-snip- actually meant to snip my post above but I can't remember what I put here.
[QUOTE=scratch (nl);44728971]Pin 13 (atleast on the official Arduino) has a built in resistor, so you can plug the LED straight onto the thing.[/QUOTE]
Pin 13 has a resistor and an LED, however the output from the Pin is the output from the processor so if you connect another LED to it you need an other LED AND another resistor, just look at the schematic
[editline]5th May 2014[/editline]
[url]http://forum.arduino.cc/index.php/topic,3616.0.html[/url]
[editline]5th May 2014[/editline]
I.e the resistor is only going to the built in LED, not to pin 13
[QUOTE=wakeboarderCWB;44719841]Bought myself an Arduino Leonardo and LCD module to play around with.
[thumb]http://puu.sh/8yO5G.JPG[/thumb]
First time doing this kind of programming and dealing with microcontroller boards.. Really fun to do.[/QUOTE]
Hey I have one of those too.
[img]http://y05.imgup.net/cuntcaa6.png[/img]
[editline]5th May 2014[/editline]
They are fun to play with...
I'd post with my 4X20 (BLAZE IT) display if it wasn't for the fact I'm in bed now :v:
[editline]5th May 2014[/editline]
Which also has
[t]https://dl.dropboxusercontent.com/u/43645231/photos/electro/2013-11-02%2015.46.58.jpg[/t]
some retarded shit going on
[t]https://dl.dropboxusercontent.com/u/43645231/photos/electro/2013-11-05%2022.49.35.jpg[/t]
So I decided to do some good old passive filter design.
Here is a 12th order Butterworth bandpass filter from 108MHz to 137MHz (aka the air band)
[thumb]http://u.cubeupload.com/Chryseus/xqoth2.png[/thumb]
[thumb]http://u.cubeupload.com/Chryseus/ijqw7V.png[/thumb]
Constructing it is going to be quite a challenge, I've determined the inductors need a Q factor of 50 or greater which isn't too hard, for the capacitors C1, C3 and maybe C5 I'll use the old twisted wire trick, the rest probably a series / parallel combination of ceramics or make my own caps.
[QUOTE=HeatPipe;44728912]Is it also possible to build LED frequency visualizer this way?[/QUOTE]
Yes it definitely is possible. That's how they did it in the old days. The main difference is you need to use active filters using opamps arranged as a bandpass filter. There's a limit on the number of bands you can use before everything gets more complicated.
Aww my 1602a needs the headers to be soldered onto it and I don't have an iron
[editline]6th May 2014[/editline]
Is there a simple way to not need a resistor for every LED? It's messy [url]https://dl.dropboxusercontent.com/u/30829668/Phone%20photos/VID_20140506_035316.mp4[/url]
[QUOTE=sambooo;44733671]Aww my 1602a needs the headers to be soldered onto it and I don't have an iron
[editline]6th May 2014[/editline]
Is there a simple way to not need a resistor for every LED? It's messy [url]https://dl.dropboxusercontent.com/u/30829668/Phone%20photos/VID_20140506_035316.mp4[/url][/QUOTE]
A resistor on ground should work, if I'm not mistaken.
Will that not cause the LEDs to dim as I turn more on?
[QUOTE=sambooo;44733671]Aww my 1602a needs the headers to be soldered onto it and I don't have an iron
[editline]6th May 2014[/editline]
Is there a simple way to not need a resistor for every LED? It's messy [url]https://dl.dropboxusercontent.com/u/30829668/Phone%20photos/VID_20140506_035316.mp4[/url][/QUOTE]
Prototypes tend to be messy. I wouldn't worry about it.
[QUOTE=Chryseus;44732143]So I decided to do some good old passive filter design.
Here is a 12th order Butterworth bandpass filter from 108MHz to 137MHz (aka the air band)
[thumb]http://u.cubeupload.com/Chryseus/xqoth2.png[/thumb]
[thumb]http://u.cubeupload.com/Chryseus/ijqw7V.png[/thumb]
Constructing it is going to be quite a challenge, I've determined the inductors need a Q factor of 50 or greater which isn't too hard, for the capacitors C1, C3 and maybe C5 I'll use the old twisted wire trick, the rest probably a series / parallel combination of ceramics or make my own caps.[/QUOTE]
Have you looked into variable capacitors and variable inductors?
