Zeitronix WB Settings in v1.5

[BLOWNZ06]

I have no way to know if that green sheet data is not right.

I know that the equation that I have fits that data.

When you are saying that my data is not right what are you checking it against? A second wideband?

Who decided that the data on the green sheet was incorrect? How was this decided? If this is the data from the manufacturer I think that it is correct.

What I do not understand is you state that you check my equation against someone logs and it does not work? How are you proving that?

I am not saying that I am that smart and everyone else is wrong, I am wondering what you are comparing this against, that is all... I just put information up for though, not as though it is law.

Also when you are saying it is off, where is it off at. Can someone mail me one of the voltage logs? [email protected]

The only areas that are crucial that the formula is really right is from 11 to 13.5 AFR.

[Magic_Chicken]

I haven't had time to log more data or do any more research or test all the new formulas. When you say:

The only areas that are crucial that the formula is really right is from 11 to 13.5 AFR.

I need mine to be accurate from at least 10 to 15 (all the way down to 10 for forced induction applications).

After talking to Zeitronix (which isn't a big company or anything, it's just 1 guy doing widebands in his spare time, like HP Tuners is 3 guys doing this in their spare time, which we all know doesn't really mean anything, because HP Tuners is the best stuff in the world even if it is a part time gig ), the data on the green sheet is correct internally to the Zeitronix, but the analog output doesn't match 100% because there is a slight delay sending it to the output. And for piping the analog output voltage back into the Zeitronix, he says that won't give a good representation of the data because the user inputs are heavily filtered because they can't tell what the user is inputing or if it's grounded properly, etc...

[BLOWNZ06]

Hmm... not sure why you would need yours to be correct at 10?

I have a FI 2002 Z06.. this is what all of my tuning has been done on.

There is no need to have it accurate to 0.1 at 10 because you would never set the car up for a final tune at 10? Again.. not sure why you would need it accurate to 0.1 at 15 either... the sensor is only being used to tune WOT and certainly anything set up at 14 let alone 15 would cause damage to anything internal to the engine.

So if in your example the sensor was off 1 at 10 but accurate to 0.2 even at 11.8 for example, I would consider this data completely accurate enough to tune the vehicle with. Again.. not telling you that you are wrong, just unsure of why you think it is necessary to see that the device is THAT accurate from 10 to 15, seems unnecessary for the intended purpose of the sensor.

Has anyone else been able to determine if another sensor brand is in fact MORE accurate than the Zeitronix is in these ranges? I would be surprised if they are.. same sensor for the most parts after all.

Just more food for thought.

[Magic_Chicken]

Want to run a little more rich than usual on the setup that I'll be tuning. Better safe than sorry.

Low compression 403ci (8.4:1)
Rear mount twin turbos (20PSI boost)
Liquid/Air intercooler
Methanol Injection
Direct port nitrous (200 HP minimum)

Sure would suck to blow a $15-20,000 setup.

So far, Zeitronix is the only wideband system that is having this issue. Unless none of the other customers with other widebands are expressing their concerns.

[BLOWNZ06]

Looking through the voltage logs that were emailed to me, I am not sure what the purpose of them is.

The formula I came up with matches the Zeitronix information. I still don't see what measurement mechanism or control you are using to show that the voltages and formula are actually off? I cannot imagine that anyone here (maybe with the exception of the HP Tuners crew) has the equipment necessary to measure and accurately log the voltage from that sensor, it is changing very rapidly. There have to be some acceptable assumptions in any setup.

I expect to purchase HP Tuners in the next month or so, I am currenlty an LS1 Edit user and used the Zeitronix with the Zeitronix logging software to tune my car, along with the very basic LS1 Edit logging software.

Once I have it, I will set it up and simply compare it to the logs with the Zeitronix software. Should be simple enough to come up with some correlation between the two of them.

