Electronics AV Autoswitch: New info, Advice requested

Has anyone seen any designs for a DIY AV autoswitch. Given the cost and lack of flexibility in commercial AV autoswitches, I thought I'd try building one.

I've seen a circuit for detecting video component signal for use in a power relay switch that I plan on adapting. Everything would be PCB mounted.

My concept was to drive signal relays by sensing incoming component signal(s) with a cascade effect. A signal from a higher prioity device would switch to it even if a lower priority device was on, e.g. my satellite reciever is always on. The AV switch would connect to SAT unless the DVD or LD is on.

Can anyone suggest pitfalls or guidance on components (type of relay, etc.)?

TIA!

Mike

 

This is the circuit for a video activated power relay using a 2N2222 NPN Transistor.

My thought was to use 4pdt relays (3 composite signals plus ground) triggered from the transistor circuit when video is sensed. I would use NOR gates (e.g. MC74HC4078) to provide the logic for priority if more than one input is powered.

Allied Electronics has 1.5V coil 4pdt relays ($2) and quad RCA pcb mounted jacks ($.50) which would be the bulk of the design.

Anyone have experience with this type of circuit? Any pitfalls I haven't considered?

TIA

Mike

 

Rather than use multiple 4pdt relays to switch the video input, is there some ic form of a rotary switch that would simplify the design?

 

Checked with Jameco and Digikey, but no luck on a ditigal matrix IC that would act as a rotary switch. I'm still hoping linear, macro or one of the other gurus has some ideas.

 

Hehe, I'm reading with great interest. I don't have a lot to contribute, though.

What do they use in the consumer-grade video switchers you can buy, the ones with the little effects generators in them?

 

You can get analogue switches designed for video, check out the dallas semiconductor website. Why not use 4 pole relays?

 

Actually one of my reasons for updating is to share what I'm learning. My last EE course was 20 years ago.

I'm aware of 2 commercial AV auto switchers -- one of which is not for sale in the US yet. The available one has 6 iputs and is only $110 (before shipping) but I wanted to do my own for the experience and learning benefits. I figure I can make a 6 - 8 device auto switcher for about $30-40. And get to say I did it.

I'm not aware of any that generates effects.

Cheers, Mike

 

Thanks for the tip. I'll check it out.

I currently am planning to use 4 pole relays (N inputs - 1 relays) but wondered if there was a more elegant solution.

My plan is for 7 inputs which would need 6 quad RCA ports, 6 relays, several NOR and transistors for powering the relays.

I would modify the circuit above to use apply the logic from the signal sensing before using transistors as current amplifiers to the relays.

Cheers, Mike

 

Based on macro's tip, I've done some research on the Dallas Semi/MAXIM site.

It appears the MAX310 CMOS 1-of-8 RF/video multiplexer may be the best choice. It would eliminate most of the logic gates and the relays as well as provide better impedence matching, channel isolation and a break-before-make contact.

I believe I could use 1 each for the 4 signals -- Video , R audio, L audio and gnd.

I should have a preliminary circuit in a few days to share.

Cheers, Mike

 

Macro, et al:

The typical operation circuit in MAXIM's datasheet shows a MAX450 video amplifier on the MAX310's output to prevent insertion loss. Is this necessary/advisable for this application?

Cheers, Mike

 

Video switching/routing is prone to losses and ghosting unless care is taken with circuit layout, impedance matching and the use of the correct components and cables. That said, since you are making this unit yourself, try it without first, if the losses are unacceptable then use the buffer. As a guideline a cable joint can easily produce a 3db loss of signal!

 

Thanks macro.

I think I'll add the amps for the 3 signals.

 

New facet:

Since I'm going with the video muxes, I now need to resolve the addressing issue in base 2.

I worked out that I could encode 3 address bits (octal) for 8 input signals with 4 4-input OR gates, 4 2-input AND gates and 4 3-input NOR gates. I'm actually using a NOR as an inverter in one case to keep the differnet number of gate types to a minimum.

I would like to just use a single chip encoder and have located 8 to 3 encoders from TI and Motorola -- SN54/74LS148. But looking at the function tables the logic is askew.

My intent was the highest priority input would be selected when it's input was on (high) e.g. input7 on would equate to 111, input 5 = 101, etc. The highest priority input would be selected regardless of the state of any lower priority input.

Code:

     Inputs       Outputs
0 1 2 3 4 5 6 7   a2 a1 a0
H L L L L L L L   L  L  L
X H L L L L L L   L  L  H
X X H L L L L L   L  H  L
X X X H L L L L   L  H  H
X X X X H L L L   H  L  L 
X X X X X H L L   H  L  H
X X X X X X H L   H  H  L
X X X X X X X H   H  H  H

H = HIGH
L = LOW
X = DON’T CARE

Is there an encoder that works like this?

 

I've only had a quick glance at this but the inputs and outputs of the '148 are active "low". So when when input 7 is low the output will be 000 giving you the priority you want. I haven't checked the mux but you can either "reverse" your logic thinking or invert the '148 outputs with a driver depending on how you want to do it.

In fact a quick look at the '310 data sheet reveals that with an address of 000 channel 1 is enabled so you should be ok

 

Thanks.

I actually didn't have time to finish the post and just put up my funtion table.

The 148 does invert the inputs which makes everything goofy.

Code:

SN74LS148
FUNCTION TABLE
INPUTS                     OUTPUTS
EI 0 1 2 3 4 5 6 7     A2 A1 A0 GS EO
H X X X X X X X X      H  H  H  H  H
L H H H H H H H H      H  H  H  H  L
L X X X X X X X L      L  L  L  L  H
L X X X X X X L H      L  L  H  L  H
L X X X X X L H H      L  H  L  L  H
L X X X X L H H H      L  H  H  L  H
L X X X L H H H H      H  L  L  L  H
L X X L H H H H H      H  L  H  L  H
L X L H H H H H H      H  H  L  L  H
L L H H H H H H H      H  H  H  L  H

I could invert all the inputs and either reverse my number scheme for the switch or invert the outputs as well.

Seems cumbersome. I was hoping there was a more applicable encoder available but I haven't found it yet.

It surprises me that this one looks for a signal to be pulled low to activate its address. I guess it's all relative.

 

Thanks for the advice, linear and macro.

I'll keep updating this thread as I learn and the circuit develops -- as long as there's interest.

It's served my purposes so far -- made me clear the cobwebs, learn something and have fun. I must have spent 45 minutes working out my own encioder with just basic logic gates. It was interesting getting back that basic level of logic flow.

Now if I can just find an encoder that isn't bassackwards.

Cheers, Mike

 

I use this a lot, maybe it will help you shop for an encoder:

the mighty GIICM

Take a stroll down the 74xx aisles, I bet you'll find the thing you need...

 

Thanks for the resource, linear. While it didn't help this time, I have bookmarked it for future refernce.

The 74147 & 74148 were the only encoders I found and they inverted of course. I'll either invert everything signal again or build my own encoder.

Cheers, Mike

 

Why complicate matters? It will do what you want now.

 

My logic must be off...

It triggers the address (inverted) when the line is pulled low. I'm looking for a line to be pulled high -- signal present.

What am I missing? I understand I can resverse the adressing scheme, but I don't understand how the encoder works as is.