Back
[00:00:29] <^kat^> naturally
[00:00:34] <rue_house> the 7474 is getting a doubel clock pulse, that causes the output to not be right for dividing the pwm, which causes an output glitch
[00:01:11] <rue_house> the 393 is feeding a 7414, which there must be a signal bounce in
[00:03:04] <^kat^> the correct approach is to eliminate the cause of the glitch, so the ramp and the clock remain sync'd
[00:14:17] <rue_bed> indeed
[00:14:45] <^kat^> is the input to the 7414 ripply and slow?
[00:14:59] <rue_bed> I noticed last night that a pwm glitch was caused by a .7v bump in the 7414 input, so dropping the input .7V cleaned it up
[00:15:09] <rue_bed> I'll look for similar bumps
[00:15:22] <rue_bed> is anything slow at 250khz?
[00:15:49] <rue_bed> I wonder what slew times are like on lvds lines
[00:16:22] <^kat^> 1000v/us or more
[00:16:40] <rue_bed> the slew rate on pin 3 of a 555 is damn good
[00:16:49] <^kat^> why is there a .7v bump in the 393 output?
[00:16:57] <rue_bed> |-|
[00:17:01] <rue_bed> I dont know yet
[00:17:28] <rue_bed> I didn't finish tracing it out, some pin numbers are missing from my schematic
[00:17:44] <rue_bed> esp where I actually stopped following the schematic and started making things up
[00:17:50] <^kat^> is the first pulse of the double-clk in the right place?
[00:18:02] <rue_bed> I dont think the 12V version of my pwm osc had this issue
[00:18:17] <rue_bed> one of them is, they are about .1us apart
[00:18:46] <rue_bed> its truely a hairs width glitch
[00:18:59] <^kat^> that's on the output of the '14 tho, right?
[00:19:05] <rue_bed> yup
[00:19:14] <rue_bed> I didn't look at what was going in
[00:19:21] <rue_bed> from what I can tell..
[00:19:31] <rue_bed> the 393 has a 220R pullup
[00:19:50] <^kat^> what if you replace the '14 with a c04?
[00:19:57] <rue_bed> that pulled up line goes to the 74ls14 which drives the RC network back to the 393
[00:20:09] <rue_bed> would a c be fast enough?
[00:20:24] <^kat^> no, that's the point, it may ignore the 393's glitch
[00:20:39] <^kat^> and the C threshold is a volt higher
[00:20:59] <rue_bed> for some reason I decided to use the '14 to drive the RC for the timer, I dont know why, its output driving capability looks awefull from the specs
[00:21:10] <rue_bed> hmm
[00:21:31] <^kat^> um, the schematic you showed me uses a transistor to drive the RC for the 393
[00:21:35] <rue_bed> yea
[00:21:41] <rue_bed> apparently I decided to change that
[00:21:52] <rue_bed> but I think I was chasing the wrong goose
[00:22:06] <^kat^> well, crap, i cannot see the correct schematic, i may as well be copy/pasting random words from a book ehere
[00:22:17] <rue_bed> I'm wondering if I should re-breadboard things and test more
[00:22:26] <rue_bed> yea, me either
[00:22:40] <rue_bed> after the signal comes out of the 393 I'm not 100% where its going
[00:22:46] <rue_bed> I know its to the '14
[00:22:56] <rue_bed> but I dont know if its thru 2 gates or just 1
[00:23:20] <^kat^> then no one can help you debug what you actually have there
[00:23:40] <rue_bed> and I dont know if I tap or duplicate it for the square wave that drives the '74
[00:23:50] <rue_bed> I'll get more into tommorow
[01:09:32] <^kat^> ok
[23:10:36] <katsmeow-afk> i just took this pic from Tiggr5's screen :
http://designerthinking.com/images/DSCF6689m.jpg
