameless plug. I just wanted to give an example. May 05 18:25:02 well nothing to be sorry , i think you just saved my day .. or saved my many days May 05 18:25:58 apart that yours is into hte sound subsystem ( which is already very cool and a way i thought of too) and mine is into iio bus, you have both rpmsg and shared io , right ? May 05 18:27:44 oh and u use both pru , very nice May 05 18:28:32 rob_w: Yes, driver uses both rpmsg (for control messages) and shared memory (for data). May 05 18:29:06 rob_w: I think it should be possible to write an IIO driver that uses RPMSG to communicate with PRU. May 05 18:29:15 i ve that May 05 18:29:52 only for pru1, pru0 is for interboard comm to other beagels of ours May 05 18:30:20 so i only need to "steel" your shared mem part , but my honest kudos to that project there May 05 18:30:32 need a job ? ;-) May 05 18:31:26 rob_w: I already have a 9-5 :) Actually pru0 is handling big banging, and pru1 is doing rpmsg. May 05 18:33:02 so pru1 still uses shared mem to arm right ? May 05 18:40:47 rob_w: Yes. Check how handle_peer_interrupt writes into stream.dma_addr in pru/main1.c May 05 18:43:16 and stream.dma_addr comes from arm pru_prepare.buffer_addr May 05 18:44:43 but that looks like alsa stuff , substream May 05 18:45:36 I'm transferring audio data, so it is natural to aligh with ALSA. May 05 18:46:09 what i miss then is the actaully memory which arm "finds" and tells the pru to write too May 05 18:46:44 i was exspection some memory addresses or such, no ? May 05 18:48:39 rob_w: ALSA host driver allocated memory. Then it passes the base physical address via RPMSG - grep for BEAGLEMIC_PRUCMD_PREPARE. May 05 18:49:56 so its not like there is a mem area specific to pru<->arm but you actully get a mem from somewhere, hence alsa, and show this to the pru to copy to .. maybe ? May 05 18:51:10 yes. all memory in the system is managed by linux kernel, so we cannot bypass it when getting memory. Even carveouts must be declared in DTS. May 05 18:51:14 pretty cool .. i think i can rip yours apart and frankenstein me my stuff May 05 18:51:34 what is a carveouts ? May 05 18:56:17 rob_w: carveout memory is physically contiguous and set aside at boot. Kernel will not allocate pages from it for userspace and kernel heap. May 05 18:56:48 i see , but u didnt use that for that previous example ? May 05 19:01:06 No. ALSA allocates physically contiguous memory from the DMA or CMA pool. May 05 19:02:16 so that really helped me to understand the "any" dma can be used thx alot dinuxbg__ , consider me to owe u some .. May 05 19:06:23 but also i love that you went the alsa way , i was considering that too instead of iio, peaking to gst and its pipes and streams .. May 05 20:43:50 @zmatt: When you ever may make time for this source again, will you please post in this chat forum? This may not be complicated source to you but I have not come across issues like this before in Python3 source. May 05 20:57:48 Okay. Do not fret. I found some more info. on w3schools.com. I will review the __init__ function more before passing out over this info. May 05 20:58:41 set_: so my fix didn't help? May 05 20:59:02 share the full error you're getting May 05 21:02:44 It is not your source. May 05 21:03:07 It runs w/out error. I cannot set up another file to use the wrapper yet. May 05 21:03:21 I can show you that error. May 05 21:03:45 set_: well the last error you shared looked like an issue with my class, hence I made a change to hopefully fix it (and told you about this) May 05 21:04:42 Yes. I saw that fix. I am trying w/ that fix so far. And no, that error was not from your source. May 05 21:04:49 what arguments are you passing to the constructor? May 05 21:04:59 It was from the use of the wrapper in my file. May 05 21:05:33 