First off, I must say I've very proud of myself. Considering I am a technological retard, I figured out how to Datalog, save my runs, and open them with EEC Analyzer all by myself. Ok, with a little bit of help from Jake (GTJake). I recorded the following during my run: Injector Pulsewidths 1 & 2 LAMBSE 1 & 2 KAMRF 1 & 2 RPM Spark MAF Kg/hr MAF Voltage Percentage of Load Actual Load Throttle Position Sensor ISCDC Now, the questions: I'm assuming that because I have my new values for MAF kg/hr and MAF voltage, I can make my own MAF transfer for my C&L 76mm MAF, correct? If so, I would want to focus on the bottom half of the curve more because it seems that the values grow more exponentially than they do towards the top half. Am I right in that assumption? Should I alter the MAF curve/transfer in the EEC Analyzer or in Cal Edit, or doesn't it matter? I put all of my cam specs into the Analyzer and hit save. How do I write these changes, or any other changes I make in EEC Analyzer onto the Tweecer, so it will reach the PCM? I know this seems juvenile, but I can't understand the relationship between the Tweecer and EEC Analyzer. I'm running 10 degrees base timing, plus the 6 I added through the Tweecer for a total of 16. Does it seem right that I'm only running 30* total? WOT in 2nd gear at 5800 RPM (5797.75 RPM) = 30.25. I thought that was a bit low, so I figured I'd ask. The limits of 93 octane are somewhere between 34-38 total, right? If anyone has any other suggestions of what to look for/at in my datalog, I'd appreciate it. Sorry for the length. Joe
You should be cause as you now know, you don't get detailed instructions on using the Tweecer. I will try to answer some of your Q's. I think you are saying that you have got the 30 point transfer from one of the Tweecer sites. Is that correct? If so you just load them in your cal file under the maf transfer function. You could do the import/export thing but if you are not sure of how to do it I would just do it the old fashion way. As you imput each cell in Caledit you will see the curve change if that is what you are talking about. You do understand that you can save one change at a time and name it differently to create a new cal file if you wish, correct? The Analyzer is just a tool to analyze the data. You can cut and past some things like maf curves but for the most part you use it help you crunch the data and make heads or tails out of the datalogs. It is like Log Analyst which I've used and like but it has more features. It has been my findings that most NA Ford small blocks like about 36 degrees total spark give or take a few. Aluminum heads have been known to like more but not in all cases. You can always bump it up a degree or two until you find the sweet spot. For wot fuel you can go at it two ways. MAF TRANSFER: At the point of the curve where you switch from closed loop to open loop you make the kg not voltage value in the cell larger to fatten and smaller to make things lean. Fuel Table: For our cars you use the fuel_table_stabilized_OL I just tweec the two highest load (0.9 & 1) rows for wot conditions. At 1500 rpm and all cells above that rpm I load the same a/f ratio like say 12.471. Joe, I know you have been around and I can tell you have a good understanding about cars but I will say this anyway +++ When working with fuel ratios dial back the spark a good bit and put the good gas in the tank for just in case you go too lean. I think a wide band is a must when dialing in your af ratio but if you don't have one you can make very small changes and get in the ballpark as they say. As for spark I see you already have a handle on how to datalog to verify your changes. Hope some of this helps! Welcome to the would of self tuning Later Grady
I've never made my own MAF function, but I thin you just enter in your numbers for the transfer. If you enter in 10 points, you can use the translater in CalEdit under the MAF Transfer Function to convert it to 30 points, which is then entered into the graph. The correct way to create a transfer function is to use the numbers for your specific meter givento you from the manufacture. This isn't so with C&L (as far as I know) but Pro-m does supply with you a flow sheet. If you can't figure it out, just try loading in the calibration supplied with CalEdit if there is one for your meter. I am using a preset calibratio on mine, and it does make the car run better. I only wish I had the sheet that came with my meter, but I bought it used... I hope this helps somewhat? Just follow your post on the tweecer board, you'll probably find the most help there.
