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.
Mostly because the now welded intake is awkward to mount in the mill, but still.
Yeah, and warped all to hell welding it together. I've welded enough stuff that I know what happens and why. Good practice and all but, the flange will still need to be trued flat. I like the oring thing pure sweetness.
That's pretty close to what length I'm going to be aiming for as well. One of the 2nd Gen SCCA racers (23racer) has a hyabusa throttle body setup with his length around 14.5 inches total and he claims really good power everywhere and also the math for sound wave reflections leads me to around 14-15 inches for a torque peak around 6-7k RPM. Can't wait to see your results
Yeah, and warped all to hell welding it together. I've welded enough stuff that I know what happens and why. Good practice and all but, the flange will still need to be trued flat. I like the oring thing pure sweetness.
Yep, true. I mounted it as flat as I could, machined the mating face flat, and then added the o-ring grooves.
Originally Posted by Lavitzlegend
That's pretty close to what length I'm going to be aiming for as well. One of the 2nd Gen SCCA racers (23racer) has a hyabusa throttle body setup with his length around 14.5 inches total and he claims really good power everywhere and also the math for sound wave reflections leads me to around 14-15 inches for a torque peak around 6-7k RPM. Can't wait to see your results
Yeah, I saw that one aswell. Cant wait myself either :-).
So the car is coming along great! :-). Didnt get a lot of pictures but I sure got a lot of stuff done!
First thing I did is finish up the intake. There are velocity stacks in that filter with the correct length, the bike injectors are replaced with ID1000's which required some spacers, and I added some studs between the intake manifold and the filter baseplate so everything stays together.
Next I fitted the worst exhaust ever. It will get a nicely tuned exhaust when its running, but for now it's a stock cast exhaust manifold, then a 2.5" linking pipe into the 3.5" exhaust that came off the 4-rotor.
Next some small stuff. Got the clutch to work, got the brakes to work. Got the shifter to work. Got the oil system filled / pressurized. Got the coolant system done and filled up. Fitted an electrical cooling fan instead of the mechanical one. Replaced fuel lines / filter and added a pressure regulator. Fitted / filled power steering system, new battery.
So at that time all of the mechanicals were pretty much done and I started on the wiring. Going through all the connectors in the engine bay, checking what I do and don't need. Removed the stock ecu and all the wiring that I don't need. The wiring for the engine, fuel pump and cooling fan is all new. Still need to crimp some connectors here and there but it's getting there. Also need to replace the fuel pump because the stock one is seized stuck. Probably stick a bosch 044 in there.
So yeah, super excited that its almosy ready to fire up! :-).
What's the ignition setup going to look like? Good progress. Makes me feel bad for not getting my TII back together this weekend when I had a lot less to do than you have completed lol
So the car is up and running
Took me most of the day, because I had to hook up all of the connectors, fix the fuel pump (I had two lines reversed ), replace the started, and then troubleshoot the EFI system.
Had the leading and trailing coils switched, the CAS didn't work with the microsquirt (I'm using a microsquirt module, which needs a 51K resistor added for the 2nd VR input to work with the FC sensor), and I had the CAS itself mounted wrong. By the time I fixed all that the car was flooded pretty badly, had to pull-start it.
Started with dialing the megasquirt in, needs some more tuning but I'll get there. I'll see if I can get some pictures up
Not sure yet about that, although the large side ports and semi-pp's would be pretty cool :-).
Some pictures of today:
Still working on getting the kinks sorted. The transmission is very noisy, and there is oil coming from that area as well, so I'll have to see whats up.
Also still messing with the megasquirt. I had this wierd problem that it would go way rich and stall whenever I touched the gas, usually when trying to drive away at the traffic lights in rush hour. Turned out that there was a fault in one of the acceleration enrichment settings that caused every enrichment action to taper to 2ms, which is a ton of fuel with 1000cc injectors.
Another (obvious) issue is that the ITB's are very sensitive, 60mph is around 5-6% throttle or so. I already scaled the VE table so there is more resolution at these lower throttle openings, but it's still pretty difficult to get right. I actually wired in a MAP sensor, so I can try the megasquirt's ITB-mode, which uses the map sensor up to 90kpa manifold pressure, and switches over to TPS after that. I need to improve the MAP signal (it's still a bit jumpy) and get everything dialed in better before using it though.
Last edited by John Huijben; 05-18-17 at 05:29 PM.
How is your MAP sensor mounted? I think with ITBs people usually use a vacuum manifold hooked into all the ITBs to get a strong enough MAP signal to tune with. But in the end you might have to just give up and go fully alpha-N lol
The map sensor is currently hooked to the rear rotor runner with a long thin tube. A common vaccuum manifold would provide a cleaner map signal, but I would like to avoid connecting the intake runners together if I can.
The car is alpha-n for now, or well, it was before this happened:
Looks like the oil leak im having is coming from somewhere near the rear stationairy gear, so I pulled the engine to take a closer look. Took about 1.5 hours to remove.