https://pastebin.com/SwsqTmLA is as far as I can get but it gives an error. May 05 21:06:33 that looks okay so any error you get from that is probably a bug in my code May 05 21:06:58 (reminder: I have no way of testing my code, so bugs are entirely possible) May 05 21:07:31 The error is in this paste: https://pastebin.com/zWVLRaG0 May 05 21:07:34 actually no it doesn't look okay May 05 21:07:47 the class is named Pca9685, not PCA9685 May 05 21:07:51 I know. May 05 21:07:54 I changed it. May 05 21:07:59 also you shouldn't import *, you should import just Pca9685 May 05 21:08:01 I am calling it correctly. May 05 21:08:05 Okay. May 05 21:08:35 hmm May 05 21:08:39 Same error. May 05 21:09:41 oh yeah that does look wrong, both in your code and in mine.. I must have miscounted the number of zeros. lemme check the datasheet May 05 21:10:13 Oh. I thought there were seven of them. May 05 21:12:08 you can simply pass 0 as address anyway since I already bitwise OR it with 0b1000000 (previously 0b10000000 but I've just fixed that) May 05 21:12:27 Also, on the wrapper, get_pwm shows assert output in range(0, 16). May 05 21:12:38 That is 17 servos, right? May 05 21:12:41 except that's also wrong based on the schematic of the ServoCape May 05 21:12:50 Okay. May 05 21:12:59 So, it should be 0, 14? May 05 21:13:07 Sorry. 0, 15. May 05 21:13:16 no with "that's also wrong" I was talking about the address May 05 21:13:25 range(0, 16) means the maximum is 15 May 05 21:13:32 Oh. May 05 21:13:55 the servocape has pull-ups on all six address lines, so the address you should pass is 0b111111 May 05 21:14:12 Oh. B/c they are already High? May 05 21:14:37 though I kind of hope that's wrong, since I'm pretty sure that's an invalid i2c address May 05 21:15:17 yeah, so I don't know if the kernel is going to accept that... it really shouldn't, but maybe it does May 05 21:15:25 I can try it. May 05 21:15:32 yet another great example of the servocape just being badly designed May 05 21:16:52 Same error. May 05 21:17:26 then honestly the servocape may be unusable without hardware patch to fix it May 05 21:17:33 Oh. May 05 21:18:05 Okay. May 05 21:18:05 Dang. May 05 21:18:09 you did change 0b10000000 to 0b1000000 in my source code? May 05 21:18:17 Oh. No. May 05 21:18:39 (or just re-download my pastebin, since I've fixed that too) May 05 21:18:55 That did it. May 05 21:19:08 Whoo Ya! May 05 21:20:03 So, me using 0b1000000 in the wrapper and using 0b1111111 worked. May 05 21:20:31 The 0b1111111 worked on the python3 file that I will be running. May 05 21:21:37 WHere did you look on the PCA9685 datasheet to figure that out? May 05 21:22:29 well 0b10000000 definitely couldn't be right since it's a 8-bit value while i2c addresses are only 7-bit May 05 21:22:48 Aw. So, i2c addresses are only 7-bit. Okay. May 05 21:23:03 Is that in general or just for this chip? May 05 21:23:46 read the protocol May 05 21:23:51 Okay. May 05 21:23:56 I will one day. May 05 21:23:57 standard I2C has 8 bits clocked it for addressing May 05 21:24:01 1 bit is R/W May 05 21:24:03 so :D May 05 21:24:18 Oh. May 05 21:24:21 there are non 8 bit I2C for extended addressing May 05 21:24:23 in general i2c addresses are 7-bit (or rarely 10-bit). for the pca9685 specifically, figure 4 in section 7.1.1 in the datasheet shows that the address consists of the six address inputs and bit 6 is equal to 1 May 05 21:24:52 i.e. 0b1000000 | addr where addr is the 6-bit value representing the address inputs of the chip May 05 21:25:05 Okay. May 05 21:25:31 a quick glance at the servocape schematic shows all six of those are 1, so you can pass 0b111111 to my class, and its final i2c address is 0b1111111 ... which is invalid, but the kernel accepts it anyway May 05 21:25:53 thought it is valid May 05 21:25:56 just