When you click on links to various merchants on this site and make a purchase, this can result in this site earning a commission. Affiliate programs and affiliations include, but are not limited to, the eBay Partner Network.
Thank you, also your graph looks very nice. I tried too to make graphs in excel but it lacks an autoscale or an easy way to have multiple Y-axis scales.. anyway I'm FINALLY getting feedback from FITECH. Every time a new staff member answers, like if no one wants to work on my case so they keep passing it to someone else.. This time the guy sent me a new version of the software and some new calibration files. I checked on my PC, and the software is new, but the calibrations are identical. everything is the same except some new values (the ones I was trying to add earlier) are now available for tuning.
The guy said
"Dario, I would like you to install this newer software into your unit. I want to get a baseline on the system to start adjusting from. Please follow the provided instructions for installing this software. Also when you are setting up the Calibration file in Write Cal to ECU please use the 6 cylinder selection as this more closely follows the ignition sequence of your engine. Once you have this installed let me know if you have the hesitation still and we can go from there."
As you may know a 2 rotor wankel engine does NOT follow a 6 cylinder ignition sequence, but I said whatever let's try it: of course it read 2/3 of the RPMs, so I put back the 4cyl one, then drive it around, had the same exact problems as before, made a new log file, and now I'm waiting for their answer again. We are 9 hours apart, so the process is very, very slow. But maybe someday I will get there, who knows.
Did this file get converted over correctly? I don't have my laptop to try and get that to work with DLL right now, curious to know if the text to columns command worked well. I haven't seen a csv (comma separated variable) file using periods before. Spaces, tabs, commas and others but not periods. Is that from the FItech computer?
uhmm... no ok I'm sorry, I uploaded the file saved from megalogviewer..(indeed it was called "modified"). Now I uploaded the original version, and should be comma-separated.
Today I made other tests, but I will not spam the thread with all my daily data. I will post something when I'll have some new development about it.
Right now I'm trying to find volumetric efficiency charts for 13b na engines to copy into the ECU. I found something and in the next days I'll try this and other stuff.
Just curious if you have made any progress on getting the values better or any help from FiTech to get it running better and consistently.
yes I actually made progress, and the car runs normally if I don't step too much on the gas, but the main problem I think is pressure/geometry related. Every time I accelerate, the drop in pressure is so great that the engine stalls for a moment, no matter what I do.
The holley had vacuum secondaries, this one is mechanical and there's no way to avoid it.
I'm studying what's the best way to keep the air speed up without losing power. I need to measure the actual CFM at full power, and build a restrictor that allows just that amount of air. Maybe that would do.
So my advice about this is that if you want something like this on a 2 rotor NA, it has to be already well over 200hp to be adequate. A stock or nearly stock engine is too small.
None of these "learning" closed-loop ECUs ever work on the rotary- I think you will have to hack the ECU to make your own closed loop maps and modify some correction maps (like TPS "Throttle Pump" map).
The biggest problem you are having right now is it appears the ECU cannot supply the correct "Throttle Pump" settings for your intake/engine combination to keep the your target AFRs on acceleration.
You open the throttle fast, intake velocity decreases, liquid fuel droplets fall out of suspension with the air and stick to the intake runners walls and the whole inside of the rotary engine (huge surface area) causing the engine to run severely lean.
The O2 sensor reads this lean condition from the exhaust and the ECU adds in a whole bunch of fuel into the engine to compensate, but actually the intake velocity is recovering by now and more of the injected fuel droplets are staying in suspension with the air and contributing to combustion so the car runs rich and floods out even though the O2 sensor is now reading rich and trying to reduce the fuel injected.
The lag by using the O2 sensor is too long to start with so you will always have an initial lean-out on throttle tip in and then on top of that you have the other driveability problem of subsequently running rich because again the O2 sensor lag is too long to pull fuel out fast enough.
The actual "Throttle Pump" correction from TPS has to be adjusted correctly.
The ECU is using O2 feedback on top of a set "Throttle Pump" enrichment/decay setting that is not enough to compensate for the rotary engine internal surface area.
When I was tuning my old Haltech on my modified 1987 turbo RX-7 I had polished the inside of my intake runners and was running all three throttle plates to open all at once. It was still a long runner intake system with injectors in the stock locations (1989 TII lower with 1993 TT upper/ TB). Still, I had to make huge changes to the "Throttle Pump" settings to make the car drive right even though I started with the correct and well proven base map for my car and injectors.
I had to add a bunch of fuel at low rpm throttle input and have the added fuel decay very quickly. This issue is going to be even more pronounced on the short runner Holley 4150 intake manifold.
One thing you can do to help mask the problem is to add in a very strong CDI ignition system that will fire the air/fuel mixture even when it is really rich instead of stumbling.
..exactly, I confirm 100% everything you said, that's why my idea was to play on the intake turbulence instead of endlessly correct the fuel values...
About the ignition, with the FC coils and direct fire I don't think I can build a much stronger ignition (without spending thousands on it).
