CAS problems not showing RPM while cranking but show if turned by hand
 

Alright this is getting frustrating here, this is what going on:

With the CAS in the engine and trying to crank now that I've swapped my dual shielded wiring in for it I get basically no RPM showing. Or if I do I now get about 30rpm ish. And nothing obviously fires up.

With the CAS out and me spinning it by hand, I can spin up to about 2000 rpm (shows on Megatune), I get pulse width to the coils. I can hear the injectors ticking, so basically showing functioning operation. So the wiring is all good, and the Megasquirt can see everything fine.

So what's happening when I try to crank it? Before I swapped the new shielded wiring I was seeing about 300rpm spikes, but from seeing things now with the shielded wiring, those spikes I think were just noise in the wire from something. And without the noise its showing next to nothing (atleast the shielding is working). Anyways if you guys have any ideas let me know.

Are you running second trigger ?

If so, it sounds like you probably have noise on the second trigger signal, which is common on rx7's. You can get rid of it by putting a .01 uF cap across the G+ and G- wires where they attach to your second trigger conditioner.

Ken

Yes I'm running second trigger signal. I thought I had it in there ... but maybe I overlooked it. I'll double check tonight again or tomorrow sometime. If i don't have it in there and it works ... I'll give you a million forum cookies :D:D:D.

I don't recall seeing a 0.01uF capacitor on your board at G+/-.

Also, remember that the shield should be grounded at the ECU end only.

The shield is only grounded at the ecu end. You should see the wiring now Aaron .. its worth a trip down just for that ;) lol

Alright I tried placing a .01 cap (didn't solder on on the board but I tossed it inbetween the 2 connections on the DB37 cable at the megaquirt. The result is now I don't even see one single RPM on Megatune ... so I take it then that the occasional 20/30rpm that I saw before were just noise from something then. When I spun it by hand it was still functioning fine. Any other clues?

Just wanted to add that the megasquirt is grounded thourgh the negative battery terminal and the positive right off there too (THis is temperarily till I get the relays in there for it).

I just wanted to add more to the frustration, if I take the cas out, crank the car, spin the cas by hand it still shows normal operation also, so its not like its noise from the starter or anything disturbing it. So the only thing changed literally is the CAS being out of it spot in the engine ...

Yeah, it's something I've been adding more recently. I have not even added it permenantly to the board design yet.

Ken

Strange, try grounding the MS to the engine.

Ken

Ya really strange ... I'l try doing that and i'll report back

Have you tried adjusting the cas pickups closer to the wheels?

Why would I need to move the pick ups closer if it picks up a signal already? For no reason I tried taking resistance while it was in and out of the engine to see if there are any variances but its all identical. So its no like its shorting or something.

I haven't tried grounding the MS to the engine as it started snowing here but when I'm free next I'll do so. Do you want me to ground the whole thing to it? Or could I just add another ground to the engine.

Now I might be wrong but what if the Megatune only picks up a signal after spinning the CAS up to a certain speed. Would that make sense? Would that be possible? Its only me trying to do everything but I'm not sure what rpm the injectors start to fire up.

Did someone else write this? It's worked for others.But you are farther along then I am with mine so I would listen to what muythaibxr suggests. Grounding correctly is important(like to the engine) as muythaibxr constantly stresses,so that sounds like a good place to start.

When I'm saying I'm not getting any rpm in that post is when the CAS is being spun by the engine. When it's lifted from the engine and I turn the gear by hand it shows RPM hence me saying "When it was spun by hand it was still functioning fine". Thats where the porblem lies, why would it work with it out side the engine but not inside.

I'm guessing because you are spinning it faster by hand and getting a stronger VR signal then when cranking.

I'm saying ground it to the engine b/c the body of the CAS is touching the engine.

The factory also grounds the stock ECU to the engine, not the battery, I believe so that anything that uses the engine as a ground has the same ground reference as the ECU.

IF you ground the MS to the battery, in some cases if some of the wiring isn't that great, you can create a ground loop.

Ken

Alright thanks for the info. Now I have to repull my ground from the inside of the car to where it was in the engine bay. Still snowing here so still have to wait to try it out.

alright I tried going back at this today with no luck again. I tried hooking the ground to the engine with some results. atleast now I'm getting rpm showing while cranking with a peak of about 350rpms, but most float around 150ish -200. I went in the start up/idle setting changed the cranking rpm to 150rpm instead of the 300 rpm but still no luck. Also I'm still not sure if those are actual rpms or just spikes neither. The reason that I'm saying that because I'll see the needle move up there but still no injectors firing and not showing rpm tach on megatune move like it would on the cars dash.

Now just to make sure as I haven't played around with the software much, when I wrote that 150rpm for start up, I clicked burn to ECU, then I closed it. Thats all I would need to do correct?

Also playing around with it some more, when i was turning the CAS by hand and cranking it, it finally chugged as in it seemed like it sparked in atleast one chamber trying to turn it over. So there's hope but seems like I'm still far from it. I really don't know what else to try.

