West-se

If there's anyone running boost I need some help with timing. I am getting some detonation around 130+ Kpa 4000 rpm + my spark table I have 17 deg between 3800 and 4500 @ 130 Kpa. Im tuning with my wideband right now so its hit and miss and detonation scares the hell out of me. I am thinking of very soon investing in an EGT.

renns

17deg under those conditions doesn't sound unreasonable. Have you verified that the advance displayed in MT matches what you see with the timing light? If not, check that first before going any further.

From the datalog you sent over the secondaries don't come on line until 140 kPa or so. Looks like you are using rpm-based staging, at 3700? I'd suggest map-based, set at 80kPa or so. Seems to me muy mentioned he'd added another input in recent code releases. Take a look, as that may be of assistance as well. Maybe Scott can chime in to give you some further guidance. As I recall he's spent a fair bit of time fine-tuning his turbo 13B.

Roger.

muythaibxr

What are you running for split in those conditions?

It also looks like you went a little lean there, from about 12.2:1 to 12.8:1 .

West-se

in my split table I have basically 15 to 205kpa except where i have 10 from 3500 - 7500 @40 - 80kpa

I was using the msq posted here that was for a t2 and after looking at the split table I am begining to wonder about its values. Basically shows 15 deg up to 205 Kpa like i states and only shows 10 deg for a small portion of low boost at high rpm's which really makes me wonder.

Sean

Rex4Life

I'd have to say that the split table you mention is not optimum but should be safe.

I haven't had time to go and fine tune the split so I put something in that should be safe. I have no doubt it could be better, but right now it's not a priority for me.

Here's a few sources of info that could help:

Also take a look at what this guy has for the Power FC: http://www.zeroglabs.com/rx7/rx7_tuning.pdf

As Roger mentioned, I'd make real sure that your leading timing is firing at the right advance and check to make sure no trailing sparks happen when they are not supposed to. Put a mark on the timing pulley opposite the red mark, make sure it's a different color. You should only see the red mark when on the #1 trailing and the new mark with #2 trailing. On my 88 T2 pulley, 1 cm distance on the perimeter is worth 10 degrees of advance--you can add a few marks and verify the advance from MS is matching with the marks.

I believe the leading timing map to be conservative. Once I get everything working well I plan to tune on a dyno, till then I went with a conservative map.

There is a LOT of info and arguments about running split--you'll have to search for it. Let me know how it goes for you. I've been up in the 8-10 psi range with the same map and no problems thus far. I've got the boost control stuff installed and will be attempting to use the EBC feature of MS.

It could also be your staging setup. Are you running lean during the transition? I haven't seen your datalog. I've had to fatten up the fuel map right at transition. I'm using 4200 rpm or 160 kpa to transition along with a fattened up 4300 rpm bin for the ve map. So far it seems to do the job well with only small lean spikes during transition.

Scott

MSI v3.0 MSNSE-029j1

muythaibxr

In his datalog there's a slight lean-spot on transition, like I said, it goes to about 12.8:1 if I'm converting his voltages right.

The staging transition code will get rid of an actual lean "spike" but it won't completely remove the lean spot. The idea was that at least the nasty stutter could be removed, and then any further leaning could be removed by richening up the bin where the transition occurs. I *could* make the primaries dropping their pulse width lag a little behind the secondaries ramping up to try to combat this lean spot. Does anyone care enough for me to try doing that?

Ken

West-se

Ok, so last night on my way home by tach stopped working and after talking to muy I knew it was my trailing's weren't working. So I got home and noticed that my LED 16 IGt-T wasn't comming on, so I resoldered it and it is now working and so is my tach. With that done, I set my timing, in MT i set my fixed angle to -5 and the light to 0, then change back to -10. Attached is a datalog and my msq from last night. I noticed better drivability up to around 4000 - 5000 where there is still detonation. I had my transition point at 4200 but noticed that my primary duty cycle was 100% before hitting so I lowered it back down to 3800 there is still a blip under acceleration when the transition occurs but its marginal. My biggest thing right now is trying to control the detonation above 4000 under load I'm runninng stock boost right now, but I have an agressive street port with full 3" exhaust and ported wastegate, it does creep I'm seeing boost as high as 165 Kpa but there's not much I can do since my wastegate is actuated with the factory spring 6.5 psi (I think) I do have a digital boost controler but its not doing any interviening right now.