Seems like you've got answers to most everything, but there are a couple I can respond to. Nope, at least not if your C&L is calibrated for something other than 19# stock injectors. Since it's lying to your EEC to get correct fuel delivery, the kg/hr you see is a big fat lie. Personally I'd experiment with one of the Pro-M curves that comes with the software (whichever one is for the same size injectors that your C&L is calibrated for), or check the Tweecer forums for someone else who's already done the legwork -- if you can get close enough that it'll run, the KAMRF will let you dial in the MAF curve. It'll take some time, but it's doable. So you're using a 6* global spark adder? The values you see in the datalog, and the ones in the table, are assuming 10* of base timing. So if you've got a global adder, that should mean that 30.25 is really 36.25. Personally I run 36* total timing at the top end, and I'm getting a bit of ping on warm days, so I'm going to back it down some -- also, Ed Curtis suggested that anything above 34 for my setup was likely not helping any more. I think I'm going to go out now and hookup the laptop to mine. I haven't done it in awhile, and I need to see how my new MAF is looking (gonna do some KAMRF logging myself ). My car feels like it's running pig rich since I did the Fox swap yesterday. My gas gauge is also dropping just a wee bit faster than I'd like. Dave
So, I just went out and did some tweecing. Not a lot, since it's getting late and I don't want to annoy the neighbors *too* badly. I did make some observations, however: 1. My MAF curve is plus or minus 5% throughout. Not bad. With the 75mm it was very nearly perfect, but that's because I entered the flow sheet. Now I'm being lazy and I'm still using the flow-sheet for the 75mm instead of the one that came with my 80mm 2. It gets damn hot under the hood. Idling, stopped, hood closed, my ACT was registering 175 degrees. 3. The adaptive control maximum ACT setting does not seem to work, I was in closed loop with 160* ACT, even though it's set at 140* max. 4. The "N" on my guage is 198 degrees. During the summer I frequently run warmer than that on the freeway. Damn. I think before next summer I'm going to get a bigger radiator. 5. The stock J4J1 Cobra calibration is *WEAK*. Just for giggles, I loaded up a stock J4J1 calibration, entered my MAF curve and injector slopes, and took it for a spin. I think I was down 20+ horsepower, it was a *DOG*. It drove nice, just had no oomph. I tried the same stunt with the stock T4M0 GT calibration, but I couldn't keep it running long enough to get out of my driveway. Anyway... I dropped my timing two degrees on the top end, to 34* total. Hopefully that'll cure my occasional pinging.
Oh, and one more observation before I forget... It does seem that even after the Fox conversion, the EEC will properly baseline the voltage from the TPS. At idle mine is registering 0.85 volts, and the EEC correctly registered when the throttle was closed, part, and WOT. Dave
You guys got a lot cleared up for me. There is too much to quote, so I'm just going to start in the order of posts. For my "new" MAF transfer, I'm going to try the Pre-defined "PROM_COBRA_24" file, and see what happens. I also changed my injector timing. After putting my cam specs into the EEC Analyzer, saving them, and then hitting calculate in the Injector timing in the box below, it suggested an injector timing of 383, as opposed to the Tweecer's stock setting of 300. Grady, you helped me out a ton. You explained it very well, and I understood every bit of it. Dave went on to tell me that I indeed am running 36* of total advance, so I'm going to leave my timing be for the time being. Thanks a lot for all your help! Dave, as always, you cleared up things I needed questions to. I took your advice, and I'm going to try messing with a pre-existing MAF transfer. Thank you once again. I've come up with a few more questions. I'm using the J4J1 calibration which uses the same values for high and low injector slopes. Would I benefit from either better economy or better performance if I set the high slope to 24, and the low slope to ~28 (24 x 1.2)? I'm also very curious to know the answer of Jake's question. The reason I am is so curious is because while I was at WOT, my LAMBSE's were 11.55, which make me think I need to add a bunch of fuel. It wasn't detonating though, and this is why I ask. Is there a reason to alter the injector breakpoint? It seems that I would only want to switch to open loop when I'm running at WOT and when I first start it. It seems that there are 4 major spark tables. Spark_MBT_table, spark_altitude_table, spark_base_table, and spark_bdln_table. Is there any sense in messing with these at the point where I am? Should I get my fuel issues sorted out first, and then start messing with the spark? Thanks for all/any input. I'm going to go run another datalog and see if my changes helped or hurt anything. Joe
