Jakesrail

1986 13-B N/A , MSII extra .well, here is what I have got since I last posted. i got the inovate LC1 wideband installed. I am running at around 11:1 AFR not the best, but not horrible. I am working on the AFR issues, but the big thing that gets me is the death at full throttle. I have this motor in a sand rail, and that being the case, alot of my driving is light on the peddle, and then mash on the gas to climb. I will be cruising along in first gear at around 2000RPMs, I will put the throttle to the floor, or atleast 75%, and the motor will rev to about 3000 RPMs, and then die. if I let off of the throttle, it will start running again as it idles down. I am wondering if it is in the map sensor. the other thing that is weird, is that every time I go to tuner studio, I have to set my 02 to the wideband LC1, it keeps resetting itself to the narrow band. and my temp calibrations are gone for the air temp, and water temp. I have a data log that shows what it is doing. I will post the data log, and my MSQ and see what you all think. thanks for any help you can give me.

and one last question. I have a temp gauge , but I need to tie it in to the motor. can I run my gauge on the same water temp sender for the MSII(the stock Mazda setup)? I want to make sure I dont interfere with the MSII.
thanks
Jake

noddleman

For the temp gauge, I took the coolant neck that runs off the front left of the motor towards to radiator, and cutomized it a little to allow the sensor to slid into the side of it. It's probably the most accurate reading of engine temp you can get. And for the full throttle thing, when you smash it does the engine want to bog down, or does it back fire really bad? In some cases, or mine usually, the "electric" OMP can send a signal to the ecu, and send the motor into "limp mode." Pretty much when this happens you cant go past 3000 rpm fpr the ecu senses that something is wrong, the engine will idle, but will not rev high. Best way to try this is to unplug your OMP, this will allow it to reset, then plug it back in. Then try reving it.

Aaron Cake

iTrader: (1)

At first glance your MSQ looks fine. Can you post a datalog of what happens?

The stock Mazda setup used two sensors, one for the ECU and one for the dash gauge. You should so the same. Tap another port into the water pump housing before the thermostat.

If I am picturing this correctly, this is probably the worst spot you can mount the sensor. The sensor should always be before the thermostat in a location where coolant is circulated. For example, on the front of the water pump housing in the middle of that little bung where the air pump bracket used to mount. Or on the water pump housing just under the alternator area.

The OP has a Megasquirt as stated in his post and as should be made clear by the fact he is posting in the Megasquirt forum. Therefore, no limp mode and no electronically controlled metering oil pump. And he mentioned his engine was an S4, which has a mechanically controlled metering oil pump anyway.

Jakesrail

thanks for taking a look at this Aaron. I did post a datalog, it is the first zip file just above the MSQ. you can see about a quarter of the way into it where the throttle hits 99% and the RPMs start a steep drop off. but once i let off of the throttle, it starts idling again. I noticed today, that it may be doing it worse when it is cold, or it could have been the fact that it was so cold that it was just having a hard time running (around 30 degrees). at first I thought it was a map sensor problem, but now it almost seems like a spark problem, I dont know, I am kind of stumped. I am not a tuner, so there is a pretty good learning curve to this. I also noticed today, that if I just slowly increes the throttle, it does it, so I really wonder if my spark table is off a little if I work the throttle right, and get it past the hesitation, I can run it right into the rev limiter (6500 RPM) with no problems, it is just that hesitation at 3000 that is giving me problems. and just so you know, the tables I am running are from muythaibxr, so at least those were created by someone with some experience.

Aaron Cake

iTrader: (1)

You posted a zipped shortcut to the datalog.

Jakesrail

Im an idiot, here is the datalog. I am hoping you guys catch something that makes sense. I am taking it out this weekend, and I am trying to get it fixed before then.

Aaron Cake

iTrader: (1)

Where in that datalog will I see the problem? I took a quick look but didn't spot anything immediate except unusually high injector PWs. Also I may have the incorrect wideband set up because all the ratios are in the 8s and 7s. What wideband are you using (which LC1?)?

Jakesrail

it does it at 128. I am using the standard lc-1, and I am using the default setting that it gives me in tunerstudio. if you see something that looks wrong, by all means, let me know, because it is most likely wrong. at this point, I am ready to pay someone to dyno tune it. it runs bad with a load on it. I am having a hard time finding someone that can tune these in utah, does anyone know of someone out here that knows what they are doing with megasquirt, and rotary? I am also posting the video of me trying to drive it at the dunes this last weekend, so if you listen, you can hear what it is doing. I am posting it to youtube, but it wont be on for an hour or so , so I will post the link tomorrow.

