V3.3.0 firmware / configuration software - for V3 hardware

Post by **brentp** » Mon Jan 12, 2009 7:31 pm

Thanks for the additional information. Unfortunately, I was side-tracked with other things. I'll try to look at it today.

Regards,

Post by **brentp** » Wed Jan 14, 2009 8:51 am

I tried to reproduce this, but I did not reproduce what you observed.

I am running with the default calibration, but after reviewing the code I don't see how changing calibration would affect things. Still, I will look at it a bit more.

If anyone else can reproduce this, showing results one way or another, that would be great!

Yvan · Post by **Yvan** » Wed Jan 14, 2009 5:49 pm

Here is screen recording of what I did (I prepared it in Linux, so I hope that you can watch it in winblows, it's DivX file):

http://www.ptt.rs/korisnici/i/v/ivica/slike/demo.avi

15.01.09 - I did some editing on the original file, and added another version to youtube:

http://www.youtube.com/watch?v=00C4QRAV3RA

Patriq · Post by **Patriq** » Fri Feb 06, 2009 1:20 pm

When connecting several outputs from the sensor in parallel, don't you affect the readings ?
I remember at least that, if you wanted to run a MegaSquirt in parallell with stock ECU - if you had both stock ECU and MS connected to WaterTemperature sensor, you needed to remove a bias resistor from the MS pcb so not to confuse the stock ECU.

Or am i completely wrong here ? I have a a gauge on my instrument panel, and intend to connect to my MJLJ in parallel for water temperature.

Yvan · Post by **Yvan** » Fri Feb 06, 2009 3:28 pm

That was my idea too, but I was told that it probably would not work, that I'll probably need to add another sensor.

BTW, did you try to connect temp sensor to MJLJ? Can you reproduce my problem? Or is it working as it should?

I again downloaded both tuning software and controller firmware, flashed MJLJ, installed tuning software on the freshly installed OS (Vista), and nothing changed.

Post by **brentp** » Mon Mar 09, 2009 3:56 pm

3.3.1 config software has been updated- restored missing engine configuration combo-box.
http://www.autosportlabs.net/MJLJ_V3_Do ... d_Versions

When I was re-testing this software with this firmware I was observing periodic timeout errors while reading runtime data, but only with the 'engine' running. This was also happening with the 3.3.0 config software.

Funny that I didn't observe this before; can anyone else with V3 hardware running this firmware verify?

Thanks,
Brent

Yvan · Post by **Yvan** » Sun Mar 22, 2009 6:41 pm

I do not have running hardware, but I can confirm that I still have the same problem with user configurable outputs.

Gauge shows temperature changes, but outputs do not change from ON to OFF (and back) at preset values.

Post by **brentp** » Mon Mar 23, 2009 7:15 pm

Yvan,

You might have done this already, but since I have not been able to reproduce on my end (yet) I would suggest re-loading the beta firmware, with the default AUX input scaling, and verify the voltage change directly on the ADC input pin on the processor. At the same time, monitor the user output state.

If someone else can test this feature that would be immensely helpful.

Thanks,

Yvan · Post by **Yvan** » Tue Mar 24, 2009 6:53 am

But if AUX input gauge shows temp change, surely that part must be OK?

Anyway, I should measure voltage change at AUX pin (and ground) on the board? Nothing there. It reads 0.00V. But I just heated temp sensor from a room temperature to the body temperature. AUX input gauge went from 60 to 47.

Yvan · Post by **Yvan** » Sun Mar 29, 2009 2:24 pm

Brent can you answer please, should I measure voltage change at AUX pin (and ground) on the board?

Post by **brentp** » Sun Mar 29, 2009 5:36 pm

Hi Yvan,

Sorry for the confusion. To clarify what I was requesting:

1. Reflash with the latest beta firmware (to eliminate any changes- especially to restore calibration data)
2. Using your volt-meter, monitor the ADC input pin right on the processor chip (pin 8 )
3. Compare the aux input gauge in the software against your volt meter- and test the user output in AUX mode- verify the voltage is changing with the gauge, and the output triggers correctly.

I will also double-check on my end. If anyone else can test as well, that would be great.

Regards,

Yvan · Post by **Yvan** » Mon Mar 30, 2009 5:21 am

brentp wrote:2. Using your volt-meter, monitor the ADC input pin right on the processor chip (pin

Just to be sure, it is this pin, circled in red:

Post by **brentp** » Mon Mar 30, 2009 5:47 am

Correct, pin 8 on the CPU.

Yvan · Post by **Yvan** » Mon Mar 30, 2009 3:37 pm

OK, I did it.

I used a cup with a cold (tap) water, and a cup with a warm (~65 Celsius or ~150 Fahrenheit) water and dipped temp sensor to each cup.

No voltage change. It stays at 0.00V.

But when I pulled out the sensor from cold to warm voltage raised to 1V and gradually went down. Change from warm to cold resulted in instant 0.00V. Again cold to warm up to ~1V and gradually down.

So I thought it may have something to do with a fact that cup for cold water was made of metal. I changed to ceramic one, and now again cold or warm it was 0.00V.

Again back to metal cup - 0.00V.

What voltage change should I expect here?

Just to mention all the time AUX gauge was changing from 70 (cold) to 25 (warm).

DannyP · Post by **DannyP** » Mon Mar 30, 2009 10:49 pm

Yvan, don't take as an insult, but was the sensor grounded when you tried to take the readings?

This gives me an idea for a warmup program on my car. With a temp sensor and the aux input, I can add a few degrees of advance to keep the rpms up until the engine is warm. Does this make any sense to anyone?