Why do I get the feeling this thread was DOA?

It certainly is an interesting challenge! And guessing what sort of system needs timing that close is also an interesting exercise. Certainly there are many semiconductor devices that can switch off rapidly enough, and carry the required current, but every one of them requires driver circuits that need to be powered. And all of those driver components will need to be adjusted to produce the correct speed and delay times.
Most mechanical relays also suffer from a bit of contact bounce, both making and breaking. Actual speed limitations can be overcome by driving harder, but that gets complicated as well.
So now questions: First, what other difficult requirements are there?? Next, how often will cycle on and off?? How much contact bounce can be tolerated?? Does the relay circuit have other requirements such as low capacitance or a wide temperature range? AND, what are the special requirements for the circuit closing??? Is there a space or weight limitation??

 

You might be able to do it with a 4 pole relay with the 4 contacts opening at nearly the same time.
10µs timing it still likely problematic, but it would be relatively simple to test.

None of the four pole relays I have checked are close and even double pole relays have significant differences between the poles. There are a few single pole relays that seem to be consistently close to others out of the box. I am leaning towards using those with a time offsets on the control side. That gets me under 10 uSec easily but I don't know if that holds over time. I thought to bring the question here before I went through with designing a calibration function to make sure I wasn't missing some easy solution. It doesn't appear so. I got a suggestion somewhere else to use a MEMS switch/relay which looks promising but they aren't mature enough in the market at the moment. Thanks for the suggestion.

 

It certainly is an interesting challenge! And guessing what sort of system needs timing that close is also an interesting exercise. Certainly there are many semiconductor devices that can switch off rapidly enough, and carry the required current, but every one of them requires driver circuits that need to be powered. And all of those driver components will need to be adjusted to produce the correct speed and delay times.
Most mechanical relays also suffer from a bit of contact bounce, both making and breaking. Actual speed limitations can be overcome by driving harder, but that gets complicated as well.
So now questions: First, what other difficult requirements are there?? Next, how often will cycle on and off?? How much contact bounce can be tolerated?? Does the relay circuit have other requirements such as low capacitance or a wide temperature range? AND, what are the special requirements for the circuit closing??? Is there a space or weight limitation??

You are correct on the issues with semiconductors. Some options change timing depending on load voltage/current which isn't consistent here. The normal switches all seem to have an issue that makes them undesirable (MOSFET body diodes, etc). Coto make a little telecom relay with a pretty clean release. They get a little dirty after breaking their rated current a few times but they still don't end up being terrible. They will work for this but I don't know if there are minute mechanical changes affecting timing that occur over time with changes in storage temperatures. To account for that I'll build in a calibration circuit and but that is a mountain of extra parts and programming so I was hoping to avoid that.

The other requirements are pretty easy. Ultra low cycles. This is almost a one-time use item. Capacitance doesn't matter and the temperature range is from 0-120F. The circuit will be closed when it is subbed into the circuits it will be breaking so that doesn't much matter either. Small and light are preferred like every application. I am hoping to pack it into about 6-8 cubic inches including the battery and anything under 12 ounces would be great.

 

Long ago I implemented a dual audio signal generator with two PIC16F84A using a common crystal (4MHz IIRC).
I forgot most of the details of the project but once started, both outputs were NEVER in phase but ALWAYS with the SAME phase difference.

This could spark some ideas to get that disconnection in sync.

 

Except for the voltage drop you could use two heavy duty triode vacuum tubes in anti-parallel as the switches, I recall that now from the very first industrial electronics course I took. It was useful at the time because some old equipment still used tubes.
Presently, some industrial power relays do work rather fast. BUT there is still the contact bounce issue