| 00:00.06 | SpeedEvil | this is the - connection |
| 00:00.06 | Sunrunner20 | biggest issue is going to be the connector matching database |
| 00:00.25 | Sunrunner20 | 8mm would be too powerfull probally |
| 00:01.04 | SpeedEvil | one surrounding 'skirt' that's conductive, and engages with the other matching electrode on the other side, with the magnet at the base |
| 00:01.06 | Sunrunner20 | and the other issue is |
| 00:01.15 | Sunrunner20 | impact resistance |
| 00:01.33 | SpeedEvil | the magnet has a thin plastic cover, on which the actual contact is. |
| 00:01.37 | Sunrunner20 | I noticed that the mac books use neodinium |
| 00:01.44 | Sunrunner20 | magnet |
| 00:01.48 | Sunrunner20 | fragile little things |
| 00:01.57 | SpeedEvil | Not if you deal with it right. |
| 00:02.00 | ds2 | SpeedEvil: I still think a linear arrangement of: [N][Contact+][Contact-][S] is simplier |
| 00:02.15 | ds2 | keeps things thin |
| 00:02.22 | Sunrunner20 | the linera arragment _is_ simpler |
| 00:02.24 | SpeedEvil | You do not want a large magnetic external field |
| 00:02.32 | SpeedEvil | as that will screw with electronics. |
| 00:02.32 | Sunrunner20 | yea |
| 00:02.41 | Sunrunner20 | SpeedEvil: not as much as you would think |
| 00:02.45 | SpeedEvil | the field must be contained as much as possible. |
| 00:02.51 | Sunrunner20 | a static field shoudl not be an issue |
| 00:02.54 | SpeedEvil | I lost a LCD backlight due to a fridge magnet. |
| 00:02.58 | ds2 | right but if it is connected, the magnets "should" have flux lines within themselves only |
| 00:03.02 | Sunrunner20 | o.0 |
| 00:03.13 | Sunrunner20 | SpeedEvil: explain |
| 00:03.16 | SpeedEvil | had a fridge magnet on a steel wall behind my laptop |
| 00:03.24 | SpeedEvil | pushed laptop against wall. |
| 00:03.28 | ds2 | it is only when it is not connected that is the issue and a steel cover should short out the mag circuit |
| 00:03.31 | SpeedEvil | Caused backlight inverter core to saturate. |
| 00:03.37 | SpeedEvil | Caused fuses to blow. |
| 00:04.00 | ds2 | SpeedEvil: they should've sheilded it with mu Metal ;) |
| 00:04.05 | Sunrunner20 | lol |
| 00:04.15 | SpeedEvil | Arguably. |
| 00:04.23 | ds2 | (and yes, I have priced mu metal) |
| 00:04.35 | SpeedEvil | But was this the absolutely only vulnerable bit of electronics out there? |
| 00:04.50 | *** join/#edev ccjoe (n=ccjoe@www.clearcore.com) |
| 00:05.05 | ds2 | make the entire case out of mu metal :) |
| 00:05.23 | SpeedEvil | sucks mechanically |
| 00:05.24 | ds2 | should help with the FCC/CE approval ;) |
| 00:05.39 | *** join/#edev RobotGuy (n=robotguy@dsl093-038-072.pdx1.dsl.speakeasy.net) |
| 00:06.01 | Sunrunner20 | I don't need either |
| 00:06.14 | Sunrunner20 | only UL |
| 00:06.25 | ds2 | no one "NEEDS" UL |
| 00:06.37 | SpeedEvil | make a connector that also absorbs RFI, and you'll be liked. |
| 00:06.48 | Sunrunner20 | I have to be nutral |
| 00:07.03 | ds2 | SpeedEvil: is that before or after it catches on fire from walking too close to an antenna? |
| 00:07.04 | Sunrunner20 | I can't affect any "noise" in the power transmission line |
| 00:07.35 | Sunrunner20 | remember, some laptops communicate w/ their bricks |
| 00:07.43 | SpeedEvil | not that way |
| 00:08.04 | SpeedEvil | I mean akin to the RF filters |
| 00:08.21 | SpeedEvil | the lossy ferrite elements which many laptops use on the power leads. |
| 00:08.35 | SpeedEvil | not actually breaking hte cable |
| 00:08.48 | SpeedEvil | Tehy only affect common mode signals. |
| 00:09.12 | ds2 | which is why we need to do inductive charging! |
| 00:23.19 | Sunrunner20 | TBH |
| 00:23.34 | Sunrunner20 | I think the best solution is injection molded magnets around connectors |
| 00:24.05 | Sunrunner20 | with a three pin configureation |
| 00:24.21 | Sunrunner20 | for regular 2 wire power supplies anyway |
| 00:24.39 | *** join/#edev ico2_ (n=ico2@modem-2088.chameleon.dialup.pol.co.uk) |
| 00:24.44 | Sunrunner20 | + - + |
| 00:24.48 | Sunrunner20 | or - + - |
| 00:25.42 | Sunrunner20 | its cheap, simple and effective |
| 00:25.52 | Sunrunner20 | I could probally fit all the manufacutreing equipment in my room too |
| 00:26.32 | ds2 | you have IJ moulder? |
| 00:27.38 | Sunrunner20 | i wish |
| 00:28.16 | Sunrunner20 | http://en.wikipedia.org/wiki/Image:PlasticsInjectionMoulderJones.jpg |