[editline]6th May 2014[/editline]
[QUOTE=No_Excuses;44726638]I was bored and decided to redesign my main speaker's crossovers because they suck. I built them many years ago following a project I found on [URL="http://partsexpress.com"]partsexpress.com. [/URL]
Old and new schematic
[IMG]http://i.imgur.com/1kSmDLJ.jpg[/IMG]
[IMG]http://i.imgur.com/gpg1czZ.png[/IMG]
Old and new frequency response:
[IMG]http://i.imgur.com/LdGkSFi.png[/IMG]
[IMG]http://i.imgur.com/OTy4wCg.png[/IMG]
There's something about analog filters that just gets me going. I love bode plots, phase diagrams, s-domain analysis and all that good stuff.[/QUOTE]
I can understand where most of these passive filters come from, but what's the purpose of the following parts?:
R10/R12 voltage divider,
C6/R11 low pass filter(?),
R8/R15 voltage divider
My need for shields has been ignited
[t]https://dl.dropboxusercontent.com/u/43645231/photos/electro/2014-05-06%2019.39.22.jpg[/t]
I must get the biggest tower
so I got a soldering iron and soldered headers to my 1602a ([url=https://dl.dropboxusercontent.com/u/30829668/Phone%20photos/IMG_20140506_185934.jpg]link[/url]) but now I can't get the tutorial from the website working
do I have an unsoldered joint there? doesn't look like it to me but what do I know?
for people who don't want to click
[t]https://dl.dropboxusercontent.com/u/30829668/Phone%20photos/IMG_20140506_185934.jpg[/t]
[editline]6th May 2014[/editline]
also what is an unsolder joint?
That doesn't look very good in general, but the pins on the far right could be the problem.
[QUOTE=Angus725;44738163]Have you looked into variable capacitors and variable inductors?[/QUOTE]
Yes, I might use a few variable caps to help trim it, variable inductors are somewhat harder especially with such small values, I can just carefully adjust the turn spacing to 'trim' the inductance which works fairly well.
[QUOTE=sambooo;44734928]Will that not cause the LEDs to dim as I turn more on?[/QUOTE]
It would, but it's either that or just accepting that breadboarded circuits tend to be a little messy at best unless you trim all the component leads down.
A resistor for every LED won't look bad at all once you get it on a proper circuit board (if you're planning on eventually making it permanent) otherwise as others have said breadboard circuits usually aren't pretty.
[QUOTE=Leestons;44738724]ay lso what is an unsolder joint?[/QUOTE]
meant to say an unsoldered joint, as in one that isn't actually connected
I checked the pins and they aren't shorting
[QUOTE=Leestons;44739552]A resistor for every LED won't look bad at all once you get it on a proper circuit board (if you're planning on eventually making it permanent) otherwise as others have said breadboard circuits usually aren't pretty.[/QUOTE]
This discussion reminds me of
[img]http://i.imgur.com/r7dpco5.jpg[/img]
big image for shock value.
Is that a cpu?
[editline]6th May 2014[/editline]
I asked for solid core wire and was given stranded without realising because I just assumed the guy knew what he was doing. Arse.
The prices in Maplin's were decent for more expensive stuff like the soldering iron I got but the markup on components was silly. Also their resistor/capacitor packs didn't list how many of each kind they had in them, they were labelled 'lucky bags'.
I'm going to buy the components I need online since those pennies really add up.
bitsbox.co.uk is your friend, great for components. Tayda isn't too bad either.
Yeah Maplin is a pretty bad supplier for most things, I personally quite like [url=http://cpc.farnell.com/]CPC[/url] as they do free and quick delivery in the UK, although they are not as cheap as Chinese suppliers you at least know you're getting real components not shitty clones.
Rapid and Conrad are also quite good for certain things.
Best suppliers are Farnell (element14), RS, Mouser and Digikey although you often have to pay a decent amount shipping so these are more for bulk orders or if you need an obscure component.
Certain things though you almost always want to go to China for, such as perf board, potentiometers, connectors, LEDs, some tools and low cost equipment (I.E power supplies), also there are a bunch of sellers on ebay that keep local stock so you don't have to wait a month for shipping.
Make sure you get a good selection of resistors, any decent pack should include the full E12 series, a selection of 500 to 1000 will usually last a good while as some will get used, blown up or lost in your carpet, capacitors are less critical, a selection of common values will do such as 10p, 22p, 100p, 1n, 10n, 47n, 100n, 470n, 1u, 4.7u, 10u, 100u, 1000u.
Values above 1u will be electrolytics and below film or ceramic.
Also since you've got a soldering iron you can start desoldering stuff, you can get a lot of nice things out of broken or unwanted stuff, particularly higher power equipment like old CRT TV's, computer power supplies, printers, washing machines, etc.
So we have our new IRL540N plugged in and outputting just about the perfect amount of amperage for us. However, when it is plugged in it emits a high pitched whine from somewhere (PWM noise?) and the MOSFET still gets quite hot even though it should not be having any problems with dissipating heat.
What can we do to get rid of this whine and if possible reduce heat dissipated?
[editline]6th May 2014[/editline]
nvm its just our multimeter making noise
[QUOTE=Angus725;44738163]Have you looked into variable capacitors and variable inductors?
