I got my GSM shield + Arduino UNO clone package for my car and a couple ESP8266 modules in the mail today. Unfortunately, the library I found for the SIM900 shield doesn't seem to work, so I'll need to write my own functions to replace them. I'm also going to play around with power saving features, I'm a little hesitant to leave everything on 24/7 draining my car battery.
School doesn't have any regular good-old logic MOSFETs, they're all weird power MOSFETs that I'm not sure how they would behave.
I'm going to just skip the transistors. It'll also cut down on the amount of crap I need to connect, and the holes I need to drill, which is always nice, because holes usually end up being ugly.
I solved the "LEDs and shit are on when nothing is controlling the pins" problem by taking an IDE cable and using that.
It'll also make the entire thing smaller.
Now I just need to... redesign my entire PCB.
[editline]2nd October 2014[/editline]
Also,
[QUOTE=r0b0tsquid;46124715]You could have solved the default-on problem by putting a pullup resistor on the 595's OE# pin, which means the outputs will float (not be connected to vcc or ground) until your code starts and grounds that pin.
[img]http://i.imgur.com/tUKjQz7.png[/img]
[/QUOTE]
I so missed that when reading the datasheet, for some reason.
If there's any lesson I hope I can pass to other people, it's that modularity is the only thing at the end of the day that really counts.
Your thing might be awfully pretty, but if I can't plug it into something and make it go, I don't care.
Having a little hole you stick a cable through is acceptable in a lot of cases, but you have to recognize when you need to just grab a standard header, that you can grab any cable for, and stick it in.
[QUOTE=nikomo;46131055]School doesn't have any regular good-old logic MOSFETs, they're all weird power MOSFETs that I'm not sure how they would behave.
I'm going to just skip the transistors. It'll also cut down on the amount of crap I need to connect, and the holes I need to drill, which is always nice, because holes usually end up being ugly.
I solved the "LEDs and shit are on when nothing is controlling the pins" problem by taking an IDE cable and using that.
It'll also make the entire thing smaller.
Now I just need to... redesign my entire PCB.
[editline]2nd October 2014[/editline]
Also,
I so missed that when reading the datasheet, for some reason.[/QUOTE]
Power FETs tend to have higher threshold voltages (Vgs) to switch them on, and will switch more slowly. (sub 100kHz really for any large amount of power).
You're not removing the shift register as well are you? That was a cool idea.
That day you realize you need a third multimeter to complete a task.
[QUOTE=pentium;46135681]That day you realize you need a third multimeter to complete a task.[/QUOTE]
Thats one of the (mainy) reasons why i have an oscilliscope :) 4 channel = 4 measurements in AC or DC, true RMS capability and so on,
then 2 multimeters for current, should i need more, i have a current shunt somewhere :)
I have noticed people seem to forget that an oscilliscope is actually a REALLY accurate volt-meter and can also do really accurate current measurements. It truely is my fav tool :)
I really want a scope, just don't need one yet.
[QUOTE=nuttyboffin;46135800]Thats why i have an oscilliscope :) 4 channel = 4 measurements in AC or DC, true RMS capability and so on,
then 2 multimeters for current, should i need more, i have a current shunt somewhere :)
I have noticed people seem to forget that an oscilliscope is actually a REALLY accurate volt-meter and can also do really accurate current measurements. It truely is my fav tool :)[/QUOTE]
I don't trust my scope around high voltages unless I have the appropriate probes.
Gee, I wish both of these fucking flukes on my bench worked.
Not really that accurate with an 8 bit ADC, then again usually you don't need to be, I use my vintage analog meter for most things.
[QUOTE=pentium;46136013]I don't trust my scope around high voltages unless I have the appropriate probes.[/QUOTE]
You can use an isolation transformer which is much cheaper than differential probes, or if you like a bit of danger you can float the scope.
[QUOTE=r0b0tsquid;46135443]Power FETs tend to have higher threshold voltages (Vgs) to switch them on, and will switch more slowly. (sub 100kHz really for any large amount of power).