| 00:28.36 | Sunrunner20 | granted |
| 00:28.42 | Sunrunner20 | there's no way in hell I would put it in my room |
| 00:28.47 | Sunrunner20 | I would outsource it |
| 00:29.03 | Sunrunner20 | and I would love ot use preexisting connectors :) |
| 00:29.05 | ds2 | they make desktop size units |
| 00:29.18 | ds2 | or have the prototypes machined out of plastic billet |
| 00:29.24 | Sunrunner20 | I could probally make one >_> |
| 00:29.48 | ds2 | one of these days I will make myself an IJ moulder |
| 00:30.07 | Sunrunner20 | the hardest part is 1. pressure |
| 00:30.09 | Sunrunner20 | 2. mold |
| 00:30.19 | ds2 | mold is easy |
| 00:30.40 | ds2 | pressure isn't too bad unless you want to have a rapid cycle |
| 00:30.55 | Sunrunner20 | and |
| 00:31.05 | SpeedEvil | A reliable mould is _hard_ |
| 00:31.13 | Sunrunner20 | I think i'm going to put the first batch in a local store |
| 00:31.26 | Sunrunner20 | also note |
| 00:31.40 | Sunrunner20 | this is going to be molding _around_ stuff |
| 00:31.49 | ds2 | SpeedEvil: just find the oldest guy in the plastics place and ask him to do the voodoo in your cad drawing |
| 00:31.50 | Sunrunner20 | the injected plastic will be the binder for all the parts |
| 00:32.06 | Sunrunner20 | lol |
| 00:32.24 | ds2 | one thing you can try is making a mold and use RTV |
| 00:32.41 | Sunrunner20 | I'm not sure what i'll make prototypes out of |
| 00:33.07 | ds2 | PE is simple |
| 00:33.09 | SpeedEvil | filled epoxy works weell. |
| 00:33.11 | Sunrunner20 | i won't be suprized if its shrink wraped w/ epoxy |
| 00:33.13 | Sunrunner20 | :P |
| 00:33.23 | ds2 | ewww epoxy |
| 00:33.26 | Sunrunner20 | lol |
| 00:33.35 | Sunrunner20 | its easy to work with ds2 |
| 00:33.42 | Sunrunner20 | you can file down extra etc |
| 00:33.46 | ds2 | it is too hard and generates heat |
| 00:33.56 | SpeedEvil | you need a mix of stuff. |
| 00:33.57 | Sunrunner20 | ds2: : these are prototypes |
| 00:34.02 | SpeedEvil | Epoxy+silicone can work well |
| 00:34.05 | ds2 | I much rather machine it out of acrylic and plastic weld it |
| 00:34.19 | SpeedEvil | and acrylic has problems with common solvents. |
| 00:34.29 | ds2 | <Sunrunner20> ds2: : these are prototypes |
| 00:34.31 | ds2 | :P |
| 00:34.39 | ds2 | acrylic is easy to weld |
| 00:34.41 | Sunrunner20 | who cares about heat :0 |
| 00:34.54 | Sunrunner20 | ds2: i know, i have some MEK in the back |
| 00:34.56 | ds2 | no - heat when hardening |
| 00:35.18 | Sunrunner20 | yea ds2, I know epoxy gets warm... |
| 00:35.27 | Sunrunner20 | but it flows well |
| 00:35.59 | Sunrunner20 | so long as one of ya'll don't jack my idea |
| 00:36.02 | Sunrunner20 | this could make me a rich man |
| 00:37.32 | ds2 | :P |
| 00:37.55 | ds2 | another way is to mechanically assembly it and make out of delrin or anther engineering plastic |
| 00:38.02 | ds2 | or ABS |
| 00:38.03 | Sunrunner20 | yea |
| 00:38.06 | Sunrunner20 | but |
| 00:38.19 | Sunrunner20 | I know I'll use injection molding for mass runs though |
| 00:38.27 | ds2 | you can probally share the same drawings for the mold and the machined part |
| 00:38.55 | Sunrunner20 | know what |
| 00:39.02 | Sunrunner20 | Steam looks really ugly under vista |
| 00:39.40 | Sunrunner20 | I need to write up a buisness plan |
| 00:39.42 | Sunrunner20 | if I get that done |
| 00:39.45 | Sunrunner20 | I'll move ahead |
| 00:40.21 | Sunrunner20 | most companies that don't start w/ a plan fail in like the first 6mo |
| 00:40.22 | ds2 | if you need protos machined out... |
| 00:40.44 | Sunrunner20 | if your a machinest... |
| 00:40.59 | ds2 | that is one of my hats |
| 00:41.06 | Sunrunner20 | lol |
| 00:41.23 | Sunrunner20 | tbh i'm most likely to use the guy at school :) |
| 00:41.29 | Sunrunner20 | yea, my uni has a machine shop |
| 00:41.44 | ds2 | 'k |
| 01:06.04 | *** join/#edev Zeta_RHIT (n=zeta@cpe-24-160-201-84.ma.res.rr.com) |
| 01:15.34 | SpeedEvil | Why would a 3.5" LCD have 60 wires? |
| 01:15.37 | SpeedEvil | http://www.ebuyer.com/product/130741 |
| 01:17.27 | SpeedEvil | I can see on the order of 30 - the PSP |
| 01:32.13 | RobotGuy | I like this one: http://store.earthlcd.com/ACX705AKM-7-ACX705AKM7?sc=7&category=193 |
| 01:37.39 | SpeedEvil | Sure - but note that the one I posted is cheaper with a PSU and stuff :) |
