Rotary EGT imbalance conundrum

I've had my RX8 up and running now with a street ported S4 13b (6 port N/A ). I'm running a 4 barrel style intake and throttle body and MS3 ECU. Fuel is handled by 450cc/min primary injectors and staged 1000cc/min secondaries. 4 FC trailing coils are lighting the fire. I was happy with my conservative base mapping and was moving on to some finer calibration to get my mileage up. I have EGT bungs about 6" down the header and I'm monitoring them with my DMM. Right off the bat my front rotor was reading really hot, 900C+ on the expressway at low 14:1 air fuel and 25-30 degrees of advance. Rear is a happy 700-800C. I suspected a bad thermcouple so I pulled it out to find the tip burned off... this wasn't that surprising as they were cheap and only rated to 400C. I replaced it a proper 1000C rated EGT thermocouple and found the same results so I tend to believe the numbers aren't lying. 
I assumed I had a clogged injector so swapped injectors front to rear but no change, front rotor 150C hotter EGT. I trimmed 9% extra fuel to the front rotor. This helped but still up to 100C hotter on the front rotor. If it's an airflow or fueling problem the error would have to be huge to be that far off.
I double checked my ignition, I verified timing with a light and I'm running zero split at low load so I'm pretty confident I'm getting same same front to rear.

I'm sure I'm forgetting some other checks I've done but I'm at a loss, what am I missing! 

Tried swapping the thermocouples front to rear?

Assuming the temperature difference is real- since you've spend a lot of effort trying to balance the two rotors via fuel, then you need to see if there's shifts in spark or air.

One thing I would very much check is that the leading spark plug is working properly.  If it's not, then the second one would be very similar to running retarded spark- which does increase the exhaust temp a lot.  Basically, I'd look for any thing that would cause the hot rotor to run retarded spark relative to the cool one.

Then there's air- if spark is identical and the fuel trimming still gives you a big imbalance, then it's air.  One rotor is getting more air than the other, and doing more work, so it's hotter.

(this all assumes that you don't have things that are there to slow down combustion, like EGR, or that the internal mixing of the two make sure that the exhaust back flow during the intake is actually the same- EGR will retard the combustion enough to cool things down)

I assumed the temperature difference was real when I saw the front thermocouple melted and the rear intact. I'll have a new thermocouple for the rear arriving tomorrow that I can swap front/rear double check the difference. Can't go assuming things and expect to solve a problem...

I've done the best I can to verify all coils are firing. ECU test mode, checked plugs briggs block, and timing light all check out.  Ignition timing seems to be a bigger knob to turn than fuel for EGTs. I've gone between running 8 and 0 deg split. If it was only firing on the trailing EGTs should have risen with more split (more retarded) but I didn't notice a change. 

On the airflow side I have one worry, I removed the aux port sleeves and devcon filled/smoothed the port transition. It's possible the epoxy came loose on the rear rotor and is choking the port. My throttle body and intake runners are huge and the imbalance is there whether MAP is 40 or 99kPa. If there's a restriction it's downstream of the intake. I think I have to get a borescope in there and take a peek.

EGR is a tricky one, no external EGR and my porting is mostly later closing intake and earlier opening exhaust. I'm showing 90psi on the rear and 75psi on the front rotor. Not great but not bad for as it's still breaking in. Front rotor being lower would suggest more seal leakage and dilution, if anything that should make the front rotor richer/cooler. I'm not seeing any blow by into the crankcase either. 

Finally got some garage time and a new thermocouple on the rear rotor after suffering through goobered threads on the bung. Both are now the same 5mm probe rated for 1000C and the temp difference is showing a little lower (rear showing  little hotter than previous) but still showing a temp difference up to 80C in 6th gear cruising at down the expressway and the extra fuel trimmed to the hot front rotor. 

I should add, I'm sure some of the elevated EGTs are due to the fact that I'm running a pretty huge Racing Beat "street port" with 84-deg BBDC (13-deg eariler than stock, see below) opeing. My header is also wrapped in some McMaster Carr ceramic wool, and cheap ebay header wrap. Not suprising heat, and noise, are in abundance. I just can't explain the front/rear difference!

A thermocouple reads the difference in temperature between the two ends of the pair of wires.  Just as a sanity check, you do have the amplifier side of the two thermocouples in the same location, right?  If one was in the engine bay and one was in the cabin, that could explain a significant difference in reading...

I wrote a long missive about the stock ping-pong intake manifold and then I saw that you're doing it right with a Holley type manifold.  I assume that all four runners are separate with no plenum.  Good job.

I'd run a pair of widebands.  I think the MS3 is capable of reading a pair of O2s that way.  You may have weird fuel trim issues for any number of reasons.

codrus said:

A thermocouple reads the difference in temperature between the two ends of the pair of wires.  Just as a sanity check, you do have the amplifier side of the two thermocouples in the same location, right?  If one was in the engine bay and one was in the cabin, that could explain a significant difference in reading...

I'm reading the front rotor on my DMM in the cabin and the rear through an AD8495 based conditioner underhood back to megasquirt. the conditioner is cold junction temp compensated within my underhood temps so I shouldn't be seeing much error. I have switched front to rear but not since the new TCs. I'll have to do that again to rule out instrument influence.
playing with fusion TC conditioner

Knurled said:

I wrote a long missive about the stock ping-pong intake manifold and then I saw that you're doing it right with a Holley type manifold.  I assume that all four runners are separate with no plenum.  Good job.

I'd run a pair of widebands.  I think the MS3 is capable of reading a pair of O2s that way.  You may have weird fuel trim issues for any number of reasons.

I really like the Holley style manifold, no plenum or open spacer, all 4 plates open together. Throttle response is incredible as is the noise. 
I worry that widebands wouldn't survive long in the header. I have heard the LC-1 (what I'm running) can stream raw data fast enough to differentiate the cycle by cycle values in a collector. I have a USB O-scope that I may have to test this with. 

I eat wideband sensors like its candy. I don't even keep one on the car anymore. Only for tuning now.

Knurled said:

 

I'd run a pair of widebands.  I think the MS3 is capable of reading a pair of O2s that way.  You may have weird fuel trim issues for any number of reasons.

My MS3X tuning interface shows room for up to 8 O2 sensors.

Well I pulled the intake off and confirmed the epoxy in the aux ports hasn't dislodged, a big relief there. This also let me quadruple check for vac leaks - again none found. I scoped my injectors and I'm getting the same pulsewidth as commanded so I'm going to rule out a driver hardware issue. I swapped thermocouples and showing no difference, front is still showing hot EGTs. I think my next step is to log both EGTs and trace the delta to conditions to see if I can find an explanation there. I'm up against the weather at this point. I don't have my heater core plumbed and they I'm hearing talk of white fluffy stuff in the forecast. Maybe I just need to go ahead with the turbo and let the chips (of apex seal) fall where they may.