View unanswered posts | View active topics It is currently Wed Apr 24, 2019 2:50 pm

Reply to topic  [ 79 posts ]  Go to page Previous  1, 2, 3, 4, 5, 6 ... 8  Next
Open Source LSU4.9 Wideband Controller 
Author Message
User avatar

Joined: Tue Jan 15, 2008 2:31 pm
Posts: 15249
Location: Home sweet home!
Re stoich, leaner, and oscillations, the OEM ECU does that to help the cat work... oscillating back and forward all on one side of stoich is totally useless.

_________________ - where Open Source means Open Source, and Free means Freedom - the open source engine management system
FreeEMS dev diary and its comments thread and my turbo truck!
n00bs, do NOT PM or email tech questions! Use the forum!
The ever growing list of FreeEMS success stories!

Sat Aug 17, 2013 2:05 pm
Profile WWW
DIP8 - Involved

Joined: Wed Jul 31, 2013 11:12 am
Posts: 24
I've added an Arduino Uno based board to the sketch, I probably still have too many non-Arduino libraries included (Teensys kit installs a bunch of them), but I can get the ATmega328 + crystal from the local Jaycar for $9 or something, preprogrammed with the Arduino bootloader on it. I haven't done a schematic up for it specifically, but the basic circuit is the same (although I think I've added the 100k tying 5v and PIN_VS as per toalans suggestion).

If you've got a form of ISP, you can probably do this circuit up a little cheaper, but this thing has been doing the narrowband feedback for my stock ECU since midday Sunday, and is being daily driven and working fine. At some point I need to wire up a display to it so I can see the lambda reading :)

Mon Sep 02, 2013 12:04 pm
TO220 - Visibile

Joined: Thu Nov 20, 2014 4:06 pm
Posts: 6
Location: Omsk,Russia
Hi All,
I'm developming my own bosch lsu controller and i have some "problems" :)
My circuit is similar to that posted on first page, i have an "virtual ground" point at 2.5V and an amplifier of pump cell current sensing resistor (62 Ohms).
When i run test for first time, i've got strange behavior:
1. I set pump current to "0" (potential on Rcal:1 point is equivalent to "virtual ground")
2. I set heater power to value, when Nernst resistance is near 300 Ohm
3. Then i've injected small portion of butane from cigarette lighter to measurement chamber of sensor (bosch lsu 4.9).
4. Voltage on Nernst cell have raised to value a little over 1V, but the same time current signal from pump cell is showing current.

Is the pump cell of sensor produce voltage signal, when reach mixture is inside measurement chamber?

I have checked points of "virtual ground" and potential on other end of pump cell and this voltages holds on their levels all time.

P.S. sorry for my english :(

Thu Nov 20, 2014 4:18 pm
Profile ICQ WWW
Wideband Wizard

Joined: Tue Jun 10, 2008 2:53 am
Posts: 241
Location: Toronto Canada
Whenever there is a O2 concentration difference between the sample chamber and the pump/nernst cell there will be a voltage generated. That is probably not the best wording, there are proper electro chemistry terms that eludes me.

The oxygen sensor is more or less a symmetrical device. Look at the 3rd diagram from the bottom of this page Referencing that picture, above the sample chamber and below the sample chamber, is made of the same materials. If you were to pump exhaust gas into the bottom (replacing the free air) and pump free air to the top (replacing the exhaust gas), you can swap the terminals and the sensor will still work.

Since the nerst voltage changes WRT O2 concentration between the sample chamber and the free air reference chamber this also means the pump cell voltage will change WRT the O2 concentration between the sample chamber and the exhaust gas.

The above is for the LSU 4.2, but it is the same idea with the LSU 4.9.


Fri Nov 28, 2014 3:32 am
Profile WWW
TO220 - Visibile

Joined: Thu Nov 20, 2014 4:06 pm
Posts: 6
Location: Omsk,Russia
Ok, thank you. Now i have some understanding of processes running in wbo ))
However, i've replaced my pump drive VOLTAGE source with CURRENT source and get much,much smooth operation near stoich point of mixture )

Fri Nov 28, 2014 4:22 am
Profile ICQ WWW
TO92 - Vaguely active

Joined: Thu Jan 22, 2015 10:20 pm
Posts: 1
Hello everyone. I'm still quite new to electronics. and do not understand many things. Please explain to me how your scheme broadband oxygen sensor works , I'm interested in what you were guided in its design ? Hell Fire could you put on github your final scheme design and possibly drawing PCB , and a list of used electronic components to be mounted on board , as I do not know how to pick them up myself. Thank U.

