Any questions you have before you begin buying, building and installing.
#62658
Izzy46 wrote: Tue Apr 25, 2023 4:27 pmThat's interesting. Given that they are all Bosch at heart, would it be possible to wire a 4 wire WBO2 sensor with the Spartan controller?
No. An LSU4.9 sensor element is different than a LSU4.2 or ADV or NB sensor element, etc. I'm not saying they are all the same, just that the LSU and ADV sensor elements are all made by Bosch.
#62660
I did some testing of the controller and sensor to see if I could isolate an issue. I focused on the heating of the element in relation to sensor output (Nernst).

Test 1)
- Controller is removed and test done directly on the Bosch sensor.
- Between heater+ and heater- I apply a constant 12v
- Nernst climbs quickly and stabilises at 4.9v after about 1 minute .
- With a heat gun I measure the tip of the sensor, the highest temperature I see is 195*C
- The Nernst quickly and consistently drops by 0.3v when a cigarette lighter is lit in front of the sensor

Test 2)
- Controller wired correctly and connected to the Bosch sensor.
- Between heater+ and heater- the voltage starts at around 5v and slowly climbs to 12v before slowly dropping back to around 5v where it stays.
- Nernst starts at around 2v volts, slowly climbs to 3.5v and then drops back down to 2.2v where it stays.
- With a heat gun I measure the front of the sensor, the highest temperature I see is 85*C
- Unfortunately both of my lighters ran out of gas and my neighbours (who are avid smokers) were also out of lighters. :? :roll:

So it seems to me that the controller is not heating the sensor properly, despite the LED heating light indicating otherwise. I suppose the sensor is never actually getting up to temp.

I guess there are three options:

1) Order controller number 3 and see what happens

2) Take the boat out on the water and allow the higher RPM to heat the sensor

3) Bypass the heating aspect of the controller and wire the heating element directly to the fuel pump circuit. I'm wondering if this runs the risk of overheating the sensor...?
#62664
With a wide band the controller monitors the heating of the sensor and adjusts the element output accordingly.
You should not be directly driving a wideband sensors heater.

Only the 4wire narrow band sensors should have the heater connected directly to a source such as the fuel pump drive or more commonly an ignition controlled output .
#62665
dazq wrote: Wed Apr 26, 2023 8:05 am With a wide band the controller monitors the heating of the sensor and adjusts the element output accordingly.
You should not be directly driving a wideband sensors heater.

Only the 4wire narrow band sensors should have the heater connected directly to a source such as the fuel pump drive or more commonly an ignition controlled output .
Thanks dazq, I've discovered some more documentation on the Spartan 2 controller including an Arduino sketch. After looking at the code it seems that the controller needs to know the temperature of the sensor to modulate the sensor pump and heater. Just as you point out above :D

I guess that rules out my plan to bypass the controller. :?

I also found that the Nernst pin should settle around 0.45v when fully heated rather than the 4.9v that I'm seeing. I guess there is a good chance that the sensor has a problem...

Could the increasing and then decreasing voltage coming from the heat- and heat+ of the controller be a result of the controller seeing high voltage at the Nernst pin, interpreting that as high temperature and then reducing the voltage at heat- and heat+ to stop the element heating?
Attachments
(43.73 KiB) Downloaded 171 times
User avatar
By PSIG
#62670
The LSU 4.9 tops at clean air. You can see below, at 12% O2 the Lambda is already at 2.42 (*14.7=35+AFR gasoline), and so the gauge reading is dependent on the controller range, not the sensor. LSU 4.9 PDF at bottom for reference.
LSU4·9_Ip_vs_Lambda.jpg
LSU4·9_Ip_vs_Lambda.jpg (178.68 KiB) Viewed 1109 times
(369.42 KiB) Downloaded 135 times
#62672
It does seem to be controller dependent, my LC-2 goes 7.35 to 22.1. It can also be programmed to use either 4.2 or 4.9 sensors, so they are not interchangeable unless the controller can switch calibration curves.
#62719
After spending half a day hunting lambda sensors (prices in this country vary from $300USD for genuine, down to $80USD for Chinese, and nothing used so far) I finally found a Chinese AFR sensor that works!!

I took the boat out yesterday for the first time in over a year with the intention of fault finding and ironing out issues. Unfortunately the babysitter called in sick at the last minute so I was a little limited in the testing I could do. After a few hours on the water swimming around and doing a few runs I found that the boat goes as well as it did previously.

The main areas to work on now are:

1) Working out how to stabilize the idle when the throttle is suddenly closed (when a skier falls)
2) Adjust AFR to target across the VE table

Idle Stabilization
In regard to idle stabilization there are probably a number of ways to do it with the VE table, idle and idle advanced settings. I'm curious on the VE table the roll of the far left vertical column and the lowest horizontal column when it comes to idle stabilization. In my case when the throttle is closed it dives to the bottom of the VE and then shoots across to the left column where it seems to be 'searching' for something?

The table I'm working for looks like this:
Screen Shot 2023-04-29 at 11.06.25 AM.png
Screen Shot 2023-04-29 at 11.06.25 AM.png (377.35 KiB) Viewed 972 times
Apollard I notice that your VE table is leaner at low RPM and mid to high kpa.
Screen Shot 2023-04-29 at 11.09.38 AM.png
Screen Shot 2023-04-29 at 11.09.38 AM.png (374.98 KiB) Viewed 972 times
Is this for starting or idle recovery purposes?

General Tuning
Going forward I'm looking for a solid tuning strategy. Starting and idle is pretty good and the boat accelerates and cruises pretty well. It's rich in some areas and lean in other areas.

Should I drive at a specific idle and adjust the VE to hit the target AFR or should I use the autotune feature on TS?
How should I think about the "tip in" where the VE table moves too quickly to capture the right AFR?
How should I address the idle stability after the throttle is closed quickly, is it VE, idle, advanced idle or a combination?

Once again thanks for all the input, this is a slow moving process but I feel I'm heading in the right direction. :) :) :)
By dazq
#62722
Someone posted info the other day on discord about OEM that fit wideband sensors, could be a cheap source.
See screenshot of discussion
Attachments
Screenshot_20230424-225432.png
Screenshot_20230424-225432.png (353.27 KiB) Viewed 964 times
  • 1
  • 13
  • 14
  • 15
  • 16
  • 17
  • 20

Hi, I am trying to assign Signed values to the x-a[…]

Vr Conditioner Noise when cranking

New version 202305 don't fix the issue. Now after[…]

blitzbox

I've finally figured out why MAP and Lambda weren'[…]

Hello, I bring news!! Let me tell you that after […]

Still can't find what you're looking for?