I'm serious, they're like the microcontrollers of analog circuitry.
[QUOTE=Chryseus;33700244]Here you go
[img]http://i.imgur.com/9xzC8.png[/img]
I assume the hard drive starts up as soon as you push the power button and that the 5V is not
available before that, if it is then I'll have to do it slightly differently.
The delay time is set by adjusting the 100k potentiometer, when the voltage across the capacitor
rises to 2.5V (the reference voltage) the op-amp switches on Q1 which supplies power the the LED.
The LED signal is applied to the base of Q3 which inverts it, so when the signal is low Q2 is turned
on and the LED lights.[/QUOTE]
Rated friendly.
Yeah, no 5V is emitted from the drive (if an LED was attached, the LED will not light up during spinup and test). Only during read/write/seek.
[QUOTE=MIPS;33711159]Rated friendly.
Yeah, no 5V is emitted from the drive (if an LED was attached, the LED will not light up during spinup and test). Only during read/write/seek.[/QUOTE]
You could do this with an ATtiny25,resistor and maybe a decoupling capacitor(arguable but recommended), Chryseus design is 12 components, compared to 3 for a digital solution, unless you want to be able to change the time.
I see what you mean by cheating, its easier, alot easier.
but if you're looking for a modern solution on the easy, then digital is your way.
[QUOTE=MIPS;33711159]Rated friendly.
Yeah, no 5V is emitted from the drive (if an LED was attached, the LED will not light up during spinup and test). Only during read/write/seek.[/QUOTE]
So you need an extra powersupply?
I'd try a capacitor with 100 uF or more, and only have that NPN-transistor. If the LED-output doesn't have a limited current, it should charge the capacitor fast enough for the LED to remain lit up for a few seconds.
You might want to look at Goldcaps, they are 1F at 5V and are pretty small.
Since when are decoupling capacitors 'not recommended'?
Yeah, sometimes you might be able to get away without them, but if you have a few 0.1uF caps on hand, it's almost always advisable to put one across VDD/VSS. Digital electronics don't have consistent current draw, most of the time they draw almost nothing but they draw huge spikes when the clock switches. You need to sure that charge is [i]right there[/i] when the MCU needs it. You can't be sure without this capacitor, long supply lines have inductive properties, which mean they don't like rapidly changing loads. It only costs you like $0.03 and a negligible amount of board space, but it can potentially save you hours of debugging later.
It's actually a bad idea in general to assume your power supply is good. I spent a few hours debugging an oscillator which was misbehaving even though it was fine on paper. Turns out it didn't like my school's bench power supplies, which were a little noisy. A small decoupling cap solved the problem right away.
I had assumed that lab hardware was going to give me a clean, constant voltage, which turned out to be entirely wrong.
I absolutely agree
-manual merge-
[QUOTE=DrLuke;33715065]So you need an extra powersupply?
I'd try a capacitor with 100 uF or more, and only have that NPN-transistor. If the LED-output doesn't have a limited current, it should charge the capacitor fast enough for the LED to remain lit up for a few seconds.
You might want to look at Goldcaps, they are 1F at 5V and are pretty small.[/QUOTE]
I should not need an extra power supply if +5 and +12 are supplied by the drive enclosure for the ST-251N.
Oh yeah, the programmer arrived. It looks exactly as pictured and seems to be quite well built. Included was a CD containing a complete software package and documentation with large amounts of diagrams and pictures.
I like this programmer so much I think I'll grab a hobby box and two more TEXTOOL sockets and make it look like a real programmer. Problem is that this thing can be customized so much, to make all the jumpers and dip switches acessible I'll have to install around 30 toggle switches. :v:
That USB port is also for power only so the main requirement to use it is a parallel port. I can drive it though off a generic wallwart 12V power supply which si preferable anyways ebcause you can program a larger variety of chips.
[QUOTE=ROBO_DONUT;33717354]Since when are decoupling capacitors 'not recommended'?
Yeah, sometimes you might be able to get away without them, but if you have a few 0.1uF caps on hand, it's almost always advisable to put one across VDD/VSS. Digital electronics don't have consistent current draw, most of the time they draw almost nothing but they draw huge spikes when the clock switches. You need to sure that charge is [i]right there[/i] when the MCU needs it. You can't be sure without this capacitor, long supply lines have inductive properties, which mean they don't like rapidly changing loads. It only costs you like $0.03 and a negligible amount of board space, but it can potentially save you hours of debugging later.
It's actually a bad idea in general to assume your power supply is good. I spent a few hours debugging an oscillator which was misbehaving even though it was fine on paper. Turns out it didn't like my school's bench power supplies, which were a little noisy. A small decoupling cap solved the problem right away.
I had assumed that lab hardware was going to give me a clean, constant voltage, which turned out to be entirely wrong.[/QUOTE]
well i did say it was arguable didn't i?