| 01:41.26 | RobotGuy | SpeedEvil: Yeah, OK, but I am thinking for a specific project. |
| 01:41.56 | SpeedEvil | Also - dealextreme.com |
| 01:42.10 | SpeedEvil | NDS front and back screens at $14 with free shipping |
| 01:42.25 | ds2 | except they take 234321423489327493827894327489327984237984237894372984712398472398 years to ship |
| 01:42.36 | SpeedEvil | 15 days or so. |
| 01:42.41 | ds2 | so far one week has pasted and it hasn't even shipped yet |
| 01:42.42 | SpeedEvil | or less if you go expedited |
| 01:42.59 | ds2 | they claim they didn't have the item til yesterday |
| 01:43.22 | SpeedEvil | I can only say I've had good experiences with them. |
| 01:43.43 | ds2 | this is my first order and I am irked at the pace they move for "instock items" |
| 01:43.43 | SpeedEvil | To the extent I've just done a $70 order with them to replace my broken meter leads and stuff. |
| 01:43.51 | RobotGuy | I have a 1.1 MB compressed initrd now. |
| 01:44.20 | RobotGuy | NDS? |
| 01:44.28 | SpeedEvil | Nintendo Dual Screen |
| 01:44.32 | ds2 | Nentendo DUal Screen |
| 01:44.32 | RobotGuy | Oh! |
| 01:44.42 | RobotGuy | The DS can do Linux too. |
| 01:45.21 | *** join/#edev redblue (i=star@ppp069.109-253-207.mtl.mt.videotron.ca) |
| 01:49.35 | RobotGuy | I wonder if the NDS could make a good remote robot console. |
| 01:54.26 | Sunrunner20 | you need the RAM expansion pack though |
| 01:55.28 | Sunrunner20 | its unfortunate that they so underpowered the DS's RAM |
| 01:56.02 | RobotGuy | Sunrunner20: Is that an external add-on rather than internal? |
| 01:56.12 | Sunrunner20 | yea |
| 02:02.08 | Sunrunner20 | if your going to use the DS |
| 02:02.19 | Sunrunner20 | get a "M3 DS Real" |
| 02:05.53 | Sunrunner20 | anybody ehre have a botonist as a wife lol? |
| 02:07.29 | RobotGuy | I don't have a wife. I don't even have a girlfriend. |
| 02:51.47 | *** join/#edev chouimat|away (n=dieu@r2351064.cidc.net) |
| 03:06.24 | *** join/#edev purl (i=ibot@pdpc/supporter/active/TimRiker/bot/apt) |
| 03:06.24 | *** topic/#edev is a channel for the discussion of hacking small consumer electronic devices, mostly ARM based; Hammer Development boards now available at http://www.tincantools.com/ |
| 04:24.03 | *** join/#edev Sunrunner20 (n=chatzill@pool-71-244-2-95.dllstx.fios.verizon.net) |
| 04:50.02 | *** join/#edev Zeta_RHIT (n=zeta@cpe-24-160-201-84.ma.res.rr.com) |
| 05:02.41 | *** join/#edev DrZimmerman (n=theo@tbaumgartner.broker.freenet6.net) |
| 05:25.17 | *** join/#edev harpal (n=harpal@124.125.77.253) |
| 07:23.23 | *** join/#edev harpal_ (n=harpal@124.125.77.253) |
| 07:59.50 | *** join/#edev magnetron (n=magnetro@unaffiliated/magnetron) |
| 08:08.43 | *** join/#edev harpal__ (n=harpal@124.125.77.253) |
| 08:19.29 | *** join/#edev dfarnsworth (n=dfarnswo@65.200.49.156) |
| 08:21.07 | *** join/#edev harpal__ (n=harpal@121.246.75.165) |
| 08:53.54 | *** join/#edev dfarnsworth (n=dfarnswo@65.200.49.156) |
| 09:11.59 | *** join/#edev lyakh (n=lyakh@p57BD2A52.dip0.t-ipconnect.de) |
| 09:15.58 | *** join/#edev rodserling (n=rodserli@unaffiliated/rodserling) |
| 09:57.24 | *** join/#edev magnetron (n=magnetro@unaffiliated/magnetron) |
| 10:52.39 | *** join/#edev harpal_ (n=harpal@124.125.77.253) |
| 11:34.10 | *** join/#edev Sunrunner20 (n=chatzill@pool-71-244-2-95.dllstx.fios.verizon.net) |
| 12:22.37 | *** join/#edev sjhill (n=sjhill@real.realitydiluted.com) |
| 12:22.37 | *** mode/#edev [+o sjhill] by ChanServ |
| 12:34.45 | *** join/#edev bbradley (n=bbradley@82-32-35-42.cable.ubr05.azte.blueyonder.co.uk) |
| 12:44.11 | *** join/#edev harpal (n=harpal@124.125.77.253) |
| 13:12.42 | *** join/#edev prpplague (n=dave@mail.americanmicrosystems.com) |
| 13:14.30 | prpplague | ho ho hum |
| 13:14.38 | prpplague | merry freaking friday |
| 13:42.32 | *** join/#edev chouimat|work (n=dieu@209.217.106.98) |
| 13:43.32 | prpplague | ho ho hum |
| 13:43.47 | chouimat|work | morning |
| 13:47.49 | prpplague | grumbles something that might be a greeting made in the morning |
| 14:08.41 | prpplague | ho ho hum |
| 14:09.05 | prpplague | wishes he could find out if patches for new systems types is still open for 2.6.25 |
| 14:10.17 | prpplague | i really wish the subsystem maintainers would post a schedule |
| 14:27.42 | *** join/#edev sunrunner20_ (n=chatzill@utdpat242007.utdallas.edu) |
