#robotics Logs

Nov 09 2012

01:48 mulletchuck i've got a little arduino-H-Bridge-stepper motor setup using a Shift Register chip to control 4 H-Bridges and 4 motors, and the motors are running around 120-130F.
01:49 mulletchuck what might cause them to run so hot? this is even if I just turn on the circuit and don't pass them any data triggering rotations. just being in idle, the motors get really hot.
02:53 Jak_o_Shadows mulletchuck: If it's stalling, it will get hot. Could be stalling due to lack of power? (i don't know about that), or maybe it's broken internaly?
02:54 mulletchuck Jak_o_Shadows: they're getting enough power. my PSU is supplying close to 2A and the motors only require 0.33A each.
02:54 mulletchuck they're not stalling, but they are being held in place when not spinning, so i guess the H-Bridge chips are keeping them in a locked position.
02:54 mulletchuck meaning DC current is running through them
09:23 rue_mohr "What fancy manipulators you have grandma"
14:36 reynmo so I'm reading that between the earths surface and 2 meters off of the earth's surface, there is a 200 volts potential difference
14:37 reynmo does that mean a conductor 2 meters long has a current?
14:37 rizlah I eat 200 volts for breakfast.
14:38 rizlah iirc I have no idea, but I assume so, hence the reason earth rods in houses are stuck that deep down
14:38 rizlah But I have no idea
14:38 reynmo just seems like if for every meter off the planet, we get 100 volts, why can't we turn that into energy?
14:38 reynmo http://en.wikipedia.org/wiki/File:NearField-Earth-eField.png
14:39 rizlah reynmo: Because if sticking shit into the ground makes energy, that means everyone can do it. If everyone can do it, people at the top of the pyramid don't get paid.
14:39 rizlah That makes these people very angry, which in turn makes politicians angry.
14:40 reynmo surely that's not the only answer
14:40 rizlah That is.
14:40 smeding potential is not energy
14:40 rizlah A lot of shit goes on behind the lines comrade.
14:40 smeding haha
14:40 smeding of course there's a conspiracy nut to answer
14:41 rizlah smeding: I'm not a conspiracy nut
14:41 rizlah FML I actually am not
14:41 rizlah srs
14:42 smeding the real reason you can't turn 100 volts into energy is because you measure energy in joules and they are not the same unit
14:42 reynmo smeding: but aren't volts just a measure of potential energy
14:42 smeding so you need current, and when you get current the voltage goes away
14:42 smeding no
14:42 smeding they are a measure of the electric field, sort of (still not quite the same)
14:42 rizlah �p�
14:42 smeding reynmo: think of it like a capacitor
14:43 smeding if you discharge the capacitor, the voltage goes down
14:43 smeding but in this case it's such a tiny capacitor that it doesn't take much
14:43 reynmo what if you moved the conductor through the air such that it is constantly in contact with new ions that produce the voltage
14:43 smeding you'd probably still generate negligible input
14:43 smeding only a fraction of what you'd put in moving it at any rate
14:44 smeding negligible power*
14:44 smeding i'd bet on fusion for a decent, clean power source :)
14:44 reynmo well what about at lets say, 2,000 meters, where there would be some 200,000 volts
14:44 smeding and probably fission in the mean time
14:45 reynmo 200,000 volts discharged would be a decent amount of power right?
14:45 smeding no
14:45 smeding P = U * I
14:45 smeding it all depends on the current
14:45 reynmo wouldn't just a piece of copper wire create the current?
14:45 reynmo with low resistance
14:46 smeding like i said
14:46 smeding it acts like a capacitor
14:46 smeding as soon as you draw current, the voltage is gone
14:46 smeding and you can't draw much before it is
14:48 reynmo I'm thinking about lightning, how its a discharge of the potential difference in the atmosphere propogated by the right conditions in the air to be conductive enough to connect it to earth
14:49 reynmo that's quite a number of watts
14:49 smeding it's not quite about conductivity
14:49 reynmo what if we had a long copper pole up into the ionosphere - it would give negative and positive ions a chance to flow to eachother which would create current
14:50 smeding it's about athmospheric conditions making that electric field big enough to the point that it overcomes the insulating effect of air
14:50 smeding and while lightning carries a lot of power, it carries a mediocre amount of energy
14:50 smeding in a hard to harvest way
14:51 reynmo ok thanks for the clarifications
14:52 smeding http://en.wikipedia.org/wiki/Harvesting_lightning_energy
14:52 smeding the last paragraph talks about what you suggest
14:52 smeding it boils down to "yes, it's technically possible, but it's not worth it"
14:53 smeding things like solar, wind and water are right there and easy to harvest
14:53 smeding and like i said, i believe in nuclear, both fission and fusion
14:53 smeding modern nuclear, anyway
14:53 smeding but let's not get into that
14:53 reynmo yea I do too, just had this thought and needed to clarify it
14:54 reynmo just seemed a bit crazy when I read that every meter off the earths surface represents 100 volts
14:54 smeding well, you ever play with a high voltage generator?