Hey guys.
What's the easiest way to measure the force that impedes the movement of a servo?
[QUOTE=Nikita;33721276]Hey guys.
What's the easiest way to measure the force that impedes the movement of a servo?[/QUOTE]
You can't. If at all, you can only determine the torque that's applied to the rotor. But to do that, you'd probably have to measure the current the servo consumes with different torques applied.
[QUOTE=Nikita;33721276]Hey guys.
What's the easiest way to measure the force that impedes the movement of a servo?[/QUOTE]
It depends on how much force you're dealing with.
My guess is you'd probably use some mechanism like a [url=http://en.wikipedia.org/wiki/Torque_wrench]torque wrench[/url] uses which measures the deflection of a cantilevered arm.
There's also [url=http://en.wikipedia.org/wiki/Strain_gauge]strain sensors[/url] and the like. You'd generally use them in a wheatstone bridge circuit.
I had a question guys, new at electronics here so don't rage too hard. If I had to connect an oscillator to the Z80 Computer I'm going to build, do I need an oscillator circuit with a crystal? Can't I just buy a crystal oscillator (one of those block-like things with 2 connections.) How would I go around wiring it up, just one end to the CLK pin and another to GND?
If it only has two connections, it's probably just the crystal itself without the oscillator circuit.
Some microcontrollers/processor have the oscillator built-in so you just need to provide the resonant circuit (crystal or ceramic), but it varies so you'll need to read the datasheets.
[QUOTE=ROBO_DONUT;33724722]If it only has two connections, it's probably just the crystal itself without the oscillator circuit.
Some microcontrollers/processor have the oscillator built-in so you just need to provide the resonant circuit (crystal or ceramic), but it varies so you'll need to read the datasheets.[/QUOTE]
From what I've read/seen, it only has one connection and you have to build the circuit yourself. Can I just sub the circuit with an oscillator?
I guess the simplest way to do it would be to get one of those programmable oscillators.
If you have an inverter on you and a few resistors you could just build this and connect the output directly to the CLK pin:
[IMG]http://www.z80.info/gfx/hctres.gif[/IMG]
Found it here:
[URL="http://www.z80.info/uexosc.htm"]http://www.z80.info/uexosc.htm[/URL]
Oh and this might even be more useful, all sorts of Inverter oscillators 8):
[URL="http://www.fairchildsemi.com/an/AN/AN-118.pdf"]http://www.fairchildsemi.com/an/AN/AN-118.pdf[/URL]
I spent 3 hours figuring out what's wrong with my parallel port echo-er and turns out I forgot a single fuse >.<
So, i created my own little program to work out my gain for my instrumentation amplifier, it uses the equation that is in here [url]http://en.wikipedia.org/wiki/Instrumentation_amplifier[/url] and an array of resistors, heres a pastebin [url]http://pastebin.com/KeXB5R4n[/url]. This could easily be changed to a different equation.
Also, any mistakes i made? if i can be bothered ill add some features, like lowest resistor values etc.
[img_thumb]http://i.imgur.com/xk6ut.jpg[/img_thumb]
After replacing the 10v zener in the power supply the scope powered right up (well, after finding the trace), beauty! Time to calibrate it!
(Yes, that's a coffee maker next to it. Had to go to the kitchen since that's the only grounded outlet in my whole house)
Nice
[QUOTE=ddrl46;33736771]Had to go to the kitchen since that's the only grounded outlet in my whole house[/QUOTE]
:S
The scope is awesome, though. It always feels good to get some good use out of something with a simple repair, especially when it would've been expensive otherwise and would have probably just introduced a crapton of heavy metals to a landfill somewhere if you hadn't saved it.
Saving the environment through hacking!
Just done some Christmas shopping
[img]http://i.imgur.com/utS0I.jpg[/img]
Who do you usually order from?
[QUOTE=LoneWolf_Recon;33737892]Who do you usually order from?[/QUOTE]
Bitsbox, ultra cheap shipping and usually fast.
Not the best selection though but eh, who cares.
[QUOTE=LoneWolf_Recon;33737892]Who do you usually order from?[/QUOTE]
[url]http://bitsbox.co.uk[/url]
Le ninja.
[QUOTE=ddrl46;33737938][url]http://bitsbox.co.uk[/url]
Le ninja.[/QUOTE]
Stop spying on me :-C
[QUOTE=ROBO_DONUT;33737523]:S
The scope is awesome, though. It always feels good to get some good use out of something with a simple repair, especially when it would've been expensive otherwise and would have probably just introduced a crapton of heavy metals to a landfill somewhere if you hadn't saved it.
Saving the environment through hacking![/QUOTE]
Repairing old gear is awesome, pay almost next to nothing and get something incredibly useful which might need some minor repairs.
Thanks for the automerge :v:.
Sorry, you need to Log In to post a reply to this thread.