| 14:27.53 | sunrunner20_ | the new Imac's have a 5pin connector on the break away |
| 14:29.14 | sunrunner20_ | and there is an indicator light on the connector |
| 14:29.37 | sunrunner20_ | I'm going to assum that the middle pin, sence it is significantly smaller is the power supply for the indicator |
| 14:30.36 | sunrunner20_ | also, the symmetrical design allows for the current supply to be divided amogst the pins |
| 14:31.18 | *** join/#edev CosmicPenguin (n=nobody@163.181.251.103) |
| 14:31.18 | *** mode/#edev [+o CosmicPenguin] by ChanServ |
| 14:40.35 | *** join/#edev chouimat|away (n=dieu@r2351064.cidc.net) |
| 14:40.36 | *** join/#edev dijenerate (n=dijenera@66.54.118.16) |
| 14:43.20 | sunrunner20_ | hey |
| 14:43.27 | sunrunner20_ | any of ya'll have a new dell XPS laptop? |
| 15:09.59 | *** join/#edev bbradley (n=bbradley@82-32-35-42.cable.ubr05.azte.blueyonder.co.uk) |
| 15:28.07 | *** join/#edev sunrunner20__ (n=chatzill@utdpat242007.utdallas.edu) |
| 16:06.24 | *** join/#edev cajaka (n=none@cpe-24-58-193-200.twcny.res.rr.com) |
| 16:09.21 | *** join/#edev ccjoe (n=ccjoe@www.clearcore.com) |
| 16:24.22 | *** join/#edev ico2 (n=ico2@modem-495.chimpanzee.dialup.pol.co.uk) |
| 16:52.48 | *** join/#edev dfarnsworth (n=dfarnswo@65.200.49.156) |
| 16:53.26 | *** join/#edev ico2 (n=ico2@modem-495.chimpanzee.dialup.pol.co.uk) |
| 16:56.43 | *** join/#edev lyakh (n=lyakh@p57BD2A52.dip0.t-ipconnect.de) |
| 16:56.44 | *** join/#edev RobotGuy (n=robotguy@dsl093-038-072.pdx1.dsl.speakeasy.net) |
| 16:58.17 | *** join/#edev dijenerate (n=dijenera@66.54.118.16) |
| 17:01.57 | prpplague | CosmicPenguin: ping |
| 17:07.45 | CosmicPenguin | yo |
| 17:09.57 | prpplague | CosmicPenguin: nm, i'm too frustrated to discuss it now, re patches |
| 17:13.46 | *** join/#edev _ccjoe_ (n=ccjoe@69.20.152.109) |
| 17:16.17 | RobotGuy | oy |
| 17:16.33 | RobotGuy | Patch frustration? |
| 17:17.07 | prpplague | RobotGuy: patch submission to be more accurate |
| 17:17.17 | RobotGuy | prpplague: Ah, OK, I understand. |
| 17:17.54 | RobotGuy | We have a good discussion going on the TCT forums now. :D I wish there were more people involved, but this can be good and bad. |
| 17:18.31 | RobotGuy | I put the word out on Trossen Robotics' forums too. |
| 17:18.37 | RobotGuy | And the PARTS list. |
| 17:20.43 | prpplague | screams at the @#$%@$%^#$^&#$%^#$%^# arm linux kernel patch system |
| 17:22.31 | prpplague | tries to decide he is going to waste any more time trying to submit this patch |
| 17:24.32 | *** join/#edev ce_geek (n=ce_geek@roger-blanchette-4.umelst.maine.edu) |
| 17:24.51 | chouimat|work | gives prpplague a guinness |
| 17:27.21 | samuelr | prpplague: What's the problem? |
| 17:30.01 | prpplague | samuelr: 1) no idea when certain types of patches will be accepted 2) you have to post to the mailing list first 3) then the patch sits there ignored 4) a thousand little formating details that aren't posted anywher 5) system maintainer for s3c24xx is complete jerk and ignores anything that isn't from him or his company |
| 17:30.40 | samuelr | prpplague: Wow. |
| 17:31.10 | samuelr | Unfortunately, that actually doesn't sound atypical :-/ |
| 17:32.11 | prpplague | indeed |
| 17:32.30 | prpplague | tries one more time just for hell of it |
| 17:32.50 | samuelr | Actually, your comments and past experiences have made me start wondering why there's no Bugzilla-type tool for patch management |
| 17:33.32 | samuelr | You start a project, set up source control, set up a bug system, set up a mailing list, start a wiki... |
| 17:33.54 | samuelr | Track everything except that patches get slopped around wherever and whenever. |
| 17:34.35 | prpplague | samuelr: the arm subsystem does have patch tracking, but the rules for submission are so convoluted, it would take an IRS auditor to figure out what the heck you are suppose to do |
| 17:35.07 | samuelr | Well, at least they have a system. |
| 17:35.08 | chouimat|work | lol |
| 17:35.37 | samuelr | The problem is that they have a system... and so does everybody else: entirely different systems. |
| 17:35.49 | *** join/#edev redblue (i=star@ppp196.108-253-207.mtl.mt.videotron.ca) |
| 17:36.18 | samuelr | So once you learn how to submit a patch for arm, that does you absolutely no good when submitting a patch for some other project. |
