Have not had any issues with the car in years and right at first start of the day a few days ago two coils failed at the same time or quick succession. One melted down so much it was nearly on fire and the other just stopped working. These were #1 and #6, produced codes for cylinder #1 misfire and random misfire. I replaced both coils with new and the car ran fine for a few days and then today the same exact thing happened again on first startup, except this time it is #3 and #6. In both cases it is #6 that just stops working and the other melts.

This is just an odd pattern and something must be killing the coils but not sure where to go on it. I was wondering if anyone has any ideas? Thanks.

 

Wow, yeah, I agree... a very odd pattern! A few diagnostic questions...
On the original failures, I assume these were OEM? Did you replace with something like HITACHI or NGK, or cheap, off-brand? When you replaced, I assume it was the entire assembly, including the tubes? If not, then you should definitely replace the tubes as well (with the spring inside).

So, if I'm reading correct, #6 has stopped working twice now, and you've had two separate coils, (#1 and #3) fail in a destructive manner (overheating to the point of melting). Really the only thing that would cause a coil to get that hot (other than a blow torch would be some sort of failure of the solid-state switch inside the potted coil body. In order for substantial HEAT to be generated internal to the coil body, it has to be dissipating POWER (watts). During normal operation, these very simple devices should NOT be dissipating much power because the internal NPN transistor is being driven hard enough to go into saturation and operate like a switch (i.e. low power dissipation). If the transistor stops operating like a switch and goes into its linear (acting like an amplifier) state, then all of a sudden it will be dissipating a lot of power across the device and probably die very quickly.

At this point, we have to ask ourselves, is there a common element that could potentially cause the transistors to not be driven into saturation and subsequently kill FOUR coils in three locations in a short amount of time?
1) It would be very difficult for the ECU to outright kill an ignition coil. Can probably rule this out.
2) My money is on a grounding problem.

I would inspect each of the harness connectors at the coils and carefully measure the resistance from the coil connector's GROUND terminal, back to the negative terminal of the battery. They should all be in low tenths range, like 0.1 to 0.2 ohms. A decent Ohmmeter is best for this, but if you only have a low-cost version (Harbor Freight, etc.), then touch the two leads together and determine what the meter "thinks" is zero. It likely won't be zero, but you would treat this number as if it was zero. For example: touch the meter leads together while in resistance mode and get 0.2. If you go measure some circuit and it says 0.3, then the real measured value is 0.1. Pretty straightforward.

If you find any coil-connector-to-ground values that are out of family or if they all seem high, like in the >0.5 ohm range, then you probably need to trace down the harness ground point at the block and clean it very well and take a critical look at the harness in general. There are two ground points right about the front valve cover. I'm not 100% but I think one of these is the coil harness ground point.

I've included the relevant schematic below for reference. Also, I think until you find the issue, it sounds like you might continue to blow coils if you just keep replacing them.
Also notice that there's a "condenser" (old terminology for "capacitor") in the circuit that's wrapped up in the harness. Its job is to smooth out (filter) the "ripple" on the 12V rail, caused by all the switching (coils and injectors). I highly doubt that a bad condenser could be a cause of failed coils, but it would be interesting to know if, when you turn on the stereo, if you're getting a lot of engine noise (whining)?

Good luck tracking it down!

 

Once again, pethelman has an excellent response. I'm not sure I can add anything except speculation.

He's right, start by checking all of the ground wires for the engine wiring harness. There should be two ground wires on the front going to the intake manifold. Check other ground wires in the rest of the engine bay as well. You may want to peel back some of the factory wire loom/electrical tape going to the coil packs and see if you can spot any split or exposed wires. The wires in our cars are getting old and brittle.

I would also pull and inspect all of your spark plugs, noting which one was in each cylinder. Check the gap for each plug too and see if any have bigger gaps than other ones. While they're out, replace them with some fresh NGK plugs unless you know they've been recently replaced.

Also, let us know if your coil packs are original OEM, and what brand you're replacing them with. If the coil packs aren't made in Japan, they're probably junk.