Thanks Troclo. Should be sorted out soon enough. The video we took sucks, car was so loud up against the garage door that it distorts the audio. Going to post something quality up once its rolling around.

 

Much respect on getting that 20b up and going you've done a hell of a build! I wish I could be brave enough maybe one day.

 

Thanks alot man. Just go for it, as annoying as some of the problems and lack of knowledge in the process seemed, it's been alot of fun learning which is the whole point in the end.

 

Do you have any videos of your drag car? I wanna check it out

 

Towed it to an empty parking lot tonight. Very hard to start this time. Pulled the plugs and they were wet with mixture of oil and gas. The plugs were cleaned and we turned the motor over with plugs out. Reinserted plugs and then tried to start again, no go, even with full cranking speed and alternating on/off fuel. So rather then oil down the runners, we tried to pull start the car with the truck. Slowly backed off the clutch around 10mph and it fired right up. Its using a ~650CCA battery, maybe it needs 1000CCA plus.

We kept it running for around 15mins. Timing remained locked at -5 and idle around 2500rpm-3000rpm.

Tried to restart again but this required pull starting. My friend moved it around the parking lot on extremely light throttle, although we forgot the timing lock was on. Ended up shutting it down to fix leaks/confirm temps. Slowly coming along

 

try again with fresh plugs before you change too much

am wondering if the adaptronic is dropping back to duty dwell and negative trigger during cranking
and this would be incompatible with some coils and explain the reported hard starts

 

I'm pretty sure the adaptronic is sending the correct signals to the coils. It sparks rather brightly/loudly with the car off and manually turning the CAS. The in-built oscilliscope shows the pulse signals being sent and the incoming signal from the CAS. I will double check for spark during actual cranking with the starter by removing a plug or two. Andy has confirmed that he is using LS2 coils on his personal car with similar settings.

I'm wondering if some of this may be a ground issue. For some reason, this car drains the **** out of batteries when cranking. It threw the alternator light on a v6 truck. When connecting jumper cables, attaching the negative to the alternator bracket rather than the chassis makes a remarkable difference in cranking speed. I'm going to upgrade the battery and run a negative ground wire from the battery all the way to the starter and/or power plant frame.

Thanks for your help. I'm sure it will get there, it just needs to be tamed and broken in lol

 

Hi guys, just to clarify, our ECUs don't switch between "duty" and "dwell". They always give out a fixed duration pulse (what you might call "dwell mode").

The Select ECUs have the option (which we use as the default) to fire the ignition outputs on the last trigger pulse before TDC during cranking. This means that we don't need a stable RPM, we just need to know where TDC is. They also have the option to "time" the ignition pulse as if the engine was running normally, but for that you need a stable RPM at cranking so it cranks for a little longer.

The 1280 ECU (which is on this 20B) doesn't have that option, it always "times" the ignition based on the angle that you're trying to fire it at, but if we get another trigger after we should have fired it but we haven't fired the output yet, then it gives a pulse of the dwell time. Eg if you have a trigger tooth at TDC, and you're trying to fire at 5 degrees before, and we get to TDC and haven't fired the ignition output, then the ECU will output a pulse on that ignition output for the requested dwell time. This only happens on the initial part of the crank though while the RPM is stabilising.

It's easy to see what's going on though using the built-in scope (it can watch the trigger inputs, injection outputs and ignition outputs, current crank angle as well as any variable calculated in the ECU), just remember that the timebase will need to be set to about 10 or 20 ms/div because the dwell is so short compared to the overall cycle at cranking speeds.

Thanks
Andy

 

There was nothing wrong with the compression of that engine when it left our shop, and certainly nothing that would prevent it from starting right up with a proper map.

Sounds to me like your timing is way off, either the marks you put on the FD trigger wheel are in the wrong locations or the timing map is not configured properly. I know that the base map I was sent was using 20* of split in every cell, I would not bother trying to run the car with this map.

