Digital switches?

speed?

different speed needs different components.

I'd say 100ms or less.

PWM is about 96 khz on most systems

probably mosfets will do the work the issue is going to be on the number of pulses you require for channel and how you going to control the open and close on the mosfet. Is your application needs a word bit programming ? why 13 channels is this a parallel 12 bit application?. what is what you whant to control?

What's wrong with using sensitive relays?

I guess a solid state relay is what I really would like to use. Is something like this acceptable:

http://www.cel.com/pdf/datasheets/ps2501.pdf

I am using some 2n3904 and 2n3906 transistors for now, but I just want something in package form.

Use an and gate. Put your PWM signal on one input ans apply your control signal through the other. If you don't have the PCB real estate for a whole IC you could use an FET and bias it off so that when the 5V is applied, the PWM signal will push it over the threshold to conduct. The gate method doesn't take any design effort though. Just connect the 2 signals and power and you're done.

Ed

use a single AND gate in the same size package as an SMT transitor, e.g.

ST Micro 74LX1G08

NXP 74AHC1G08

IC type CD 4016 (CMOS quad bilateral switch) or equivalent should do your job. You can look up the details of the IC from its data sheet. You get four switches per IC, so you can work out the number of ICs needed.

You really don't provide enough information to make a good recommendation. It could be that a simple 2 input AND gate could do the job. The 4016 recommended above has considerable "ON" resistance and may not suit. The 4066 is an improved version of the 4016 with much lower "ON" resistance but still, since we don't know much about what you are doing it may not be appropriate. There is also an "open collector output" 2 input AND gate that might suit, or multiplexers (digital and analog). Tell us more and perhaps we can zero in on what would work the best.

As Enauman says an AND gate (NAND, OR or NOR) will all do or quad tristate buffers like say 74???125 (??? choose your technology LS HC or HCT) or if you want common control what about octal tri state buffers like 74HCT244 (or 240 or 241).

hello.

the most effective form to do it is trough MOSFET transistors.

though, those transistors need extra care in assembly, cos the gate is easily destroyed by electrostatical discharges.

the easy form is trough BJT transistors, if you dont have any kind of problems of supply power...

all depends ...!! if you want a super efective drive with low comsumption, or a easy and very reliable sistem.

personally ,for that kind of voltage, I would use BJT....

and again it depends in many tings.

ex (PWM freq)

First of all, thank for all the replies. Second, let me apologize for the lack of detail. I will try my best to explain my situation more thoroughly.

I'm trying to trigger the buttons on my wired Xbox 360 controller. They are capacitive buttons and the signal slightly resembles a pwm (not exactly though, it's just similar…I really don't know what it is). It's a continuous string of three short, decaying pulses. Perhaps that's just for debouncing, but I don't know.

If I connect the leads on the front, the buttons trigger fine. However, the points are vias and are almost impossible to solder. I just traced the leads on the back and found points that I could solder to. When I connect the points with all the external wires, the buttons trigger just fine too. The problem comes when I try using a transistor to switch them on or off. Some of the points have a common negative lead. When I try to turn on certain buttons, it triggers two instead of just one. The other button triggered is the one that shares the common negative point. This happens for several different sets of buttons. I tried several configurations for the switch with no luck.

Originally, I was trying to connect the controller to my Saitek X45 Joystiq, but I am more interested in using it with a microcontroller at this point.

Here are some pics and a video:

http://www.flickr.com/tineras

http://www.youtube.com/watch?v=zbt6eF59FGc

Please be patient with me. I'm a student and I'm just trying to do projects here and there to get some more practical knowledge. I will gladly answer any further questions to try and clear things up. Thanks for your input!!

It sounds like a scanned matrix. You should use either analog switches such as the 4066 or some newer form. TI has a new analog switch that operates well on 3V (TS12A45X, X=1 or 2). Yes you could also use FET's but it would probably be easier to use the analog switch since it already has all the FET's arranged for you.

well...

(i did not tink that the sistem was a complex digital machine.I tought that you were in a project of a power source of some type)

I dont know aswell what kind of coding and electrical hardware they use .

I would start to determine the exact code that the pushbutton sends.

meanwhyle it would be a good idea to determine the exact polarity for all the terminals.

datasheets for the components in use if available, etc...

if the joystick use comun electrical source from the xbox, thats a ting...

(but my guess is that you have tryed that already)

if not , you have another situation.

and to be honest, if it is to repar the joystick it could be a good idea just to buy a new one.

but if you intend to made a new one from the scratch, that is a intire project for it self and that´s a load of work...

My own experience on this subject is:

1)quad CMOS gates are simple to use,and are available on SMD S014.
CD4066 have a far lower on resistance than the slightly less expensive CD4016,hence a wider bandwidth,about 5MHz under 5 volts for CD CMOS.
faster CMOS such as 74HC and HCT should allow a bandwidth of 20MHz (rise and fall time of 25ns)

2)inexpensive JFETs such as BF245,BF256,2N4416 can be driven as fast switches,even analog in synchronous detection.
A usefull trick is to drive the gate through a megohm resistor(to limit charge injection to the channel)with a reverse connected low cost rectifier such as 1N4001,which behaves as a low cost variable capacitance diode

3)on an inexpensive game supplied under 5 volts,I use low cost plastic NPN with a floating base driven through a 100 kiloohms resistors to the digital signal.

At low speed,as such transistors are extremely low cost,this may be the least expensive and simplest solution

I temporarily solved the problem by adding a diode to the negative lead of the offending button. This seems to have corrected the problem. I will still look into the other solutions that everyone has posted. I just happen to have this equipment on hand and a bunch of time off right now.