Not knocking you but it is a bit off to build a "$15,000 to $20,000" setup and then tune it to 10:1 and leave 100+ HP on the table, only to then add a 200 shot of nitrous to it. Put a 50 shot of Nitrous on the car and tune the car to 11.5 to one and you would get the same amount of power. Running that kind of boost and that kind of nitrous shot is not simply going to be safer because you are running it at 10:1. The cylinder pressure present during a 20psi run with a 200 shot nitrous setup is going to loose head gaskets on your car... not running it at the correct AF ratio. You want to make that setup safer, run less timing, not run the car WAYYY FAT. Again, this is off topic, but your logic of putting that much money into a setup and then running it at 10:1 is uhh...

I will wait to get my copy.. I do like the constructive conversation though.. something is lacking as LS1 Edit gets older. The software is not fancy but it works. This Enhanced interface is the ticket for those of us looking for more than just a simple tuning tool.

Have a good one.

[Magic_Chicken]

Sorry I didn't specify exactly what the plan was. I know there is power to be had by leaner AFR, but would much rather START pig rich and slowly lean it out than to guess at an AFR and go for it. But, like you said, this is off topic.

[Keith@HPTuners]

I do not have a Zeitronix wide band so I am unable to test them...

You can tie the Zeitronix output to 2 of the EIO inputs and log 1 as raw voltage.

If you guys cannot iron this out and figure out the correct formula across the entire voltage range we will have to pull Zeitronix support.

In your betas I have put in all the Zeitronix formulas I was told existed... and it seems that you each have a different formula that works for you..

We contacted Zeitronix for permission to support their wide band interface in our software and they agreed and gave us the formula that we had implemented originally... and now it seems as if they are not sure what formula to hand out, or do not wish to assist with the situation we are in.

Here is what we need.. All of you can create a user defined input derived from whatever formula can come up with.. so realy there is no need for us to put more betas on the webserver unless you have verified it to work using a user defined input.

If you can come up with a user defined input that works derived from information received from Zeitronix that backs up the voltage->AFR you are receiving then we can support it.

We will not support Zeitronix if we have to build our own equation that comes close to what the AFR should be.

Unless we know with great certainty that the AFR we are displaying is accurate based on information gathered from Zeitronix themselves, it is a risk to have it in our shipping version of software that we will not take.

It saddens me that Zeitronix does not want to chime in here to help us iron out the problem. Certainly they must have some insight.

Also, if you guys can come up with a curve for formula that works or is off by a constant across the entire voltage range, check your grounds. Unplug the WB sensor from the Zeitronix unit and measure the voltage that EIO displays... A good ground is very important to accurate readings.

With the development of 2.0 it is very hard for me to dedicate the time needed to help iron the Zeitronix support out, and quite frankly I do not think its a problem with HPT or HPT's users... All of the other wide bands we support show no signs of these problems.

We will be adding a "is wide band" option to the user defined inputs that will allow you to chart your data in the histogram... but if a agreed upon formula/table cannot be completed in the VERY near future we will just scrap Zeitronix support as 1.6 is just about ready for distrib.

[Magic_Chicken]

We will be adding a "is wide band" option to the user defined inputs that will allow you to chart your data in the histogram... but if a agreed upon formula/table cannot be completed in the VERY near future we will just scrap Zeitronix support as 1.6 is just about ready for distrib.

With this addition, speaking for myself, I don't care if there is native Zeitronix support. As long as user definable inputs can be charted in the histogram, native support is irrelevant (to me).

Thanks, Keith!

[BLOWNZ06]

With this addition, speaking for myself, I don't care if there is native Zeitronix support. As long as user definable inputs can be charted in the histogram, native support is irrelevant (to me).

Thanks, Keith!

I can agree with this as well.

[2000Firehawk]

Me too..
No need for Zeitronix support , if it might cause a problem for even ONE, uninformed HPTuner customer...

Dave

[Dirk_Diggler]

Has anyone stopped to think that maybe the EIO cable is somewhat loading down the Zeitronix output.

If this is the case you will need to find out the difference between it unhooked (with the car off) measure the output voltage and then measure the output voltage with the output hooked to the cable. YOu should then see how much of a difference is between the two and correct for it. This may account for all the variations everyone is seeing.