14:54 reynmo no, I'm a programmer :)
14:54 smeding http://en.wikipedia.org/wiki/Van_de_Graaff_machine
14:55 smeding it's pretty trivial to generate high voltage, it's just not generally useful
14:55 smeding power is the key metric for doing anything useful with it
14:55 reynmo which depends on current and voltage?
14:56 smeding yeah
14:56 smeding well, i guess energy is the key metric
14:56 smeding which depends on current, voltage and time :)
14:56 reynmo so if the current is basically the voltage over the resistance, the resistance is the problem?
14:56 smeding it's not that simple
14:56 smeding it depends on current and voltage, but when you draw current, voltage can change
14:57 smeding a computer power supply only outputs a constant 12V, 5V etc. because it keeps pushing power into the system
14:57 smeding with feedback to push more power in if it's below 5V, less if it's above
14:58 smeding so if you're not constantly putting as much power in as you're drawing, like in athmospheric electric fields, the voltage changes
14:58 reynmo so the problem is that even if the atmosphere has 200 volts between two points, it doesn't maintain the voltage between those two points when a conductor is introduced to allow current to flow
14:58 smeding yeah
14:58 smeding that's what i meant by saying it's like a capacitor
14:59 smeding you can charge it, but if you don't keep charging it while discharging it the voltage drops
14:59 reynmo and when the voltage is super high, the current is instantly very high as well but that is too hard to store>
14:59 smeding depends
15:00 smeding in the case of thunderstorms, for example, yes it's high, but the problem is that thunderstorms are pretty unpredictable
15:00 smeding weather is complicated
15:00 smeding and lightning strikes don't last very long because as soon as the current flows, the potential equalizes and then no more current flows
15:01 reynmo until a potential exists near it and another lightning strikes
15:02 smeding lightning doesn't strike twice in exactly the same spot
15:02 smeding in quick succession
15:02 smeding it's a complicated process and i'm not even sure scientists even fully understand it
15:03 smeding potential is pretty much the presence of a charge imbalance
15:03 smeding just a few more electrons than positively charged particles
15:04 smeding a wire allows the electrons to move easier, so they move in the direction of the charge imbalance because it exerts a force
15:04 reynmo I just wonder if you waved a conductor such that it taps into multiple places of potential differences in the atmosphere if it would have some kind of reasonable constant current
15:04 smeding oh, i just thought of a good analogy
15:05 smeding if you plot gravitational potential energy versus distance from the surface
15:05 smeding you'd see pretty decent amounts of potential too
15:05 smeding if you drop a weight from high up, it gains kinetic energy
15:06 smeding of course it's different
15:06 smeding but it shows that once the weight has fallen, you stop getting energy -- it's a transient thing
15:07 reynmo so we need a constant potential, to get a constant kinetic
15:07 reynmo but couldn't we get a constant potential energy by moving the conductor constantly?
15:07 reynmo a car driving on the interstate with a large conducting rod above it
15:08 smeding well
15:08 smeding that's going by the assumption that you get a sensible amount of initial energy
15:08 smeding power, i should say
15:09 reynmo just because 200 volts are available doesn't mean you will get 200 volts applied in the current
15:09 smeding 200 volts are not 'available'
15:09 smeding as soon as you change the environment by introducing a metal thing, it changes
15:09 smeding briefly, charge flows and you get current, making power
15:09 smeding if you move around, you get some more power, maybe
15:10 smeding but the energy you get is /negligible/
15:10 reynmo 200 volts is quite a bit though, right?