I am having a similar problem with one the car I am working on. I am going to check all ground on the car. I will let you know how I make out, going to try tomorrow.

I have tried grounding at the engine, at the battery, on the chassis, so really unsure. What is yours doing exactly?

the exact same has yours. I removed everything and tried it on the bench and it worked perfect. So I have no idea what it could be. But I going to take everything out and start over. I used the same setup in my other car for 2 years with no problem. Remove and place it this car and can't get it to work. So it must be a grounding or power issue with this car.

The other one that I might try doing is shielding the power cable as Muythaibxr has suggested previously.

I had the exact same problem! I had actually soldered my VR pots in back to front and was limiting the signal with them turned anti clockwise. Everytime I turned it by hand it would register but the engine couldn't crank fast enough to get a signal through. I just mucked around with the pots until I had a brain wave to turn them back all the way the opposite direction. Then I got spark.

is that both pots on the Megasquirt or just the one?

I had both pots around the wrong way. However I don't know if it is whats causing my idle problems. I have heaps of noise at idle and the tach is jumping from 800 to 8000 rpm. I just assumed the pots are a variable resistor and it doesn't matter which way they are positioned, it just changes the direction required to increase or decrease it. Am I correct in assuming this?

I'm not sure...

It would certainly rule them out if you put them in the right direction, and then turned them fully counter clockwise.

Ken

so i made sure that the pots are turned counter clock wise. MAx value that I see out of them is about 550ish ohms does that sound about right? Also I just noted something, I have one red light in the bottom right that says not Synced? Anyclue what that could be?

I lied, the not synched light only show up when voltage are lower than normal. Once I had the battery charger on it again it was showing synched when trying to crank it. So it shows synched, says ready, and WUE all in green, but doesn't want to highlight the cranking one when I'm doing it still. :( ...

And RPM shows up?

Does RPM show up and then not change while cranking?

That is also indicitive of noise.

I've been able to start a car in that condition by adjusting the noise filter (ms2/extra 1.0.2).

Ken

Hey,

Ya it seems to be doing that now. Show's one pulse of rpm and holds there. I tried turning the noise filter on in the megasquirt. Not sure how exactly it works but once I turned it on it wouldn't show any RPM's at all, so I switched back to having it off.

Taking your word on noise through the power wire can cause some problems, I have relocated the power wire and now it seems like that aslong as I have a good charge on the battery the Cranking light will come on, so atleast its a step forward in the right direction.

Another thing that I have done is relocated the ground from that is now going to the engine to a different location, that isn't as close to the alternator if that would of made a difference.

I'm going to try to relocate a few wires here an there to maybe try to get as far away from noise signals. I really wish I had an ossiloscope at home to scope things out, would of probably made a world of difference.

Any other things keep them coming.

Thanks

Come to think of it, I don't think I've seen your msq... Could you post that? I don't want to cause you to do a lot of extra work if it ends up being a bad setting causing your problems.

The Noise filter should be set at around 500 usec at crank down to about 10-20 usec at redline... You should decrease linearly from the crank value to the redline value.

Ken

Hey,

By seen your MSQ, what do you mean by that? Is that the saved file from Megatune?If so I can post one up when ever. My settings are set like H4inf or atleast for the info he has posted on there.

As for the noise filter I will try to do so. I didn't take a close look at it on how to adjust the noise level for different rpms but I will try to figure it out. I just saw it and thought I'd give it a shot by just turning it on.

Thanks for all the support!

Sam

Well I'm alot happier right now :D, I did your trick of the noise filter in Megatune and set it up from 500-8500rpm to go from 500-14.7usec. Also I did tried to shorten the CAS signal cables as much as possible at the same time and redid the power wire location again. Now atleast the RPM is showing real, real-time RPM, and fluctuates with the engine pulses and goes down when the battery starts to die out at bit. It fired up for a few second, but its running pig rich. Need to adjust the table better. And also I think that my CAS angle isn't proper neither right now so that doesn't help neither. I need to find my bulb holder for my timing light.

I'll let you know if I get anymore problems. Thanks for all the support though couldn't of done it without you guys.

I can't emphasize enough to make sure your grounds are in good shape as well.

I just cleaned up all the grounds on my rx7 today, and now I don't even need the noise filter... it just fires right up no problem.

I strongly recommend looking at the FAQ for the link to Aaron Cake's ground cleanup HOWTO, and using it. That's what I did and as I've told everyone with this problem, the cause is almost always bad grounds.

Ken

I have not fully reworked my grounding system, but I do understand that and I think that Aaron can voich for that one. Me installing a 0 guage wire from the battery to the chassis :D. I also plan of hooking one up straight to the engine bay area with a 4 guage or so and rebranch it from there in multiple locations.

All that isn't necessary unless you're using your own harness completely from scratch.

I just cleaned up the stock grounds on my GXL.

