mileage computer for a carbureted engine

[TheTestPilot]

Post 186 concludes it's good for a fuel guage only ,nothing else.
Yes , My Vehicle has fuel cut off , my drive is uphill one way with 6km of flat freeway of the 35km journey.
I look forward to your recoding for the paddle wheel sensor ,before I proceed any further with my flow sensor.

[TheTestPilot]

Hi Superturnier
What you have described with the vacuum sensor (GM map) is sitting on my bench ,although the problem with this system it does not take into account power valve enrichment or Secondary barrels + enrichment if so fitted & as you mention the choke,so who knows what the outcome will be?

What we are tallking about here is not the end product, this is just part of the development process,which has the potential to be a stinker at the end.

I have used Zemco's since the mid 80"s & as you note that ball sensor was just unreliable ,Have also tried other sensors too ,but most of these not linear & at low flows simply do not measure anything,which also causes problems with accuracy.

[Quezacotl]

Originally Posted by Quezacotl

I just got few flow meters, and i'm gonna attach them to my car soon to test how it works.

Two inexpensive flow sensors, 4011-logic-ic to make some logic to in/out flows and MPGuino.

Until now, it sounds easy, but we'll see

Now i have read the entire thread. So no real results from this idea...
Like someone said, it may be tricky to get proper reading at MPGuino. If i don't have enouh skill to modify the code, i'll make a pulse counter myself.

I actually have an electronic injection in my car, but i want to test this too, just for fun. And besides i can put two meters side to side so i can monitor how much they differ.

It works! Atleast in the test bench.
I hooked the device to MPGuino, and it really counts the pulses, and looks realistic.
The MPGuino counts the injector pulse durations, but in this case it apparently counts the time how long it takes the metal piece in the flow sensor to pass the hall sensor.

Now i just need to get some pipe somewhere to install this in the car

[AndrzejM]

The problem is that you have to install two flow sensors and count the difference between those two readings. That will give you an amount of consumed fuel. But changes can be very slow so there's no use to just count the pulses for instant FE readings. To do that you have to measure time between the pulses on both flow sensors and that will give you more accurate reading of instant FE.

[Quezacotl]

Originally Posted by AndrzejM

The problem is that you have to install two flow sensors and count the difference between those two readings. That will give you an amount of consumed fuel. But changes can be very slow so there's no use to just count the pulses for instant FE readings. To do that you have to measure time between the pulses on both flow sensors and that will give you more accurate reading of instant FE.

It does just that. The circuit subtracts the outgoing from ingoing fuel. MPGuino gets just one signal. But we'll see what is the result

[AndrzejM]

That's great. You have a nice setup then. Didn't get it before that you have anything else but flow sensor, and you have an additional unit to count the difference between flow sensors.
Could you please share some more details on that?

[Quezacotl]

Originally Posted by AndrzejM

That's great. You have a nice setup then. Didn't get it before that you have anything else but flow sensor, and you have an additional unit to count the difference between flow sensors.
Could you please share some more details on that?

It is just logic-ic, 4011, NAND gates in a row. I made it like this, truth table:
00 0
10 1
01 0
11 0
Meaning it just subtracts outgoing fuel from ingoing.

I have it installed now in my car, and seems to work properly, no crazy jumping with instant reading, it's stable enough.

Until now, it seems to work, but i will test it with couple of tanks to see if there's something.
I will have over 500km driving at this weekend, so that is a good testing time


But funny that when i turn the key, but not start, it shows that fuel is used. At the same time when i can hear a short "whrr" from under the hood.
I did'nt know that the car pumps a little amount of fuel every time i turn the key. But that's logical, you can't start sparking with no fuel..

[AndrzejM]

Originally Posted by Quezacotl

It is just logic-ic, 4011, NAND gates in a row. I made it like this, truth table:
00 0
10 1
01 0
11 0
Meaning it just subtracts outgoing fuel from ingoing.

I have it installed now in my car, and seems to work properly, no crazy jumping with instant reading, it's stable enough.

Until now, it seems to work, but i will test it with couple of tanks to see if there's something.
I will have over 500km driving at this weekend, so that is a good testing time


But funny that when i turn the key, but not start, it shows that fuel is used. At the same time when i can hear a short "whrr" from under the hood.
I did'nt know that the car pumps a little amount of fuel every time i turn the key. But that's logical, you can't start sparking with no fuel..

From the truth table it seems that pulses from the second flow meter are completely ignored.
A bit of theory...

Let's say that engine is using the amount of fuel corresponding to one pulse per minute, but fuel pump is pumping 6 pulses per minute. So during a minute of driving you should read 6 pulses from the first flow meter and 5 from the second one, right?
So in one minute the table of pulses would look like below

seconds	FM#1	FM#2
0	0	0
10	1	0
12	0	1
20	1	0
24	0	1
30	1	0
36	0	1
40	1	0
48	0	1
50	1	0
60	1	1
total	6	5

6-5=1 pulse FE

And with your truth table

seconds	FM#1	FM#2	pulses for MPGuino
0	0	0	0
10	1	0	1
12	0	1	0
20	1	0	1
24	0	1	0
30	1	0	1
36	0	1	0
40	1	0	1
48	0	1	0
50	1	0	1
60	1	1	0
total	6	5	5

total pulses sent to MPGuino 5

So counted FE is 5 times bigger!

I think that simple NAND logic is not enough to do the job. It has to be a time related function. IMHO best way to do that will be to count time between pulses on both FMs. Then just counting the difference.

[Quezacotl]

Oh, you are right! I calculated it wrong. I was thinking that if the first FM is rolling about continuously, and second almost continuously, then the second would be subtracting a pulse that happens to be in first FM at that time.
But i never thought that they might be rolling like you described.

But the simple logic could still work. We need a circuit added to this what stores the FM2 pulses, and releases them at that time when FM1 gives pulses. That should work. So it stops the random bit banging.
Right?
Shift register chip might be the one. But i've never done any other than the basic AND, OR, NOT. But i'll search for what would be the chip

[AndrzejM]

Right, but that won't me much of use to measure instant FE with a decent accuracy. But for the start you can try such setup. Everything depends on how many pulses you'll get per minute from those flow meters, the more you get the better for an accuracy of the whole measuring unit.
Anyway I think that it would be much easier to change MPGuino code to read pulses from two flow meters than build a net od NAND gates to get right amount of pulses.