Jump to content

AFR Reading


Recommended Posts

So I connect up my laptop to my ECU and the Dashboard AFR Gauge is red and off the scale, go into the AFR Table and it's where I expect it to be, plus the car is running just fine. Have loaded older tunes, created new project and even reloaded the firmware but the gauge on the dashboard is still way off. 

Any ideas what's causing this?

Darren

afr.jpg

Link to comment
Share on other sites

Hello Darren

                      Is your your Lambda sensor controlling the ECU?(closed loop etc) if not the reading could be faulty? As the ECU is just putting what the map says(I think!)

I still need to sort out my low speed running and stumbling and erratic fast tick over!!!!! which stops when ignition is powered off and on?

(still its a long winter?) and a new engine to build! plus I have a noise that may be a o/s driveshaft U/J(All new and bearings and shaft!)B*****r

Plus do not mention the diff clunk(worn inner drive shaft splines)plus gearbox oil leaks!

Still its a well use Triumph(just back from Northern Ireland tour) 1200 miles

 

Roger

ps and I get a smell of petrol but I have changed the breather setup?

Link to comment
Share on other sites

44 minutes ago, Nick Jones said:

HI Darren,

IIRC you have to "re-calibrate" the O2 sensor whenever you update (or just change version of) the firmware.  I've been had this way myself.  I don't remember the procedure for doing this but it's not difficult.

 

Nick

 

Hi Nick,

I'm using the Spartan 2 Controller, this sends out 2 calibration voltages during start-up and you use these to calibrate. With the latest MSExtra firmware it does give you the option to retain calibration or not during the flash process, I have tried both ways and neither work. The gauge in TS displays the 2 calibration readings, it's just after that it goes off the scale.

I'm leaning towards controller or sensor failure, but only because I can't think of anything else it could be. Next time I get an opportunity I will check voltages out of the controller to see if they are in the expected range.

Darren

 

Link to comment
Share on other sites

8 hours ago, Nick Jones said:

Does the Spartan allow direct serial connection (as the LC1 does) so you can read directly from a laptop?

ISTR than I had to load a calibration table to get mine to work after firmware change (but was a few years ago now....)

 

Nick

No direct connection AFAIK. Calibration is just by inserting the linear values.

Controller supplier has replied overnight and seems to think it's sensor failure.

Darren

 

Link to comment
Share on other sites

This is the response I got from the controller manufacturer in the US:

The fluctuations are limited to low engine load at low RPMs. I do not know specifically about MS, but it seems to be trying to close loop control the AFRs to near stoich during light engine loads to reduce emissions and reduce gas consumption.

Not exactly sure what he's trying to say tbh, but I guess he's referring to the EGO settings below, I haven't touched them so are what you get when selecting single wideband.

ego.jpg.7d5659f900a4581c88e0781b383ea6fc.jpg

Darren

 

Edited by Dazman1360
Link to comment
Share on other sites

Does anyone know where I can get a cheap & cheerful Wideband AFR Gauge, just so I can see a reading other than the one in Tunerstudio? They all seem to be part of a kit, which i don't need. If I get the same readings on a gauge then it is definitely either the controller or sensor rather than any controller to ECU problem. The warranty return for my controller is in Canada, so a bit of a faf and costly to get it returned, so want to make sure before I go that route.

Darren

Link to comment
Share on other sites

The reply from the manufacturer doesn't make much sense to me........  although, seeing as you've given the controller 15% authority (ie it can change the VE values by up to 15%) I'd have thought you might have noticed something as it will richen the map by 15% throughout it's operating range trying to get back to it's notional target.  15% probably wouldn't spoil the running very much but I think you'd notice the extra fuel use.

Within Tuner Studio it's possible to display the amount of correct being applied (may need to tinker with your virtual dashboard) and this would give an important clue as I'd think it would be permanently pegged in the rich direction if the ECU really thinks you are running at 20:1+

I've never found a reasonably priced AFR meter either singly or as part of a package.  There's no good reason for this as they are basically only a poshed-up voltmeter.  Checking the actual voltage output would also be interesting. Guess you'd need to connect a pull-up resistor across the output  (1kohm?) and measure across that.  Given the readings you are getting in TS you'd expect to be getting the full 5v!

Food for thought......

No doubt you've already read this?

https://www.14point7.com/blogs/news/16539352-maximizing-accuracy-between-spartan-2-and-megasquirt

 

Nick

Link to comment
Share on other sites

11 hours ago, Nick Jones said:

The reply from the manufacturer doesn't make much sense to me........  although, seeing as you've given the controller 15% authority (ie it can change the VE values by up to 15%) I'd have thought you might have noticed something as it will richen the map by 15% throughout it's operating range trying to get back to it's notional target.  15% probably wouldn't spoil the running very much but I think you'd notice the extra fuel use.