| 17:37.12 | samuelr | On the other hand, once you've used svn or bugzilla a bit, using the same tool for a different project misses that un-translatable learning curve. |
| 17:37.19 | *** join/#edev cajaka (n=none@cpe-24-58-193-200.twcny.res.rr.com) |
| 17:39.27 | RobotGuy | The jerk needs to be replaced then. |
| 17:40.27 | RobotGuy | Well, I have to exit to go to an appointment. I'll be back sometime this afternoon |
| 17:59.14 | ds2 | there should be a checkpatch script to point out hte formating details |
| 18:09.16 | prpplague | ds2: that requires you understand the rules, and the only people who do are high unlikely to waste their elitist valuable time on something so mundane |
| 18:30.40 | ds2 | prpplague: eh? AFAIK, all kernel patches should pass the checkpatch.pl validation |
| 18:30.59 | prpplague | ds2: LAK has its own rules |
| 18:31.42 | ds2 | prpplague: didn't seem like it. but then I send stuff to OMAP and OMAP pushes it peice meal up |
| 18:32.18 | ds2 | I do wish things under arch/arm would be more consistant though |
| 18:32.25 | prpplague | ds2: omap seems much more flexible than the s3c24xx |
| 18:32.33 | prpplague | ds2: amen |
| 18:32.53 | ds2 | possible. |
| 18:48.04 | *** join/#edev dijenerate (n=dijenera@66.54.118.16) |
| 18:56.18 | *** join/#edev ico2 (n=ico2@modem-495.chimpanzee.dialup.pol.co.uk) |
| 18:58.13 | *** join/#edev landley (n=landley@adsl-68-93-84-18.dsl.hstntx.swbell.net) |
| 18:58.13 | *** mode/#edev [+o landley] by ChanServ |
| 18:58.39 | prpplague | landley: greetings |
| 18:59.04 | landley | prpplague: hiya. |
| 19:02.20 | prpplague | landley: you get my email |
| 19:12.45 | landley | checking... |
| 19:15.43 | landley | prpplague: I hath gotteneth your email, sir. Thanks. |
| 19:15.51 | landley | (There's an "indeed" missing from the above.) |
| 19:16.07 | prpplague | landley: np, hopefully that explains your questions |
| 19:35.40 | prpplague | landley: you've got spam |
| 19:52.38 | prpplague | landley: hey, are you using your nc to do a serial terminal for the nail board? |
| 19:54.50 | prpplague | goes to get his kids from school |
| 19:54.52 | prpplague | bbiaf |
| 20:02.47 | *** join/#edev emil|k (i=emil@f054014174.adsl.alicedsl.de) |
| 20:30.58 | *** join/#edev ico2_ (n=ico2@modem-3305.chameleon.dialup.pol.co.uk) |
| 20:36.05 | prpplague | landley: i have the main tct hammer support patch posted on LAK |
| 20:36.24 | prpplague | landley: but bjdooks says he won't be looking at it until monday or tuesday |
| 20:37.50 | prpplague | landley: i don't expect there be anything wrong with the code, but between bjdooks and rmk , i'm sure i'll need to stand on one foot, juggle three kittens, and sing god save the queen |
| 20:40.13 | landley | prpplague: yes, I'm using my netcat. I have a scripts/minicom.sh that does the tty setup stuff. |
| 20:40.41 | landley | prpplague: I tweaked the patch a bit locally, I'll check LAK in a bit. |
| 20:40.51 | prpplague | landley: if you having trouble getting to a prompt on apex, i'd suspect that nc might not be passing the ctrl-c thru to apex |
| 20:41.00 | landley | I can do the signed-off-by dance once I've seen the differences between your patch and mine. |
| 20:41.18 | landley | It's passing through the ctrl-c, the problem is the apex I have is booting after a delay of 0 seconds. |
| 20:41.29 | landley | I never upgraded apex from the version I was sent. |
| 20:42.02 | landley | I suppose I can try to hit ctrl-c repeatedly right after rebooting, but I still haven't got a script here that'll reboot the board through the jtag without reflashing it. |
| 20:42.03 | prpplague | landley: hmm, thats odd, i suspect that the boss may have put a different copy of apex on your nail board for testing |
| 20:42.11 | landley | shrugs. |
| 20:42.25 | prpplague | landley: no, if its set for 0 it won;t prompt you at all |
| 20:42.30 | landley | It highlights the problem that if somebody ever _does_ set the delay to 0, how to the set it _back_? |
| 20:42.41 | prpplague | landley: you have to reload apex |
| 20:42.53 | landley | Should be able to set envvars with the jtag. :P |
| 20:43.22 | landley | If the envvars are in the apex image, I might whip something up to tweak 'em before a reload. |
