pgmfi.org

Rise from the dead!

just some new info and some more research. i believe that op amp is a voltage follower or a buffer. i was messing with a MCP602 (microchip dual op amp) and configured it for a voltage follower with a low pass filter of 10hz. with Vcc at 5.5v i would get up to around 4.3v input and output would jump up to 5.35v. i did a little bit more ready and found that once the input gets to a 1.2v of Vcc, it will max out. this paticular op amp would go to 5.35v with a Vcc of 5.5v. i upped the voltage to 6.5v and now it has a full swing from 0-5 volts with an average voltage difference of around .03v. it is possible that at a 5v input with the op amp used in the honda ecu that the limit is 3.8v which is 1.2v of Vcc.

Just to be sure, all you really need to do this mod is an extra 10uf cap?

No, you must bypass the op-amp (or make other modifications to remove its effect) - read the post previous to yours!

The cap is simply a hack to get around the fluctuation problem which is a seperate issue.

There is a resistor (r5) you remove that basically cuts the op-amp out of the circuit. Then you feed the (10uf) + CAP lead through the input of r5 (Before the op-amp) and use the excess of the lead to make a jumper to the output of C3(after the op-amp)
and connect the negative cap lead to a ground plane.

like here:
http://www.pgmfi.org/twiki/bin/view/Library/O2InputMod

Sorry, I should have been more clear. I meant parts-wise, there isn't any additional parts needed except for the cap.

It seems that the 10uf cap is too large.

If we look at the low pass filter with 10k and 10uF, you have a -3db frequency of 1.5 Hz. That's a result that may give you poor response times.

lets say you need to detect a .5 AFR change in 1/10 sec as a reasonable start. that's a .25 V change over .1 seconds.

.25V = 5V sin(f*.1)

f = 28 Hz

I'd recommend a .1uf (that gives a 160 Hz response, that's a .5 AFR change over .02 seconds or 2.75 AFR change in .1 sec) or no cap.

The stock ECU has a response of 800 Hz.

If the engine is at 9k rpm, for .1 sec you have 7.5 engine cycles, just to understand why you need the fastest response times possible.

Thanks for that analysis.. will have to see if it makes any difference logging-wise..

I agree 10uf is large, and excessivley so when you crunch the numbers but this was required to dampen out the fluctuation i was getting from the PWM of the wideband's heater. If you take that noise out of the equation then something much smaller is required, as you have proven. I wouldnt suggest everyone uses that value just because i did.

As it happens, even with the heater noise i was getting if you had a reasonable hit count when datalogging the AFR would average out almost the same anyway - either way you are still dampening the readings whether it's done as a hardware hack or in software. The fact you need to average x readings makes a snap reading of the AFR impossible. I believe you have to step back and look at things in context, we arent tuning competition systems with massive tables and endless decimal places of accuracy though, its a relativley basic setup with the position sensors being driven via a piece of rubber & nylon.

The only true solution in my case would be to eliminate the heater noise completley, i need to spend longer tracking it down. I've noticed recently that the LM1 can be upset a lot by changing the cable length, even if you use foiled & braided mil-spec stuff for the cell and seperate supplies for the heater (although at the time i was using the stock cable). I guess there are some neat tricks going on inside the LM-1 firmware.

sorry, I didn't mean that your circuit was wrong, it was just a suggestion. It's great you took the time to create the wiki on it, thanks.

I reason I was worried about the cap is not the steady state with noise from the heater etc. but the dynamic change. If you have a AFR change from the engine that takes .5 secs longer to react you could get the wrong value in the datalogging, i.e 15 instead of 14.5. While the 10uf or even 100uf would work because the tables are just averaged anyway it pays to be as accurate as possible if your tuning ignition or different fuels. I think of the problem as just adding more delay serially into the system, yes you have delay from the wire, IC's, datalog trans rate but why should you add more.

Did you test the heater noise when the engine was on and at higher rpm's? The alternator my increase the voltage and current supply removing the heater's draw.

Did you record a voltage drop at the LM1 power supply when the heater goes on?

Hi everyone,

i have a stupid question. Lets say we make this o-5v mod, how ecu will react on it with the closed loop? I mean if i do this mod, and will be daily driwing with stock O2 sensor it should be fine. Or i am wrong? What about connceting wideband output and driving on closed loop? it should go crazy in my opinion...

So this mod is only good for dataloging and driving on open loop?

Sory for my stupidity, but i just want to be sure what i am doing

thanks

you can't connect the wideband and run closed loop unless you figure out how the ECU utilizes the o2 input and re-scale it for the wideband signal (Which has a positive slope when narrowband has a negative slope.

You can leave the mod in place with narrowband because it doesn't use anything over 1 volt anyways..

Thank you, naw it make sense.

I use my lm-1's wideband emulation all the time when I'm not tuning. If I need to tune I just go into logworks and change back to the wideband slope.
I haven't done this mod yet would it cuase a problem with the way I'm using narrowband o2 emulation with the lm-1?

It shouldn't affect anything

Cool I'll Try it out then.