has a special meaning May 05 21:26:08 ds2: no, reserved for future use May 05 21:26:30 Right, if I was going to put add-ons. May 05 21:26:41 oops, i was thinking of 0 May 05 21:27:10 I will go to google groups and fix my post. May 05 21:27:27 I was making a post about this idea but I had the incorrect ideas down pat. May 05 21:29:50 ds2: 0000xxx and 1111xxx are reserved by the protocol, valid addresses are 0b0001000 - 0b1110111 May 05 21:32:03 zmatt: 0000000 is valid as in a broadcast address May 05 21:32:16 not valid as in a device should make it its only address May 05 21:33:14 ds2: right, I didn't mean to imply you can't perform any i2c transactions with addresses in that range, but they're not valid device addresses May 05 21:34:17 Oh cool, zmatt is here. Difference in timers and pwm on BBB? I'm assuming its just that timers are input or output and pwm is strictly output? May 05 21:34:23 I have wondered, how common are write only I2C devices May 05 21:35:05 biggi_45: timer, eCAP, and eHRPWM can all be used both as timers and as pwm outputs, but they're different types of peripherals and have different features May 05 21:35:07 For LEDs, probably common. May 05 21:36:22 biggi_45: timers and eCAP can also be used in "capture" mode (i.e. timestamping signal edges on their pin in input mode) May 05 21:36:57 set_: got a P/N for a I2C LED?? May 05 21:37:26 Gotcha. I'm working on this encoder finally and didn't know if I should use just a timer or if I needed to use an encoder eQEP setup May 05 21:37:26 Yes. May 05 21:37:31 PCA9685 May 05 21:37:42 biggi_45: eCAP and eHRPWM run at a higher clock frequency (100 MHz instead of 24 MHz) May 05 21:37:46 If I run the encoders single eneded, I can run them on the eqep I believe May 05 21:37:50 Oh. You mean, the i2c LED. May 05 21:37:53 Not the chip? May 05 21:38:07 Not off the top of my head. May 05 21:38:28 They're quadrature encoders, this is why you had mentioned the differential to single ended chip, but I think I can just run them single ended straight out of the encoder @zma May 05 21:38:38 biggi_45: for encoders you'd typically use a quadrature-modulated signal and use eQEP to decode them, although eQEP can also be configured to do simple pulse-counting May 05 21:38:44 ds2: Let me look. I was rummanging the other day and found some ideas. May 05 21:39:20 Roger that. eQEP "timers" won't interfere with me doing PWM out correct? I know some boards say you can either do one or the other on an internal timer May 05 21:39:54 biggi_45: you can, the main benefit of using them in differential mode with a differential receiver was because it gives better noise-immunity, which can be useful when having an encoder near noisy stuff (electrically speaking) like a motor May 05 21:40:38 eQEP has nothing to do with PWM May 05 21:41:24 Right; so this is my encoder (built into motor) https://www.omc-stepperonline.com/download/17HS19-1684D-PG5-E1000.pdf I would put EA+ and EA- into one set of inputs on the chip and the EB+/EB- into the other then run the two outputs to the BBB eQEP? May 05 21:41:42 Chip I'm using for reference (the one you recommended): http://www.ti.com/lit/ds/symlink/am26lv32.pdf?ts=1588618460471 May 05 21:42:14 ds2: I found one addressable by UART so far. I am not sure if I made it up or I came across one. May 05 21:42:39 It may have been the UART addressable LED I was thinking of. Let me keep looking. May 05 21:43:15 (just don't want to blow anything XD) May 05 21:44:04 https://www.sparkfun.com/products/8579 is an inexpensive version. Yikes. $12.00. May 05 21:45:39 That led is addressable by i2c. So, it could in theory be used w/ the PCA9685. May 05 21:46:33 biggi_45: I doubt I "recommended" that chip, at most it's probably the first random search result I found as an example May 