Bosch LSU4.2 wideband o2 sensor

markiii · October 3, 2006, 20:19

ok lets take it from teh top

a wideband sensor by itself is useless. you need something to condition teh output and something to read teh result.

The profec reads teh result but you need a controller to make the signal readable by the Profec, both the LC1 and LM1 have one, yours apparently doesn,'t so you need to to get one.

once you have one you can do 2 things (either or both)

1. output the (narrowband emulation) signal to the ECU instead of the stock 02 sensors. (this is pointless with your setup, unless your stock 02 sensors are dead, in which case it saves you money.

2. And why you really wanrt it. Connect the controller/conditioner to teh profec via teh signal harness you bought to show AF on teh Profec as a warning for your only. However in addition to connecting it you need to calibrate teh voltages so that teh Profec displays wsomething that makes sense, i.e what voltage is equal to what AFR/

The latter isn't hard but I'm not the person to tell you how as I've not done it yet.

any clearer?

kanujunkie · October 19, 2006, 17:31

well i finished fitting the wideband just now and after calibrating the wideband unit with the external guage through my little box of tricks i went for a quick drive, the main problem is that the sensor is just too damn quick but overall it does give a general idea of whats going on. Good news is that overall its running in the optimal range, only going to rich on the overun and lean when you lift off and the BOV fires, i need more time with it but at the mo i'm really happy and feeling a bit more confident on the install. time will tell on the rest

Anonymous · October 19, 2006, 20:43

Glad it's all ok Stu. From your install pics on the other thread it looks great.Congrats.

philster_d · October 19, 2006, 21:52

Quote from: "kanujunkie"2nd, Phil i'd have thought you would need the wideband controller, the sensor on its own wont allow for setup unless it can be done on the profec

it must be returning a signal between 1 and 0 like all the others

kanujunkie · October 19, 2006, 23:18

Quote from: "mr-s_turbo"Glad it's all ok Stu. From your install pics on the other thread it looks great.Congrats.

cheers mate, it works a treat s:D

s:D . I had to build a voltage divider to drop the output voltage down from 0-5v to 0-1v but thats only a couple of resistors so no wowsers, only issue now is its highlighted after a longer run that my first thoughts were wrong, its running lean!!, obviously not that bad but definatly losing power on what it could be

Quote from: "philster_d"
Quote from: "kanujunkie"2nd, Phil i'd have thought you would need the wideband controller, the sensor on its own wont allow for setup unless it can be done on the profec

it must be returning a signal between 1 and 0 like all the others

yes but you need the controller so that the signal can be calibrated within that 0-1v range. Its a bit like takeing the sentence below

"THE MUDDY BROWN FOX RAN OVER THE FENCE"

Without calibration that would be written as

"BROWN FENCE OVER THE MUDDY RAN FOX THE"

basically theres no order to the uncalibrated sensor output, the controller does this and then sends the signal on to the profec or guage. Basically your sensor is useless without the controller

hope that makes sense s:? :? s:?

Anonymous · October 20, 2006, 09:31

Quote from: "kanujunkie"cheers mate, it works a treat s:D s:D . I had to build a voltage divider to drop the output voltage down from 0-5v to 0-1v but thats only a couple of resistors so no wowsers, only issue now is its highlighted after a longer run that my first thoughts were wrong, its running lean!!, obviously not that bad but definatly losing power on what it could be

Are you using the grey wire direct to your ecu for the narrow band input? If not you could of just used that as that is 0-1v.Save the hassle with the resistors.

markiii · October 20, 2006, 09:38

narrowband input only give rich/lean readings as oper teh stock ones though surely?

if you want an actual AFR ratio you need the wideband outputs

Anonymous · October 20, 2006, 10:10

That's totally right mark. Just reading what Stu said, that he has used resistors to change the voltage from 0-5v to 0-1v as i assume that is what his new guage input requires. I'm just making sure he realises that the controller box already has a WB and a NB voltage output.

I assume he's got the guage more for show and the digital controller box is what he'll fine tune with.

markiii · October 20, 2006, 10:13

his AFR guage is sipposedly a true wideband, but for some reason needs inputs ina lwer range than teh sensor gives out

Anonymous · October 20, 2006, 10:17

Well if the range is 0-1v then it's a narrow band AFR guage s:? :? s:?

Tem · October 20, 2006, 10:51

Quote from: "mr-s_turbo"Well if the range is 0-1v then it's a narrow band AFR guage s:? :? s:?

Not necessarily, I've had a 0-1V wideband gauge myself.

There's no "rule" that wideband has to be 0-5V, though it usually is. s8)

s8)

Anonymous · October 20, 2006, 11:49

There is no rule to WB, i know, but can't see the point of a guage with an input of 0-1v and then spending all the money on a WB controller just to input what a std NB can supply anyway.

Anyway this is off what we are talking about here. Stu has a guage that it sounds like he is using the WB output wire(white) and altering it with resistors to get the required voltage, when he could use the NB output wire(grey) that is already being simulated by the WB control box.

Tem · October 20, 2006, 12:12

Quote from: "mr-s_turbo"can't see the point of a guage with an input of 0-1v and then spending all the money on a WB controller just to input what a std NB can supply anyway.

It's not the range, but the contents.

Here's the content of normal narrowband:
m http://www.not2fast.com/instrumentation ... vs_afr.gif m

As you can see, it's very close to 1V at anything below 14.7 and very close to 0V at anything above 14.7. It's very, very hard to get any usable reading from that, other than pulsing around the 14.7.

