|
blazer status: put about 30 miles on it yesterday. we still have a WOT backfire through the carb but it is running better and better as I get braver and advance the timing. Tonight we pull off the air horn (top cover of carb) and double-check because I think I set the floats a little low, and it may be going lean when I open the secondaries. this is a dumb center float holley weirdo quadrajet style carb and nothing is "the holley way" on it. it takes weber jets ffs also started a rear cargo area/sleeping platform in Sketchup. may end up sweet as hell
|
# ? Feb 17, 2014 18:50 |
|
|
# ? Apr 26, 2024 14:56 |
|
Jonny 290 posted:blazer status: put about 30 miles on it yesterday. we still have a WOT backfire through the carb but it is running better and better as I get braver and advance the timing. what's WOT precious?
|
# ? Feb 17, 2014 18:58 |
|
wide open throttle, i.e., pedal to the metal
|
# ? Feb 17, 2014 19:04 |
|
web of trust
|
# ? Feb 17, 2014 20:45 |
|
actually it's war on terror and pretty much sounds the exact same CYLINDER 7 WAS AN INSIDE JOB
|
# ? Feb 17, 2014 20:49 |
|
Jonny 290 posted:actually it's war on terror and pretty much sounds the exact same The pushrods and rockers were in the conspiracy
|
# ? Feb 17, 2014 23:09 |
|
i went and saw the lego movie and then this... happened
|
# ? Feb 18, 2014 08:19 |
|
|
# ? Feb 18, 2014 08:29 |
|
pagancow your fan thing works good
|
# ? Feb 18, 2014 10:23 |
|
yah i'm thinking imma make one for myself
|
# ? Feb 19, 2014 01:33 |
|
Mathhole posted:pagancow your fan thing works good Did you mean me? Or did pagancow post a thing and I missed it?
|
# ? Feb 19, 2014 03:13 |
|
us file avatar people just merge into one mega poster, with 13 shades of poo poo in every hour
|
# ? Feb 19, 2014 03:40 |
|
brb getting my aladdin stuffit icon sorted
|
# ? Feb 19, 2014 03:54 |
|
echinopsis posted:us file avatar people just merge into one mega poster, with 13 shades of poo poo in every hour it's hard enough with that renaming thread shiiiit
|
# ? Feb 19, 2014 10:08 |
|
i wrote a touch gui w/ qt on top of my satools library: https://github.com/thismachinechills/AwfulPy there are a bunch of bugs and its p much an homage to dr bronner right now. also made a module that takes a username and pops out a trigram poster based on their post history. metal pink babble's generated text was funny. salisbury shake fucked around with this message at 05:28 on Feb 22, 2014 |
# ? Feb 22, 2014 04:55 |
|
So until today I didn't know WTF a biasing resistor was, so for my fume extractor I just wired the transistor's base straight into the MCU, and then wired the transistor's collector straight to +5V and the emitter straight into the fan motor. Apparently this is bad or something, IDK. Also, after a few minutes the power circuit's voltage regulator is almost too hot to touch, and the power circuit's reverse voltage protection diode is too hot to touch. So something is hosed up, somewhere, but hell if I know what. Checked really closely and there don't seem to be any shorts to ground, but when I used a little standalone 5V regulated PSU that I made it's voltage regulator was barely warm. The transistor itself is barely warm, and nothing else seems warm. Damnit. I'm tempted to just drill some holes in the enclosure and call it good...
|
# ? Feb 23, 2014 03:22 |
|
Doc Block posted:I'm tempted to just drill some holes in the enclosure and call it good... Part number of the diode you are using? You might need to heatshink those to the side of the case to dissipate the heat. It looks like one of those two part might be a TO-220 package so you need to make sure to insulate the tab from the grounded case. The tab is typically electrically hot and will be shorted without an insulator.