I forgot to add, my wb02 (innovate lc-1) display is reading about 1 point higher than MT is. I am looking into this but haven't found any answers yet.

muythaibxr

but WHAT is the innovate reading?

Megatune displaying the wrong value could just be because you have your custom.ini set up wrong, or it could be that you haven't programmed in the proper voltage -> afr scale into the lc-1, OR it could be that you haven't run a reference ground from the lc-1 to the megasquirt's ground.

No matter which it is, I'd like to know what the LC-1 says it's reading, as that's the accurate number. If it's reading something like 13.2-13.8:1, then you're running too lean, and you need to richen things up.

I would also suggest making the staging transition happen at 100kPa or lower.. even if the duty cycle isn't that high yet, you're less likely to damage something with the momentary lean spot (even if it's not a lean spike, and just a little lean, you could damage something), and you won't feel any kind of hesitation.

EDIT:

Just looked at your msq. You have 0 split for everything on boost. That could also be helping to cause your detonation.

You should change the staging second parameter kPa to 100, and change the delta to around 10, as this will keep you from staging on boost.

Your second trigger active edge should be set to rising... This could be causing your detonation around 4000 rpms, as when tofuball and I tried the original code that first included the "falling (like IRQ)" setting, we saw all kinds of weirdness at 4000 rpms, but as we weren't on boost, we didn't see knock. PLEASE change that setting. It's in the wheel decoder settings.

I also noticed that you're running closed-loop. First of all, if the megasquirt and the LC-1 are getting different readings, you should not be running closed loop, second of all, it looks like you're also running closed-loop on boost, which due to the nature of how closed loop works, I don't recommend. (with closed loop, you'll get oscillations in the AFR, of as much as .3-.5 AFR points... if the oscillation takes you too lean, kaboom).

Judging from what you've told me of your AFR problems, it looks like you might have set up custom.ini for a Techedge wideband. I saw that because the voltages I'm seeing at the AFR's you have in your wideband match up for the techedge curve. If I'm right, megatune is reading what you think is correct, or around the 11.0-11.4 range, but the techedge is reading around 12.0-12.4, right?

Ultimately, I think the cause of your 4000 rpm detonation is probably the wheel decoder second trigger setting.

Last, I noticed that you have an ego ignition events per step set to 200, but you're still using ignition pulses instead of mSec in Advanced Code Options in the Code Config menu. This won't cause your problem, but should probably be changed.

For tuning on boost, I'd recommend you take a more methodical approach. Turn off closed loop on boost, and then tune to 100kPa, then get on the gas pedal a bit harder, and tune to 115kPa, and keep going like that while manually tuning. Just make sure you start out with values before you start tuning that are way over rich to be safe.

Ken

West-se

I've got my settings.ini set to the Innovate LC-1 default 0-5v. I've also got my grounds from the LC-1 at the same location as the MS grounds. I've also got the outputs on the LC-1 set to as suggested in the ms forum, 0v - 10:1 5v - 20:1 .The LC-1 is reading on average about 1 point leaner than what MS is logging. I did some reading on the ms forum and it seems that the LC-1 runs too quick for ms to keep up.

From MS forum.
So I will change the update freq on the LC-1 and im sure that will help. Also, im my AFR table i've got it set to 12.5 until the boost gets to 140 Kpa then I drop to 12:0 is this considered safe ? I have it even richer for higher boost

muythaibxr

hrmm, when I looked at your AFR table I was seeing values that were 1 point richer... than what you say, probably because I'm set up for the techedge. I've never heard of this problem with the lc-1, but it makes sense.

