#garfield | Logs for 2016-09-01

Back
[00:00:21] <rue_house> did you hear about hte china shutdown for the g20?
[00:00:48] <katsmeow> no
[00:02:53] <rue_house> hmm
[00:03:08] <rue_house> 20V is, suspiciously, close to 5V over 12V
[00:03:59] <rue_house> I wonder if I should, for the base drive of the fets, pump a 12V by 5V
[00:04:00] <katsmeow> you can use that to your advantage sometimes
[00:04:19] <katsmeow> 5v is not much gate drive
[00:04:37] <rue_house> no, but 17 is pretty good
[00:05:00] <katsmeow> 17 gate drive might break them
[00:05:01] <rue_house> esp if your fet is so tiny you dont have to worry about excessive capacitance
[00:05:07] <rue_house> they have 20V gates
[00:05:11] <katsmeow> k
[00:05:12] <rue_house> "surprise"
[00:05:25] * katsmeow aims for 10-15
[00:05:50] <rue_house> yes, however, I'm using a high-side N channel on a 6V power rail
[00:06:09] <katsmeow> you can always use a switched cap on the 12v line to high side 12v systems to 10v gate
[00:06:16] <rue_house> so I'm somewhere between 17 and 11 as it switches
[00:06:38] <rue_house> bootstrap it
[00:07:02] <rue_house> its for a pullup resistor
[00:07:04] * katsmeow nods
[00:07:06] <rue_house> hmmm
[00:11:06] <rue_house> hard to mentally model that, so much happens at once
[00:11:39] <katsmeow> stretch your brain, ye must
[00:11:44] <rue_house> but I dont see it not working
[00:12:07] <rue_house> tho it does add 16 diodes, and capacitors to the system
[00:12:14] <rue_house> I dont think the caps need to be big
[00:12:22] <rue_house> those are tiny fets
[00:12:31] <rue_house> prolly even 0.1uF would work
[00:14:01] <katsmeow> and tiny fast diodes too, like 1n4148
[00:14:29] <katsmeow> iirc, they are like 500 for $1
[00:15:44] <rue_house> I'll put one togethor and see that it works ok
[00:16:53] <katsmeow> 2n2222 and 2n2709 have the cheapest-lowest Vsat for a to-92
[00:17:13] <katsmeow> er, 2907
[00:17:20] <rue_house> ok, these motors pull more than an amp tho
[00:17:32] <katsmeow> but the gates don't pull one amp
[00:17:44] <rue_house> which I assume they cant maintain for long, being such small motors
[00:18:17] <rue_house> and I need really low votlage drop, less than 0.7V from the 6V supply
[00:18:21] <katsmeow> but the gates don't pull one amp
[00:18:36] <rue_house> with those tiny fets, I'm at 0.4
[00:18:42] <katsmeow> but the gates don't pull one amp
[00:18:49] <katsmeow> but the gates don't pull one amp
[00:18:50] <rue_house> indeed, no
[00:19:00] <rue_house> but too much votlage drop for this
[00:19:24] <katsmeow> how can a drop of 100mv at the gate be too much for anything??
[00:19:44] <rue_house> the bipolar?
[00:20:06] <rue_house> an H bridge with 2222 and 2907 drops more than 0.7...
[00:20:14] <katsmeow> yeas, bipolar as the boostrapped gate drive
[00:20:20] <katsmeow> fuck
[00:20:37] <rue_house> oh, its ok, I'm just gonna use a 2308 with a 1k pullup
[00:20:47] <rue_house> works great
[00:21:10] <rue_house> sorry, I worked till 1:30am last night and havn't recovered
[00:21:35] <rue_house> pomg
[00:22:24] <rue_house> I cant find a src for the shift register based open collector load drivers that has a reasonable price
[00:22:41] <rue_house> 74595+uln2308 in one
[00:23:30] <rue_house> I think they come up alot in inkjet printers, but I cant get datasheets
[00:55:29] <rue_house> yay, my tiny13 adc to pwm code works
[01:00:16] <katsmeow> are you bidding on thos 4538's?
[01:00:45] <rue_house> ... no
[01:00:52] <rue_house> remind e what they are?
