Running way too rich w/ Motron 80lb Injectors

[hardrock21]

I attached a config. file. I could fix the histogram to display the commanded Lambda but the error wouldn't work for me. I think you need to try deleting the custom AUX input and re-set it up. Then check to see if the new name for the aux in put shows up when you look at the Lambda_err Aux input. I hope that makes sense. Or if you could give me your transfer function for your wideband I can finish it. I don't think you will be able to have it work for the previous scans but you have the data.

Regarding the flow rate I was asking for the what your stock tune value was at 0 kPA. What type of vehicle it this? I can download a stock tune. I need to look at the starting point to calculate the new min fuel mg value. There is a post on here where Greg touches on adjusting the min fuel mg.

Focusing on what was changed I really think your problem is coming from effects of the min fuel mg and the min PW floor (also based on the min PW tables). Looking at the scan, the trims are fine until it is trying to go below 2.2 msec which is telling me you were hitting the min PW floor. The computer is maxing out the negative trim and it still can't get to stoch. Then as you move up in load the trim is carried up and it is too lean. You can see it in the attached pic where that is happening.

If you keep it out of closed loop for now that might be more apparent but you need to get the injectors to go to a lower PW. It is way rich. I would lower the min pw and default pw more. I have 0.593 in mine and this works out fine with the addition of the offset.

Lastly, if your previous VE table data was known to be good I would load that back in.

It's a 00' Trans Am WS-6 6 Speed.

I'm trying to follow what you are talking about with the histograms. By transfer function, do you mean the equation for the wideband? If so, it's (Volts/9.756) + (.68)

The previous VE table before the injectors was fine. It's been altered trying to get these injectors lined out though. Are you referring to the VE table I had w/ the 42lb inj?

I changed the transient mg, default PW and Min Pw to what you listed and I will try those out. Thanks for being so dedicated to helping a rookie out. I get overly frustrated when I just can't grasp concepts of different things and it just drives me nuts how over my head I feel most of these parameters, variables and algorithms are.

[hardrock21]

Good reading here......

I've seen the 80's on a flow bench tested with J1832 test methods....believe me they have horrible flow characteristics at pulse widths below 4.00 ms...

http://forum.efilive.com/showthread....acked-up-to-be!

Howard

I think you and everyone else has reassured me that I am fighting a losing battle w/ these injectors. Thanks for the reading information, I'm scrolling through it all now!

Paul

[Redline MS]

Unfortunately for quite a while it was the only big injector out there that had the flow capability to feed big power. Now that has all changed so the options are much better.

With the range that companies like Injector Dynamics have with good data it's just doesn't make sense to beat your head against the wall.

Howard

[hardrock21]

Unfortunately for quite a while it was the only big injector out there that had the flow capability to feed big power. Now that has all changed so the options are much better.

With the range that companies like Injector Dynamics have with good data it's just doesn't make sense to beat your head against the wall.

Howard

Ya, the ID's are nice and in retrospect, I wish I would have ponied up and bought those. Unfortunately I didn't know these were as bad as they are when I bought them. So now I'm just crossing my fingers I can get these lined out with the help of the people on this board. If not, it looks like I might be coming up with $1k for a set of ID1000's.

[homebuilt]

It's a 00' Trans Am WS-6 6 Speed.

I'm trying to follow what you are talking about with the histograms. By transfer function, do you mean the equation for the wideband? If so, it's (Volts/9.756) + (.68)

The previous VE table before the injectors was fine. It's been altered trying to get these injectors lined out though. Are you referring to the VE table I had w/ the 42lb inj?

Ok, I found a stock file and it didn't change the number much. By this method it results in a 0.012 min fuel mg.

I setup the conf. file with your WB. This one should work for new scans going forward.

Yes, I would go back to known good VE data. This should in theory not matter which injectors you have so if you think the VE table was right when you had the 42 lbs injectors then I would go back to that data. Not data adjusted based on trims. You might be better off just staying in open loop mode.

[hardrock21]

Ok, I found a stock file and it didn't change the number much. By this method it results in a 0.012 min fuel mg.

I setup the conf. file with your WB. This one should work for new scans going forward.

Yes, I would go back to known good VE data. This should in theory not matter which injectors you have so if you think the VE table was right when you had the 42 lbs injectors then I would go back to that data. Not data adjusted based on trims. You might be better off just staying in open loop mode.

You figured out the right formula for the Lambda_error just as I did, lol. I set that based on Greg's Calibrated Success DVD and it was really throwing me off. I'll plug .012 into the min fuel mg. I should be able to post back with results at some point tomorrow!

Paul

[hardrock21]

I would like to just take a minute and sincerely thank everyone in this thread for their help in trying to figure this issue out. Homebuilt and nodnarb, your help is/was greatly appreciated and this board is lucky to have helpful members like you guys! I would like to extend a special thank you for the efforts of blueblownz06, this guy took it upon himself to get in touch with me and take an active effort in trying to help me get this lined out. He has had some experience with these injectors and as it turns out, with the right injector scaling data, these things can be lined out and controlled just fine! He went above and beyond to help me out and it was greatly appreciated! Mad props to him!

Again, thanks to everyone for their efforts, thanks to all of you I don't want to light this car on fire and push it off a cliff anymore!

Paul

[homebuilt]

I'm glad to have helped. I would love to know the details of what solved your problem. Please , do share.