You're not removing the shift register as well are you? That was a cool idea.[/QUOTE]
Here's the current (and last, probably) revision of the board, I already did the UV stuff today at school, but we have a 4-hour day, so I'm going to do the acid etching on Monday.
[img]http://i.imgur.com/cRXKnDe.png[/img]
I also switched the MOSFETs controlling the RGB LED to a simple NPN transistors (BC547C, school has like truckloads of them).
The 40-pin header is just something that a standard ATA connector plugs into.
On the other end, I drilled a hole so the one extra pin on the Pi (when compared to an IDE connector) will fit, and on the other end, the connector is weird little 90 degree angle metal things, so I just ripped it out, so the cable can be standard on one end.
You can also just buy 40-pin headers, and crimp your own cable, it's entirely standard, you can get a bag of the headers for under 2€ on eBay.
The design ended up being slightly clever, I have to have resistors on the bases of the NPN transistors, so I just used them to jump a gap and simplify it a bit.
Looks good. You might still want to add a resistor of about 10k between the OE# pin on the 595 and your positive rail, to keep the outputs floating until your code starts.
The idea is that I plug the board after I have control.
But I'll stick a through-hole resistor in there, since it's easy to do.
[QUOTE=r0b0tsquid;46140861]Looks good. You might still want to add a resistor of about 10k between the OE# pin on the 595 and your positive rail, to keep the outputs floating until your code starts.[/QUOTE]
You mean to prevent it from floating, right?
[QUOTE=DrDevil;46141366]You mean to prevent it from floating, right?[/QUOTE]
To keep the OE# [I]from [/I]floating, and to keep the outputs [I]floating [/I](i.e. hi-Z) until you ground it
[editline]3rd October 2014[/editline]
To keep the LEDs off (it's a common anode display)
Also, I just got informed that my school is eligible for the sample thing that Microchip has (that I didn't know about), where you can get 9 samples a month for free (3pcs x 3 different products).
You guys got suggestions? I won't be even learning about microcontrollers for a while, but, you know, free stuff, and it's once a month, so why not?
School has a few PICkit 3 programming flasher thingies, so something that works with those would be nice.
[QUOTE=nikomo;46141685]Also, I just got informed that my school is eligible for the sample thing that Microchip has (that I didn't know about), where you can get 9 samples a month for free (3pcs x 3 different products).
You guys got suggestions? I won't be even learning about microcontrollers for a while, but, you know, free stuff, and it's once a month, so why not?
School has a few PICkit 3 programming flasher thingies, so something that works with those would be nice.[/QUOTE]
Well, stocking up on PICs would be a start...
And wifi modules. I take it your school supplies stuff like resistors and capacitors and all that jazz? Can never go wrong with spare voltage regulations.
Aww, they don't give samples for any of the WiFi related stuff.
And I'm limited to 2 free sample orders right now.
I'm just going to check the model number for the 8-bit PICs that they use at school, on Monday, and at least get some of those.
New toy after a fuckton of cleaning!
[url=http://i.imgur.com/TPxbwOt.jpg][img]http://i.imgur.com/TPxbwOtl.jpg[/img][/url]
I've got something else that's really nice as well but I'll show that once I've fixed that.
Lovely, think I just fried a 10W audio amp that took a month to get here from china. You'd think they'd have some kind of protection against reverse voltage, especially when they don't tell you which pin is positive.
If all the chip does is get really hot, does that mean it's probably done for?
I finished most of the hardware for my SMS controlled remote car starter. I'm just waiting on a buck converter from China so I can actually power it with the car. Once I have everything, I just need to find a case for everything.
[t]http://i.imgur.com/oZ2TmYv.jpg[/t]
The temperature sensor appears to work:
[img]http://i.imgur.com/0vh1fbA.png[/img]
On the software end of things, I have all the features I want implemented using the shield. I just need to bring the power consumption down for peace of mind. Without putting any code in to reduce power consumption, it seems to idle at about 0.07 amps. If I shut down the GSM module, I can get it down to about 0.03, though I might even be able to bring that down by sleeping the Arduino.
