Well it's in a Truck that didn't have an o2 sensor to begin with, it's the help fine tune the Carburator and it's more for fun then anything, currently it's hooked up to a multimeter and as you can imagine it's a little hard to read since the numbers always bounce of course and the LCD display on the multimetter isn't quick enough to keep up so it's a blur.

it's just for fun and to have something to do.

BTW I figured out my pin out fromt he diagram.

Thanks.

 

Sorry to bump a way old post.

I built the circuit and it worked fine for about a month, now a strange problem has come up, the LED's don't light properly and some flickers and beahve strangely.

I narrowed it down the the Voltage regulator I added (12 volts to 5Volts)

Bassicly when the unit behaves strangely all I do is short the ground and output pins on the regulator and the unit will work fine again until the 12V power source is cut off (unit is inplugged or truck is turned off) Is the regulator bad? anything else that comes to mind on what might be happening?

 

Maybe the regulator oscillate.Have You capacitors (100nF) across the input and GND and output and GND? (Nearby the pins as possible)

 

No no capacitors in the circuit. No easy way to add them either. It will be hard to remove the circuit board from the project case, almost better to start over. I didn't really need the voltage regulator Maybe I will try to remove it altogether. it may be simpler.

The regulaotr is a 7805A

http://www.radioshack.com/product.as...%5Fid=276-1770

 

Any other ideas? Basicly I'll probably restart from scratch, just wierd that it worked leaving my bench and for about 1 day in the truck and now it's bombed.

BTW shorting the ground and output pin on the regulator no longer causes the unit to work.

I've anaged to pull the circuit board from the case and tested all the connections and they are fine. I guess the problem is either with the voltage regulator or the display driver. I will rebuild the circuit without the voltage regulator and see what happens.

 

0-1volt at 0.0000001 amps of currents coming from the o2 sensor.
Not much current comes from the o2 sensor,wich can be robbed by some circuits.

Replace to 1meg resistor with a 4.7 meg one.

I have found that the 1 meg resistor will drop the oxygen sensors voltage too much.
Use a 4.7 meg one instead.
You will get the true voltage if not using the 1m resistor at all, but when the oxygen sensor is cold, the circuit freeks out, and neads a resistor.

if not using the 1meg or 4.7 meg resistor, all the leds will light,untill the o2 sensor warms up.


Led brightness will steel voltage from the o2 sensors current and give false readings.You must then adjust for the lower o2 sensor voltage.


Even a digital multimeter will effect the reading slightly.
So even calibrating the circuit with a multimeter, then taking the multimeter off, will give different readings.Good Luck.

I don't use any resistors except for the 4.7 meg that replaced the 1 meg one.

 

Ok thanks I'll try a bigger resistor as suggested.

But I suspect it may not work, the device was working fine on my test bench and in the truck for a few days and all of a sudden no longer works on the bench or the truck,

It was bench calibrated, I generated a .200v, .500v and .750v signal and adjusted to the proper LED there was no mulimeter connected once the voltage generator was set.

Again thanks for the help (and it worked great and was a good use to us to determine the why the trucks power band was flattening out in the upper RPMS(it was running rich when we thought it was running lean, we rejetted and it's awesome now))

 

A friend of mine and I just developed a much better 21-LED version of this, using two LM3915's, one in "reverse" so that the logarithmic values line up with the output of the stock sensor. Obviously in no way can the LED's be assigned actual air/fuel ratios, but it can be quite handy for tuning purposes. The meter is setup in a center-weighted bar graph fashion too, so it's very easy to see whether you're rich or lean, and by how much. We also included an op-amp as a high-impedance buffer as not to load down the O2 sensor. The stoich area isn't even measured; the center LED is acutally ON all the time, so that if you're in the stoich area, that single LED is lit. Only the rich/lean areas of the lambda curve are actually measured by the LM3915's.

I'll try to post schematics and maybe even a board layout soon, if there's any interest.

 

Ken any updates on his circuit I'm curious... I just started rebuilding mine.

 

You will notice that when the o2 sensor gets hot~very hot, it will not give proper voltage to air-fuel ratio above the 14.7:1 ratio.

Mounting the sensor as far as possible from the exhaust manifold may be benificial.

Attached Images

voltage_vs_af.gif (8.7 KB, 1007 views)

 

Well sorry to bump this again.

I rebuilt the circuit as posted in my fist post, I did not add a voltage regulator this time. and used a 4.7k ohm resistor instead of 1 ohm.

It worked for about .5 seconds then something went wrong. I rechecked all my traces and found some were grounding. After fixing this the circuit does not work.

All the traces are good.
There is 12.2 volts to pin 3 and to the LEDs
The Input pin (pin 5) shows whatever Voltage I'm sending to is (say .200v)
The output Pin on the LM3914 show 10.58Volts(all of them) That's not normal is it

Any ideas what's going on? any information you need to help me more. Pretty much going to scrap this project and just buy a damn guage.

 

Use a 4.7 meg (4,700,000 ohm) resistor !!!, Not 4.7k (4700 ohm).
It will not work if using 4.7k
or, dont use one at all, but then you will have to wait for the sensor to warm up before the sensor will start reading.

dont loose faith.

you dont even nead any resistors in the whole circuit, as the chip is programed to read 1.2 volts when no components are connected to it. eg

Remove all components except for the Leds.
connect pins 2,4 & 8 to ground.
connect pins 3 & 9 to positive.
connect pin 6 to 7.
connect O2 sensors positive signal to pin 5.
connect O2 Sensors negative signal to your circuit's ground.

Attached Images

scanimageo2sensor.gif (95.4 KB, 885 views)

 

Thanks Screetch, I abandonned it for the weekend but I'm going to get back at it tonight, will a 10 Meg Ohm resistor work as well?

 

10meg is the same as not having a resistor at all with this particular circuit.
With a 10meg resistor, the meter will work only when the O2 Sensor has warmed up, ( I know because i tested it).

Any less then 4.7 meg, and it will leak some voltage from the O2 sensor to ground. (meter will show less voltage)


You previously used a 4.7 Kay resistor, and it made all the voltage leak to ground, thats why zero voltage was detected on your air-fuel ratio meter.

HINT FOR ALL: The O2 sensor will ground it's self to it's ground wire when it is warm to hot. the meter freeks out when it's cold because there is no ground. Thats why we put a high ohm resitor to ground(xMega- ohms).

good luck with it Gotrek.

 

Thanks again screetch at Least now I understand it. Thanks a bunch. Sometimes it sucks just being a soldering monkey.