MS2/Extra 2.0 beta 21 (20080210)

6040solder · Post by **6040solder** » Fri Feb 15, 2008 3:21 pm

I wouldn't be embarrassed if I were you. At least one person in every beta release thread has done that since they expanded the memory past the official limits.

So, you're not alone!

vlopsahl · Post by **vlopsahl** » Fri Feb 15, 2008 4:52 pm

TPS lagfactor not working anymore?

It has bin a while since I have been using my car and played with Megasquirt. I was running perfect with beta 12. Upgraded to beta 17b to play with pwm-closedloop. When I started the engine it was running very rich because accel enrichments was turning on and of. My TPS have some noise and I have to use much lagfactor to make it stable. The quick solution then was to turn off TPS-accel enrichments. Today I upgraded to beta 21 but the problem is still there, jumpy TP and TPSdot and it does not seam to matter what lagfactor I use anymore. I will test some more tomorrow and try to make a log of my problem.

Thanks
Vegard

JeroenBosma · Post by **JeroenBosma** » Sat Feb 16, 2008 4:34 am

Just download the file, unzip, run the file "download-ms2-firmware.bat" and follow instructions.
After that: run "copyini.bat".

That will set your MegaTune up for the new firmware.

vlopsahl · Post by **vlopsahl** » Sat Feb 16, 2008 5:08 pm

Sorry. Seams my TPS is playing with me again. I don’t think there is a problem with the TPSlagfactor. Tried a few older betacodes today, but my noise problem was still there no matter what factor I use. Maybe I will have to use a shielded cable for my TPS signal…

Beta 21 seams to work as it should on my simple Audi 1,8T with 60-2 trigger and waste spark.

Thanks
Vegard

Keithg · Post by **Keithg** » Thu Feb 21, 2008 9:35 am

Finally loaded this version. I was running a pre-version with the multiple PWM idle frequency selections.

All I can say is awesome. I am finally back with a drivable car that I can have valeted or loaned to my wife without fear of repercussions. I am running 100ms loop time and 30hz PWM frequency. It adjusts nicely to load. It is winter, now, and I am sure I will have to adjust the PID constants in summer as a tweak, but it is very nice with my Bosch 2 wire valve.

I noticed one thing with the pre-version that it would hold open the AIC valve every so often following a hard accel when throttle was dropped. It would adjust for this quickly. I will need to log this to be able to figure out exactly what is going on or even if it is apparent with the 2/10 version.

Thanks!

Keith

AbeFM · Post by **AbeFM** » Thu Feb 21, 2008 11:10 am

Ah, I'm going to try it myself, I'd read above that the code was "broken", but if not then it's on my list.

I've found out that the Miata uses a very high 488 Hz, which would be a 16 on your scale, but it's much quieter at least on 244.

Keithg · Post by **Keithg** » Thu Feb 21, 2008 2:06 pm

Abe,

Keep in mind that the frequency is inversely proportional to the number of steps. I needed more steps to actually allow the PID controller to have enough 'room' to operate. This is truly a balancing act between number of steps and noise due to lower than desired frequency.

Hopefully, there will be a delicate hardware/software modification in the not too distant future to have true PWM control for the idle for those of us running Hi-Z injectors. That way you can have both higher freq and reasonable step size.

KeithG

AbeFM · Post by **AbeFM** » Thu Feb 21, 2008 2:14 pm

Uh huh, just putting it out there was all. As I said, it appears to be working well.

Keithg · Post by **Keithg** » Sat Feb 23, 2008 6:28 pm

Ken,

There is something odd about the new idle code. It does not seem to be in the logic as that has not changed and it is predictable and acts normally. It acts normally in normal driving. It is just when throttle is dropped abruptly from a higher RPM and decent load. (hard accel then drop throttle at 4krpm or so.) RPMS hang and drop as though the AIC percentage was not at a value that the MT DC% shows. My guess would be that if the 'dashpot value' was 37%, it was probably at 50% or so when the idle control engages.

Traffic was terrible on Friday, so I was not able to get a log, but I really do not think that MT reflects this, best I was able to see on teh laptop real time. It should be apparent on an oscilloscope trace, though. I now have a jimstim and a spare MS2 board. As soon as I get the board modded, I'll be able to see what I can see on the bench, but I thought I'd mention it here, now.

most of the time, you would not notice this. But it can surprise you in some situations.

KeithG

Post by **muythaibxr** » Sun Feb 24, 2008 2:27 pm

I've tried to duplicate that on the bench with the scope and can't get it to do anything weird... I wonder if your valve is just getting stuck or something b/c of the low frequency (I've seen that happen).

Ken

Keithg · Post by **Keithg** » Sun Feb 24, 2008 4:18 pm

Could be, I guess. I'll have to see if I can log it on the car and see if I can get my MS2 chip set up on my bench as well.

KeithG

muythaibxr wrote:

I've tried to duplicate that on the bench with the scope and can't get it to do anything weird... I wonder if your valve is just getting stuck or something b/c of the low frequency (I've seen that happen).

Ken

oskar · Post by **oskar** » Fri Feb 29, 2008 8:31 pm

6040solder wrote:... since they expanded the memory past the official limits.

as I am trying to learn ... can you be a bit more detailed, please?

Thx
Oskar

6040solder · Post by **6040solder** » Sat Mar 01, 2008 3:31 am

Yes, the chip is the C64, and they are treating it like its the C128 (which it effectively is) however, the bootloader *thinks* its the C64, so there is bootloader code built into the firmware to handle that.

Detailed enough? :-) There is more if you search.