 

Thanks for the input Chris. I aligned the motor to the leading mark on the 20b pulley and dropped in the CAS verifying the correct position with the cap off. I then removed the 20b pulley, installed the FD pulley and then marked the FD right below the pin.


I'm glad to to know the engine has good compression. Surely we will get the kinks worked out soon. Even though I've only heard it idle a few times, it surely sounds great.

 

You are aware that the Cosmo pulley mounts opposite of the FD, correct? Meaning it mounts over the hub, whereas the FD sticks out from the hub. If you mounted it on the hub the same as the FD pulley sits the timing would be off substantially.

Now if you did everything correctly, it is the timing map. To tell if you have the same one I was sent, go to the "Page 22 spark Split screen" if it is 20* everywhere, then it is the same, and I would recommend you stop trying to run the car.

Break out the compression tester and verify for yourself, so that you have a baseline.

I do have some constructive criticism. From the pics you have posted I noticed a LIM bolt missing and a block off plate bolt missing.



Also you should take a die grinder and remove all the excess material under the BOV, there is no reason to restrict the airflow down to a 1/2 hole when you have a 25mm flange.

 

Yes, I'm aware of that. I mounted it facing the inwards, the correct way.

I do not suspect compression is the issue if you are confident it should have enough to crank on its on.

Yes, the rotary engine split timing map reads 20* everywhere. The main timing map looks relatively similar to my AEM map from my 13b.

Why would I stop trying to run the car? I was going to change the split to 5-8* at idle working up to 10 at 0psi and 12-15 above that. Would split timing really affect basic idling and driving to that degree? I mean 20* means the trailing just fires 20 degrees behind right?

I also believe it may be running lean. I will add fuel to the lower part of the map.

The bolts have been replaced, but thanks for point it out. Should the block off plate have a gasket behind it?

Ill grind the BOV hole at some point thanks

 

Having dripping wet plugs is not an indicator of running lean.

RTV the BOP.

 

Yea the 1280 Looks Awesome.... I'm definitely going to be running one on my next build.

 

Well, things changed with the charge piping attached last night. AFR was around 14.0 holding the idle at 2.5k. Going to adjust the VE table and reduce the split timing and try again.

 

I dont have any good videos everytime we took good passes we got excited and never thought about recording. Soon next year though

I would put my focus more on the ve table if your getting those plugs with that kind of results. Its being pita with the amount of fuel thats going in the chambers.

Maybe a steady idle will be what im looking for 1st then start working toward the ve table. I would suggest if there a mix trim option to utliuze that to see what the cars asking and adjust accordingly. Split timing on 13b or 20b stock would be 15* not 20*@ idle.

 

Monsterbox your issue is exact to a fd i recently built a customer.
Andy and i worked for many weeks trying to figure it all out. I did manage to finally get the car to start on its own now but it still does not start the way i want it to.
Fwiw I adjusted the crank enrichment table to finally get it to start on ita own. It still cranks with too much fuel it seems but it wont get any better no matter how much its adjusted.
We are now working with Elliot at turblown also. We are going to put all new ev14 style injectors in and in hopes that the old injectors are the cause. Im wondering if older injectors are harder for the adaptronic to control or run on a different type of pulse or frequency.
Keep us updated

 

I had similar issues with my car when I first got it. Sometime It'd leave me stranded bc it would just crank and crank and crank without firing. I basically tuned the starting enrichment map until it would fire and flood. Then dialed it back down.

 

Hey guys, thanks again for the input. Whether its cranking fuel, older style injectors, or anything fuel related, I don't see how that relates to zero popping or sounds of life w carb cleaner etc

I checked for spark by hand by spinningCAS. There is output to all 6 plugs. However, if I spin slowly, injectors click but inconsistent / absent spark and no scope signals on ecu. If I spin fast I instantly get signals on ecu and spark fires more frequent/consistent. So, that tells me that the CAS must hit a minimum rpm on this ecu to fire the plugs.