Within Tuner Studio it's possible to display the amount of correct being applied (may need to tinker with your virtual dashboard) and this would give an important clue as I'd think it would be permanently pegged in the rich direction if the ECU really thinks you are running at 20:1+

I've never found a reasonably priced AFR meter either singly or as part of a package.  There's no good reason for this as they are basically only a poshed-up voltmeter.  Checking the actual voltage output would also be interesting. Guess you'd need to connect a pull-up resistor across the output  (1kohm?) and measure across that.  Given the readings you are getting in TS you'd expect to be getting the full 5v!

Food for thought......

No doubt you've already read this?

https://www.14point7.com/blogs/news/16539352-maximizing-accuracy-between-spartan-2-and-megasquirt

 

Nick

Hi Nick,

The car is going great, certainly haven't noticed any change in how she drives or a drop in economy but if it's only just happened maybe I wouldn't have. It's a good point about seeing what the correction is doing, will go for a drive with the laptop plugged in to see what it's says. Until such times I have this issue fixed, would it be better to disable the sensor or give it 0% authority?

I have done the done the maximising accuracy routine, the differences are quite small so I don't think would account for how far out it is, but have always set it up that way.

11 hours ago, rogerguzzi said:

Hello Darren

                      I have a meter that may work and a Innovate LM meter and sensor that you are welcome to try for cost of posting

I will look tomorrow as to what they are?

Roger

Hi Roger,

Thanks very much for the offer. Is that the LD Performance meter from the controller kit you bought? IIRC they use a different Bosch sensor with a wider reading than the 10-20AFR that the LSU 4.9 does, so I'm guessing it wouldn't be accurate if used with the Spartan Controller. I wanted to bypass the ECU/Tunderstudio readings and use a gauge just to ensure there wasn't something random happening with the connection there, with a gauge connected it's displaying something direct from the controller, so if still wrong then it's definitely controller or sensor.

I will see if I can get a voltage reading from the controller, that may just tell me that it's kicking out the full 5v.

Will report back when I've had a chance to take a look.

Thanks to both.
Darren

 

Edited by Dazman1360
Link to comment
Share on other sites

Everything I've done so far was with the car stationary and idling in the garage, so to see what the EGO correction was doing I had the laptop plugged in as I was driving. I could see that whatever the problem is, it's intermittent, as every now and then the AFR gauge would go back into the range you'd expect to see for a bit, then way off the scale again and then repeat this process for the whole journey.

The EGO correction gauge was trying to richen the mixture when the AFR was reading incorrectly as expected, I also found a gauge to show AFR input voltage which mirrored what the AFR gauge was doing, so reading at 5v when the AFR gauge was off the scale and then around mid point when working normally.

I'm now thinking that if this were a controller or sensor fault it would be wrong all the time, maybe it could be something as simple as connection or wire breaking down intermittently? Just a thought and a cheap fix.

Darren

 

 

Link to comment
Share on other sites

Except...... pretty sure that you need the higher voltage (5v) to get the high reading.  So that makes it unlikely it's down to a wiring fault between controller and ECU.  Possible wiring fault between sensor and ECU or intermittent fault within one of them would be my guess.

Nick

Link to comment
Share on other sites

Hi Nick,

The AFR gauge had a max of 19 and the EGO volts a max of 5, so I extended both the gauge ranges and got readings of 19.6AFR  & 4.99v, I also probed the o2 cable from controller to ECU and got 5v, so would seem TS is displaying correctly. Wiring looks sound and some jiggling around of cables at key points made no difference, so I'm back to thinking a sensor or controller fault.

I think a new sensor is my best bet, just feeling a bit mean at the moment so hoping to find one cheap. If that doesn't sort it, then I'll have to return the controller for testing.

Do you think I should disable the o2 sensor for now or turn down the ECU authority % right down in the EGO Control settings? 

Darren

IMG_20171002_182446588.jpg

Link to comment
Share on other sites

Darren,

 

I am far from an expert in such matters, but I don't see the dilemma?

If the sensor/controller/wiring is faulty, and giving a bad signal, then the only option is to remove it from the equation. The O2 sensor is not crucial to operation, it is added extra. Which right now is making things worse.

Reducing the authority doesn't change this, you simply alter how much worse it can make it, surely? So disablement is the only option for the time being. When you replace/repair what is wrong, then running on reduced authority makes sense until you have full confidence in the system again.

 

Cheers,

 

Phil

Link to comment
Share on other sites

7 hours ago, Dazman1360 said:

thinking that if this were a controller or sensor fault it would be wrong all the time,

 

I have twin wideband sensors reporting to an Innovate LM2, one in each exhaust branch so that I can log what each carb is doing, in the last month or two one of them has started playing up occasionally. Symptom is that both warm up and report expected AFR's, then at some non specific time afterwards (could be 2 mins could be 20 mins) the front one goes doolaley, it could suddenly report AFR's as rich as 7.5 (more usual) or sometimes an AFR of 22, or just switch to an E8 error state, once gone into weird mode it just stays there, however just flip the power to it so it has to go through its reboot cycle and it comes back up normal. How long it stays like that is as random as the initial failure. I did swap the sensors around on the channels into the Innovate but as the problem moved with it, I'm 99% certain its a sensor that is failing.