15:10 smeding no
15:10 smeding you're still equating voltages with energies
15:12 smeding energy, voltage, current, power are all different things
15:12 smeding voltage is a measure for how much charge is around (that's not canceled out by opposite charge, anyway) compared to a reference
15:13 smeding current is how much of that charge is flowing along a path
15:13 reynmo in the presence of high voltage, a low resistance conductor would create high current?
15:13 reynmo current = voltage / resistance?
15:14 smeding yes, instantaneous current
15:14 smeding but by having charge flow from a point to another point, you end up with less charge at the former point of course
15:15 reynmo so a 9 volt battery that provides a constant 9 volts can power small things
15:15 reynmo how can 200 volts not power bigger things?
15:15 smeding because there's chemical reactions that take place inside the battery
15:15 smeding and supply the energy to keep it at 9V
15:16 smeding eventually, the chemicals deplete and the voltage on your battery drops
15:16 smeding for your sky pole, the voltage would drop pretty much instantly
15:17 reynmo so purely theoretically, if I'm moving fast enough through the atmosphere to get a constant average of 100 volts between two points connected by a copper wire, I'd get decent power
15:17 smeding yes, now calculate the speed you'd need ;-)
15:17 reynmo in practice I know it wouldn't be that simple, just trying to conceptualize this stuff
15:17 reynmo hehe yea
15:17 smeding at least, that's my understanding, i'd need to think about it more to be sure
15:18 reynmo well I'm going to go strap a 6 meter copper pole to my car and hit the highway, I'll let you know
15:18 reynmo ;)
17:42 Centauri If memory serves.. they strung a pretty long cable (5 miles?) in space (behind the ISS or a shuttle) to cut the earths magnetic field lines for a power generating test
18:02 chris_99 Centauri, just found http://en.wikipedia.org/wiki/Electrodynamic_tether
18:03 Centauri chris_99: Yes, thats it
18:03 chris_99 seems a nifty idea
18:03 Centauri Talk about a rather *large* coil ;)
18:04 Centauri but this exact same effect can be seen at home
18:04 Centauri pass a straight length of wire past the magnetic field of say, some supermagnets scavenged from old dead hard drives
18:05 chris_99 oh yeah i realise it's just induction, but the idea that it uses earths own field is quite nifty
18:07 chris_99 would it be possible to use the kinetic energy of the satelite
18:07 chris_99 somehow
18:08 Centauri I dont beleive that was the point of that particular experiment.. tho there are talks about "skyhooks" which does
18:09 chris_99 cool
18:10 Centauri space is my other "thing" besides robotics :)
18:14 Centauri heh, actually, part of what I am doing is for (or could be for) space anyway
18:23 chris_99 nifty, can you talk about that?
18:24 Centauri Yes, because I do it on my own, as a hobby, not for a company *sigh*
18:25 chris_99 so what's your project?
18:26 Centauri chris_99: truely, my eventual goal would be to turn loose a whole small horde of robots.. that are just smart enough to scrape the lunar surface.. and be able to navigate to the drop off point and go right backj where they were to resume digging ( the drop-off point would then be running a solar forge to refine the anothite) - this would also begin to make a good sized hole - rather like a strip mine.. so at some point.. th
18:26 Centauri ye could dig sideways, some 10-20 meters down.. to dig out a habitat for us air breathers
18:28 Centauri for this reason, I have been pushing on navigation code and obstacle avoidance, and, when I get better, repeatability for these "farming patterns"
18:29 chris_99 sounds funky, so have you done any prototype kind of stuff
18:30 Centauri yes, I am working on my exploratory bot to iron out free navigation, as this is a home-based project, it does take time
18:32 Centauri this bot has ultrasonic (for now) but I am looking into IR distance sensors.. I already use a 3 axis gyro for course correction
18:32 Centauri plus whiskers or bumpers will be in the works as well
18:32 Centauri this particular bot is really a learning platform for many aspects.. not a diggerBot
18:33 Centauri the last pic on this page is what he looks like right now: http://www.centaurihome.net/gallery2/main.php?g2_itemId=153&g2_page=5
18:34 chris_99 ooh that looks really cool. What chips are powering it?