Ken

can i know more about this noise filter in the megatune? i have been fighting a drop in signal at 6200rpm and cant work it out with the dashpot. any help would be great

it's not a feature of megatune, but a feature of the ms2/extra firmware.

Basically you set it up so that at specific RPMs the firmware won't "accept" input pulses shorter than the value you set. most "noise" pulses are very short, so if you throw out all the short ones, and keep only the real ones, then you end up with a much smoother-running engine.

Most engines shouldn't need this, but rx7's with stock harnesses are problem-prone in this area compared with other cars I've worked on.

Ken

The weather was finally right to retrieve the file so here it is:

I would like to point out that a not uncommon source of a no-start in distributored 2.5 Chevy trucks is the distributor body corroding so badly that it is no longer grounded to the engine block.

When that happens, crank signal is lost. Adding a ground wire between the distributor body and the engine block fixes it every time.

I believe Ken may have picked up on something here. The CAS only gives a signal to the MS when it is not touching the engine.

I wonder if the wires to the CAS got crossed and something that shouldn't be grounded is, and something that should be grounded isn't, so everything works fine when the CAS is floating free, but when the CAS body is grounded to the engine, the signal disappears in a direct short.

Just a hunch.

In most of the cases where the engine won't start, and works fine turning the CAS by hand but not while cranking in the engine, a push-start fires up the car every time... so I'm not sure that the CAS touching the engine or not makes the large difference.

As I was saying before, the largest difference I've been able to see was just from cleaning up all the stock ground points, possibly using new cables when necessary. If that didn't work, then usually replacing the starter helped.

Ken

Ken,

Would you be able to double check my msq?

possibly today... If not then Wed I'll look at it.

Ken

I took a quick look at your MSQ. Everything looks good at first glance. Can you do a datalog of your startup-stall event so we can get a look at what's actually happening and what kind of vacuum the engine draws?

I'll take a look as soon as possible as well. The more eyes we have looking at this the better.

A datalog as Aaron asks for would also be helpful.

Ken

Ken,

Curious if you had a chance to look over it yet?

Thanks

Not yet, but will look today. What specific problem are you having now... looking back at your last posts it looks like you had it working?

Did you mean to run Alpha-n with a secondary fuel table and no ignition load parameter?

Normally you'd want speed density or % baro for the primary load with no secondary fuel load.

You also didn't have a primary ignition load set. That should also normally be set to % baro or speed density.

You'll probably want to set your MAP, TPS, and RPM lag factors higher to get better response.

Your rev limiter is set pretty low (set to the defaults it looks like).

Your Ignition Input Capture setting is going to depend on whether you have VROUTINV wired to TSEL or VROUT wired to TSEL. If you have VROUTINV wired to TSEL you'll want falling edge, otherwise Rising edge. The MS build instructions on www.megasquirt.info say to use VROUT to TSEL with rising edge.

Your cranking RPM is set WAY too low. That should be set to 400 or so (you have it on 100).

You have staged injection turned off, so you're probably squirting fuel on primares and secondaries, causing the engine to almost flood. This is assuming of course a 2nd gen 13b that actually uses staged injection.

That's all I've found so far.

Ken

Engine is a 12a with ITB's and only running one injector per rotor (680cc). I'll answer what I know which I know isn't much.

The Cranking RPM is set at a 100 because it whenever I crank it it only normally shows about 250ish max. So 400 wasn't going to start it up. Is there any disadvantages to getting it that much lower?


Where is the lag factor changed from?

I'll try to get the Ignition load set. What kind of percentages are we talking about when compared to baro?

Rev limiter I'll change as soon as I can get the car to idle atleast ... that's just fiddles :D


For my ignition input capture I believe that is setup right. I have it setup like H4inf's setup.

The first 2 question not really much sure about that ... Maybe Aaron Cake can pitch in on that.

Thanks alot for your time Ken Really appreciated.

Sam

That explains it then. I wasn't sure how you were doing things there.

yeah, when you set it lower than the rpm you see while cranking, you never use any of the cranking settings. You have to set it slightly *higher* than the actual RPM you see while cranking. I use 400 most of the time for rotaries. Having it set to 100 is going to make it harder to start, if it'll start at all.

The lag factors are changed from the menu that says "General, lags"

I'm not sure I understand the question. % baro is a type of load calculation. Instead of just using the absolute manifold pressure, it uses ((current MAP * 100) / baro to give you a percentage of barometric pressure for load. This is good for people who drive up and down mountains on a regular basis.

Alright, that should be ok then. If you have it backwards it can result in all kinds of odd problems that's all.

No problem. Hopefully some of that helps.

I still don't know what problem you're having though.. is it starting up then dying?

Most likely to get it to stay running, you'll need to tinker with req_fuel until it'll stay running, then reset your req_fuel, and adjust the whole fuel table higher. Then once it idles, you can work on tuning. You'll definitely want to turn off the secondary fuel load stuff until you get it running well with just one table though.

Ken