|
# ? Feb 23, 2014 04:54 |
|
The diode is a 1N4001. The project case is plastic, and yeah, nothing is touching the heatsink fin on the 7805 voltage regulator (the TO-220 IC at the bottom; the TO-220 IC at the top is the transistor). I'm measuring ~230mA current draw across the entire thing, but since the fan is getting PWM'ed that might not be super accurate (fan itself is rated for 260mA, so the whole thing is probably drawing way more than 230mA). edit: I also just realized that I don't have a back-EMF protection diode in place for the fan motor. edit2: I added the onboard power circuit at the last minute to save space; I'd already made a standalone one and it ran the whole thing just fine, with the voltage regulator and the diode barely getting warm. For the onboard PSU, I used a regulator with a much wimpier heatsink fin from a different manufacturer (so for all I know its absolute max current is 300mA or something), and the position of the filtering caps in the circuit is decidedly non-optimal. Doc Block fucked around with this message at 05:59 on Feb 23, 2014 |
# ? Feb 23, 2014 05:42 |
|
you can use an heat sink calculator to figure out if the one you have is adequate or not. but if the diode gets hot as well it's pulling way more current than it should. 1N400x does 1A forward current so there has to be a problem somewhere
|
# ? Feb 23, 2014 09:30 |
|
i'm wondering if the diode is just poo poo at heat dissipation. 9.2V (coming in from the wall wart) @ 0.23A is about 1.5 watts if i'm not mistaken. i would think that if there was a short somewhere my multimeter would show it drawing way more current.
|
# ? Feb 23, 2014 17:42 |
|
if its a reverse polarity diode why is it conducting at all
|
# ? Feb 23, 2014 19:02 |
|
i'm not sure about your ps problem but from your description of how you have your fan wired it sounds like this: if so that probably explains the 0.7V drop you saw, transistors turn on when Vbe voltage between base and emitter is roughly 1 diode drop so when your micro pin pulls the base up to +5 there's a 0.7V drop from base to emitter. the transistor ends up with Vce=Vbe if you want to reduce the drop try wiring low side, but then you have to have the base resistor: the Vbe drop will still want to be 0.7V, so the other 4.3V needs to drop across the resistor. you don't want too big a resistor for this so you get enough base current to turn all the way on (4.3ma, check the current rating on your micros outputs but it'll usually be 10-20ma). if it works like expected the current through your power supply will increase, though. more heat... (i sort of assumed the 2nd before)
|
# ? Feb 23, 2014 19:12 |
|
Bloody posted:if its a reverse polarity diode why is it conducting at all It's a reverse polarity protection diode. Basically, it doesn't allow current to flow if I accidentally hook the power up backwards. Useless now (unless I go and try to hook up a wall wart whose power plug isn't center-positive), but cheap insurance when I was testing the circuit. Doc Block fucked around with this message at 20:48 on Feb 23, 2014 |
# ? Feb 23, 2014 20:39 |
|
Base Emitter posted:if you want to reduce the drop try wiring low side, but then you have to have the base resistor: the Vbe drop will still want to be 0.7V, so the other 4.3V needs to drop across the resistor. you don't want too big a resistor for this so you get enough base current to turn all the way on (4.3ma, check the current rating on your micros outputs but it'll usually be 10-20ma). if it works like expected the current through your power supply will increase, though. more heat... or just use a mosfet, jeez. no one uses bjts for this sort of thing any more mr. base emitter
|
# ? Feb 23, 2014 20:43 |
|
though with a name like that you could easily be aware of some effect that i have no concept of and i'm just talking out my rear end
|
# ? Feb 23, 2014 20:45 |
|
Was messing around with it again. If I plug in power via a battery (I've got a 9V battery pack with a barrel jack), the voltage regulator and diode are both cool to the touch. Only when power is coming via a wall wart do they heat up. When I tested it with my little standalone PSU, I was using the battery pack. Hooking the standalone PSU to the wall wart and delivering power straight to the fan results in the standalone's voltage regulator also getting hot. The amount of current being drawn is the same, though vv So I think for Fume Extractor 2.0 I'm going to see if I can find a beefier diode and place the filtering capacitors a little better, and put a back-EMF protection diode in place for the fan motor.