I could be off my rocker here, but what about using the oil injector holes as a vacuum signal tee'd together with high temp hoses? Would there be a consistent enough vacuum signal there?
I could be off my rocker here, but what about using the oil injector holes as a vacuum signal tee'd together with high temp hoses? Would there be a consistent enough vacuum signal there?
I tried that on the 4-rotor, and it worked pretty well. For now I just hooked a vaccuum rail to both primairy runners, and go to the brakes, fpr and map sensor from there.
Also modified the breather hose, fixed the oil leak (The rear stat gear o-ring was bad) and added a few EGT sensors. planning to put the engine back into the car on friday!
Yeah, off the shelf. I pulled it off a faulty pneumatic valve unit at work ,
That 's right, that fuel rail is the one that came with the motorcycle ITB's. I had to put spacers under them to fit the ID1000's, but now it works fine.
So I got a lot done tonight , First I prepared the "new" transmission. The one that was in there worked, but it was pretty noisy, and I had another one, so in it went. Swapped the mount, sensors, and speedo drive over, changed the fluids, and fitted the clutch and transmission onto the engine. Then I fitted the engine in the car. All in all it took me about 3.5 hours.
Started right up, the way I have all the vaccuum stuff hooked up is better like this. MAP-signal is good. Took it around the block, and it drives ok. It is missing on power though, so I'll go over everything again tomorrow. Re-check and zero the ignition first, and I got another wideband sensor and controller, I think the current one is bad.
Last edited by John Huijben; 06-02-17 at 06:58 PM.
Checked ignition timing with the stock -5 and -20 marks again, which were good, so I thought maybe there is a mismatched pulley/hub set on there, so I set the engine at TDC (Using a transparent plastic hose hooked up to the leading plug hole with some premix fluid), marked TDC, and checked that, which was spot on aswell. So the ignition timing is good. Was running 16deg BTDC max, Increasing it to 26 which helped a bit but not that much.
Biggest issue is that I can't get a wideband lambda to work. I was using a bosch LSU with an LC-1, and it read rich all the time, so I swapped the sensor and the controller for a new 14point7 spartan 2 one, and it does the same thing. It reads the richest it can (10:1) all the time, idle, throttle, rpm or no rpm, it doesn't matter. Only on fuel cut the AFR starts reading some random number.
When I try to lean the tune out it dies before the AFR's even begin to go up. The spartan 2 has this output sequencer to check if the megasquirt sees what the lambda controller is putting out, and that's spot on. The sensor is new and came with the controller so that's not it either. I don't think there's something mechanically wrong, when I manually adjust the VE table I can get the car to run nice and smooth.
Maybe I need to make a tubular header for it instead of the cast one, and see if hooking the sensor up to either the front or the rear rotor makes a difference
Just from reading around, it looks like the PP engines and semi-p engines do not like any back pressure and it's very difficult to tune by AFR because of the intake reversion and the idle misfires. You may need a header to split up the exhaust pulses to keep some of the reversion down and tune AFR to make the engine run as well as possible. The raw fuel and air could be tricking the gauge.
Split header with 2 EGT probes would be pretty beneficial for tuning that engine I would think. I'm super impressed you did all that work you described in 3.5 hours though. Seems like I spend that much time just getting all the tools out I need for any small task...
Yes, I agree with all of that :-), Best thing to do probably is to put the car on the dyno and adjust the ve table so it makes the most power. But before that I want to change a few things. Currently working on this:
I'm planning to change the ecu from the microsquirt to an ms3x, so I soldered an v3 board together, and its up and running on the bench :-). This allows stuff like sequential injection (microsquirt does split timing or sequential injection, but not both), dual o2 and egt sensors, onboard logging and lots of other stuff.
The second thing I want to change is the exhaust header. Already machined the flange and Ive got pipes and bends coming in tomorrow :-).
I was wondering how long it would take for project creep to set in You know you're supposed to be spending this extra time and money on the 4-rotor, right???
Yeah... I know, but when I get the 4-rotor ready again, I want to run it using this MS3, so already assembling the ecu, mounting it in the car, and getting everything calibrated and all the kinks sorted will help a lot later. I'll make the wiring loom so it fits on the 4-rotor aswell.
