Has anyone successfully set up this combo yet for A/F ratio logging? My datalogit should arrive later this week and the AEM unit looks nice, and cheaper than the prebuilt OZ unit now using the Bosch sensor.. I assume it will work fine since it just spits out 0-5V readings and the software does the conversion for you.

Matt

 

Those guage-displays look cool. Too bad the needle only goes down to 11 AFR. Our rotaries run richer than 11 AFR.

 

I ordered one of the AEM units yesterday, should be here next week. I got the box single channel version, no need for some other gauge in my car, all I care about is logging to tune it. Now that I finally got onto the datalogit forums last week there are a few guys with this wideband + datalogit working already.

 

I am using the AEM with my datalogit and have been doing so for about 2 months. You DO need a polynomial as the AEM is NOT linear as some have posted in the past. I have the polynomial. Send a check to me for $50 and I'll send you the polynomial. J/K. I'll try and post it later when I have my laptop in front of me

 

NP.. let me know your fax number and I'll send the check over.

 

I'm using the AEM/datalogit combination also. Granted, this is the first time I've looked at wideband info, but I'm getting a huge amount of "noise." When I get readings fluxuating 2 or 3 AFR points for a single cell, it makes it difficult to tune. I don't know if maybe something in the engine is actually causing this variation, or if maybe I'm looking at a bad voltage source or ground (it's grounded directly to the battery). Still trying to track down the problem....

Haven't gotten a chance to verify it's readings, but hopefully I'll be on the dyno Wednesday and can check it out.

Anyway, not bad mouthing the AEM. None of this is necessarily it's fault, but just don't be like me and be all disappointed if it isn't immediately perfect for you!

 

Is grounding to the battery really a better spot than a good chassis ground?

 

here is the latest polynomial I found on the AEM forums:

AFR = .0295X^4-.1795X^3+.4877X^2+.5333X+8.4422

where X = voltage (0-5V)

 

Edit:

Seems we can't use this equation as datalogit only has inputs for a,b,c.. this equation has too many variables.

 

Just wanted to follow up. I did some testing today by pulling the sensor out of the bung and letting it dangle in free air while the engine ran. It suddenly started reading a steady free air voltage. So it looks like it's an engine issue and not anything wrong with the UEGO.

The instructions say to take the ground directly to the battery. It's very clear about that. The chassis always has some voltage in it due to ground loops.

The polynomial for the datalogit is out there. It's on the AEM forums and also the datalogit forum once you get access to it. I don't have it available right now, but it's easy to find.

Good luck with it!

 

I converted the polynomial to a 2nd order. Someone may want to peer review my work though. The first graph shows the full range from 0 to 5V. The second one has the AFR range from 10 to 14.

 

Your calculation looks good, the only issue I have noticed is that datalogit is calculating the polynomials incorrectly? After entering the equation the numbers listed under the 'value' column do not match what the calulated solution is in excel.. i.e. the voltage values 0 and 5 do not match what their calulated A/F is in excel.. anyone else notice this?

 

That's real close. The polynomial that I've seen posted in various places is 0.4533x^2 - 0.7119x + 10.051. That's the one I'm using. It was posted on both the AEM and datalogit forums.

 

I would use tailhappy's polynomial then. They are all approximations anyway. I don't know how the above polynomial was developed maybe someone can research it. Mine is an approximation of an approximation (the 4th order equation posted above).

Who knew all of that math would come in handy someday?

 

Here is an example of what I am referring to.. I input two points along the AEM voltage chart that comes with the kit, then input the polynomial listed above.

 

Now here is the auxiliary window, you can see that the voltages do not match up with the AEM chart

1.716V = 10.12AFR
4.056V = 14.64AFR

The values calculated in datalogit look way off.. is this the case with yours as well?

 

That is strange. What version of the software are you using. I do not have a wideband yet so I haven't been able to check it first hand. I have another spreadsheet to use for any 2nd order polynomial. If anyone would like it PM your e-mail. I am also going to put it on the yahoo group.

 

V1.10.. I just reverted back to 1.09 to test and same results.. very odd.

 

I'm not totally sure I understand what you're saying, but if you're referring to the picture in poly 2, those numbers don't mean anything anymore, as indicated by them being grayed out. They're only used if you want to use a linear representation instead of a polynomial curve. However, if you're talking about going to your logs and doing the calculation from there and it's wrong, that's a different story......

 

So you have to manually calculate the AFR from the voltage from the logs? I thought the Datalogit software was supposed to calculate the AFR for you since you put it the polynomial. If you have to manually calculate it then why do you even put in the polynomial?

 

Yeah, sorry, it was late and I probably wasn't clear. Once you put in the polynomial, the datalogit software converts the raw voltage to AFR for you. I was just meaning that if mjw was saying that when he went to his logs and hand calculated the AFR using the polynomial in Excel, and it was different than what the datalogit had calculated in the log, that would be an issue. But if he was just concerned that the data in the aux screen is wrong, it's grayed out and not meant to be pertinent after you put in the polynomial.

 

not to bring this back from the dead but were you all happy with the gauge type AEM wideband afterall?

 

I'd like to know as well

 

Well.. its tough to give an answer on it seeing as I have had little time to tinker with it and tune. The #1 problem I am having with it (as is another guy on the datalogit list) is a lot of excessive noise in the voltage signal. I have it installed and grounded as per the instructions, but there are some good fluctuations in the readings.. I have to get a lot of samples in one cell to get average readings that seem believable. I'm not sure how much this is the case with other widebands. I have mine set up with the delta box checked and the extra ground into AN2, otherwise without the extra ground my WOT readings seem 1-2 AFR too lean which is not the case. Like I said I haven't had time to really troubleshoot yet as my free time has been about nil, lately I have been fixing a nice fuel leak anyway.

This is with the box type wideband just used for logging, not the gauge type although they use the same sensor.

 

the gauge type states it does not go below 11.1 AFR. Im hoping this is a problem with the gauge and that the output wire that you datalog from will allow you to see values lower than 11.1 care to comment on your findings?