18:36 Centauri heh, I have quite a few, as you can see on the sensor layer: the main processor is an arduino uno R3 and I also have 2 MCP23017 chip (16 GPIO pin expander chip I2C) and a couple of analog mux chips: CD4067 and CD4097
18:37 Centauri IN addition, there is an electronic compass and a wixel RF serial comm board so he can talk to the "mothership" (my unix server)
18:38 Centauri the multi-layers screwed me up for a time, as this was the 1st time I messed with more than 2 layers total, so wire passing amongst the various layers required some rework over the past 2 months
18:39 chris_99 yeah, you've got a lot of boards there, must be tricky to wire them up
18:39 Centauri indeed it is :)
18:39 Centauri but I have a LOT of potenial IO I can do now, with very few pins used on the arduino
18:40 Centauri the CD4067 is a 16:1 analog mux chip.. so by addressing that chip.. I can get 16 analog reading on one analog pin on the arduino
18:40 chris_99 yeah i've bought a load of those analogue recently for a little project
18:40 chris_99 *mux chips
18:41 Centauri the 23017 chip is actually 2 8-bit ports - so I use one port as inputs, and one port as outputs (which I use for addressing on the CD4067/4097)
18:41 Centauri I send serial comm thru that CD4097, as I have 2 serial control boards a servo controller, and the dual motor controller
18:43 Centauri the software is as much the complication, as the wiring really, when chaining things this deep
18:44 chris_99 i guess you're using C for that?
18:44 Centauri yup, all 2500 lines of it LOL
18:44 chris_99 heh, that's a lot of code, what it do atm with that
18:45 Centauri Well, I can give him a command (for now) like mfu (move forward until obstacle) and he will differentially speed one wheel or the other based on the gyro input.. to track straight, while looking at sensors, to know when to stop
18:46 Centauri he can turn on a varying radius (like turn left wheel forward less than the right wheel), plus what I call a "timid mode" to back away if the ping sensor sees that something is too close
18:47 Centauri but he isnt yet running the patterns and such.. I am still removing glitches
18:47 Centauri I basically started on this bot Mid May after 8 years off
18:48 Centauri most of the code deals with all the lower level things.. the beginning of a subsumption architecture
18:52 Centauri ON top of that, I have some "personality" type functions: explore, curious, back_away, seek_recharge, sentry_mode, etc
18:53 Centauri much of the work has been what I would call the "sub-concious" or pure body functions, upon which I can build up higher levels of behaviour
18:54 chris_99 i like the idea of it finding it's charger
18:55 Centauri heh, yeah, I was pondering some sort of inductive charging setup, where all he has to do is "park" on top of a speical mat, that had the coil built inside of it
18:56 Centauri for now.. his battery is 4.5Ah, so he runs quite a while, given the draw is only 150-200ma even with the drive wheels at 30% use
18:57 Centauri I can basically play with it all night, then let him charge while I am at work
18:58 Centauri when the navigation stuff gets better.. he should be able to get at least pretty close to the charger, and shut down the other sensors, to save power (for now)
18:58 Centauri much of that code is already in place
18:59 Centauri the bot can control the gate of a series of MOSFETS (to-92 case) that are 500ma each, to completely shut off specific hardware (this is dual purpose, as it allows the bot to hardware HARD reset sensor that are not playing well)
19:00 Triffid_Hunter that's a neat idea
19:01 Centauri for some of my sensors.. it was *required* as the compass module gets into stuck conditions in my office.. I have a rather large hard iron source that messes it up (my beer fridge)
19:02 Triffid_Hunter how about using magnets for the charger? get a hard drive magnet with contacts near it so it pulls the bot in when it's close enough, use electromagnets to detach
19:02 Centauri so its a good test.. if he can escape my office, the REST of the house is easy
19:02 Triffid_Hunter then stick an IR beacon on top so he can find it easily
19:03 Centauri Triffid_Hunter: I have pondered many way to "jack-in" automatically.. even something that looks similar to space docking ports like this: > > but that is farther down the list
19:03 Triffid_Hunter easiest way is of course dodgem style
19:04 Centauri I considered a pair of cones.. but on this bot.. I just havent spent much time on re-charge yet, as I am still working other basics
19:04 Triffid_Hunter what battery chemistry?