Little bit of progress on the 2-rotor. I got the car tech inspected and it passed so it's street legal now, so I took it for a drive. At first I wasn't really getting anywhere, couldn't really tell where it was rich, or where it was lean, no AFR reading whatsoever. Manually adjusting the VE table while driving really is a nightmare. After I while I tried swapping the injector plugs. The ITB's have 4 injectors fitted, but I was only running the 2 center ones, that go to the semi PP and the primaries. (I have ID1000's fitted, so one injector per rotor is enough). So instead I hooked up the outer 2 ones that feed the secondary ports, and it did made a difference! Not night and day, but it was idling and driving better. After a bit of more fiddling I got it to drive ok, so I took the car on the freeway, so that I could keep the RPM and load more constant, and after a few miles I started getting an AFR reading! I think the tune was just so far off before that it kept fouling the wideband sensor or something. After a while I was even able to use the ecu's autotune function to adjust the VE table. With the ITB's and porting it is a bit picky, but after 80km's or so I got the map reasonably dialed in up untill 6000 rpm or so. AFR is around 1:13 everywhere. Still got some more tuning to do, and the power output isn't where it should be right now, but I think the cast header is mostly to blame for that. Also the IAT's are too high, around 130F all the time. So I really need to do the header, and maybe add some ducting for fresh air or something.
I have been watching your build for a long time. Some interesting stuff on your current project. I am also interested about your ECU choices. I have just got my 20b FD on the road and it has a Microtech LTX-12 fitted which I am currently learning the limits of. I have been told many times to ditch if for something better. Megasquirt, ms3 and haltech have all been suggested. I want to get good value for functionality and ease of use. Megasquirt looks good value but you ate swapping out so I wander why? Would love some feedback. Not sure I would be able to build my own ECU. But I gather you can get them already built.
anyway.. Keep up the good work.. Regarding your air temp that sounds really high? But I guess you don't have an intercooler so it is drawing engine bay heat straight in? I assume you have checked sensor is reading right?
Yup, I would make the airbox with hose to the front area as well installing the header before finalizing the tuning as I have found you have to tune it quite rich for when the AIT is low like after pulling off the freeway since rotary can run so much leaner/smoother with the high AITs.
Once it gets cooler air the leaner tune that is currently on it will cause surging idle in my experience. Once you add a bunch of fuel it will just have a higher idle rpm in these lower AIT situations.
Congratulations on getting the car street legal again.
I have been watching your build for a long time. Some interesting stuff on your current project. I am also interested about your ECU choices. I have just got my 20b FD on the road and it has a Microtech LTX-12 fitted which I am currently learning the limits of. I have been told many times to ditch if for something better. Megasquirt, ms3 and haltech have all been suggested. I want to get good value for functionality and ease of use. Megasquirt looks good value but you ate swapping out so I wander why? Would love some feedback. Not sure I would be able to build my own ECU. But I gather you can get them already built.
anyway.. Keep up the good work.. Regarding your air temp that sounds really high? But I guess you don't have an intercooler so it is drawing engine bay heat straight in? I assume you have checked sensor is reading right?
I'm pretty biased to megasquirts when it comes to ecu's because I've installed quite a few of them. The tuning and logging software is good, and there is a lot of information about it online, so a lot of the problems can be solved by googling.
Im swapping from a megasquirt 2 to a megasquirt 3 because its much more advanced. The ms2 can run a 2-rotor, but without sequential injection (if you use split timing), limited spare in or outputs, no onboard logging ect, so its basic. I got the diy kit (v3.0 board), but you can get an assembled unit aswell, for a 20b you need the expansion board, so its the assembled ms3x v3.57 one you want. If you want a nicer housing with better connectors and even more stuff you can get a ms3pro, but it is pricey.
Originally Posted by BLUE TII
Nice progress!
I don't miss that street tuning!
Yup, I would make the airbox with hose to the front area as well installing the header before finalizing the tuning as I have found you have to tune it quite rich for when the AIT is low like after pulling off the freeway since rotary can run so much leaner/smoother with the high AITs.
Once it gets cooler air the leaner tune that is currently on it will cause surging idle in my experience. Once you add a bunch of fuel it will just have a higher idle rpm in these lower AIT situations.
Congratulations on getting the car street legal again.
I found that aswell. during warmup, when the engine reaches normal temp it still runs lean, untill the entire intake is all heak soaked and hot, and than its fine again. I'll get it sorted. started on the header tonight:
Not perfectly happy with fitment yet, so will need to modify it tomorrow. Ive added a flange near the rear diff where the single pipe splits for the 2 mufflers, and I want to align that flange with the square one on the header so that everything in between them is straight. that way I can modify stuff more easily if needed, as I have no idea what length my primaries need to be. All the info about that is confusing, even in the sae papers. For starters Im trying around 50 inch primaries, into a collector, into a 3" premuffler, into the y-pipe and dual 2.5" mufflers.
Just need to heat-wrap it and put it on the car :-).
Right now it's 43mm ID primaries, about 1100mm long, then a 30degree collector that taperd down to 56mm, then a cone to 76mm (3"), a premuffler, then the pipe splits into 2x 63.5 (2.5"), into 2 huge mufflers.
Not really sure if the header length is right. Its one of those things where the more you try and search for info about, the less you end up knowing. I added a few flanges so I can try different lengths more easily.
04-21-17, 05:47 PM
), replace the started, and then troubleshoot the EFI system.
I don't think there's something mechanically wrong, when I manually adjust the VE table I can get the car to run nice and smooth.