Aaron Cake

iTrader: (1)

Your pulsewidths go from 3.75 @ 125 to 4.492 @ 126 to 6.658 @ 127 to 4.453 @ 128.

That is a horrid amount of fuel and most certainly causing your problem. I think you should turn down your AE and see what happens. Set it as 100% TPS and then cut your values in half, then check the result.

I noticed that you are using duty cycle as a stage parameter. I have no experience with those settings but perhaps such high AE values are causing the primaries to go to 80%, thus staging the secondaries?

muythaibxr

As far as staging, no matter how you set it to stage, the total amount of fuel injected should match what would've been injected by the primaries.

As far as the best way to configure staging goes, my new table-based staging method is the easiest to get tuned for smoothness.

Ken

Jakesrail

ok, I will be bringing this back to the house this weekend, and I will look through all of that. in the mean time, here is the link to the video of my trying to run this out at the dunes. it may give you a little extra insight as to what it is doing.
http://www.youtube.com/watch?v=Nus3ruocQoE

elturbonitroso

check your a/f they should be in the 12.8 to 12.5 area at wot.

pmrobert

Are you sure your fuel delivery (pump/pressure/regulator/plumbing/filter) system is OK? Can you check fuel pressure while under load?

Jakesrail

i will see what i can come up with to check my fuel pressure. I have been at a bit of a stand still, but that should hopefully change this week. I had a bad radiator on my last trip, and it finished itself off by splitting along the plastic side tank. so I got the new radiator, and i just have to make a few modifications to the mounts, and this weekend, I will start tuning again.

Jakesrail

OK, I tried changing my AE settings to 100%TPS, that did nothing. does anyone know a good running AE configuration, or even the whole MSQ for a totally stock 1986 13b N/A?? I have another trip coming up, and willl be heading out to the dunes with this, and after last time, I want to shut up some of the people that have been giving me crap about running a rotary, and megasquirt. i just really need to get this thing running like it should I will post my datalog(I will pull the MSQ from the MSII tonight to make sure i post the right one). it is still the same problem. if i roll into the throttle real easy, it accelerates fine, if I hit the throttle just a little bit quicker, it bogs down. in the datalog, look between 140 and 175, I held the throttle open the whole time, it died (as you can see by the RPMs dropping) and with the throttle still held in the same place, once the RPMs got a bit lower, it fired back up and accelerated fine. PLEASE HELP ME!!!!!!!

Jakesrail

ok, here is another little thing that I cant figure out. it will sit and idle no problem, and then for about 2 seconds the pulse width will o from around 2.5 to around 4.5, the idle will drop like it is about to die, and the AFR will jump up to 21:1. then, it will idle just fine again forabout 15 seconds, and it does it again. I dont know why it is doing it, but also, I cant figure out why when the pulse width goes up in value, why my AFR pegs the meter at 21:1. it seems to me that it would drop down on the AFR you can see what it is doing in the datalog at 1180-1217 thanks to everyone for the help.
Jake

Jakesrail

at wide open throttle, it pegs the meter at 21, I just cant figure it out!

muythaibxr

You're losing sync a LOT when you get down towards idle... The lost sync count goes up a lot in your log, and you're showing several different lost sync reasons.

I believe this is also the cause of the stumble behavior you're seeing on idle and could be the cause of the full throttle issue you're seeing too (although I couldn't find anything in the log you sent on the full throttle part, so that could also be that you're just running WAY too much fuel and confusing the wideband). How did you do the wiring for your VR and grounds? You're getting several lost sync reason 11's which means that there's noise on the 2nd trigger. You're also getting lost sync reason 17 which usually means there's noise on the primary trigger.

The usual causes of this noise are:
1) Bad grounds. Did you ground the MS to the chassis or the engine? The correct way to ground is to ground the MS to the engine, then the engine to the chassis and the chassis to the battery. You can also ground the starter to the chassis.
2) Incorrect shielding on the VR wires. Did you shield them? Is the shield grounded on only 1 end?
3) Sensor gaps in the CAS are too wide or the CAS is faulty.
4) Incorrect VR conditioner setup. For the 2nd trigger, did you ground pin 14 on the lm1815? Did you put in a .01uF capacitor across G+ and G-? For the primary trigger on the v3 board, did you turn both pots fully counterclockwise?

As far as using MS vs any other standalone, really you'd have these issues there too depending on settings and wiring.

As far as settings go, here is what I've found:

1) Assuming you're running stock high impedance injectors or low impedance with a resistor pack, you should not run any PWM current limit (set it to 100%). This could also cause noise.