19:04 Centauri Even tho this started in Mid may.. I only spent a certain amount of time on this.. and sometimes Henry sits for days and weeks
19:05 Triffid_Hunter heh yeah I know that one all too well
19:05 Centauri Li-ion
19:05 Centauri 2x3x0.75 Inch battery pack with 4.5Ah capacity
19:05 Triffid_Hunter you mean lipo? li-ion isn't rechargeable
19:05 Centauri in the picture.. my battery is that blue thing
19:05 Triffid_Hunter those lipos are awesome
19:06 Centauri its not lithium Polymer
19:06 Centauri its this: http://www.jameco.com/webapp/wcs/stores/servlet/Product_10001_10001_2129351_-1
19:07 Centauri Since I never really draw even 1 amp max, this pack serves me well
19:07 Triffid_Hunter interesting, I have a 3S lipo for mine
19:08 Triffid_Hunter I run the motors from the high current cable, and the electronics from the balance plug.. the ESR is so low that I never have any voltage dip problems even when doing wheelies
19:08 Centauri Well, my bot draws only 105-145ma idle... each motor draws 100ma free turning (no load) each, and about 200ma at stall current
19:09 Centauri whole bot weighs in at maybe a pound to 1.5 total
19:09 Triffid_Hunter mine's a modified R/C car with ackerman steering.. the drive motor takes 5-6A stalled, needs slightly more careful design :)
19:09 Triffid_Hunter had an off-the-shelf driver that emitted fire recently, just waiting for some parts to finish one I designed
19:10 Centauri for a sense of scale.. those acrylic hexagone are 7 3/4 inch when measuring from one parallel side of the hex to the opposite parallel surface
19:10 Centauri :O
19:10 Centauri Ooooo no fire! easy enough to have other probs!
19:11 Centauri this bot is small, so I can keep many parts light in operation and use
19:11 Triffid_Hunter hm I don't visualise inches too well.. around 200cm?
19:11 Triffid_Hunter 20cm I mean
19:11 Centauri yesa
19:11 Centauri 20 cm almost exactly
19:12 Centauri wheels are 12cm diameter.. and max speed is about 60 cm / sec
19:12 Centauri but I usually run him about at half that
19:13 Triffid_Hunter yeah most of the sensors don't have enough range to go fast, can't respond in time
19:17 Centauri actually, he is pretty fast, as far as his sensors,, especially with the untrasonic being able to see to to about 365cm
19:19 Centauri part of the inner core of the navigation is a tight loop of course_correct() then a ping and test.. plus snging the ultrasonic left-center-right on a 60-90-120-90-60 back and forth swing to look at distance constantly as he is moving
19:20 Centauri this is because of the 30 degree width of the ultrasonic
19:21 Centauri but I am pondering some of those Sharp IR sensors for the 2-18 inch range (6-45cm)
19:22 Centauri a friend posted me a great youtube vid about bots, tho its 2.5 hrs long
19:23 Centauri subsumption architecture and such, I think it must be older because he is using the motorola 6811 that was common long ago
19:23 Centauri bu tthe emergent behaviour, by combining basics is good stuff, I was doing a good part of it, but got better ideas after watching it
19:24 Centauri http://www.youtube.com/watch?v=8CXReb7f0Eo&feature=plcp
19:26 Triffid_Hunter Centauri: yeah always fun doing the layering
19:27 Triffid_Hunter those sharp sensors are great, can't see anything that doesn't reflect IR though. use any camera to see the spot
19:27 Centauri Triffid_Hunter: Luckily, I code that way anyway, so when I have ot change hardware.. its generally not too painful, as I have at least 3 layers
19:27 Triffid_Hunter garbage bags are almost totally transparent to both IR and sonar if you want a challenge for your 'bot ;)
19:27 Centauri LMAO
19:28 Triffid_Hunter Centauri: also if you want fast scanning as well as decent range with your sonar, look up EERUF
19:28 Centauri If I threw a bag over Henry, He'd think he was trying to escape a condom ;)
19:28 Triffid_Hunter haha no you'd think that, he would think his motors were stuffed
19:28 Centauri LOL point made
19:29 Triffid_Hunter either that or you'd left him in the middle of a carpark
19:29 Centauri my office is soo cluttered, he really has a hard time navigating in here at all
19:30 Triffid_Hunter time to get your monoslam on then
19:30 Triffid_Hunter see http://www.youtube.com/activevision
19:30 Centauri Well, I dont mind the challenges.. like him seeing a clothes pile.. and getting around it
19:31 Centauri This bot isnt something I am tryting to bring to market, for say, planetaryresources.com
19:32 Centauri they actually have a tougher issue (almost no gravity to hold the bot down) which I wasnt counting on for lunar diggerbots
19:34 Centauri I'm just glad I am now past 90% of the "money-spending" part, now that i have what I need to do all this testing, most of the rest is software, and a bit of hardware and mechanical engineering
19:35 Centauri Robotics is that special mix of all 3 (SW, HW, Mech engineering) - I had someone at work ask me about a cheap simble robot. I had to laugh.. I said there is not "simple robotics" and certainly not "cheap"
19:39 Centauri Triffid_Hunter: I saw a page about EERUF, and I solve this by only having ONE Usonic sensor which I swing around mounted on a servo, with a 2nd servo tilting that whole thing
19:40 Centauri software is sooo much faster than whats going on physically, that it really reminds me of "tron"
19:41 Centauri I have never had issues where the CPU cant keep up with the "real world" most of my stuff can sample easily 200 times a second
19:51 Triffid_Hunter Centauri: EERUF applies to one sensor just as much.. your ping doesn't stop bouncing off stuff after it hits the first thing!