Why not the regular PROM_24? It's not really important which calibration you have right now, just what Pro-M calibrated it for. Or am I understanding you wrong? I just now installed EEC Analyzer . Nice program... all I have to do is find my detailed cam specs -- I know the easy stuff off the top of my head, but I don't have the valve events memorized. Low slope is only used when the injector pulsewidth goes below 2 ms. This will be primarily when you're idling, or coasting in gear with closed throttle. If your idle is okay, I wouldn't worry much about the low slope for now. If your MAF curve is dead on, in theory LAMBSE should correspond exactly with what a wideband would see. For most people that's not true, which makes LAMBSE essentially useless for tuning, it might be two points off of reality. Get the wideband . As I understand it, the only one that really matters is base. Altitude is not used because our cars don't have a sensor that can detect altitude changes. The bdln table is load-based -- Darren has a mention of it in his documentation, but I didn't research it in any detail. I look forward to seeing your progress. I'm going to datalog some more on the way home today, and see how hot the car gets on a 90+ degree day with the A/C running . I also am getting some erratic readings in my datalog, looks like corrupted samples -- it's been awhile since I surfed the TwEECer forums regularly, I need to start staying up-to-date on that stuff again. Dave
Joe I'm pretty sure I've seen the 30 point curve for the C&L meter on the Tweecer or Yahoo Tweecer site. Many are using the same meter and like Dave said, just benefit from their efforts. The final spark you see in the datalogs is derived from base, altitude, borderline, & mbt tables. Not only are there four tables but there are some functions & scalars that can add to or remove from your final spark. Very good info that explains it better than I could can be found on the Tweecer sites. I can tell you that I killed the borderline & mbt tables by making all cells have the value of 55 which made it much more simple to get the results in the datalogs I was looking for. Your inj slope question would keep me at this keyboard for I don't know how long to try and answer in some way that would help you but I will touch on the big points. Lurk around the Tweecer site to fill in the gaps. As for dialing in your inj's I will share some things I've found based upon the time (it is way more than I want to even think about) I've spent datalogging and lurking on the Tweecer site searching for my questions. You check your average kamrf's by driving around using light throttle to stay in CL under all kinds of conditions which will give you pw's from about 1 to 7 ms. Throw the dlog into EEC Analyzer and it will crunch the data for you. The thing I would have you to understand is the time needed for GOOD VALID data. On a warm motor after a pcm reset you will need about 5 mins of drive time before the K's start to come alive. Drive another 15 mins under various conditions but trying to stay in CL. Pull over but don't kill the motor and switch to LOG TO FILE which will put your dlog at the 15 to 30 min time period after your pcm reset. You do this over and over until you start to get closer to your goal. When you get to within a couple percent of perfect (1.0) kamrf's then you need to obtain a second dlog at the 60 to 75 min time period cause you will see that things change more slowly the closer you get to your final goal. The J4J1 uses the same values for the high & low slope so the breakpoint becomes null & void. The following things effect how the kamrf's will look when dialing in your inj's: high slope low slope min pw breakpoint inj offset vs batt voltage A rule of thumb is the curve supplied buy the meter maker is close as you get it from them so you use it to start with. For a good tune in CL & max power, as I understand it, the steps are as follows: You go through the inj dial in process for good CL operation. Tweec a few points of the maf curve if needed and if it is needed the changes should be kinda small. The fuel ratio is confirmed with a wide band for wot max power by dumping its output data into your dlog. To adjust the final fuel ratio some use the upper end of the maf curve and some use the fuel tables. It has been said that if your maf is on the money and you have a wb to verify you can make the wb and lambse match which would mean that you could disco the wb and just use the lambse to achieve your desired a/f ratio. See what I mean about a quick & easy answer! I was trying to keep it short. Hope this helps and as they say your results might not be the same as mine but the stuff above is working for me. Later Grady
