Back
[00:01:13] <katsmeow> a coil slot?
[00:01:16] <katsmeow> coin
[00:01:24] <rue_house> yea the computerised sorter
[00:11:01] <katsmeow> hmmm, an ant species with 4 sexes, and who the queen mates with will determine what kind of babbies she will have
[00:12:28] <rue_house> hmm, sounds quite human
[00:22:45] <katsmeow> you look funny, who in their right mind would walk only on their rear legs?
[00:24:52] <rue_house> he who paves the ground to make it work
[00:34:14] <rue_house> robot holding can of pop: "its broken" human:"hu?" "I tried to put it back in the charger" "charger?" "it wont recharge" "?!?!?!?!?!?"
[00:35:04] <katsmeow> cells useto be only cylindrical, there's some lead acid car bats with round cells
[00:35:22] <rue_house> yup
[00:35:31] <rue_house> resists plate flaking
[00:36:52] <katsmeow> but then with all it's stated advantages, why doe sit cost 2x as much as a regular battery?
[00:37:55] <rue_house> your taking to someone who recently watched a video where a guy described how to buy things from aliexpress, "brand" them, and sell them for up to 20x as much on ebay cause theywere "branded"
[00:39:15] <katsmeow> with any brand?
[00:39:34] <rue_house> "freds" yea
[00:39:48] <katsmeow> some China sellers are openly ststing their usa-branded parts are generic clones and rebrands
[00:40:19] <rue_house> the guy said for a few cents ea, you can get china to do the branding for you
[00:40:33] <rue_house> they just laser on your logo
[00:40:44] * katsmeow nods
[00:43:11] <rue_house> what made me ill, is that people buy the stuff and pay the amount for the "branding"
[00:44:02] <katsmeow> 100 pcs IRFZ44 $16.81 free s&h
[00:44:17] <rue_house> I have 200 currently...
[00:44:29] <rue_house> (rue went overboard)
[00:44:44] <katsmeow> have you *looked* at the soa curves on thos, or the IRF1404 ?
[00:45:28] <katsmeow> with 20v across them, the curve says they can handle 5amps , but that's 100 watts, way more than a to-220 can handle
[00:46:11] <rue_house> the to-220 dissipation numers are way higher than the to-3 I could find
[00:46:12] <katsmeow> and at 5 v across them, they can handle 100amps dc ?! , their lil pins won't handle that
[00:46:20] <rue_house> which left me really confused
[00:46:33] <rue_house> there is a * on the first page
[00:46:41] <katsmeow> the to-3 has 4x the heatsink contact as a to-220
[00:46:51] <rue_house> ifyou can find the text for it,they rate the pins at like, 40A?
[00:46:57] <rue_house> or soemthing stupid
[00:47:22] <rue_house> yea, but when I was looking at power dissipation, the to-220 were higher, I think cause of thermal resistance
[00:47:31] <rue_house> die is closer to the heatsink?
[00:47:40] <katsmeow> and at 20amp the resistance goes up to 0.13 ohms? that's worse tham most bipolars
[00:48:26] <rue_house> yea, fets are only good in low voltage stuff
[00:48:33] <katsmeow> mosfets have been around for decades, and i am still unsure where i'd use them
[00:48:43] <rue_house> zippo:/files/programming/c/DAlib# ohm -r 0.13 -i 20
[00:48:43] <rue_house> Wattage is: 52.000000
[00:48:44] <rue_house> Current is: 20.000000
[00:48:44] <rue_house> Voltage is: 2.600000
[00:48:44] <rue_house> Resistance is : 0.130000
[00:48:44] <rue_house> You have new mail in /var/mail/root
[00:49:35] <katsmeow> the olde TIP36 will pull 15amps under a volt C-E, course it need an amp of base drive to do it, but it's only a volt too, not 10 volts
[00:49:59] <katsmeow> it's wrong, you have 2 emails
[00:50:54] <rue_house> no thats root emails, I get system messages in the form of local emails that are not worth figuring out how to access
[00:52:02] <katsmeow> o
[00:52:03] <rue_house> the last one says that posting my house IP address to ruemohr.org was successfull
[00:52:11] <rue_house> 3!