19:52 Centauri Triffid_Hunter: true enough.. LOL I had one interesting circumstance.. where I saw EXACTLY why the door of the beer fridge is "stealthy"
19:52 Centauri the bot was facing it at a 45 degree angle off direct facing, and it was pretty close
19:53 Centauri so at 45 degrees to the other side.. (90 total) as I moved my hand closer to the fridge door.. the bot would back up.. a perfect example of the distance measuring, with the reflection
19:54 Centauri currently the pong software will take the microsecond time from the first receiveable ping.. not any subsequent echos
20:00 Centauri Triffid_Hunter: I also use multiple sensors, and I believe the ones least likely to error.. like I beleive the gyro over the compass and I beleive the whiskers/bumpers over other DME
20:01 Centauri some kind of wheel turn encoding is next
20:07 rue_mohr heh, silly audio rangers
20:08 Centauri Well, they do have their issues, but each type of sensor has an areas its good at, and bad at
20:08 Centauri thats why many is good
20:08 Centauri but building a heirarchy of which take precident, is important
20:09 Centauri once of my basic rukles in course correction is this: if the compass says I am making a left turn, but the gyro sees a right turn.. *believe* the gyro
20:10 Centauri the gyro is NOt affected by hard iron, like the compass is
20:10 Centauri and all that sort of thing is needed all over, as you add sensors, and take readings from many
20:11 Centauri what happens whe the wheels are spinning, and the shft encoders SAY they are spinning, but the compass, gyro, and accelerometer say "no way Jose"
20:11 Centauri classic stuck condition, where the wheels are slipping and spinning, but no movement
20:23 ace4016 Centauri, could use image references to tell when movement takes place
20:23 ace4016 like an optical mouse
20:23 ace4016 some satellites will use star maps to sort of tell where they are
20:26 Centauri Stars dont move enough to use as an accurate refernece in our solar system.. however.. the position of the earth, moo, venus, and the sun and other nearby object would
20:26 Centauri moo= moon
20:26 Centauri but on the ground, there does need to be some hierarchy of what things can be trusted over others.. a type of fuzzy logic
21:33 rue_mohr the cow jumped over the moo?
21:34 rue_mohr Centauri, I'm gonna try putting sensors on the castor wheel of my new mowerbot
21:34 rue_mohr for tracking motoin/ knowing when stuck.
21:35 Centauri rue_mohr: that may or may not work.. depends on rotation. lol as far as MOO - umm I run a MOO based MUD, and thats where I tested and simmed a lot of things before I got back into robotics
21:43 Tom_itx bond was good
21:44 Centauri rue_mohr: truely, I find the more monitoring one can do in a bot, the better, better to have too many sensors - than too few.. you dont HAVE to read them ;)
21:46 Tom_itx you do have to power them and carry them along
21:47 Centauri my own routines do things like - read the ambient light once each 100 times the routine is called, but only check the battery level each 10,000 times (done in the sensor read routine) I read some sensors a lot, and other not so much, although they all have the potential to be read each and every time
21:47 Centauri Tom_itx: this is true.. however.. an LM35 is cheap power wise.. and I do power down my compass and gyro when the bot is in a minimal night time sentry mode
21:57 Centauri Tom_itx: I actually do a fair amount of power management as best I can, given the number of devices :)
21:57 Centauri and the fact that only so much can be done, without adding more than its worth, to manage things :)
22:09 Centauri Tom_itx: actually, I do watch what the additional current draw is or specific devices.. and think about its cost on my battery budget