MPGuino release one workspace

[dcb]

For now, yes. There's a few glitches yet to sort out before we can calibrate it with a fillup, but it shouldn't take more than a fillup or two to be able to compute an accurate microSecondsPerGallon value. And it shouldn't take too many users reporting the values they use for their cars before we have a database of parameters, and maybe some fancy schmancy online lookup tool or whatever to help new users get started.

But we here are on the bleeding edge of this thing

[awillard69]

As a followup to the LCD issue. I don't think it's the LCD.

I was able to reproduce the problem, and with tracing on, here is what I got:

instant: 0,0,0,0
current: 0,0,0,0
instant: 0,0,0,0
current: 0,0,0,0
instant: 0,0,0,0
current: 0,0ü

Notice the funky character - the Guino apparantly crashed. I've repeated the experiment multiple times with the same basic result - inoperative unit.

I'll continue to investigate.

[dcb]

Do you have an ICSP figured out? I should send you my hex file and see if it is an ubutu/linux thing or something.

[dcb]

here is a hex file I just created from post 1, if you have iCSP figured out you can try that. If not then we'll figure something else out.

[ttoyoda]

My power supply is a 130w power supply from an old PC, so it should be nice and clean. I put some taps on the front for the 5v and 12v and it works just fine. A good cheap trick for an already engineered power supply with multiple voltages!

I always though that these supplies needed a significant load to actually give the correct voltage. Have you had a chance to measure the voltages when the duino is the only load?

[ttoyoda]

While we are on the subject of how harsh this environment is for electronics may I suggest LM2940T-5.0 as a replacement voltage regulator with the proper filtering. I too also agree with opto's maybe 4n35 or at the very least a transistors on the input. 3 or 4 bucks extra will go a long way in possibly saving to replace Atmega etc, just my 3.26 cents

http://www.national.com/ds/LM/LM2940.pdf

A nice choice, what with data sheet info saying this is specifically for "vehicular" environments, and that it has overvoltage shutdown in case of higher voltage spikes. It can withstand 60 volt spikes in certain versions. I don't say this is definately better, but I would be tempted to use the 9 volt version, and use that to feed the 5 v regulator already on the board (or at least on the decimellia I got). That way we would share the thermal load of the voltage drop between the two regulators. Might get away without a heatsink.

[ttoyoda]

We delivered over 200 [SuperMID] unit in the field, and we never heard any failures on the interface circuit.

That is very, very impressive. Much better than we manage at work. Have there ever been any failures at all that you were able to find a cause for other than that old standby "user applied corona gun to circuit board"

[MilesPerTank]

Originally Posted by dcb

For now, yes. There's a few glitches yet to sort out before we can calibrate it with a fillup, but it shouldn't take more than a fillup or two to be able to compute an accurate microSecondsPerGallon value. And it shouldn't take too many users reporting the values they use for their cars before we have a database of parameters, and maybe some fancy schmancy online lookup tool or whatever to help new users get started.

But we here are on the bleeding edge of this thing

Is the software operating under a condition that would need the microsecondspergallon value of one injector to be divided by the number of injectors?

I'm trying to base my initial value off something semi factual as I have seen multiple things in the software off by factors of 2. It makes me want to avoid manipulating things to fit here and there.

[dcb]

Originally Posted by MilesPerTank

Are you really just randomly adjusting it to "look right"?

I should clarify, I've had other mpg gauges in my car and various fillups so I already have a good idea what mpg it should be @50mph. The adjustment isn't technically random, you adjust the vss and the microseconds/gallon up or down by the percentage that the display is off. Same technique will apply at fillup, except it should be possible to get the computer to help out a little.

Originally Posted by bockwho

...it is caused by the horn .. after some testing today no matter where the unit is in the car if you sound the horn it several H's and M's will go full black..

Someone remember to test their horn

[dcb]

Originally Posted by MilesPerTank

Is the software operating under a condition that would need the microsecondspergallon value of one injector to be divided by the number of injectors?

Yes, if there are other injectors squirting fuel then it may just be as simple as that. A larger microseconds per gallon will give you better mpg, a smaller one means it takes less time to squirt out a gallon. So I think your example left like 350 seconds/gallon after deviding by 4. There are still a lot of variables like fuel pressure, but you will be within an order of magnitude. I'd guess 500 seconds a gallon. But remember to get the distance/speed sorted out before adjusting the fuel.