[00:52:14] <katsmeow> that's how i figure you feel about emails i send too
[00:52:52] <rue_house> a buddy has an led tv that the backlight driver isn't workin gon
[00:53:15] <rue_house> "I tested the the LED power line P801, and while on standby it was
[00:53:15] <rue_house> measuring 107V, when I turn it on it floats between 216 and 230V, the
[00:53:15] <rue_house> schematic says it should be 157V for 42". I'm wondering if one of the
[00:53:15] <rue_house> mosfets is bad. I tested mosfet at Q801 and one of the legs shows 666,
[00:53:15] <rue_house> and the other just goes to 1 (which I think is OL on this meter). I
[00:53:16] <rue_house> did test the mosfet while still attached to the board, is that an
[00:53:18] <rue_house> issue?"
[00:53:21] <rue_house> I'll think about that one later
[00:53:31] <rue_house> lets see what kat says
[00:53:57] <rue_house> I dont think memory can be measured that way, BUT
[00:54:11] <rue_house> hmmm
[00:54:56] <rue_house> given the right system, I'm sure 8k would do it
[00:55:03] <rue_house> links and logical operator
[00:55:04] <rue_house> s
[00:55:19] * katsmeow nods
[00:55:23] <rue_house> ah, per word
[00:55:38] <katsmeow> but read the 2nd email
[00:56:01] <rue_house> instead of words, what about idea
[00:56:02] <rue_house> s
[00:56:17] <katsmeow> ideas can be expressed in words
[00:57:03] <rue_house> I think there is irony in the fact these letters differ by 1 idea
[00:58:08] <katsmeow> sharing?
[00:58:31] <katsmeow> i figured i had to explicitly state it, hence the 2nd email
[00:59:39] <rue_house> I wonder how many things differ byt eh way their components are assembled (in the mind)
[01:00:04] <katsmeow> ditto "is, am, are, be, being, been" could use up 6 8k blocks, but they prolly need only one
[01:00:36] <rue_house> but I think the organic brain stores things in 'messes'
[01:00:46] <katsmeow> phrases too
[01:01:15] <rue_house> 'messes' are hard to describe,
[01:02:13] <rue_house> its like a network that consists of links that do everything they need to do as a set of layers, without things happening to cross over
[01:02:50] <rue_house> I want o experiment with them
[01:03:05] <rue_house> cause they have a large degree of redundancy
[01:03:44] <katsmeow> so grab a mess of attinys, plug them into each tehr in layers of serial connectons, and see if it evolves
[01:04:11] <rue_house> I'm trying to understand the low level operation of a neuron,
[01:04:23] <rue_house> there is a fundamental concept I'm not getting
[01:04:41] <rue_house> I know this when i was reading up on neural plasticity, how neurons learn
[01:04:58] <rue_house> there are two differnent triggerings, I dont understand what both are
[01:05:12] <rue_house> the timing of them determines the trainign of the neuron
[01:07:48] <rue_house> the brain is digital
[01:07:58] <rue_house> with an analog bias
[01:08:33] <rue_house> I'm starting to understand digital-centric analog systems
[01:08:38] <rue_house> long way to go
[01:09:57] <katsmeow> ok
[01:10:44] <rue_house> my latest pwm genorator is a great exampls
[01:11:04] <rue_house> usually, I would make a triangle or sawtooth osc, and then use a comparitor
[01:11:27] <rue_house> but instead, I used a square wave, ran it thru a low pass, and then did the comparitor
[01:11:35] <rue_house> its symantics
[01:11:43] <rue_house> and I love symantic shifts
[01:11:51] <katsmeow> why a lowpass?
[01:14:27] <rue_house> W
[01:14:35] <rue_house> generates the ramps
[01:14:53] <katsmeow> ok, not as a gate then, gotcha
[01:15:22] <rue_house> I was having problems with freq control when using a schmitt on an RC
[01:15:47] <rue_house> with the low pass, I can prettymuch get whatever I want out of it
[01:16:13] <rue_house> well, its 1 hour past my bedtime!
[01:16:16] <rue_house> gnight :)
[01:16:23] <katsmeow> gnites
[22:47:28] * katsmeow wavers
[23:29:39] <katsmeow> so,, making a h-bridge with 100 MJ13007 (and a couple MJE5852 to make Szlakai pnp with some MJ13007) ?
[23:40:15] <katsmeow> 50amp peak h-bridge for ~$20 to make sine 115vac from the square waves coming in
[23:41:40] <katsmeow> the trick has been to get the best price per amps @ 25v drop across the CE
[23:41:54] <katsmeow> and not need gobs of base drive circuits