| 20:43.29 | prpplague | landley: really just need to be able to reset to the defaults which you can do via jtag |
| 20:43.33 | landley | By the way, the script I have reloads all three. No way to reinstall _just_ apex. |
| 20:43.55 | landley | How do you reset to the defaults? Zero out some area of flash? |
| 20:43.56 | prpplague | landley: yea, just tweak the script to only erase the blocks for apex |
| 20:44.02 | prpplague | landley: yea |
| 20:44.21 | landley | It says it erases 127 blocks. Dunno what the granularity there is... |
| 20:44.23 | prpplague | landley: one sec and i'll post two scripts |
| 20:44.27 | landley | ok |
| 20:47.13 | *** join/#edev gandhijee (n=root@host-66-202-34-165.spr.choiceone.net) |
| 20:47.31 | gandhijee | hey can someone help me setup openOCD |
| 20:47.55 | prpplague | gandhijee: generally that help is over on #openjtag |
| 20:48.05 | prpplague | gandhijee: but you can ask here and see if someone helps |
| 20:49.53 | gandhijee | cool |
| 20:49.54 | gandhijee | thanks |
| 20:51.20 | Mirell | Hmm... |
| 20:51.28 | Mirell | landley: 'allo |
| 20:51.33 | Mirell | also waves at prpplague |
| 20:51.40 | prpplague | Mirell: greetings |
| 20:51.47 | Mirell | So I got pulled over today, and got a ticket for failing to maintain Texas registration. |
| 20:52.08 | Mirell | I pointed out to him the car was still in my mother's name, who lives in Arkansas, but he didn't seem to care. |
| 20:53.17 | prpplague | landley: bottom of the page: http://www.elinux.org/Hammer_Board_Software_System |
| 20:53.33 | prpplague | landley: you'll see the two ocd scripts, plus you can download the current apex.bin file |
| 20:54.11 | prpplague | Mirell: yea, they are pretty sticky about registration and such in texas |
| 20:54.28 | prpplague | Mirell: i once got a ticket because my street name was misspelled on my DL |
| 20:55.12 | Mirell | prpplague: Yet, according to the rules, I need to present a title in my name to get the vehicle registered... |
| 20:55.16 | Mirell | So...uh... |
| 20:55.16 | Mirell | Yah. |
| 20:56.33 | samuelr | prpplague: So... they fined you because the DMV messed up? |
| 20:57.07 | prpplague | Mirell: or you have to have a limited power of attorney with respect to the vehicle |
| 20:57.33 | prpplague | Mirell: you get a copy of the title and have your mother sign a vehicle power of attorney |
| 20:57.58 | prpplague | samuelr: yes and no, the ticket was dismissed after i showed proof that the address had been corrected |
| 20:58.18 | prpplague | samuelr: but it was a pain in the ass, since i had to go get it fixed, then show up for court |
| 20:58.31 | prpplague | samuelr: had to wait for like 4 hours to see the judge to have it dismissed |
| 20:59.16 | samuelr | prpplague: Amazing. |
| 20:59.18 | prpplague | Mirell: http://www.expertlaw.com/library/estate_planning/limited_power_of_attorney.html |
| 20:59.31 | prpplague | Mirell: there is a free one specifically for vehicles |
| 20:59.48 | samuelr | If you treated anyone in the government with the same contempt and disdain they showed you, you'd end up in jail... |
| 20:59.50 | Mirell | prpplague: NoNo, I have the title now, because last week, my mother signed it to me, but didn't indicate a date. |
| 21:00.11 | Mirell | prpplague: But I meant, for the ticket itself. |
| 21:00.21 | Mirell | prpplague: Because I was going to end up doing that stuff here soon anyway. |
| 21:00.49 | prpplague | Mirell: ahh well in most counties you can get the ticket dismissed by showing that you got the registration |
| 21:00.58 | prpplague | Mirell: they usually charge you like $10 for a dismissal |
| 21:01.09 | Mirell | Ahh |
| 21:01.24 | Mirell | At least he didn't give me a citation for failing to yield for a left turn |
| 21:01.41 | Mirell | That I would have been fine with, well, at least from a guiltyness perspective. |
| 21:01.54 | Mirell | Because yah, I was totally at fault there, no question. |
| 21:02.23 | landley | Mirell: hiya. |
| 21:02.34 | Mirell | 'awwo |
| 21:02.47 | landley | Mirell: what state is the license plate in? |
| 21:02.54 | Mirell | landley: Arkansas |
| 21:04.24 | landley | You might be able to get the ticket dismissed if you try. |
| 21:04.31 | Mirell | Yah |
| 21:05.51 | ds2 | Hmmmm |
| 21:06.28 | ds2 | prpplague: do I need to add any passives to use the USB slave functionality on hammer's second USB port? |