It has been a while since I put a in the ballpark tune to the a/f ratio after I put on the tb & maf but some points of the maf curve were kinda close and some points were as much as 1 point off. At that time I just pulled the spark back a good bit and watched the wide band, rpm, & maf values in CalCon for the data to get me close my desired a/f ratio at wot. I've learned some things since then and the method above can be improved upon so I don't think my answer to you is valid or of any good use. Later Grady
Joe I knew I had seen the maf data somewhere and I just found some C&L maf stuff on Clint's site. If you have ?'s just ask him cause he is a nice guy. I had a question about his program and he gave me an answer within a couple of hours. http://eecanalyzer.webhop.net/ Later Grady
I datalogged earlier on today, and I haven't really dug into those numbers yet, but it does seem that the car ran a little better, and didn't smell as fat, so I do believe I'm making progress. My KAMRF is starting to slowly decline from the 1.2x and seemed to be hovering around 1.12-1.18 during this datalog. Dave, I used the "PROM_COBRA_24" MAF Transfer rather than the regular "PROM_24" because I was afraid of going too lean too quick. The Cobra Transfer was in between the stock setting and the PROM_24 setting, which is why I chose it. Grady - Thanks for the novel! I really appreciate you typing all that. I left the battery cable off overnight, figuring that resetting the PCM would be a good thing. I'm going to go out tomorrow and check out my KAMRFs to get a true understanding of where they're at. I'd love to use the Tweecer site, but everytime I try to post, it tells me "Invalid session" and everytime I try to search, it lets me go through one or two pages, then tells me there is no data or something like that. It's done that to me for the past 3 weeks, so I've given up for the most part. I'm assuming I'm still a tad lean while cruising in CL. I'm getting closer to the "perfect" 1.00 KAMRF, but still not quite there. I'm assuming I should reduce the high and low slope scalars to take out fuel, right? I'm going to try emailing Clint to see if he knows where I could get a MAF Transfer for my meter. I would prefer not to mess with fuel settings until my MAF Transfer is close. Thanks for all the help. Joe
Adjust the MAF curve until KAMRF hits 1.00, or close to it. EEC Analyzer apparently can help with this. Or you can do it by hand. The short version -- multiply your existing curve by KAMRF at each datapoint that you have reliable logging for. Then clear KAM and go driving again. Eventually you'll get it dialed in so that you're not far off 1.00. I think you can probably fix the curve you have now, since you're well within 25%. Definitely a good idea to get the MAF dialed in before tackling other settings. Dave
Dave and Grady, you guys are the bees knees. Unfortunately, the GDMF )) hurricane isn't allowing me to take my car out of the garage. My baby can't get wet. I'll probably do some more datalogging on Sunday and now that you guys steered me straight, I should be able to get my KAMRFs right at 1.0. Thanks again! Joe
So if you're gettin a .96xx Kamrf, that means your running rich and the computer is trying to cut the fuel PW? Jake
The LAMBSE is the "mirror" of the A/F ratio, over 14.7:1. If my Lambse is targeting 11.55, that means the car is "supposedly" running somewhere around 17.9:1. So it's targeting 11.55, to try and bring the A/F back down to the stoichiometric equilibrium of 14.7:1. Joe
Then why am I getting a 12.8 lambse but a .96xx kamrf. Wouldn't that mean I'm "supposedly" running 16.6:1, but the kamrf is trying to lean it out. And it is obviously rich, because when I let off at high rpms it does the gurgling backfire deal. Never used to that... EDIT: kamrf is at .98xx while the lambse is targeting 12.8 in the upper rpms. Jake