[editline]6th May 2014[/editline]
I can understand where most of these passive filters come from, but what's the purpose of the following parts?:
R10/R12 voltage divider,
C6/R11 low pass filter(?),
R8/R15 voltage divider[/QUOTE]
A speaker appears as an inductor and a resistor in series as you probably know, at least in its most basic form. Modelling a filter to take into account that inductance becomes overly complicated and even if you do, the frequency response becomes non-ideal.
You can see this in the original frequency response where the band-pass-filter (yellow) doesn't even have an upper rolloff. It appears to be more like a high-pass-filter. This is because the original designer didn't take into account the inductance (reactance) of the speaker.
The best solution is to put something called a 'zobel network' in parallel with each speaker. The zobel network consists of a capacitor of some value and a resistance equal to the DC resistance of the speaker. The capacitor works against the inductor, cancelling out it's reactance, making the whole lump appear as a fixed DC resistance to the preceding filter network.
[IMG]http://i.imgur.com/LbhSF4D.gif[/IMG]
Before and after the zobel network is implemented.
The voltage dividers (L-Pad attenuators) are there to match the SPL of each speaker to the woofer. In this case, the woofer has an SPL of 88dB (1W @ 1m), midrange of 89dB and tweeter is 90dB. So you need -1dB and -2dB attenuation respectively, to match the woofer. (Unfortunately I don't think it's as simple as that for various reasons. I'm going to do some quick calculations and report my findings.) One can use two resistors instead of one in series because if you do the math, the total resistance of everything stays constant and keeps the filter happy.
[editline]6th May 2014[/editline]
With that last part on SPL (sounds pressure level). SPL measurements are an average over the speakers usable frequency range since SPL varies wildly with frequency. When designing crossovers, you reference all your speakers SPL to the speaker with the lowest SPL. Thankfully that tends to be the woofer, so you have only have to attenuate the smaller speakers which by nature consume less power, so you have less losses.
SPL is also always specified at a certain voltage or power, as well as distance from the speaker. 1W of power, with the measurement taken 1 meter from the speaker is fairly common, as is 2.83V @ 1m. The latter of which is much more useful actually.
The SPL of the woofer I've chosen is 88dB (1W @ 1m). Say we want to pair it with a tweeter which has an SPL of 92dB (1W @ 1m). 92dB - 88dB = 4dB attenuation. Easy! But there's more...
When the datasheet says the speaker's SPL is measured at 1W, it's basically full of shit. A speaker's impedance (like the SPL), varies wildly with frequency. Maintaining a constant 1W over the entire measured range would be a pain in the ass and give a less useful result than just specifying a voltage instead. So when they say 1W, what they really mean is, "The speaker's nominal impedance is 8 ohms, therefore we applied 2.83V when measuring the SPL. Oh, and 2.83^2/8 = 1W". If it was 4 ohms, they would have applied 2V.
So basically if two speakers are specified to 1W and have different nominal impedances, you can't directly compare SPL levels without adjusting them. But if the impedances match, or the two speaker's SPLs are measured at the same voltage then you can directly compare SPL's and attenuate accordingly.
[QUOTE=No_Excuses;44742264]
[img]http://i.somethingawful.com/forumsystem/emoticons/emot-words.gif[/img]
[/QUOTE]
Ahh, very cool.
What methods do you use find what impedances and capacitances you need?
grah I can't keep this MOSFET cool for the life of me.
that and our voltage regulator also seems determined to overheat
[QUOTE=Angus725;44743259]Ahh, very cool.
What methods do you use find what impedances and capacitances you need?[/QUOTE]
From the speakers datasheet you know the resistance and inductance. Say it's 4.5 ohms and 0.2mH. We therefore want to end up with a constant 4.5 ohms.
First just find the frequency where the impedance of the inductor matches the impedance of the resistor:
XL = 2*pi*f*L
R = 2*pi*f*L
4.5 / (2*pi*.0002) = f
f = 3581Hz
Then use C = 1 / (2*pi*3581) to find the cap value, which is 44.4uF
It's really clever how it works. At f=3581Hz, you have 9 ohms || 9 ohms = 4.5 ohms
As f-> infinity, you end up with 4.5 ohms || infinite ohms = 4.5 ohms.
As f->0, you get the same thing. infinite ohms || 4.5 ohms = 4.5 ohms.
[QUOTE=paindoc;44743367]grah I can't keep this MOSFET cool for the life of me.
that and our voltage regulator also seems determined to overheat[/QUOTE]
How much current you pulling through it and what do you have for a heatsink? Any pics?
Wow thanks for the close up bitsbox!
[t]https://dl.dropboxusercontent.com/u/30829668/Screenshots/Screenshot%202014-05-07%2014.46.36.png[/t]
[editline]7th May 2014[/editline]
I think I figured out why my LCD wasn't working, and it's really stupid. I might have been treating the rails on my breadboard as if they aren't split in the middle without even thinking about it. I'll have to wire it back up later and see if it works.
Sorry, you need to Log In to post a reply to this thread.