[hardrock21]

Well, I'm sure he will see this thread and post the answer to that if he wants to. I think he does some tuning on the side so not sure if he wants all his tricks of the trade out on the table. He is a super nice guy though, I wouldn't hesitate to contact him if you had any questions!

Paul

[blownbluez06]

Good reading here......

I've seen the 80's on a flow bench tested with J1832 test methods....believe me they have horrible flow characteristics at pulse widths below 4.00 ms...

http://forum.efilive.com/showthread....acked-up-to-be!

Howard

Unfortunately for quite a while it was the only big injector out there that had the flow capability to feed big power. Now that has all changed so the options are much better.

With the range that companies like Injector Dynamics have with good data it's just doesn't make sense to beat your head against the wall.

Howard

It's as if everyone ran around complaining about how horrible their car ran before these ID's came out. The ID's are good injectors and I personally have run a set of their 1000's for 2 years. I've used these 80lb injectors plenty and I don't have any customers complaining. I would love to see a blind test that involves a car being properly tuned (by me or anyone I trust) with these 80's and a set of ID's and see if the driver could figure out which is which.

[blownbluez06]

I would like to just take a minute and sincerely thank everyone in this thread for their help in trying to figure this issue out. Homebuilt and nodnarb, your help is/was greatly appreciated and this board is lucky to have helpful members like you guys! I would like to extend a special thank you for the efforts of blueblownz06, this guy took it upon himself to get in touch with me and take an active effort in trying to help me get this lined out. He has had some experience with these injectors and as it turns out, with the right injector scaling data, these things can be lined out and controlled just fine! He went above and beyond to help me out and it was greatly appreciated! Mad props to him!

Again, thanks to everyone for their efforts, thanks to all of you I don't want to light this car on fire and push it off a cliff anymore!

Paul

It was my pleasure sir.

[vortech]

It's as if everyone ran around complaining about how horrible their car ran before these ID's came out. The ID's are good injectors and I personally have run a set of their 1000's for 2 years. I've used these 80lb injectors plenty and I don't have any customers complaining. I would love to see a blind test that involves a car being properly tuned (by me or anyone I trust) with these 80's and a set of ID's and see if the driver could figure out which is which.

I agree with this statement 100%. Been tuning 60's, 80's, 95's and even 120's with no data-and lots of happy customers. In-fact even when the data for some injectors did come out-had to revert to other methods to get them to run properly.

So all this Rah Rah that if you don't have the data and don't have a set of ID's is starting to sound like a load of BS to me.

[jakebdb56]

So all this Rah Rah that if you don't have the data and don't have a set of ID's is starting to sound like a load of BS to me.

But all the cool kids waste money on ID's

Honestly we're lucky to have the minimal data we can find for a lot of these injectors on the market. The Bosch 220's have no "real" data, but run like butter. If it ain't broke don't fix it right?

[eficalibrator]

So all this Rah Rah that if you don't have the data and don't have a set of ID's is starting to sound like a load of BS to me.

If that were the case, why would you waste time on useless stuff like fixing the MAF curve, VE table, or base injector flow rate? There are plenty of "other ways" to make it richer/leaner too, right?

[Pekka_Perkeles]

If that were the case, why would you waste time on useless stuff like fixing the MAF curve, VE table, or base injector flow rate? There are plenty of "other ways" to make it richer/leaner too, right?

Right, but then the story probably continues anyway with things like "min fuel milligrams".

I wonder how many parameters there needs to be changed to get any 80 lb injectors working - even with those having correct info from your DVD?

[homebuilt]

Good reading here......

I've seen the 80's on a flow bench tested with J1832 test methods....believe me they have horrible flow characteristics at pulse widths below 4.00 ms...

http://forum.efilive.com/showthread....acked-up-to-be!

Howard

Going back to Howard's contribution, I generated the following plots to see what sort of Lambda variation exists as a function of the injection PW in an engine test. I also made a histogram showing just the distribution at 2.1 msec which is where my idle is currently. My VE table was calibrated in steady state on the dyno and the mean Lambda_err is pretty close.

You can see additional variation at lower pulse widths for sure. To confirm if this is coming from the injectors I'm going to lower my fuel pressure down to the original FMS calibration pressure which will move my minimum pulse widths to a much lower variation point on the plot (entering the correct modeling for the new pressure of course).

If this proves to help it would be another work-a-round with these injectors (if one has enough flow rate to spare). I don't need 1000HP on the street so I am willing to trade for a better calibration until I have the funds to consider another injector swap.

[sarg]

I will be curious to see the results. I have poor man SD 80 lb injectors I purchased mainly because I could not afford the ID injectors.

[homebuilt]

I'm still working on testing this idea. It took me a while to figure out how to convert the Ford injector data into gm data at the original calibration pressure. I will share the spread sheet after i test it on some additional published data.

[homebuilt]

I wanted to provide an update just in case anyone is waiting for the results of this experiment.

Long story short, the injector data that I derived during this process solved most of the problem so I no longer have a reason to test the injector variation at higher pulse widths (via lower fuel pressure). This is really best addressed with an injector bench anyway.

The spreadsheet I created and my tune is attached. If anyone wants to calculate the GM format values from different Ford data just replace the data on the first tab in yellow.

[DSteck]

FYI, anybody that wants IDs... I'll help you out. The price of admission is steep, but it's a much better strip club than the others out there.