Thu Jan 22, 2015 10:39 pm
TO220 - Visibile

Joined: Thu Nov 20, 2014 4:06 pm
Posts: 6
Location: Omsk,Russia
Hi All, I have reassemble my LSU 4.9 controller and have tested it on car. And i have strange results.
I put sensor deep inside exhaust tailpipe of my 06MY Subaru forester, warmed up engine and see measurements while engine idling.
On OBDII Scanner mixture shows lambda 1.0 (AFR is 14.7), but my controller shows AFR 17.2-17.3. And same value i've got while testing on other vehicle.
Is any assuptions of this results? Maybe target resistance of nernst cell is wrong? nernst voltage is about 0.5V
Could you tell me wich values of voltage and current of heater in steady state of your sensor?

Tue Jan 27, 2015 5:57 pm
Profile ICQ WWW
TO220 - Visibile

Joined: Thu Nov 20, 2014 4:06 pm
Posts: 6
Location: Omsk,Russia
Okay, i have found a bug in my mcu firmware, now my controller shows right values for AFR. How it works:

1. Standart heater control with MOSFET and PID regulator. Resistance of nernst cell is measured by pulsing AC coupled signal to cell, than amplifier and peak detector used to measure level of pulse.
2. Nernst control done by relay-control of current:
MCU sends current to pump cell in both direction with current source realized with OpAmp, Nernst voltage feed to MCU's analog comparator, comparator level is set for 0.5V for "positive" current flow, and 0.4V for "negative" current flow. So, mean voltage of nernst cell is kept on level of 0.45V
Then time of "negative" and "positive" current pulses measured with timer and "duty" cycle calculated. This duty cycle shows mean current, flowing in pump cell. Than table search-up made for lambda calculation.
3. Board has an USB interface for PC communication.

File comment: Controller board
gcFMq7nM3cg[1].jpg [ 268.46 KiB | Viewed 6231 times ]
File comment: Engine idling
d46XVfvI4f4[1].jpg [ 223.89 KiB | Viewed 6231 times ]
File comment: Sensor warm up
aKoVzdSyrds[1].jpg [ 191.09 KiB | Viewed 6231 times ]
Fri Jan 30, 2015 7:46 am
Profile ICQ WWW
QFP80 - Contributor

Joined: Wed Oct 01, 2014 7:06 am
Posts: 39
Wish i had found this post back in '13.....

Hell-fire, I am going to port your shem over to KiCad as it is what i use mostly these days.

I have a couple of ideas, possibly a UNO shield using mostly through hole stuff to test with and a smaller SMD board using a pro mini for real world use.

A few questions....

1, are you still using this?
2, have you tested it to see how accurate it is?
3, I am a huge teensy 3.1 fan, however it is a 3.3v device, how hard would it be to make this hardware/software work on a 3.3v chip?

The Teensy 3.1 has an on board DAC that might be useful, and it is quite a bit faster than standard arduino 8 bit boards.

Thu Jun 25, 2015 11:27 pm
QFP80 - Contributor

Joined: Wed Oct 01, 2014 7:06 am
Posts: 39
First proto boards are done for use with a pro-mini, ittle be a couple weeks at least before I can get the boards assembled, then assuming I didn't make any huge errors on the schem and board layout I can start testing.....

image.jpg [ 84.62 KiB | Viewed 5697 times ]
Tue Jun 30, 2015 10:05 pm
Display posts from previous:  Sort by  
Reply to topic   [ 79 posts ]  Go to page Previous  1, 2, 3, 4, 5, 6 ... 8  Next

Who is online

Users browsing this forum: No registered users and 1 guest

You cannot post new topics in this forum
You cannot reply to topics in this forum
You cannot edit your posts in this forum
You cannot delete your posts in this forum
You cannot post attachments in this forum

Search for:
Jump to:  
Powered by phpBB® Forum Software © phpBB Group
Designed by ST Software for PTF. ColorizeIt.