Someone with a DMM could check this fairly quick

[Keith@HPTuners]

I have the user config "is wideband" option completed. Should be able to use user defined inputs for the histogram now.

E-mail me your customer ID # to request the download..

If you can, I'd like this to be tested ASAP... for obvious reasons.

[Magic_Chicken]

Tested and working fine! The level of support around here is FREAKING AMAZING!

[black02ss]

Thats great to hear Magic Chicken! I believe Keith is right about it being a grounding issue. Where do all of you have the ground for the Black Box comming from? Also, might be a dumb question, but do you have the Brown wire hooked up as well?

[2000Firehawk]

KEITH - EMAIL SENT FOR SERVER DOWNLOAD...

All my grounding, [brown wire too], for both Zeitronix & HPTuner is at a double nutted long stud that mounts the steering column at the floorboard.

I felt that if ALL Zeitronix/HPTuner, grounding/returns , were at the 'same' common grounding point, that it would be easier to troubleshoot , if anything went wrong.

My Zeitronix 'box' is mounted above the brake petal area.

Dave

[black02ss]

I am going to go back on a statement I made earlier in regards to BLOWN's post. I got to thinking about a few things at work last night and decided to take another look at this. I have replotted the values and made a spreadsheet (available below) that shows the differences as well as the calibration points for each. After further looking into it, it seems that his formula is actually right on the money. If anyone wants to look at the data below and put in their thoughts, I would appreciate it. If this is in fact it, it will make it easier for future people as the homework has already been done.

http://12.222.118.155/Zeitronix(1).xls

[BLOWNZ06]

I am going to go back on a statement I made earlier in regards to BLOWN's post. I got to thinking about a few things at work last night and decided to take another look at this. I have replotted the values and made a spreadsheet (available below) that shows the differences as well as the calibration points for each. After further looking into it, it seems that his formula is actually right on the money. If anyone wants to look at the data below and put in their thoughts, I would appreciate it. If this is in fact it, it will make it easier for future people as the homework has already been done.

http://12.222.118.155/Zeitronix(1).xls

I have ordered my copy today... so I will be able to test it out as well. I went back through the calculation and could not find where it was off.. so I went ahead and ordered a copy to just test it myself.

Seems you already did that.. but I will double check it as well once I have mine.

Thanks for the updated information.

[Magic_Chicken]

Hey Nick, just curious, how or where did you come up with that formula? With the little testing I did last night to test the new "is wideband" feature, I used your numbers and they looked pretty good.

[BLOWNZ06]

Here is what I did.

1. Plug in the Zeitronix data from the green sheet... checked OK
2. Made sure I could duplicate the curve they advertised... checked OK
3. Took the data and plotted a curve with V on the X axis and then AFR on the Y axis, basically the opposite of what the advertised curve is... checked OK
4. Took the advertised numbers from the green sheet, made sure my output from my forumla matched (within reason) the AFR that was output on the Green Sheet for each of the values... this also checked OK (I posted this plot)
5. Now took my formula from step 4 and used the values necessary for HP Tuners.. 0, 0.1, 0.2,. etc and calculated AFR for those values.

That is where I got the table from you are using. I am glad you guys are having success with it. I did not want to have to buy a new wideband either. The curve is VERY close as you can see from the plot I posted here.

I think the biggest reason that other peoples formulas were not working is two fold.

1. A second order equation is just a good enough fit for data. There are three of these shown in the second chart I plotted.
2. I think that some people tried to interpolate from the data that they had instead a "best fit" 6th order equation. This would also not give good results as the line is not linear enough for interpolation.

I think this same method can be applied for pretty much any wideband that is out there. Let me know if you want further information.

I am no mathemetician, but I did stay at Holiday Inn Express last night.

[Magic_Chicken]

I am no mathemetician, but I did stay at Holiday Inn Express last night.

LOL

Maybe you need to let Zeitronix know of your findings. Tell them this formula works best for their output.