2) You're running closed loop fuel, so if you end up going too rich for the sensor to read, and then it reads lean, the O2 algorithm may cause issues by enriching when it should go lean. Also, you have the authority set to 70%. This is likely way too high. I would recommend no more than 10-15% if you're first tuning, and if that isn't enough, adjust your VE table by hand.

3) Your accel enrichment settings are WAY too high as far as the amount of enrichment added. When I was using regular accel enrichment, I was using 1, 2, 3, 4 ms with similar rates to the ones you're using. This could be the bogging you're seeing when you stomp on the throttle.

I didn't see anything wrong with your ignition settings, so the lost sync issues are likely to be caused by noise.

Ken

Jakesrail

1. the MS is grounded to the engine, the engine is grounded to the chassis, and battery.
2.I used the stock shielded cable, and it is grounded on the MS end only.
3. the CAS looks perfect.
4.here is the circuit diagram http://stuff.pw.cxMegasquirt%20MS2%2...azda%20CAS.pdf
(sadly, I didn't know about this site until after I had built the MSII) I think I placed a cap between G+ and G-, but I will check.

1. running low imp with resistor pack, PWM current limit is set to 100%
2. yes, I have been changing it back and forth, I reset it to 0 for tuning.
3. you kinda lost me on the enrichment, are you talking about the rates on the chart? do you have some specific values to put in the AE settings, or maybe a screen shot of some good AE settings. I am tuning from scratch, and I have no idea what I am doing, so if i can get some decent values, I can start playing with them a little and start figuring out what I am doing. thanks

Jakesrail

not sure what happened to the link, i will try that again
http://stuff.pw.cx/Megasquirt%20MS2%...azda%20CAS.pdf

muythaibxr

It looks like pin 14 on the lm1815 is not grounded in that pdf.

I would suggest grounding that pin and then taking another datalog. In my experience that can help a huge amount.

When I looked at your msq, the PWM current limit was set to 75, which will not work. It *should* be set at 100 though.

Ken

Jakesrail

just to be sure that i am in the same menu, i clicked on basic set up, injector characteristics, and PWM current limit % ??? I am not sure why mine would show 100 and you are seeing 75, unless i screwed something up somewhere, and then fixed it after I sent the MSQ. I will ground pin 14, and see what happens. thanks
Jake

Jakesrail

OK, I pulled the ECU out, and grounded pin 14, no more losing sync, it idles pretty smooth now, so that is one problem fixed. I reloaded the MSQ that I got from you (muythaibxr) several months back, so i knew I was starting with something that should at least be close. I havent had a chance to mess with it much, but now it is a little harder to start, it backfires a few times before it fires up, and I have to work the throttle a little to get it to idle up to 1000RPMs. if I dont work the throttle, it stays around 400-500RPMs, and then dies. I cut the acceleration enrich settings in half like you had mentioned in the past. after it warmed up, I took it for a couple of laps up and down the road. if I eased into the throttle(in first gear), it would stutter a little, but not completely die. I noticed that in second gear, some times, I could hit the throttle fast, and it would respond great, but other times, it wold bog down pretty good. I will do some more tuning tomorrow, if you have any suggestions, let me know. I will post my current msq with the accel enrich setting changes.

muythaibxr

The backfiring thing makes me think the timing might not be right.

What kind of AFR was it running when it wanted to idle at low speed?

Ken

Jakesrail

just before you posted, I found that in the tune you gave me, for smoe reason, the ignition settings were rising edge, and 0 degree offset, I just set it to 60 degree offset, and falling edge, and it runs a lot better. it still has a bit of a hesitation, but it almost feels like the timing is off, but I am no expert. one thing that I was thinking about is that this tune works fine for you, but it just doesn't respond that well with my setup. could it possibly be because I am running this in a sand rail that weighs about 1200-1400 pounds? is the weight difference throwing off the fuel, or spark settings? just an idea. I will post the data log that I just took. you can see what I mean between 189-209. the RPS go erratic as they climb, it will stumble a little, as it climbs, and go down 100 RPMs, then up 300, then down a little, then up a little more. one thing that I noticed in the datalog is that the TPS settings are off, Im not sure why they keep changing me. so where it says 35%, it is actually 0% I will fix that. also, I have been using "LC1 with default values" as my 02 sensor settings, and it was always giving me values around 11:1 AFR but I just changed that in the datalog, and now I am seeing that my AFR is closer to 8.5 at idle, and going into the 7s when accelerating. so I will be doing some more tuning next weekend, and see what I can figure out. also what should my required fuel be with my current MSQ 9.0? thanks
Jake