[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: [SIGMusic] circuit



On Sat, 2007-02-24 at 22:35 -0600, Stephanie Bayani wrote:
> i talked to a friend a little about the circuit we're making.  
> 
> he basically said it would work.  however...we run the risk of killing
> lots of circuit boards.  as well as the solenoids.  using a pic to
> provide power is basically bad design.  but whatever, we don't care.
> and his question was can you even find a transister that will pull
> up..what is it?  15mA - 50mA (or 75mA)?  or .83V - 24V?  i am assuming
> we did find transisters.  
> 

I thought the point of the transistors was that so we *aren't* using the
PIC to provide power. We just use the pic to switch on and off the
transistors -- which I suspect would be pretty normal. The ones that I
speced are the 2N3904, which support up to 40V/200mA and are only $0.10
each (not to mention the ACM has a ton that we may or may not be able to
use).

(We also need a few 2N2222A transistors, which are 40V/800mA, for the
500mA solenoid valves... see below).

http://www.kpsec.freeuk.com/trancirc.htm#chip talks about choosing a
transistor. I'm not sure if we're ok by those standards or not (I'm not
sure I understand the bits about current gain). Also, the section above
it talks about needing a diode across the solenoid to protect against
voltage spikes. Those should be another cheap part that we can get from
the ECE store (if ACM doesn't have 50 of them lying around).

> i'm not entirely sure that we need the volt regulator, unless we can't
> control the voltage output.  it's only amps, right that we can
> control.  
> 

We need something to get 5V for the PIC from the 24V supply. The voltage
regulator is like $1.50 and gives us 5V out for up to 30-ish volts in.
It has a zener diode or something in it to make us less likely to fry
the chip.

> he suggested using flip flops (D).  and something about Darlington.
> whatever.  
> 

> if i'm reading the specs correctly, the amps / volts out of a single
> pin changes depending on what pins are in use / or on.  that sucks ass
> and i hope i read that wrong.  
> 

Yipes. Where is that in the spec? That could be scuzzy, but if I'm
reading the data sheet for these transistors correctly
( http://www.fairchildsemi.com/ds/2N/2N3904.pdf ), anywhere from 1-5V
will turn them on.

> eh, i left my backpack in my friend's car, and i don't know when i'll
> get it back because of the weather.  i'm just gonna write down
> everything that we supposedly know.   
> 
> 5 I/O ports 
> PortA : 6 input/output
> PortB : 8 input/output
> PortC : 8 input/output
> PortD : 8 input/output
> PortE : 3 input/output
> 
> Port A+B+E has a sink/source max of 200mA.  
> Port C+D has a sink/source max of 200mA.  
> 
> each indidvidual pin has a sink/source max of 25mA.  
> 

As long as the transistors draw less than 25mA each, we're ok. I don't
know how to find this out. (Wit might know).

> the pic needs 5V (high).  actually this can go to 5.5V, but..that
> would probably be risky as far as frying stuff goes.  
> 
> each pic will handle 2 instruments. 
> 
> that's 11 solenoids and 1 solenoid valve controlled by 1 pic.  
> 
> all 12 outputs will each have a single transister.  
> 
> the solenoids need 24V / 50mA.  
> the solenoid valves needs 24V/ 75mA
> 

Two of the solenoid valves draw 75mA, and two draw 500mA. It's what we
could find on ebay. :-p

> the TRISx register will denote whether a i/o pin is input or output.
> this is done with the pic assembly and is part of the setup for the
> ports.  (TRISA for porta, and TRISB for portb...)
> 
> i'll draw a circuit diagram later.  it might be paper format, since i
> suck at drawing on the computer.  
> 
> i'll look at teh stupid diagrams, however, the graphs don't mean a
> whole lot to me.  i have a vague memory of "sharing" graph diagrams
> with former classmates.  
> 
> steph
> 

-- 
Jacob Lee <jelee2@xxxxxxxx>