From my last post I didn't have much time to work on the car and now it's stored in my garage for winter (I'll probably use it no more than once a month). Anyway, I'm at least satisfied on the reliability compared with the carburetor: now it starts immediately even after months of being parked out in summer and autumn. No more flooding, no more cleaning and rebuilding the carburetor for deposits...
The engine doesn't need that much CFM and I can't make the intake longer right now so here's the idea: now that I work with sheet-metal I can build something like a restrictor plate to create turbulence and raising a bit of intake vacuum: my idea is two disks with many holes, one disk on top of the other, fastened with the air filter screw. Turning one of the disks will gradually shut the holes so I can find the perfect balance without making a hundred different plates. my only indecision is between few larger holes or many dozens of small holes. I don't have time for calculations so I guess I'll make both and try what's better. I'm not in a hurry to use the car so I'll probably do this experiment it in the next few months.
I planned to update the thread anyway in the next days because I found a rear spoiler on ebay and I received it last week. I guess I'll do it now:
It's not the classic one but I like the idea that it's raised so rain don't flood the tail of the car.
It's made of foam, and it had previously been painted. I don't know if i'm gonna repaint it or just strip the paint.
I gently put the spoiler on the body with some paper underneath to see what it looked like:
I can't find many pictures of this kind of spoiler on the internet. I don't want to mount it right now, but these parts are so rare that I can't afford NOT to buy them..
Any further tinkering with the FiTech? I just stumbled across some parameters that are suggested for a Megasquirt system. I see the displacement is set at 2600cc... Maybe that might help getting it to run appropriate AFRs through the bands?
Any further tinkering with the FiTech? I just stumbled across some parameters that are suggested for a Megasquirt system. I see the displacement is set at 2600cc... Maybe that might help getting it to run appropriate AFRs through the bands?
yes I did experiment further, and I tested my theory about airflow and intake vacuum, and with a restrictor behind the air filter the engine revs smoother without bogging when I suddenly open full throttle. I didn't have time yet to test vacuum on the road under load but I think I'm on the right track. the intake was just too large and short for my poor stock engine and couldn't mix air and fuel properly.
to answer your question yes i tried many levels of displacement, from 1300 to 2600, 5000.. and over but it runs best with the real geometric displacement. It's tricky to work with self learning ECUs because some values are only a base number and it doesn't matter that much once it's calibrated. when the car will be on the road again (soon hopefully) I will test again other values with the restricted intake and I'll let you know for sure
I'm curious why you couldn't install bushes inside the throttle body holes.
the size that they come from Fitech are approximately 850 cfm, makes me wonder if you could have reducer rings machined on a lathe and then either interference fit or bonded to the inlet of the throttle body.
diameter dictates CFM, so why not "choke" it a bit to improve response
Last edited by wankel=awesome; 07-30-18 at 03:05 PM.
Finally I manage to have a free afternoon to to this: I made a video about this build and basically it's a recap on everything I wrote in this thread. I left out a lot of things (exhaust system, radiator etc...) But I think this 30 min. long version is boring enough already... I don't have much time to do these kind of things anymore, but I had fun while doing it so I will probably do other (shorter!) videos about specific things on the 1st gen rx7. Sorry about my bad accent and all the mistakes but it's difficult to speak another language, drive, and thinking about the video at the same time!
BTW, while on the subject, I see that I left you hanging about the CFM, restrictor plates, and vacuum.. I tried different restrictors but in every istance I had loss of power. Right now I still have the problem (and I show that in the video) but everything else is almost gone, and idle, taking off, flooding etc are a thing of the past. That took another "session" with the infamous customer service: I updated the firmware, they sent me like 3 wrong ones until I had the right one, and I still had to modify the files a little, but the new one has more options and I can manage the tricky parts. For all the people that will suggest to set higher values in the "accel pump", I already tried. I doesn't work as expected and I always end up with a rich (10:1) AFR during acceleration if I do that.
I saw some guy installing the new two barrel version vertically on a VW beetle engine, and I think if I do the same I can probably fit a gsl-se intake manifold, or something like that, but right now I don't want to invest too much money on the engine...
That was a pretty good watch! I like hearing about how folks got their car, and what it took to get it to this point.
If nothing else, I enjoyed hearing your 7 as you drove it around.
That was a pretty good watch! I like hearing about how folks got their car, and what it took to get it to this point.
If nothing else, I enjoyed hearing your 7 as you drove it around.
thanks man, I really appreciate that. I like to watch videos like this too, and that's why I did it!
Could you post your settings? I have an 84 That i have installed the 400hp fitech kit on. I drove it to work and it ran great but when i left i experienced bad bucking, car would idle way down then jerk forward and repeat. I had an rpm noise code which i am working to fix but would be interested to see what your settings are at.
Could you post your settings? I have an 84 That i have installed the 400hp fitech kit on. I drove it to work and it ran great but when i left i experienced bad bucking, car would idle way down then jerk forward and repeat. I had an rpm noise code which i am working to fix but would be interested to see what your settings are at.
of course, no problem, tomorrow I'll download it right from the unit so you'll have the most updated version possible. did you update the firmware? otherwise I will send you that also.