NB by occasionally playing up, I now mean almost certain to do it within 10 mins or so.

Alan

  

 

Link to comment
Share on other sites

11 hours ago, thebrookster said:

Darren,

 

I am far from an expert in such matters, but I don't see the dilemma?

If the sensor/controller/wiring is faulty, and giving a bad signal, then the only option is to remove it from the equation. The O2 sensor is not crucial to operation, it is added extra. Which right now is making things worse.

Reducing the authority doesn't change this, you simply alter how much worse it can make it, surely? So disablement is the only option for the time being. When you replace/repair what is wrong, then running on reduced authority makes sense until you have full confidence in the system again.

 

Cheers,

 

Phil

No dilema Phil, just it's not working and I want to get it resolved within the warranty period of the controller. I appreciate the o2 sensor isn't critical as the car has been set-up on a RR, but it's a useful diagnostic tool if nothing else.

10 hours ago, Nick Jones said:

If you set the authority to zero you have effectively disabled it - and this is what I would do.

 

Nick

:thanks:

8 hours ago, oldtuckunder said:

I have twin wideband sensors reporting to an Innovate LM2, one in each exhaust branch so that I can log what each carb is doing, in the last month or two one of them has started playing up occasionally. Symptom is that both warm up and report expected AFR's, then at some non specific time afterwards (could be 2 mins could be 20 mins) the front one goes doolaley, it could suddenly report AFR's as rich as 7.5 (more usual) or sometimes an AFR of 22, or just switch to an E8 error state, once gone into weird mode it just stays there, however just flip the power to it so it has to go through its reboot cycle and it comes back up normal. How long it stays like that is as random as the initial failure. I did swap the sensors around on the channels into the Innovate but as the problem moved with it, I'm 99% certain its a sensor that is failing.

NB by occasionally playing up, I now mean almost certain to do it within 10 mins or so.

Alan

  

 

I have a new Bosch sensor coming from across the pond for $63 delivered (hoping it sneaks through customs without any import charges :unsure:), so hopefully that will sort it.

Thanks to all
Darren

Link to comment
Share on other sites

44 minutes ago, Dazman1360 said:

No dilema Phil, just it's not working and I want to get it resolved within the warranty period of the controller. I appreciate the o2 sensor isn't critical as the car has been set-up on a RR, but it's a useful diagnostic tool if nothing else.

Hi Darren, 

I have just re-read the post, I have misunderstood your question. I had read it as suggesting you were considering reducing the authority to a smaller value, which is what I was confused about! A more careful perusal and I realise you meant turn it all the way to zero as an option for essentially disabling the sensor until you get the replacement. Which makes a lot more sense!!

My apologies for the misunderstanding, I shall await with interest to see how you get on with the new sensor.

Cheers

Phil

Link to comment
Share on other sites

Hi Darren,
I bought a O2 sensor from 14point7 and it didn't read correctly, a bit like yours but consistently. Sent it back but he said it was OK. On rewiring I realised that I had misread/remembered the wiring diagram. Connected it up the way the diagram said (instead of with the colours I thought should be positive and negative) and she has worked well ever since!

Just saying..........  (I am often pretty stupid at electrickery)

 

Cheers,
Doug

 

 

Link to comment
Share on other sites

  • 2 weeks later...
On 10/5/2017 at 8:52 AM, dggt6 said:

Hi Darren,
I bought a O2 sensor from 14point7 and it didn't read correctly, a bit like yours but consistently. Sent it back but he said it was OK. On rewiring I realised that I had misread/remembered the wiring diagram. Connected it up the way the diagram said (instead of with the colours I thought should be positive and negative) and she has worked well ever since!

Just saying..........  (I am often pretty stupid at electrickery)

 

Cheers,
Doug

 

 

Pretty confident the wiring is OK, has been working fine since conversion.

35 minutes ago, Chippy63 said:

Have you used silicon in the inlet system or the up stream exhaust anywhere lately? It will kill the sensor btdt

I have used silicone spray in my garage, but not anywhere around the engine bay.

I do use an additive: http://www.millersoils.co.uk/products/vspe-power-plus-multishot/624 but I did contact Millers to ask if it was safe with Injectors & o2 sensors, which they confirmed it was.

New o2 sensor is due for delivery today, so should know by early next week it it was the problem.

Darren

 

Link to comment
Share on other sites

Good news is that the new o2 sensor has done the trick and everything is working again.

Not so good news is that I did have to pay some import charges which meant it was only a small saving on buying from a UK supplier, but no big deal.

A bit curious as to why it only lasted a few thousand miles, will be more curious if this one does the same....

Darren

Link to comment
Share on other sites

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now
×
×
  • Create New...