05 21:47:10 Probably: it's just a differential to single that can take 5V on the input and outputs the 3.3V for the BBB May 05 21:47:16 ah I see.. broadcast only to specific registers May 05 21:47:43 biggi_45: ah right, you need 5V to 3.3V level shifting as well, so a direct connection to the BBB wouldn't be possible anyway May 05 21:48:11 and if you're going to need some electronics anyway, might as well use a proper differential receiver May 05 21:48:14 Yup; it's wonky. I could run them through a level shifter if needed instead of this, but if this works I'll stick with it. May 05 21:48:45 Just wanted to make sure that the EA+/EA- go to one input EB+/EB- to the 2nd and then the 2 outputs are what I sent to eQEP. May 05 21:48:53 yep May 05 21:49:11 Awesome. I've dealt a tad with steppers before, but not a lot with the encoders. May 05 21:49:30 Jack of all traits; master of none type person. May 05 21:50:47 Truly appreciate it; I'm going to hook it up tomorrow and see if I can't get it counting pulses May 05 21:51:12 The 3.3V logic has been a pain tho, these drivers use 5V optos for all their logic, so having to set everything up with BJTs haha May 05 21:52:32 optos are current-driven anyway, hence not a great match for low-voltage low-current CMOS outputs May 05 22:03:56 Yup; BJTs to the rescue! May 05 22:06:33 yeah either FETs or BJTs with a large resistor on the base... we drive some relays with BJTs and I noticed even the current from the internal pull-ups (100KΩ ±50%) was sufficient May 05 22:33:00 I'm just using 4.7k with a 2n3904 May 05 22:33:46 4.7k at the base then a 4.7k going from the collector to 5V and that goes to the opto input (5v logic) then grounding the emitter May 05 22:38:39 if that works then the opto input doesn't actually require 5v but also works on 3.3v (although it may require more current than the BBB is okay with) May 05 22:40:34 since if the voltage drop across the opto input were large enough (say > 3.3V) then your BJT would no longer be forward-biased from base to emitter May 05 22:41:55 it works, but the opto has it's own power supply line May 05 22:41:59 i'm just having to pull it low May 05 22:42:25 that's not what you just described May 05 22:42:37 oh it was May 05 22:42:44 ehh May 05 22:42:53 Page 6 https://www.omc-stepperonline.com/download/DM332T.pdf May 05 22:42:59 not really, but I see now that you meant to say that, but described it in a really confused way May 05 22:43:12 "4.7k going from the collector to 5V" is plain wrong May 05 22:43:26 Lemme draw it May 05 22:43:47 you meant 5V -> opto input -> 4.7k -> collector of BJT presumably May 05 22:44:13 brb changing computers May 05 22:44:14 no May 05 22:45:35 back; had to swap to desktop May 05 22:47:52 zmatt https://imgur.com/sPcGnUv May 05 22:48:58 that's a weird topology for driving an opto input May 05 22:49:21 it works? :shrug: May 05 22:49:24 this is something you'd more typically see for something that's voltage-driven May 05 22:49:34 yeah I guess it might May 05 22:49:43 It's what I had here to work with so I made it work haha May 05 22:49:56 All I really need to do is to ground the direction/enable pins May 05 22:50:20 So it's at the 5V when i'm not driving the bjt, then when I drive the BJT, the signal gets grounded May 05 22:51:06 yep, like I said it's a topology I'd expect for something that's voltage driven (e.g. 5V CMOS input) May 05 22:51:19 Is the driver not voltage driven? May 05 22:51:23 Am I missing something there? May 05 22:51:32 well you said it was opto-isolated May 05 22:51:44 an opto-input is current-driven, not voltage-driven May 05 22:51:57 since you're basically driving a LED May 05 22:52:10 Just making sure we're on the same page, did you look at the driver diagram? May 05 22:52:14 yes I did May 05 22:52:18 