|
# ? Feb 23, 2014 20:49 |
|
Sagebrush posted:or just use a mosfet, jeez. no one uses bjts for this sort of thing any more mr. base emitter IIRC BJTs can switch faster than MOSFETs, so if you're PWM'ing stuff wouldn't a BJT be better?
|
# ? Feb 23, 2014 20:51 |
|
Sagebrush posted:or just use a mosfet, jeez. no one uses bjts for this sort of thing any more mr. base emitter some of us still have boxes of parts from the 80s to get through (that was agreed upon before but he's still using a bipolar) oh new post: Doc Block posted:IIRC BJTs can switch faster than MOSFETs, so if you're PWM'ing stuff wouldn't a BJT be better? this will vary depending on device but unless you are using a really high frequency it shouldn't really be a big deal
|
# ? Feb 23, 2014 20:56 |
|
"faster" is relative. you're not switching at RF frequencies. even the slowest mosfet is plenty fast enough for pwm voltage control
|
# ? Feb 23, 2014 20:57 |
|
Doc Block posted:IIRC BJTs can switch faster than MOSFETs, so if you're PWM'ing stuff wouldn't a BJT be better? i've used mosfets for pwming motors at >20khz (out of audio range was the goal) and i don't think that's even approaching the theoretical limit
|
# ? Feb 23, 2014 20:58 |
|
Ok then. So for a MOSFET in low-side configuration, I can just hook the gate to the MCU and a pull-down resistor and be fine? Like this: edit: I should mention that the MOSFET I have has a 2.5V gate threshold, so I shouldn't need a charge pump or whatever, right? Doc Block fucked around with this message at 21:43 on Feb 23, 2014 |
# ? Feb 23, 2014 21:20 |
|
well thanks i think
|
# ? Feb 23, 2014 21:33 |
|
Doc Block posted:Ok then. So for a MOSFET in low-side configuration, I can just hook the gate to the MCU and a pull-down resistor and be fine? yep, that looks right to me. while it is an incorrect use of the terminology, i remember that the drain is connected to the power drain (ie., the load), the source is connected to the power source (the vcc or the ground), and an N-type goes on the negative side of the load and a P-type goes on the positive side. i bet this oversimplifies the activity of a mosfet in some way that will lead to problems in one of my circuits down the line but for now eh it seems to work and make sense.
|
# ? Feb 23, 2014 21:46 |
|
ooooh one of those people who draws circles around your transistors
|
# ? Feb 23, 2014 21:48 |
|
Base Emitter posted:ooooh one of those people who draws circles around your transistors
|
# ? Feb 23, 2014 22:20 |
|
Doc Block posted:It's a reverse polarity protection diode. Basically, it doesn't allow current to flow if I accidentally hook the power up backwards. Useless now (unless I go and try to hook up a wall wart whose power plug isn't center-positive), but cheap insurance when I was testing the circuit. ya but why not put it so its reverse biased across the supply under normal conditions and conducting if you connect your supply backwards
|
# ? Feb 23, 2014 23:05 |
|
Bloody posted:ya but why not put it so its reverse biased across the supply under normal conditions and conducting if you connect your supply backwards because then if and when the supply is backwards its effectively shorted and the diode is carrying the full current the supply will generate in series the diode has no current when the supply is connected backwards, and normal current when connected correctly.
|
# ? Feb 23, 2014 23:11 |
|
Ordered some TO-220 heatsinks. Gonna see about getting a PCB made too, so I can have a big ground pour for heatsinking. edit: at $5/square inch maybe I'll wait on the PCB. Doc Block fucked around with this message at 01:31 on Feb 24, 2014 |
# ? Feb 24, 2014 00:46 |
|
https://gist.github.com/limnick/9204663 flapmmo hax
|
# ? Feb 25, 2014 08:54 |
|
|
# ? Apr 26, 2024 14:56 |
|
i got really bored and started writing a basic VERB NOUN text adventure parser in pythonquote:E:\Programming\Python\textadventure>python textadventure.py
|
# ? Feb 25, 2014 23:46 |