I then plugged CAS back in and cranked car w timing light. I didn't see any spark. Also tried removing leading plug and grounding it, no spark this time during cranking.

So, even though it sounds to me like good cranking speed, I'm starting to believe its simply not turning fast enough for this ecu.

My friends s2k does the same. Stock ecu fired right up. Haltech unit cranks and cranks unless you have the perfect speed. As soon as you cross threashold it fires right up.

Monday ill install an Optima 1000ca and a direct ground from battery to block from the trunk and try to increase the cranking speed to rule out this issue.

Thanks

Zach

 

if i am reading it right then i think andy is trying to tell us the 1280 will need to see a stable Ne tooth signal

try closing up the CAS gaps to the low end of spec ( 0.2mm to 0.6 mm )
and see if you can adjust the filters down a bit

beefing the battery and the heavy earth cable to get consistent crank speed may also help a bit

the only other thing i can offer is maybe picking a different trigger tooth number and altering the angle vector to suit ( to effect same timing position )
it was a quirk i have seen before on some other early ecus and may be worthy of a try

be aware you have to ensure a minimum window angle number that is larger than the total timing advance the engine sees

ie.. do not choose a trigger angle less than 45 degrees

PS
it is common for many ecus ( not all ) while cranking to drop to duty dwell
.. hence the bad manners of the D585 coil cooking off,, even on rx8 ecus

and most of ecus drop to non sequential operation while cranking ( ie batch fire injectors and leading spark only )

on the haltech the rpm threshold alters the spot when the ecu reverts to sequential coil and injector operation
and its default setting was often too high
( hence the observation for many about the prolonged cranking )

some ecus also may use the home ( G rotor ) for trigger during cranking
.. or others may hold off for several revolutions before adding spark
a feature that is intended to avoid the notorious habit of some haltechs and microtechs
occasionally letting go a backfire or bump on key on , or kicking back on crank

 

Iv ruled it out now to weak CAS signal.

Installed Optima. 2ga positive from trunk to switch in dash out to 4ga to factory fuse box and out to the starter w factory harness.

2ga ground to spare tire well, 6ga ground to transmission. 2X 10ga grounds firewall to bell housing.

Cranking speed now sounds pretty good. Before the new battery, rpm signal on cranking read 20rpm. Now, it jumps from 20rpm to 200ish rpms in/out. This unreliable signal is surely the cause of all the issues.

I checked the wiring to the cas. No shorts or exposed wiring. CAS teeth clesrance visually looks super tight. Shielded wire running down UIM over the coils to factory plug where I have it spliced and soldered. Maybe its interferring w the coils? The wire is adaptronics shielded

Keep in mind, if I remove and spin cas fast it all works perfectly

 

sometimes you get a CAS that will not work at low crank sppeds when in the car,, but will when hand spun. out of it

usually that issue is a small short where the wires pass through the grommet on the CAS body

else it may just be that your filter settings are a little too high , i expect the boys will have a few tricks to increase the sensitivity

 

That is precisely the issue!

It was suggested that I install the FFE Digital Trigger as this ECU supports it and its a drop in replacement with greater accuracy and less scatter. I'd like to save the money and make this CAS or a replacement CAS work first, however, if it comes down to it I guess this would fix the issue? Digital CAS shouldn't matter the speed right?

 

if the trigger kit is a hall effect unit then the signal is cleaner at low rpm

if the trigger kit just uses reluctor senders like the FD, then there is only an improvement in accuracy ( no CAS slop deflection )
but will not increase the signal intensity while cranking

im thinking you need to check that grommet , and lower the filters or increase the gain

these settings should be there somewhere for you to play

but if you wish to increase accuracy for 500 + HP , by all means the FFE hall effect kit can only improve things

 

I might add i did put new crank angle sensors in the customers car i was working on as well as adjust those settings.
I never did mess with the sensitivity and it might be a option.
I also suggest going with a FFE hall kit.
I will be running a FFE kit on my 20b as well.