| 21:09.12 | prpplague | ds2: i'm sorry i did not understand the question |
| 21:09.43 | ds2 | prpplague: to use it as host, your appnotes/howtos say I need a pair of 15K pull downs... do I need anything like that for slave functionality? |
| 21:10.18 | prpplague | ds2: yes, you can look at the reference schematic for the hammer eval board |
| 21:10.29 | ds2 | prpplague: where do I get that? |
| 21:10.55 | prpplague | ds2: *cough* http://www.tincantools.com |
| 21:11.09 | ds2 | uh... WHERE? I been through there |
| 21:11.29 | ds2 | docs goes to elinux so that's not it and tutorials only had the note on the 15K pull downs |
| 21:12.04 | ds2 | nevermind |
| 21:12.18 | ds2 | they are found as part of the product, not under support/docs :P |
| 21:13.03 | prpplague | http://www.tincantools.com/product.php?productid=16135&cat=0&page=1&featured |
| 21:13.17 | prpplague | where it says "documents" it has "schematic diagram" |
| 21:13.39 | ds2 | found that AFTER deciding to look at every single link :P |
| 21:13.55 | ds2 | thought it would have been under tutorials/support forums/documentation |
| 21:14.55 | ds2 | prpplague: what's the inverter on the D+ line for? |
| 21:15.23 | prpplague | ds2: thats so you can have a gpio controlled physical disconnect |
| 21:15.44 | prpplague | ds2: basically a way to have a software controlled disconnect of the usb cable |
| 21:16.31 | prpplague | ds2: if you don't want the gpio control, just tie the 1.5k to +3.3v |
| 21:16.59 | ds2 | I see. I would have expected the UDC to take care that |
| 21:17.56 | ds2 | eh? is it legal (per the specs) to tie anything on D-/D+ to a local power source? thought that would violate the backdrive prohibition? |
| 21:19.37 | prpplague | ds2: i'm sorry i did not understand the question |
| 21:20.08 | ds2 | prpplague: from what i understand, a device cannot send power back through the data lines. what you suggested seems like it will do that |
| 21:20.58 | prpplague | ds2: you mean for the 1.5k pull up? |
| 21:21.03 | ds2 | yep |
| 21:21.59 | prpplague | ds2: no that is part of the USB spec, if the SoC or usb slave device is a +5v based device the line has to be pulled up via a 7.5K and if it is a +3.3v based device it is pulled up via a 1.5k |
| 21:24.08 | ds2 | prpplague: we are talking about 2 different things here. I was reading your suggestion of pulling them up via 1.5K as connecting D+ to the local 3.3V source (pin 39) via 1.5K resistor |
| 21:24.33 | ds2 | that would mean if the host decides to shutdown the port, the D+ line would receive power from the hammer. |
| 21:24.39 | prpplague | ds2: ok one sec, do you have the schematic open? |
| 21:24.43 | ds2 | yes |
| 21:24.57 | ds2 | I see the same issue with the schematic now that I look at it more |
| 21:25.48 | prpplague | ds2: ok, now, we are talking about resistor R7 |
| 21:25.52 | prpplague | ds2: is that correct? |
| 21:25.59 | ds2 | yes |
| 21:26.53 | prpplague | ds2: now, according the USB specification, D+ for a +3.3v device has to be pulled up to +3.3v via a 1.5k resistor |
| 21:27.12 | ds2 | prpplague: and I am not disputing that. |
| 21:27.21 | ds2 | what I am questioning is: |
| 21:27.46 | ds2 | PWR_JACK supplies DC power regardless of the host. |
| 21:27.57 | ds2 | U2 converts the 5V to 3.3V |
| 21:28.22 | ds2 | and in the case of the schematic, this means U4D is powered |
| 21:28.23 | prpplague | ok |
| 21:28.29 | prpplague | ds2: one sec phone |
| 21:28.39 | ds2 | okay, let me know when you are back |
| 21:33.47 | prpplague | ds2: one more sec |
| 21:34.16 | prpplague | ds2: ok back |
| 21:34.31 | prpplague | ds2: let me look at the scheamtic |
| 21:35.08 | prpplague | ds2: right u4d is powered down and the 1.5k resistor is pulled to ground |
| 21:35.27 | prpplague | ds2: well more accurately its tristated |
| 21:35.31 | prpplague | ds2: hi-z |
| 21:35.47 | prpplague | ds2: what next? |
| 21:36.06 | ds2 | okay |
| 21:36.33 | ds2 | U4 has a tristate? |
| 21:37.02 | prpplague | ds2: well when its powered off it is hi-z, when it is powered on, it follows either ground or +3.3v |
| 21:37.36 | prpplague | ds2: when gpc15 is low, the 1.5k is pulled to +3.3v and when gpc15 is high it is pulled to ground |
| 21:38.16 | ds2 | why would U4 be powered off? |
| 21:38.22 | ds2 | <-- must be missing something here |