About the rpm noise, delete the error and try again: keep an eye on the errors in real time while you're driving. it's especially important that the spark is consistent during the initial learning, otherwise you'll never get around it.. so check the spark plugs, cables etc. the IAC is one of the trickiest things to set just because of the low air speed and volume in the 13b (compared to a 400hp V8) and I found out that the best strategy is to aim a little bit high with the IAC settings so that you'll have a little bit of a "happy" idle that slows down under 1000rpm only after some time, with the IAC value then sitting between 4 and 10. keep a timing gun and a thermometer gun with you during these first drives.. so you can find if and which spark plug is occasionally not working, and also find if one of the rotors is starting to flood (lower exhaust temperature).
Also always remember that the ecu stores new learned values into the map when you turn the key off, after 30sec. so, if the next time you turn on does something strange, maybe it had learned something "bad" and it's trying to correct. once you understand this you can use it to your advance, like a save button in a videogame: you try some idling for few minutes than turn off, 30 sec, and try again..
On the other hand let's say you had some spark problems and you don't want to store the values from your last run, just disconnect the battery right after turning off the engine.
... also are you using an OMP? do you have an holley manifold? are you experiencing the things that I show in the video?
I'm happy to see other fitech rx7, so we can finally compare settings.
BTW just to remain on topic, I'm arranging to do a dyno test on Saturday morning, to measure a baseline before some mod ideas that I'm planning to do.. of course I'll make a video for the occasion!
Thanks. I am not running OMP. I have the racing beat holley manifold. The car ran great...once. I did experience a "bog" at WOT but only momentarily. After my experience with the bucking on my way home from work i had to leave the car in a parking lot overnight until i could work on it. I seperated the tach signal wire the next morning and drove it back to work where it has remained for about a month. The idle is hunting badly and requires feathering of the throttle to keep it alive. Hoping to start tweaking again soon but i rebuilt the engine and tranny in my truck (also running fitech but with great sucess) which is my daily so i havent had the time to play with my rx7.
Thanks. I am not running OMP. I have the racing beat holley manifold. The car ran great...once. I did experience a "bog" at WOT but only momentarily. After my experience with the bucking on my way home from work i had to leave the car in a parking lot overnight until i could work on it. I seperated the tach signal wire the next morning and drove it back to work where it has remained for about a month. The idle is hunting badly and requires feathering of the throttle to keep it alive. Hoping to start tweaking again soon but i rebuilt the engine and tranny in my truck (also running fitech but with great sucess) which is my daily so i havent had the time to play with my rx7.
Here's the google drive link. https://drive.google.com/open?id=1Lp...MMr14GY6D2_2J3
t's just a backup of the entire memory, so, if you already have the upgraded software (198i) you just need the "backup 1" file in the cal_files directory, and dump that in to the ECU when you are connected to the car. do your own backup first!
so, yesterday I went to a repair shop to do a dyno pull, because I want to do some experiments in the future, and this is the "before" test:
I need to also check my ignition timings, and probably you can guess it from the noise, above like 6000rpm, but I think it will need a lot of attempts to get it perfect.
I'm waiting for an S5 intake manifold from ebay, let's see what happen next!
Awesome, i already have the 198i software now ill just have to get the handheld to talk to my computer. Did you have to download software for your handheld and laptop to communicate? I havent been able to get it to work with mine.
you don't need any software on the computer. when you connect it, choose usb storage and access to the SD card. first of all do a backup copy of the SD on your computer, then drop the .cal file I told you before in your cal folder of the SD.
when you go back to the car, select write cal file on ecu, and choose that one.
You can PM me if you want to continue this conversation, because I'm really curious to see other fitech rotaries!
So.. I publicly said before that "I didn't want to spend more money and time on an engine that runs well enough" but.. you know.. we're all RX7 owners here.. you know how that thing works.. so as you probably guessed from my last post, I ordered a complete S5 intake manifold, and I will try to adapt the fitech to it. the actuators are not working (the rotating assemblies are there, but not the actuators) and I'm planning to use electric ones, using RC servos or more probably, door lock actuators.
I ordered the Manifold on ebay, came from texas (bought other parts from him in the past, like the lower front protection for the engine bay) and didn't cost that much. Then I proceeded as my usual, to convert everything to a digital 3D model, to play around with it, and check the fitment in the engine bay.
I
I asked Fitech directly if they would provide a 3D model of the throttle body unit, but it will take some time to answer for them, so in the meantime I made a mock-up version just for fitment, and right now I'm in the process of engineering the flange to connect the two:
And I got immediately worried about the hood clearance because it seemed very tall, but from my calculations, I'm still clearing:
So, fortunately I work in a company where we have lots of sheet metal and laser cutters, so I will cut all the blocking plates, and I will fabricate the adapter flange. it will not have fancy shapes inside, just a cylindrical chamber.
I will mount the port valves back in and I will test the variable geometry intake, moving it manually, and from there I will decide what to do about the actuators: I think the cheapest way would be to just buy a cheap "shift light", disconnect the lamp, and connect it to a door lock actuator. In this way I can easily open and close the actuator above a certain RPM. The parts are so cheap that I can even use multiple of them, one for each actuator.
If you know other solutions of course let me know!
04-12-17, 02:13 PM
