[QUOTE=ddrl46;42231997]It's easier to see a change in voltage / current on an analog meter than a digital one.[/QUOTE]
That's a pretty general claim, personal preference and experience is probably more relevant for that.
Bah, I'll keep using cheap meters, unless for situations that downright [B]DEMAND [/B]absolute precision.
[I]THEN [/I]I might dig out my Fluke meter.
[QUOTE=Van-man;42233878]Bah, I'll keep using cheap meters, unless for situations that downright [B]DEMAND [/B]absolute precision.
[I]THEN [/I]I might dig out my Fluke meter.[/QUOTE]
I'll just use my oscilloscope as a multimeter!!
[QUOTE=DrDevil;42234288]I'll just use my oscilloscope as a multimeter!![/QUOTE]
Pfft, I've got some CERN-quality O-Scopes have resolution down to 1 picovolt, get on my level. :v:
[QUOTE=DrDevil;42231894]If you prefer a manual ranging meter, then your auto ranging meter is crap.[/QUOTE]
But you should know what to expect to read?
And if it comes to changing voltages, I wouldn't use a multimeter anyway. Then I'd use my scope.
[QUOTE=BuG;42237504]But you should know what to expect to read?
And if it comes to changing voltages, I wouldn't use a multimeter anyway. Then I'd use my scope.[/QUOTE]
Not always, sometimes you have an unknown power supply and you want to measure it. Also voltages can slowly change over time, like measuring a battery as it discharges.
I am probably getting too ambitious but would it be possible to make a robot with voice recognition that responds to certain phrases with sound, lighting leds in a certain color, an output on a lcd screen and movement? not without effort and learning of course but without having to have a master in ee and spending a lot of money? maybe even with an ai.
[QUOTE=Desuh;42237763]I am probably getting too ambitious but would it be possible to make a robot with voice recognition that responds to certain phrases with sound, lighting leds in a certain color, an output on a lcd screen and movement? not without effort and learning of course but without having to have a master in ee and spending a lot of money? maybe even with an ai.[/QUOTE]
Short answer: No
Long answer: The electronics part is really easy, the super hard thing will be the software. In essence, you'll need a multidisciplinary team of geniuses to pull that off. Especially without spending a lot of money.
I'd approach it by using an existing speech recognition engine (Maybe Google's with the API) and keeping it simple, <call> <keyword> <parameters>.
"Chubby, Move Forward"
Using such a simple protocol, in c#, I'd create a list of delegates which were handlers for each keyword, they would be passed the parameters and return the result.
I'd run it on a Pi and use the GPIO pins to interface with the electronics.
At that level, at least, anyone with a reasonable programming background would be able to pull that off. The AI is where things get tricky, accounting for the unexpected.
[QUOTE=DrDevil;42237956]Short answer: No
Long answer: The electronics part is really easy, the super hard thing will be the software. In essence, you'll need a multidisciplinary team of geniuses to pull that off. Especially without spending a lot of money.[/QUOTE]
Well then maybe something easier. Just have the robot calibrated on one voice and certain commands. Like shout "move" so he starts moving. Or "stop" so he stops moving. Would that be possible with an arduino?
[QUOTE=Desuh;42238155]Well then maybe something easier. Just have the robot calibrated on one voice and certain commands. Like shout "move" so he starts moving. Or "stop" so he stops moving. Would that be possible with an arduino?[/QUOTE]
No.
[QUOTE=Desuh;42238155]Well then maybe something easier. Just have the robot calibrated on one voice and certain commands. Like shout "move" so he starts moving. Or "stop" so he stops moving. Would that be possible with an arduino?[/QUOTE]
Digital signal processing can't really be done on limited resource systems such as arduino or individual AVRs/PICs.
I'd recommend something with a stronger processor such as a BeagleBoard or a RPi.
Hell, if you are feeling adventurous, you can probably get away with voice recognition on a MSP430 by writing your own software with breaking down input signals and using the rather nice amount of memory to solely process a single voice command. The MSP430 has more memory and better signal processing resources than the Arduino by a longshot.
But for starting out, I'd just pass the audio file to a computer and let some precoded software decode it and send commands back. Try looking into RoboRealm for starters
[QUOTE=LoneWolf_Recon;42239684]Digital signal processing can't really be done on limited resource systems such as arduino or individual AVRs/PICs.
I'd recommend something with a stronger processor such as a BeagleBoard or a RPi.
Hell, if you are feeling adventurous, you can probably get away with voice recognition on a MSP430 by writing your own software with breaking down input signals and using the rather nice amount of memory to solely process a single voice command. The MSP430 has more memory and better signal processing resources than the Arduino by a longshot.
But for starting out, I'd just pass the audio file to a computer and let some precoded software decode it and send commands back. Try looking into RoboRealm for starters[/QUOTE]
Doing voice recognition on an AVR is certainly possible, but it will be the only thing it'll be doing.
[QUOTE=DrDevil;42239732]Doing voice recognition on an AVR is certainly possible, but it will be the only thing it'll be doing.[/QUOTE]
Alone yes but it'd need considerably better ADC resources that are fast with conversions and have fast/large memory to store the data so probably a ATmega644.
This PSoC4 kit must be one of the best eval boards I've ever had! It even has a tiny hook to connect the ground of your oscilloscope probe to, and they put on little rubber feets on the bottom. For 30€ this is absolutely brilliant. You can get it on Farnell.
[img]http://i.imgur.com/wwIfuGj.jpg[/img]
I have a 2222 Transistor hooked up to a 12v power supply, I connect the drain to the light strip on each of the 3 channels R, G and B. I then used the 2222 transistor connected to the Arduino 9th pin and using PWM and then the other pin on the transistor connected to the ground on the power supply, yet I get some funny results
Any ideas?
