09 Mazda 3 Auto Trans delayed engage, No Codes yet,AT light came on for a second

2009 Mazda 3 2.0 it is the F VIN type with a 4 speed automatic transmission. History, about three years ago on cold start the car started to display on the gear indicator LED 1st gear briefly, then 2nd gear, then 1st gear. If it showed 2nd gear for too long the car would go into limp mode. Shutting off the engine and restarting everything seemed to clear up and you could drive as normal for a few days. The code it showed when it would go in limp mode was P0894. The problem started happening more often so I took the car in and had the transmission rebuilt. When I got it back everything was good for just over a year (warranty on transmission was 1 year) then the original issue came back where I could see the gear indicator flashing between 1st and 2nd gear but so far it has not caused a code. This has been going on for about a year, due to covid work from home, I have not put many miles on the car during this issue. It keeps getting worse to where it was a delay engage type issue for a while. I took the car to a person that was supposed to work on car computers and he said he could not find anything wrong but went ahead and re-flashed the PCM. The other night the AT light came on briefly while the gear indicator showed 2nd gear for a second or so, but still no code. I have ran Forscan on it and monitored the solenoid PID’s. From what I have read Solenoid 2 (ssb) should 100% and Solenoid 3 (ssc) should be 100% for first gear. When I put it in Drive from a cold start Solenoid 3 (ssc) goes straight to 100%, Solenoid 1 (ssa) starts to go up to about 20% then falls to 0% and usually during Solenoid 1 (ssa) descent or when it hits 0% is when Solenoid 2 (ssb) goes up to 100% and gear indicator shows 1st gear. During the time when Solenoid 1 (ssa) and Solenoid 2 (ssb) flip flop it shows 2nd gear on indicator. I believe this is because Solenoid 3 (ssc) at 100% and others at 0% is second gear. Once it locks into first you can move to neutral and back to drive with no delay. When I say a cold start this is relative as I live in South Florida so never really cold, just meaning the car sat for a couple of hours. I can manually actuate solenoid 2 (ssb) with a scan tool and you can hear it clicking as I change the percentages. I was thinking it was a bad PCM possibly (admittedly I was firing a parts cannon at it) but the rebuilt PCM shows same issue. I do not have a real oscilloscope only one of those cheap pocket ones from amazon not sure it will even work for automotive use. The speed sensor seems to display correct value on scan tool as the car is sitting still when all this happens. I do get intermittent CAN codes but they have been going on for a long time GEM has a U1900 and Instrument cluster has a U2516, but they are CAN error codes. If needed I have some of the forscan recordings saved that I can attach. Any help would be greatly appreciated.

I want to thank everyone, I am not a mechanic by trade this is my personal car so my terminology may not always be correct, please bare with me.

I am not sure why the image is showing twice and I do not know how to fix it.

I included an image of the pins on both PCM and Transaxle connector I was testing.

I decided to check the wires between the PCM and Transaxle. This continuity test was done with both connectors disconnected from PCM and transaxle. I just wanted to test the wires for short or open. I know voltage drop with load is better but this was just quick and easy test.

When I did a continuity test 1G on the PCM connector to D on the transaxle connector I had continuity which seems to be correct based on wire color Black with red stripe.

When I did a continuity test 1H on PCM connector to I on the Transaxle connector it was open, again based on wire colors this seemed to be correct pins both Green with red stripe. Seems like a break in the wire.

I then did a continuity test between 1H on PCM and to D on the transaxle and I got continuity. Like there is a short between 1H and 1G in the wiring harness?

I tested Pin I on transaxle to ground, Pins 1H and 1G it is open. Like it is going no where. From what I can see the transaxle pin I is ground coming from the PCM which would be pin 1H.

It looks to me like when the PCM energizes 1G, instead of going through the pressure control solenoid it is bypassing the pressure control solenoid back to PCM for ground on pin 1H. And from watching the video’s here, this seems to be a circuit the computer wants to monitor that is why it controls both positive and reads negative resistance? This will make computer think there is no resistance as the pressure control solenoid is by passed?

Would this give me the delayed engagement and trying to go into 2nd gear when put into drive on cold start? Why wouldn’t this always happen not just on cold start? Or am I over looking a reason why these two pins should be connected together, is there another component between that could cause this and it be normal operating procedure?

Cut wires and added new wires to remove the short between the two wires. I believe the problem is still there as the Diagnostic scanner still shows solenoid 1 (ssa) going to 100% and back to zero while solenoid 2 (ssb) starts to go to 100% after solenoid 1 (ssa) starts to come back to zero. I do not know if the gear indicator showed gear as 1 then 2, then back to 1 as I was looking down at the diagnostic scanner. Will look in morning when I try another cold start.