Cause I could be plainly stupid :) May 05 22:52:32 I just saw this as a way to make it work haha May 05 22:52:38 With what I had in my toolbox May 05 22:52:47 hey if it works it works :) May 05 22:52:52 And it's similar to what they had on the diagram May 05 22:53:08 I'm not even sure I need the 4.7k going to 5V tbh May 05 22:53:15 that may just be pulling it up May 05 22:53:55 you have specs for the input you're trying to drive May 05 22:53:55 ? May 05 22:54:15 I'm driving the inputs on that driver May 05 22:54:22 So what you see on page 5-6 is what i have May 05 22:55:12 Page 3 might help? May 05 22:56:24 right, figure 2 makes more sense May 05 22:56:46 so yeah the pull-up to 5V is nonsense May 05 22:57:12 So since you haven't driven me wrong yet, how would you do it? Just out of curiosity? May 05 22:57:25 My thought was just get a "switch" of sorts (the bjt here) to ground the pins May 05 22:58:00 figure 2 shows exactly what I described earlier: 5V -> opto-input -> resistor -> collector of BJT May 05 22:58:23 (the 5V -> opto-input part being shared across the three inputs) May 05 22:58:28 Right; that's what I have except the 5v pullup May 05 22:58:37 Which I don't need May 05 22:58:47 correct May 05 22:58:54 I think that's where we got crossed at :) May 05 22:59:02 it just causes a reduction in current through the opto May 05 22:59:14 Yup; I see that. May 05 22:59:27 (by creating a parallel circuit) May 05 22:59:34 I'll pop that out tomorrow and see if it still works May 05 22:59:51 It should May 05 23:00:21 Been too long since I did this stuff May 05 23:01:49 Did quite a bit of this at my first job 8 years ago. Then my job after that and this one has mostly jsut been embedded programming and some hardware interfacing, but most of the support drivers are already on board/etc May 05 23:07:41 Thank you again; makes me feel a little beter about what I'm doing haha May 05 23:07:48 :) May 05 23:08:21 Always nice to have someone to make sure I'm not totally off my rocker XD May 05 23:08:36 Especially when working from home and told "Don't spend any money at all!" May 05 23:13:10 Right. Spend frugal-a-little-ly. May 05 23:13:25 ServoCapes on the house! May 05 23:13:30 Just kiddin'. May 05 23:15:29 I am sorry but i have gotten the source to work but I have not gotten it to work w/ the BBBW so far. May 05 23:15:36 I cannot make that darn servo move any. May 05 23:15:57 is there some peculiar part I am missing? May 05 23:16:20 Do I need each function in the wrapper to make my source work? May 05 23:17:40 This is basically what I am coming across. I am really just trying to call my functions from the class in the wrapper w/out making them new functions in the new python3 file. May 05 23:17:55 Is that possible? May 05 23:28:20 Okay, Okay. On the wrapper, on the write_regs function under class Pca9685, what are values? What does values mean? Is that in the datasheet or should I know about the value already? May 05 23:29:31 or... May 05 23:29:47 Do I use hexadecimal characters for these so-called values? May 05 23:32:17 Should I have a "data" argument in the functions that you listed under the wrapper? May 05 23:32:23 <<< shooting in the dark May 05 23:35:58 https://pastebin.com/bXWhdrzA is the error I am coming across. It seems that smbus2 is complaining about the length of data as an int b/c int has no length. May 05 23:40:45 Okay. I got past that section. May 06 00:27:25 https://pastebin.com/rip5J4TT is my source to make this wrapper work. Now, it does work but it does not in the same respect. May 06 00:27:36 The servo must hate me. May 06 00:28:23 Send ideas! May 06 00:45:31 I have a couple more code tricks and then it is blah for me. BBL! May 06 01:04:34 Is there a way to find the register address for the SP1 onboard the ServoCape? May 06 01:30:21 question May 06 01:30:39 an RS 232 port May 06 01:30:48 that is a regular serial port May 06 01:30:57 like I could UART communicate with it May 06 01:30:59 ? May 06 01:31:35 MattB0ne: you'd need to hook up an rs232 transceiver to a uart May 06 01:34:54 any way for a BBB to communicate with a rs 232 port May 06 01:35:08 preferably simple May 06 01:35:15 hook up an rs232 transceiver to one of its uarts May 06 01:35:46 ok are they cheap May 06 01:35:58 probably May 06 01:36:50 MAX3232 I think it's called? May 06 01:37:48 https://www.sparkfun.com/products/11189 May 06 01:38:03 that it May 06 01:38:06 ? May 06 01:41:45 that looks like a max3232 breakout yeah May 06 01:46:18 So I also want good headers if i want thick headers that still fit May 06 01:46:25 what should l look for May 06 01:47:14 i need a specific pitch right ? May 06 01:54:35 jumper wires and a header to plug into that board you want male to female jumpers to hook the BBB to the board that would include ground and power May 06 01:55:09 as for what you are hooking to that would be up to you, as no one person knows what you want to hook it to but you May 06 01:57:51 Hey! May 06 02:02:44 (y) Hi Guys =D May 06 02:03:43 Hello! May 06 02:04:01 I got issues, my board is telling the Cape to not listen to me. Just kiddin'. May 06 02:04:38 greetings kenunix May 06 02:04:52 Sounds serious :b May 06 02:05:28 Hey GenTooMan hows your internet access on BBB?X) May 06 02:06:17 GenTooMan: what if I want something more permanent May 06 02:06:38 like what May 06 02:08:12 Installed Debian 10.3 on a VM tonight. Couple of bugs May 06 02:09:33 Can you say RAID! Kills bugs...X) May 06 02:12:02 Are there any other good nodes (channels) on freenode ? May 06 02:12:19 Nope. May 06 02:12:23 None. May 06 02:12:26 Ha. May 06 02:12:40 There is #hal! May 06 02:13:01 MattB0ne a good question you have a few options such as a serial cape but the real question is what are you trying to do May 06 02:13:53 KenUnix: What are you doing w/ your BBB? May 06 02:13:59 kenunix I suppose it's fine I've been able to update stuff on the board such as using pip3 to install stuff. May 06 02:16:46 set_ too busy doin other stuff lile elinux May 06 02:17:19 Somebody had a 'time' issue? May 06 02:17:25 Oh? May 06 02:17:28 Nice. May 06 02:18:06 Set_ E(embedded)LINUX May 06 02:18:28 Oh! May 06 02:19:04 Awful lot of outdated / obsolete pages May 06 02:19:22 I know. May 06 02:21:52 set_ going to the page for example capes have you ever clicked on the discuss tab? Apparently you can start a discussion on an isses. If you click on a cape photo next to the uploaderd name is a talk button. Apparently someone can pose a question to the creator. Neet! Why not give it a try? May 06 02:22:14 This dog-gone server still does not want to move! May 06 02:22:15 Okay. May 06 02:22:23 server = servo May 06 02:23:00 I would but I can't post questions to myself May 06 02:23:08 Ha. May 06 02:23:53 Use some 3-in-1 oil =D =D May 06 02:24:42 I do not think it is the mechanics of it all. It is my source compared to what was given to me. I am using a wrapper. May 06 02:25:28 apparently there is a lot more to elinux than simply looking. You can for example go to the cape and in the left column download page. It brings text and photos. May 06 02:25:48 Yep. May 06 02:25:53 You can click on links too. May 06 02:26:00 Photos can be clicked, also. May 06 02:26:27 Looking at the ones I did and a few come down with empry boxes =( =( May 06 02:27:34 wrapper - Who Cool-J ? May 06 02:27:45 Nope. Cool-Mode-D. May 06 02:28:40 Did you meter the pins to see if any voltage change happens 8) 8) May 06 02:28:59 https://en.wikipedia.org/wiki/Kool_Moe_Dee sorry. May 06 02:29:06 No. May 06 02:29:08 Not yet. May 06 02:29:19 The cape is attached to the BBB. May 06 02:30:13 I have a 5v 1A wall adapter in the BBBW, usb networking, and the 5V_vdd goes to the Cape 5v in and everything is grounded. May 06 02:30:31 It's like the relay cape. You can pulse it too fast and nothing happens cause the relay itself takes about 25ms to operate. May 06 02:30:40 Oh. May 06 02:30:42 Okay. May 06 02:30:55 Maybe. There is not too much info. so far on this Cape. May 06 02:31:05 It works by way of i2c instead of GPIO/PWM. May 06 02:31:12 I think? May 06 02:31:31 I2C is a different interface, yes? May 06 02:31:35 Yes. May 06 02:31:58 It works like Master/Slave but you can have many slaves (they cannot talk to each other though). May 06 02:32:05 Many Masters too. May 06 02:32:20 Wait. The Masters cannot talk. May 06 02:32:22 SOrry. May 06 02:32:48 Well, get 1 servo to work first. Then move on. May 06 02:32:52 I was reading up on ideas and then things went goofy on end. May 06 02:32:53 Right. May 06 02:32:57 That is what I am doing. May 06 02:33:10 I am trying w/ S1 first. May 06 02:33:22 Supposedly, that is the first servo connection on the Cape. May 06 02:35:10 People don't understand this 'puter stuff can get real complicated May 06 02:36:35 Me! May 06 02:36:54 I barely understand anything I do 100% of the time. Yeea. May 06 02:36:55 * GenTooMan hits set_ with the angry sock puppet award. May 06 02:36:57 I found out by default if you do sh xxx it executes in dash. That can be changed so sh xxx would execute in bash May 06 02:37:11 Hmm. May 06 02:37:17 I won? May 06 02:37:28 GenTooMan: Gives out the award! May 06 02:37:40 bash is a ;ot more powerful than dsad May 06 02:38:00 right dash hopes to be bash? May 06 02:38:08 another wired sell is tmux May 06 02:38:17 another wired shell is tmux May 06 02:38:35 weird lesdyxia issues? May 06 02:39:36 ah well nobodies perfect so I don't know nobody. May 06 02:40:03 just wierd don't use it. If you like bash. you can as sudo cd /bin and del sh then ln bash sh May 06 02:40:42 so you hard link sh to bash... gotcha May 06 02:41:16 hum gotcha ? May 06 02:41:40 I comprehende vous? May 06 02:41:49 Hey! May 06 02:42:08 Practicamos! May 06 02:43:04 Is working w/ i2c and servos easy...no. May 06 02:43:28 ok then cd /bin and cp bash sh but first mv sh to sh.old so it's not gone. May 06 02:44:00 One of your computational thinkers know things, man. Give this "puppet award" winner some hints. May 06 02:44:26 Is begging asking? May 06 02:45:10 check it with a voltmeter to see if anything changes May 06 02:45:34 Okay. May 06 02:45:37 Good idea. May 06 02:45:53 So, run a loop and see if anything is read as feedback? May 06 02:46:46 If the software allows a slow loop then monitor for changes May 06 02:47:25 I was going to put a while True: loop in as a source and then figure out how long I have before I can touch the BBBW and ServoCape. May 06 02:48:50 Isn't there a 'sleep' or 'pause' you can put in the loop? May 06 02:50:15 Hey. May 06 02:50:21 You are not going to believe this fact. May 06 02:50:27 The servo moved. May 06 02:50:38 I am not sure why as of yet. May 06 02:51:22 The source was running and I was checking the amount of voltage at the pins powering the Cape and on the headers for the servo connections. May 06 02:51:34 Odd. May 06 02:51:55 KenUnix: Yes. there is a time built-in on python. May 06 02:55:22 You may be asking it to move too quickly May 06 02:55:22 This makes no sense. May 06 02:55:44 I understand but why did it move...maybe you are right. May 06 02:55:59 how long should I wait? May 06 02:56:19 I waited 30 seconds and I am still waiting. May 06 02:56:40 Is that too long. It seems to only move when I examine the servo headers w/ my DMM. May 06 02:57:27 Maybe the servo 5v and GND need sys_5v instead of vdd_5v? **** ENDING LOGGING AT Wed May 06 03:00:05 2020