Wideband, even if it has the same 0-1V range, has a linear voltage/mixture ratio. Or it can be programmed to be anything else too, but generally it's linear. So you have 10.0 at 0V, 11.0 at 0.1V, 12.0 at 0.2V and so on. It's very, very easy for a gauge to show the real value with that. s8)

s8)

QuoteAnyway this is off what we are talking about here. Stu has a guage that it sounds like he is using the WB output wire(white) and altering it with resistors to get the required voltage, when he could use the NB output wire(grey) that is already being simulated by the WB control box.

I know it was bit offtopic, but just wanted to clear that. s8)

s8)

Also, reversing my text above, you can't divide the voltage of 0-5V wideband output to 0-1V and make it appear as narrowband. It won't work, even if you really wanted to do that. You HAVE to use the narrowband simulated output.

kanujunkie · October 20, 2006, 13:47

Hi guys, you lot have been busy talking

yes i am useing the white wire (WB output) and yes you can divide the output voltage from 0-5v to 0-1v. The guage i have IS a WB Guage and the maths behind it is

Output voltage = R2/R1 + R2 x Input Voltage

I had to use a variable resistor for R1 which also gave a method of calibration to the guage

resistor values are

R1= 4K (this is done with a 4K7 variable resistor)
R2= 1k

The digital output on the WB controller and the external guage readings match spot on and thats all i want and need.

Anonymous · October 20, 2006, 13:59

Cool. Just a question though, why not just use the Grey NB wire that is already been converted from WB to 0-1v?

markiii · October 20, 2006, 14:00

because it onlt alternates between teh 2 extremes to show rich or lean

Anonymous · October 20, 2006, 14:06

Quote from: "markiii"because it onlt alternates between the 2 extremes to show rich or lean

Oh yeh, i see. s8)

s8) Keep forgetting it's a WB guage not a NB s:roll: :roll: s:roll:

Anonymous · October 20, 2006, 18:18

Quote from: "kanujunkie"I had to use a variable resistor for R1 which also gave a method of calibration to the guage

resistor values are

R1= 4K (this is done with a 4K7 variable resistor)
R2= 1k

The digital output on the WB controller and the external guage readings match spot on and thats all i want and need.

Good stuff Stu. I would suggest that once you have set the value of R1, replace it with a fixed resistor rather than a variable resistor as the variables have a tendency to drift or may end up being moved by accident.

Richie

kanujunkie · October 20, 2006, 18:28

Quote from: "Richie"
Quote from: "kanujunkie"I had to use a variable resistor for R1 which also gave a method of calibration to the guage

resistor values are

R1= 4K (this is done with a 4K7 variable resistor)
R2= 1k

The digital output on the WB controller and the external guage readings match spot on and thats all i want and need.

Good stuff Stu. I would suggest that once you have set the value of R1, replace it with a fixed resistor rather than a variable resistor as the variables have a tendency to drift or may end up being moved by accident.

Richie

thanks, i would have done that if i could but you try buying a 4K resistor! they dont fall into the basic manufactureurs values. Plus it does need a minute amount of trimming. The maths and the real world were slightly out and i did have to trim it a bit.

The pot was set before install to 4k useing a calibrated meter BTW

Anonymous · January 22, 2007, 10:16

Having read this lot, I'm not ashamed to say that I am now totally confused s:lol: :lol: s:lol:

I was planning on buying this one

m http://www.mkivstore.com/index.php?targ ... ct_id=1641 m

and running a fead to the ECU so only one sensor used (scrap oem ones)

but now it seems I can run a feed to the Profec in some way and consequently wouldnt need the gauge unit?

you clever sods and your techy chat do my head in, give me grease, oil and pipes any day s:lol: :lol: s:lol:

spit · January 22, 2007, 10:33

You're not alone in your confusion Perry s:wink: :wink: s:wink:

Quote from: "perry190"I was planning on.......running a fead to the ECU so only one sensor used (scrap oem ones)

Has anyone here managed to feed the ECU with the simulated NB output from a WB sensor and monitor WB at the same time?

I've had the simulated NB working really well with the ECU (confirmed with a voltmeter), but it makes the WB gauge go do-lally. Innovate suspect it could be an earthing conflict but I tried allsorts and can't get both systems to work together. s:oops: :oops: s:oops: Any suggestions?

Anonymous · January 22, 2007, 10:38

I think Ian did if i recall correctly?

I also noted this heat sink, which might be ideal for where the sensor is nearer to the turbo?

m http://cgi.ebay.co.uk/Innovate-HBX-1-O2 ... dZViewItem m

Anonymous · January 22, 2007, 15:06

I had only 1 O2 sensor and this was the WB. I fed the NB signal to the ECU and then the feed from the WB control box went to the profec to display A/F on there. You just need the optional input harness for the profec.

kanujunkie · January 22, 2007, 19:44

still not convinsed that haveing the sensor so close to the turbo is a problem, mines about 3" from the turbo exit and i've seen nothing to worry about

does the innovative stuff have simulated NB then?

spit · January 22, 2007, 19:56

Yep the controller (LC-1 at any rate) has two analogue outs alongside two for gauge/laptop connection etc. You can tweak 'em with the software to adjust switching points and hi-lo voltages - if you want to!

Ditto on the temperature thing - it may affect longevity but doesn't seem to affect accuracy while its working ..... I hope s:oops: :oops: s:oops: