Wide Band Sensor

[Sam]

Hi folks, I'm no whiz when it comes to auto electrics so I'm hoping someone can help me here.

I have a 14poiont7 DIY SLC wide band Oxy sensor control unit that I built .
It bench tested ok, so the next stage is to wire it into the car.

The problem is that I'm not sure where to get the power from.

I think there must be a 12v rail I can tap into at the fuse box.
The unit has it's own regulator and fuse, so I'm just thinking of the Fuse box because I can't think of anywhere else where I can tap into 12v where the amps would probably be low enough to use.

Is this the right way to do it?

Any ideas?

Sam

[AR116]

Just wondering what type of 02 sensor you are using? and I presume you are using the wideband kit from jaycar.

obtaining 12v power supply can be source anyway from inside the car. Try some of the wires near the ignition switch, there should a constant 12v supply somewhere. You can also get it from the fuse box, which may be easier.

Invest a few dollars in a test light, as it will be your best friend in finding power supply etc.

[Sam]

Hi AR116
thanks for the reply

I'm using a 14point7 SLC DIY Lambda Controller (http://14point7.com/) with a Bosch 17014 wide Band sensor. They are highly recomended on the Megasquirt forums.
I wasn't aware of the Jaycar unit until after I bought this one, otherwise I may have gone for that instead.

I'm unsure of taking power from any wire or device, I did this many years ago & wound up poping fuses when both devices were drawing too much current for the fuse / wire combination,  hence why I'm paranoid this time around and asking for help.
I know the dash clock has a constant power source, maybe I can splice into that?
I thought about a feed from the fuse box, I'll have to take a look tonight.
Do you know where in the fuse box I should look  ?

Thanks in advance
Sam

[AR116]

sam,
       if the clock has constant power then that should be fine.
From memory I have worked on a few alfetta's, there should be a thick red wire that goes into the fusebox to supply power for the appropriate relays, maybe you could also look at that?

regards;

marco

[Sam]

Hi Marco
Thanks for your help,  I'll see if I can find the wire at the fuse box, otherwise I'll tap into the clock.

BTW, I don't suppose you are the same Marco from Abruzzo Motors?
I was thinking of going there on the weekend to ask Marco for his advice !

are you the same Marco?

[jimnielsen]

The wideband Lambda sensor is really just a very sensitive voltmeter that measures the voltage output of the Bosch sensor. The voltage readings are very susceptible to current draw from other components on the same circuit, its best to connect the ve+ side as close as possible to the source (though you probably want it switched by the ignition for convenience). Connect the -ve side to a purpose made chassis connection or direct to the battery neg terminal. If you are connecting a datalogger to the sensor you need to connect the ground of the logger to the same point as the sensor.

[Sam]

thanks Jim

It probably is better to connect it close to the source, but what i can't figure out is where that is.

So far I think I have 2 options

1) the controller has an inbuilt regulator, so I could run a separate wire from the battery to the controller (through a switched relay etc) that's about as close to the source as I can think of,  but the battery is 30+amps, so the normal 10 or 15 amp wires would fry. I assume I would need a very thick wire to carry that many amps safely, which would be overkill. Besides it wouldn't fit in the Molex connector.

2) I could do what I discussed with Marco in the previous posts, but as you say it may not be close enough to the source. Further more I'm not sure if splicing into the clock +ve lead will draw power away form the clock , blow fuses,  whatever (sorry if that's a dumb assumption)


3) Other options???


Sam

[spoynter]

Do what Jim said.

1) the controller has an inbuilt regulator, so I could run a separate wire from the battery to the controller (through a switched relay etc) that's about as close to the source as I can think of,  but the battery is 30+amps, so the normal 10 or 15 amp wires would fry. I assume I would need a very thick wire to carry that many amps safely, which would be overkill. Besides it wouldn't fit in the Molex connector.

You're not quite right there. The current is 'drawn' by the device, not 'pushed out' by the battery. The controller will only pull an amp or two I would imagine, check the value and install a suitable fuse in series (edit- sorry just noticed that it's already fused). I can't imagine that you would need a relay.
HTH

[Sam]

Hi Spoynter

thanks for for help, I think I'm getting closer to it now.

I think the controller expects about 3 amps, not much at all.

The only reason I wanted to use a relay is so I can turn it on when I turn the ignition key to Aux, although I
suppose I could go straight from the ignition switch ....

If the controller draws the current , rather than the battery pushing it, then I can go with option #1.


To all you folks that replied - Thanks - much appreciated.
Sam

[Fylnn]

The other obvious point in the engine bay is where the battery cable comes up from under the floor and terminates in a junction box on the firewall passengers side.  Unclip the plastic cover and you have 2 large ternminals ready to use.

[Sam]

Thanks for everyone help .... here is the latest ....instalment :-)

The unit is built, installed and running ...... but I do have a problem.

The unit is acting like there is a lot of noise on the line.

While the car is idling at 1000rpm, the unit is showing more than 9000 RPM!

I can't tell if its the unit at fault, or just noise on the line.

I tried filtering but it didnt work . Although I'm not sure If I did it right.

Any ideas?

Don't suppose anyone knows what the signal from the coil to the tacho is like?

Is it high frequency, noisy etc anyone else had issues with it?

you can see my posts on the 14point7 forum for more details

http://14point7.com/forum/index.php?topic=798.msg6229

Sam

[MD]

Sam,

Can we go back to the basics. What car is your device being installed into?

[Sam]

Hi MD
thanks for your replay,
It's installed on a 1979 Alfetta.
I want to convert it to EFI , so this is the first stage, but it's still on carbs at the moment.
I wanted to measure how the current set-up is going.
The unit it completely installed , i.e. Wide Band Sensor in exhaust, all wired up etc ... The main issue I have is viewing the RPM.
Sam

[MD]

Well that's clearer. I was beginning to wonder why you would do that for a carburreted engine. Anyhow, with respect to the coil feed to tacho, it is a very dirty signal with some high peak voltages as well so I would not go anywhere near that. It's the peculiar reason why the Veglia tacho works and it doesn't work off clean signals-figure that out!

[Sam]

Yes the signal from the coils if dirty, but where else can I get it from ?

I cracked open my spare Veglia Tacho yesterday and noticed a bunch of  capacitors (amongst other things). I suspect these are there to filter the signal.
Based on this and advice from others, I put together a bypass filter with the hope that it would clean it signal up.
At at one stage, and for a very brief time, I was getting readings close to the Veglia RPM. (hooray!),  then it just went crap and crashed the display software.

From then onwards it was no longer working correctly so I eventually gave up.

Before calling it a day I decided to re-test the filter components and discovered that the 5k Variable Resister wasn't working.

I don't know if I fried it when I soldered the bypass filter together or maybe I damaged it when I accidentally touched it against the cars positive battery terminal while I was re-testing it (many sparks !)

Anyhow I'm off to Jaycar tomorrow for a new Variable Resister and capacitor.
I'll try it again to morrow night.

Sam