| 21:38.33 | prpplague | ds2: powered off on startup |
| 21:38.58 | prpplague | ds2: after power to the system is applied, u4d is powered |
| 21:39.13 | ds2 | let's assume it is powered, i.e. the PWR_JACK has power |
| 21:39.18 | prpplague | ds2: ok |
| 21:39.29 | ds2 | and let's also assume GPC15 is low |
| 21:39.35 | prpplague | ok |
| 21:39.35 | *** part/#edev Zeta_RHIT (n=zeta@cpe-24-160-201-84.ma.res.rr.com) |
| 21:39.51 | ds2 | at this point, if the host wants to turn off the device, it would drop power, correct? |
| 21:40.32 | ds2 | however, when this happens, power is still being applied to D+. that is what appears to be against the spec |
| 21:40.41 | prpplague | ds2: the host has no phyical method for turning off the slave device, only via a software request |
| 21:41.03 | ds2 | bad choice of words on my part |
| 21:41.20 | ds2 | let's say the host drops power on the +5V pin of J3 |
| 21:41.54 | prpplague | ds2: irrelevant, since the hammer announces to the host device that it is self powered when the device enumerates |
| 21:42.41 | ds2 | this is where I remember some spec saying the device must NEVER power the D+/D- pins if the host drops power on +5V |
| 21:43.17 | prpplague | ds2: i've never heard of that, and our s3c2410 devices are USB-IF tested |
| 21:43.34 | ds2 | 'k |
| 21:43.40 | prpplague | ds2: and its a rare occasion that the host pc controls the+5v |
| 21:44.00 | prpplague | ds2: most host controllers only turn +5v off if there is an overcurrent condition |
| 21:44.01 | ds2 | okay |
| 21:44.53 | prpplague | ds2: i'll double check the specs over the weekend |
| 21:45.04 | ds2 | most circuits I think derive the pull up voltage from the +5V line |
| 21:45.28 | prpplague | ds2: you can't do that if the device is a +3.3v device levels |
| 21:45.47 | ds2 | *nod* which makes it very tricky |
| 21:45.50 | prpplague | ds2: if you are a 3.3v i/o you have to pull up to 3.3v |
| 21:46.36 | ds2 | if you believe microchip: |
| 21:46.41 | ds2 | http://ww1.microchip.com/downloads/en/DeviceDoc/MCHPFSUSB_FW_UG_51679a.pdf |
| 21:46.49 | prpplague | looks |
| 21:46.53 | ds2 | Page 17 (as shown by the PDF reader) |
| 21:47.02 | ds2 | page 13 as printed |
| 21:47.02 | prpplague | one sec pulling it up |
| 21:47.21 | prpplague | ds2: ok there |
| 21:47.50 | prpplague | ds2: what am i looking for? |
| 21:47.53 | ds2 | On that page, the section on "USE_USB_BUS_SENSE_IO" |
| 21:47.59 | ds2 | or search for "back drive" |
| 21:48.24 | ds2 | this seems to suggest what i was saying |
| 21:49.22 | prpplague | hmm, that is interesting |
| 21:49.27 | prpplague | thats the first i've seen of that |
| 21:50.09 | ds2 | I am not sure how well implemented is that |
| 21:50.53 | prpplague | ds2: well i can tell you, 99% of the cheap usb hubs only have their primary +5v power connected in a gang configuration |
| 21:51.07 | prpplague | ds2: so it has no control over turning individual ports on and off |
| 21:51.17 | ds2 | prpplague: no need to even say that... I've seen the notes for USB power switches and hub chips |
| 21:51.35 | prpplague | ds2: but is definetly something i'm gonna test on monday |
| 21:51.37 | ds2 | it isn't just the cheap usb hubs, it seems to be the defacto standard despite what the spec says |
| 21:52.06 | ds2 | the reality of all this is - no one has fried anything yet so it'll probally be fine 99% of the time |
| 21:52.35 | prpplague | ds2: but it might explain a usb error we have seen in the past |
| 21:52.45 | ds2 | oh :) |
| 21:52.51 | prpplague | ds2: i can recreate a specific error with some of the dell laptops |
| 21:53.09 | ds2 | hmm power management related? |
| 21:53.14 | prpplague | ds2: yea |
| 21:53.40 | prpplague | heads home |
| 21:53.42 | prpplague | later folks |
| 22:02.12 | *** join/#edev redblue (i=star@ppp087.109-253-207.mtl.mt.videotron.ca) |
| 22:11.19 | *** join/#edev ccjoe (n=ccjoe@www.clearcore.com) |
| 22:49.29 | *** join/#edev ico2__ (n=ico2@modem-2618.cougar.dialup.pol.co.uk) |
| 22:49.51 | *** join/#edev cajaka (n=none@cpe-24-58-193-200.twcny.res.rr.com) |
| 23:00.46 | *** join/#edev dijenerate (n=dijenera@72.51.90.206) |
| 23:50.07 | *** join/#edev Crofton (n=balister@206.196.151.254) |
| 23:59.41 | *** join/#edev agaffney_ (n=agaffney@gentoo/developer/pdpc.active.agaffney) |