[QUOTE=papkee;46153210]Lovely, think I just fried a 10W audio amp that took a month to get here from china. You'd think they'd have some kind of protection against reverse voltage, especially when they don't tell you which pin is positive.
[b]If all the chip does is get really hot, does that mean it's probably done for?[/b][/QUOTE]
Yes but it's probably repairable.
[QUOTE=chipset;46154538]Yes but it's probably repairable.[/QUOTE]
Well it's one of [URL="http://www.ebay.com/itm/321377011603"]these[/URL], so I have neither the skill nor the tools to be able to fix any of those smd components.
for a uni project we gotta work with PLC's and servo's, which results in an issue revolving around the used voltages. The PLC's outputs send a signal of 24 V while the servo's run between 4.8 to 6 V (we're going with 5V as we've noticed that the servo's twitch a bit too much at higher voltages and it's a default).
What is a good way to convert the voltage level? We've been thinking of using voltage dividers, but our teacher pointed us to logic level converters. Are they (LLC's) more appopriate or are there even other solutions which work just as fine and maybe easier to use?
Eagle is seriously driving me mad.
[t]http://i.cubeupload.com/iZeKHA.png[/t]
The ground side of C7 is floating, it don't even let me go in and wire it..
[QUOTE=papkee;46153210]Lovely, think I just fried a 10W audio amp that took a month to get here from china. You'd think they'd have some kind of protection against reverse voltage, especially when they don't tell you which pin is positive.
If all the chip does is get really hot, does that mean it's probably done for?[/QUOTE]
Oh hey, you're in the same boat as me now with my Spectrum+.
Fuck you Sinclair. It was one diode. ONE DIODE!!!
[QUOTE=Chryseus;46159998]Eagle is seriously driving me mad.
[t]http://i.cubeupload.com/iZeKHA.png[/t]
The ground side of C7 is floating, it don't even let me go in and wire it..[/QUOTE]
Did you connect it to ground in the schematic? Double check if the wire really is connected to the cap, or of it's just sitting on the same spot without being connected.
Use the show tool and click on any ground connection on the board, your pad should light up. If it doesn't, it's not connected to GND.
Got a weird issue going on here.
I need to replace some resistors but I don't have matching components both in rating and wattage.
To make a 2.2K 1/2W resistor I took two 2.2k 1/4W resistors and put them in parallel. To make a 33k 1/2W I took a 22k, a 10k and a 1k in series and tied a second set in parallel.
In both cases the value of the substitute was half. The new 2.2k was 1.1k and the new 33k was 16.5K.
[QUOTE=pentium;46170839]Got a weird issue going on here.
I need to replace some resistors but I don't have matching components both in rating and wattage.
To make a 2.2K 1/2W resistor I took two 2.2k 1/4W resistors and put them in parallel. To make a 33k 1/2W I took a 22k, a 10k and a 1k in series and tied a second set in parallel.
In both cases the value of the substitute was half. The new 2.2k was 1.1k and the new 33k was 16.5K.[/QUOTE]
Pretty sure that's how resistors work
Basic resistor theory? :v
[QUOTE=pentium;46170839]Got a weird issue going on here.
I need to replace some resistors but I don't have matching components both in rating and wattage.
To make a 2.2K 1/2W resistor I took two 2.2k 1/4W resistors and put them in parallel. To make a 33k 1/2W I took a 22k, a 10k and a 1k in series and tied a second set in parallel.
In both cases the value of the substitute was half. The new 2.2k was 1.1k and the new 33k was 16.5K.[/QUOTE]
Yeah, if you put resistors in parallel, the total resistance is lower. Imagine it like a 4-lane highway being bottlenecked by a single lane highway vs. a double lane highway
Sorry, you need to Log In to post a reply to this thread.
