Need Circuit Design -- Cheap...

If there's a moderate amount of interconnect wiring it dioesn't matter where the diode is placed, if there's lots "as physically close as possible to each relay" is the wrong place to put it.

This was demonstrated in another thread last week.

Exactly. But in the usual case, putting the diode next to the coil cuts down on RFI.

how does it do that?

Really? I don't see how (or why?).

Great News!! The parts all showed up today so this weekend build is on.

I redrew the schematic with the pins on the chips where they really are to be a little closer to an "as-built", as before I think I got all the connections right, parts, polarity, etc...

I know it is asking a lot but perhaps you or others could give it a quick once over before I assemble it this weekend. I figure there was less chance of transfering it to the protoboard if the drawing I use is the same as the parts.

If not, still thanks Jim and to all that helped.

Link to your schematic -- Only change is I ran the unusd inputs to ground thinking it might use less power...

Here is the link to what I am calling my as-built...

For those joining late, here is a picture of how this circuit fits in to the overall project:

Thanks

I*A

True, but that doesn't mean the wire inductance isn't enough (to blow the driver).

Critical damping.

I'm fond of having NO diode, but add a circuit to turn the switching device back on to hold the back EMF at 2*VCC.

Figured out that trick when I needed to drive a relay with CMOS... flyback diodes are a NO-NO because of substrate currents.

...Jim Thompson

Well smoke is the last thing I hope to see, I usually use my bench power supply and start with the current down to nearly nothing and then slowly turn it it up the first time until I reach by best gestimate on the total draw. This does not always keep the smoke from escaping but the power supply has a light over the current knob that goes from off (I am supply all the cuurrent the circuit want to very bright, it wants a LOT more, this is usually a good tell. It is a 20 year old bench supply I got from a surplus place, one of the best buys I have ever make.

BTW -- Did you know that all electr>Bob Thomas wrote:

Well smoke is the last thing I hope to see, I usually use my bench power supply and start with the current down to nearly nothing and then slowly turn it it up the first time until I reach by best gestimate on the total draw. This does not always keep the smoke from escaping but the power supply has a light over the current knob that goes from off (I am supply all the cuurrent the circuit want to very bright, it wants a LOT more, this is usually a good tell. It is a 20 year old bench supply I got from a surplus place, one of the best buys I have ever make.

BTW -- Did you know that all electr>Bob Thomas wrote:

It eats the E spike instead of letting it radiate from the PCB trace going back to the switch.

Indeed.

I haven't analysed the logic, but other than that, it looks fine to me.

Well, that's the same for all of us. ;^)

Always a good idea.

I've always built my own bench supplies, but buying a bought one is a good thing.

Indeed. ;^)

PS: Please try to reply at the bottom of the post, rather than the top - it's good manners.

What's the cause of this E spike? there's no sudden change in current, the diode will see to that.

What you do get with the diode next to the relay is a sudden change in current in the loop that feeds the relay, and that's going to make it radiate.

dI*A ---- dt

Indeed :)

^ +---+----------------- | | | | | | | |+ `----------[Lcable]------------[Lcoil]-. | --- | | | --- --- | H | /_\\ | | | | | | | .--------------------------------------' | | | | | [transistor] | | | v +---+--------------

^ +---+----------------- | | | | | | | |+ `----------[Lcable]------------[Lcoil]-. | --- | ^ | | --- --- / \\ | H | /_\\ /---\\ | | | | _/ \\_ | | | .--------------------------------------' | | | | | [transistor] | | | v +---+--------------

Switch turns off; current flows in the coil through the free-wheeling diode.