[t]http://i.imgur.com/alxwB79.jpg[/t]
[t]http://i.imgur.com/qkpnR8J.jpg[/t]
[t]http://i.imgur.com/9wUB854.jpg[/t]
As you can see, some of them go purple, if I disconnect the red channel they stay on but are very faint. If I just use the ground from the arduino just 4 leds come on.
It's an m331 transistor you knob.
[QUOTE=DrDevil;42257710]It's an m331 transistor you knob.[/QUOTE]
Jamie: 2N
Jamie: 2222
Jamie: H 331
is what is says on the transistor
[QUOTE=jamie1130;42258341]Jamie: 2N
Jamie: 2222
Jamie: H 331
is what is says on the transistor[/QUOTE]
A schematic would help, also do you have some resistors somewhere in there? Preferably b/w the arduino and transistor as well as b/w the transistor and the lights
[QUOTE=LoneWolf_Recon;42258907]A schematic would help, also do you have some resistors somewhere in there? Preferably b/w the arduino and transistor as well as b/w the transistor and the lights[/QUOTE]
There was no transistor, I just put one in after the drain to see what happens and it just turns on 1 LED on the blue channel. Haven't really got a schematic, all it is an arduino, led strip and that transistor.
So what is keeping the transistor from pulling more than the like 20mA the arduino pins are rated for?
Nearly any schematic you will ever use will need a resistor somewhere. For us to fully understand how you have things hooked up in order to troubleshoot you should learn to quickly fashion up a basic schematic in eagle or something.
Errm drain ?
The 2N2222 is a NPN bipolar junction transistor not a FET, as a result it requires a base current limiting resistor.
Have a read of this:
[url]https://mega.co.nz/#!ipx03QQL!GgIfAQjbLEqX7obQLm-kc0WsYF91prtAzE_hZXzJWgs[/url]
So for my Electronics coursework, I've been given the task of creating a circuit that counts the number of visitors to a museum or something. I've decided that the counter should be two digits, and the current count should be output on two seven segment displays.
I've got a few ideas on how I could go about doing it, but I'd appreciate some input on what would be best/any other ways I could do it.
1) A combination of BCD Decade counters, BCD Decoders and an AND gate (The first circuit increments when the second circuit increments past 9, controlled by the AND gate).
2) Use a PICAXE chip to handle the counting/output of the signals to the seven segment displays.
3) Create a shield for an Arduino and have the seven segment displays/input PTM switch soldered to that, this is probably my favourite idea since i've got a couple of Arduinos and I've used them before.
#2 and #3 would be the easiest, particularly if you want to implement any fancy features.
#1 would require the most components but on the upside it'll show you really know what you're doing.
Old style BCD counters are still very much used in industry due to their high reliability.
[QUOTE=Chryseus;42272427]#2 and #3 would be the easiest, particularly if you want to implement any fancy features.
#1 would require the most components but on the upside it'll show you really know what you're doing.
Old style BCD counters are still very much used in industry due to their high reliability.[/QUOTE]
Both grade A examples i've been shown use PICAXE chips so it's possible to get high marks by using solutions like #2/#3. I found a blog post where somebody used shift registers to output a number to 4 seven segment displays and there only needs to be minimal changes made to the code to make it work.
This is the circuit diagram for a "High A grade" - [url]http://puu.sh/4xtym.png[/url]
Diagram for a "Low A grade" - [url]http://puu.sh/4xtBZ.png[/url]
This is the circuit diagram I found in the blog post (pretty much exactly what I want) - [url]http://puu.sh/4xtHn.png[/url]
I had to do a 3 bit 8x3 decoder to a 3x8 encoder for a 7 segment led for a homework problem.
My result looks like something that would come from a 4 star italian chef.
[t]http://i.imgur.com/0q3MIs8.png[/t]
Am I an electrical engineer now.
I won a new in the box Fluke 77-IV for $110 with shipping on eBay, seems like I got a pretty good deal on it but I feel pretty pessimistic about something being wrong since I got it for that price.
I won by $0.50, which was a bit of an adrenaline rush.
edit:
Shit, I think I got the wrong thing. It seems to be for very high voltage since it can't measure under 6V, could someone give me a definitive answer on if I bought the wrong multimeter?
[url]http://www.fluke-direct.com/shop/itemDetail.do?itm_id=307930&itm_index=0&item=77-4[/url]
[QUOTE=Cittidel;42277811]I won a new in the box Fluke 77-IV for $110 with shipping on eBay, seems like I got a pretty good deal on it but I feel pretty pessimistic about something being wrong since I got it for that price.
I won by $0.50, which was a bit of an adrenaline rush.
edit:
Shit, I think I got the wrong thing. It seems to be for very high voltage since it can't measure under 6V, could someone give me a definitive answer on if I bought the wrong multimeter?
[url]http://www.fluke-direct.com/shop/itemDetail.do?itm_id=307930&itm_index=0&item=77-4[/url][/QUOTE]
Thats the lowest voltage range on dc volts (so it doesn't autorange below that), there should also be a milivolt range on that meter.
So no, you didn't buy the wrong multimeter.
[QUOTE=ddrl46;42279271]Thats the lowest voltage range on dc volts (so it doesn't autorange below that), there should also be a milivolt range on that meter.
So no, you didn't buy the wrong multimeter.[/QUOTE]
Thanks, you're right. [url=http://coast-coast.co.za/Fluke/Digital%20multimeters/F-77-IV_01a_h_72dpi_778x1024px_E.jpg]I can't believe I missed that,[/url] I guess I just freaked out.
Sorry, you need to Log In to post a reply to this thread.
