Help needed building voltage controlled single pole changeov

General project help for Adafruit customers

Moderators: adafruit_support_bill, adafruit

Please be positive and constructive with your questions and comments.
Locked
Marmotta
 
Posts: 10
Joined: Fri Mar 15, 2013 2:25 pm

Help needed building voltage controlled single pole changeov

Post by Marmotta »

I need to build three single pole changeover switches which will change status depending on whether the signal is high (5V) or low (0V). I have had a search online, but can't find any schematics that match what I'm looking for, only soft latching power circuits and the like. Any help would be most appreciated.

User avatar
zener
 
Posts: 4567
Joined: Sat Feb 21, 2009 2:38 am

Re: Help needed building voltage controlled single pole chan

Post by zener »

I don't have a lot of specifics so I can't give a specific answer, but I am thinking some kind of relay or contactor. What is voltage in and out and the current you are switching?

Marmotta
 
Posts: 10
Joined: Fri Mar 15, 2013 2:25 pm

Re: Help needed building voltage controlled single pole chan

Post by Marmotta »

I think I've found what I'm after - a Triple (or Quad as they seem to be far more common) SPDT CMOS switch. I've knocked up a simple picture of how I think it should work:

Image

My YPbPr cable will have the two black pins on the AV output connected within the cable, so that when it's connected it will create a bridge, supplying the voltage to the MAX333. The RGB cable won't have the two pins connected, so that input is 0 and the MAX333 will output the RGB signal. Let me know if that's correct.

One thing I'm not sure about is how pins 5, 6 and 16 should be utilised, as other suitable switches I've found don't include them, but the MAX333 is the only one I could find that comes in a DIP package so I can prototype with it.

http://datasheets.maximintegrated.com/en/ds/MAX333.pdf

Edit: The voltage doesn't have to be 5V - I can regulate it to pretty much anything.
Last edited by Marmotta on Tue Oct 07, 2014 6:33 am, edited 1 time in total.

User avatar
zener
 
Posts: 4567
Joined: Sat Feb 21, 2009 2:38 am

Re: Help needed building voltage controlled single pole chan

Post by zener »

You should put a pull down resistor on the control line to keep it from floating when the jumper is not connected. The V+ and V- are the signal power rails. The signal must stay within these levels to keep from being clipped. The ground is your system ground, and is the ground reference for the control pin(s).

You could also use a pair of CD4066 probably. Turn one on and turn the other off.

Marmotta
 
Posts: 10
Joined: Fri Mar 15, 2013 2:25 pm

Re: Help needed building voltage controlled single pole chan

Post by Marmotta »

The system voltage goes through an LM7805, so I've got 5V and GND to power everything. For the V+ and V- do you think it would be better to connect it to the 10V DC lines before it goes through the LM7805 or use a TPS60400 voltage inverter or similar to supply 5V+ and 5V-?

Should I tie the rest of the pins on the MAX333 to ground or do you reckon it's safe to leave them floating? The datasheet doesn't appear to mention it in the safety measures.

User avatar
zener
 
Posts: 4567
Joined: Sat Feb 21, 2009 2:38 am

Re: Help needed building voltage controlled single pole chan

Post by zener »

What is the voltage range of the signals you are switching, relative to ground?

I would probably tie the unused input high or low, or just switch it along with the rest.

User avatar
zener
 
Posts: 4567
Joined: Sat Feb 21, 2009 2:38 am

Re: Help needed building voltage controlled single pole chan

Post by zener »

There can be some other issues trying to switch video signals this way. From some web searching I found this advice:

"you need to have the incoming video terminated into 75R and then AC couple it with a 100uF 16V electrolytic ( low ESR one of course) into the chip, with a pair of 10k or so resistors to bias the input to half rail. You do need a video buffer on the output with a high impedance input and Ac coupled with a clamp diode to drive any cabling afterwards, but most will do the job easily. The trick is not to try to pass any current through the switch, just the voltage. If needed the buffer can be just a darlington pair with the emitter connected to a 75R resistor and develop the output video across this straight. Works well off a 5V rail single ended with the 74HC version (4066). Slight loss and there will be a small amount of bleedthrough from the unswitched channels ( only noticeable on a input that has no signal where you will see some noise on a scope but no video signal) but good enough to switch cameras for a surveillance system."

Marmotta
 
Posts: 10
Joined: Fri Mar 15, 2013 2:25 pm

Re: Help needed building voltage controlled single pole chan

Post by Marmotta »

OK, next revision of my plans - looking a bit neater this time round:

Image

Does that seem about right? I'm not sure on the necessary values for the resistors and capacitors - I've done a little bit of reading on AC coupling and lots of circuits I've found use capacitors closer to 0.1uF

User avatar
zener
 
Posts: 4567
Joined: Sat Feb 21, 2009 2:38 am

Re: Help needed building voltage controlled single pole chan

Post by zener »

We have reached the limits of my expertise on switching video signals. I think your schematic needs a little improvement but it is getting close. Hopefully the resident experts can give you some advice. I think the terminating resistors on the inputs need to be ahead of the caps. You may want the 10K resistor dividers to bias the inputs as was described in the info I copied. You may also want buffer amps on the outputs, or emitter followers as described.

User avatar
adafruit_support_mike
 
Posts: 67446
Joined: Thu Feb 11, 2010 2:51 pm

Re: Help needed building voltage controlled single pole chan

Post by adafruit_support_mike »

Hi - I'm late to the party, but it looks like you still need a bit more information.

The reference you quoted describes a circuit that looks like this (showing a single switch for clarity):
video.jpg
video.jpg (15.54 KiB) Viewed 520 times
The 75 ohm resistor at the end of the input cable is a terminator. Without it, you'll get the electronic equivalent of echoes in the cable as the signals come in. (as an aside, they exploit that to find things like broken power lines.. send a pulse in at the substation and measure the time it takes the echo to come back. Dividing that by the propagation rate tells you how far along the cable the break is located) The impedance of a video coax cable is roughly 75 ohms, so as far as the signal is concerned the resistor looks like more cable.

The capacitor blocks DC voltage but sends AC signals through. The two 10k resistors set a no-input voltage of VCC/2, and the signal from the cable pushes that up and down. The cap is so large that the RC time constant between it and the two 10 resistors is basically "forever" as far as video signals are concerned, so you won't see any signal degeneration because of current leaking into or out of the cap through the resistors.

The output side of the circuit needs some kind of buffer (that's the triangle) with another 75-ohm resistor for impedance matching to the output cable. A pair of NPN transistors arranged as a Darlington will give you a whole lot of output current while pulling very little current from the input.

There's one potential gotcha: the input cable and 75 ohm terminator form a voltage divider, and reading the signal at the point where the two meet cuts the input voltage in half. Most circuits to drive cables use a gain-of-2 amplifier as the output buffer to boost the signal back up to its original level.

That may not be a problem if the next circuit along the line has an input amplifier. It's one of those YMMV things.

Locked
Please be positive and constructive with your questions and comments.

Return to “General Project help”