I think it should be OK, but I've heard of people running richer than that on boost. I'm sure doing some research could find a good answer.

Last, you shouldn't rely on the AFR table to tune everything for you. It shouldn't be used at all on boost in my opinion. You should be manually tuning everything above 100 kPa. If you're relying on the AFR table to set your AFR for everything, and not changing your VE table, that also is not good, and is not the purpose of the AFR table.

Rex4Life

Here's what I'm running with right now. Obviously the entire table hasn't been tuned but a major portion of it is good.

West-se

I'm not relying on my afr table to do much of anything since I seem to find its not doing much of anything really. After making the adjustment to my LC-1 I also selected the other Innovate option in MT configurator that was using afr not lambda my reading match up. I took another drive tonight this time running some higher advance numbers in the boost range and seem to have better results but still something is not right... I am second guessing my initial timming because I shouldn't be running 20+ advance at 140+ kpa and 4000+ rpm but it was feeling good. I did notice I was leaning out during my boost runs so I will go overr my ve table again tonight. I can post my logs if anyone wants to see.

West-se

does anyone know the calculation off hand to convert the voltage from the LC-1 to an AFR reading ?

muythaibxr

It's important that you change your second trigger to be on the rising edge of the pulse instead of falling edge in the wheel decoder settings. Rising edge will show problems at around 4000 rpms give or take depending on advance.

Again, I also noticed that you're running 0 split on boost, I'd change that to 10-12.

To convert voltage to AFR with the LC1 set up for 0-5 10-20, it's (voltage * 2) + 10.

Ken

West-se

well you fixed it muy, big thanks to you !!!! falling edge 2nd trigger was the problem I guess making it just too far retarded on that firing cycle. I reloaded the T2 maps I had from here and changed that and she goes like a banshee on low boost. Thank you again. I am running out of fuel when I tried on high boost though, I need bigger secondaries... my 700's just can't do it.

The other thing I noticed is that my req_fuel is alot higher than I've seen anyone else here I know im running 550/700 and I guess its my big street port that moves alot more air at idle since I've got it at 10.5 to get a smooth idle.

Its soo nice to have some power again. During my commute today I also noticed a hesitation at 4000 at highway speeds when I give it more throttle there's a hesitation then it goes, do you think this is fuel or timing ? or maybe both ?

Did I say thanks ?

muythaibxr

It's probably fuel causing that one if you're on cruise. It might be running a little lean or something.

Also, now that you switched the 2nd trigger problem, you may want to recheck your timing with the timing light.

13bdarren

what do you set your 2nd trigger to and where is it. I am have the same problem as him I have trigger pos A 1 return pos b 3 trig pos b 7 trig return pos 9 My car won't run on the setting you posted in the faq

muythaibxr

you have to have your CAS set right and timing set right for the car to run. The settings I posted in the FAQ depend on having that set right for your CAS stab. They are a guideline, not a rule.

2nd trigger settings are in the wheel decoder settings with 027 and above I believe.

13bdarren

So what should they be set at. west se what setting are you using.

muythaibxr

If you are using 1 3 and 7 9, you should set the engine to the yellow timing mark (5atdc) and set it to tooth #3 after the second trigger tooth... and start the car.

Set the MS's fixed angle to -5, and then adjust the CAS using a timing light until the yellow mark lines up with the marker.

After that, reset the fixed angle to -10 and your timing will be set.

If you're still getting knock after that, you need to retard your timing/add more fuel. Basic tuning really.

Also, if you haven't already, please read my wheel decoder howto.

13bdarren

Yeah I did that but I think I went by the wrong timing mark. Mine no longer have colour on them.

muythaibxr

going off the wrong one would retard your timing 15 degrees. You just wouldn't have a lot of power...

I guess it's possible that if it's extremely retarded, AND your trailing is retarded too, and you're not running much advance, the trailing could be firing on the wrong face of the rotor...

Ken