[01:01:07] <rue_house> buffer isn't it
[01:01:27] <rue_house> monostable, huh
[01:01:34] <katsmeow> no, dual cmos one-shots
[01:01:39] <katsmeow> retriggerable even
[01:04:37] <rue_house> they dont say what vref1 and 2 are
[01:05:08] <katsmeow> if i parallel 5 MMBT2907 to get an amp of capacity, it will cost me 5.76 cents, and be soooooo much easier to drive their bases vs fet gates
[01:05:27] <katsmeow> the data sheet does
[01:06:06] <rue_house> whats evil about eh fet gates?
[01:06:06] <katsmeow> i still prefer the olde '221, '123, '122
[01:06:20] <katsmeow> large voltage and high capacitance
[01:06:31] <rue_house> aI assume the ttl is fater too.... does that even apply?
[01:06:44] <katsmeow> fatter how?
[01:06:53] <rue_house> faster
[01:07:02] <katsmeow> yeas
[01:07:32] <katsmeow> for motor driving it prolly doesn't matter, but for video processing, it matters
[01:07:42] <rue_house> hmm
[01:08:43] <rue_house> they dont say what the ref voltages are for the comparitors
[01:08:46] <katsmeow> or for making more clock phases
[01:08:50] <rue_house> I just dont see it in here
[01:11:07] <rue_house> k, well, gnight
[01:11:17] <katsmeow> wait the 74hc4538 ?
[01:11:38] <rue_house> I'v not needed a monostable in a long long time
[01:11:41] <katsmeow> what Vref ?
[01:12:02] <rue_house> the datasheet shows 2 comparitors, they each hook to a vref
[01:12:16] <rue_house> but nowhere in the datasheet do they say what the vrefs ARE
[01:12:32] <rue_house> vcc/3 or anything
[01:13:11] <katsmeow> we have different Vref, the compataor inputs on my datasheet are all internal
[01:13:21] <rue_house> I just used a microcontroller to do 2 0-5V -> pwm channels
[01:13:36] <rue_house> yea, thats what I mean
[01:13:52] <rue_house> so you dont know the trigger votlages without knowing what the comparitors are working against
[01:14:30] <katsmeow> it's a 74HC part, run it ay 5 to 6v and forget it
[01:14:42] <rue_house> :)
[01:15:04] <rue_house> these little microcontrollers are handy tho
[01:15:33] <rue_house> I'm gonna use that analog->pwm for the new motor on my power wheelbarrow
[01:15:36] <katsmeow> configurable building blocks, just slower than dedicated circuits
[01:15:53] <rue_house> I just build a little water-proofable slider a few days ago
[01:16:24] <rue_house> not by much, the uC is 4.8Mhz, the pwm is 600Hz
[01:16:36] <katsmeow> i used a '221 as a pwm , used a transistor for the pullup instead of a resistor
[01:16:38] <rue_house> the adc is faster than 600Hz
[01:17:12] <rue_house> so for all intents and purposes, for that, the tiny13 is as fast as it gets
[01:17:34] <katsmeow> k
[01:17:57] <rue_house> I suppose it is buffering 1 pwm cycle, which a dual op-amp in an 8 pin dip wouldn't
[01:18:08] <rue_house> but I think thats just latency
[01:18:51] <rue_house> gnight!
[01:18:58] <rue_house> good talkin to ya
[01:18:59] <katsmeow> gnites
[08:50:49] <jennie> hello
[09:18:18] <rue_house> 867 4309 jenny?
[09:18:23] <rue_house> 5309
[09:18:29] <rue_house> 867 5309...
[09:49:02] <jennie> hi rue_house q
[09:49:04] <jennie> how are you
[09:50:10] <rue_house> I dont know how it is that I am, the odds have been so low
[09:52:03] <jennie> Can you please tell what size is this DC jack? http://i.imgur.com/7hYhhS4.jpg
[09:57:33] <Tom_itx> jennie is new here...
[09:58:15] <Tom_itx> 2.5mm?
[11:12:19] <jennie> Tom_itx: how did you find that its 2.5mm?"
[13:20:35] <Tom_itx> looked at your ruler
[13:51:56] <jennie> Tom_itx: how about the dc jack diameter? is it 6.5diameter for 5.5 plug?
[20:11:59] <rue_house> use the common size that is easily available from china