Hey Phillip

I'm running a small hp laptop running windows 10. Works good. I had purchased a small micro tablet running windows 10. It kept running out of memory, and Tunerstudio would hang. I believe it needs more memory.
A lot megasquirt guys are using the Raspeberry Pi. You can hardwire the system
Directly in the car. It allows you to create custom dashboards. The system can be programmed to boot as soon as you start the car, and shut down when you shut the ignition off. Might not suit our cars from a period correct standpoint.

I use linux tunerstudio but how to get timing on one of the ts gauges?

Now that I want to get my cam hall signal a bit better is it possible to run the engine without ic in the way to adjust hall gap? I just teied it but the kpa showed 2? Maybe I had my map sensor wires detached? Speed density goes by map so it should start. And if my map was working it should read about 100 kpa atmo right?

Yeah you can run it with no intercooler without a problem.

Sent from my SM-G950U using Tapatalk

smurf,

With the engine not running and MS powered up, MAP should read close to 100kpa depending on your altitude.

You can add another gauge(or replace an existing one) to TS using the Designer feature. I believe the function is called spark advance in the pick column.

I would have thought spark advance would be one of the default gauges? I also need to figure out how to add a fuel pressure gauge in there too.

Is there a possibility to add more gauges too on the screen?

Ok, managed to get the timing gauge and it was advance in designer outputs. Got my frp gauge to show too but it seems to be misconfigured and read low. Now I need to figure out correct 0V and 5V values for it. Its a generic sensor if anyone has one configured let me know the values.

Designer also has options to add as many gauges as you like on your TS screen.

Got my EBC installed today. What is best method for setting this up? Open Loop or Closed Loop? Open Loop looks simpler. I am currently running an .8 bar spring, and would like to limit the EBC to 1.1 bar. I find the Closed Loop setup a little confusing. Any tips would be appreciated.

You will want to tune open loop first before you try to use closed loop. You will use the initial values from the open loop map for closed loop.

Sent from my SM-G950U using Tapatalk

So start with open loop. Zero out Boost Control Duty Table. Add 100%DC to column 1 to help spool up the turbo. Add 10% DC to top row and continue to add 1 to 2 % at a time until boost target is reached.

Once the Boost Control Duty Table is populated these values can be copied into the Boost Control Bias Table under Closed Loop Algorithm. The Boost Target Table also needs to be populated.

Let me know if I have missed anything.

Cheers

John

John,

First thing to do is set your Overboost Protection. I set my max at 15kpa over target and a hysteresis of 25. I used Spark cut because Fuel cut would send me to the windshield if activated.

Next thing to do is determine (datalog) your boost pressure in kpa with just wastegate spring pressure (EBC off). The EBC will be controlling boost above this level. You need to populate the Boost Control Targets table. I used 40% TPS as the minimum row populated with a value just above the spring pressure value. From there fill in the rest of the table with increasing values to your max boost at 100% TPS.

In Boost Control settings, input the max DC for the valve. It does not have to be 100%DC. I use 60% for faster dome venting, but I am going to increase it to 70% as I learned some things in the last few days about the slider. Set your Boost Lower Limit Delta to 40kpa. This is the point where full DC applied to the dome is switched to a value in the Boost Control Bias Table. So if your target is 200kpa, at 160kpa the code switches to PID control using the DC% value from the Boost Control Bias table and adjusting that as needed. Note that this does not apply to open loop (Setup mode).

Next select Setup mode. Populate the Boost Control Duty table with values of DC to control the EBC valve. I used 100% at low RPM's then at 3000RPM's I had 28% up to 40% at max RPM at WOT. Then decrease these values slightly as you go down in the table. Then go out and datalog boost(MAP) boost duty and see what you get. Adjust DC values as necessary to control boost. Once you get the Boost Control duty table functional, you transfer these values to the Boost Control Bias table when you select Basic mode.

In Basic mode go do more datalogs and adjust the bias table as needed to control boost. A note on the slider. I just found out from a member on the MS forums that the slider does not do much until it over 475. I set mine to 478 and it made a huge difference. Need to go up some more to 48x. If you get it control boost without overboosting in Basic mode, there is no need to go to advanced mode. In Advanced mode you tune the PID parameters. We can talk more about that once you get there.

Thanks Dennis

Has anyone tried injection timing / fuel trims? Default values are 360deg flat. I have low torque timed 964 elgin grind so is there any advantage to run tuned map for injection timing? What I gathered numbers should go upwards as rpms rise with end of squirt.

Got tired of hunting idle so I sniped a bmw PWM IAC. I hope it works ok.

smurf,

There was a lot of discussion of this feature when it was first introduced with MS2. The theory is to inject fuel before the intake valve opens. I have SC cams which start to open about 400* BTDC. I use "end of pulse" for injection. I set my low load areas to 400* and increased to 440* as RPM;s increase to account for injection lag as engine speeds increase. But at high boost, high RPM's things change. Assuming the intake valve starts open at 360* BTDC, for an injector duty of 80% at 6000 RPM's there is only 16ms available for injection; 80% of 720* is 576*; at 6000 RPM, 360* x 6000 = 2160000 degrees/min =36000*/sec = 1/36000sec/degree = 0.0000277778 sec/degree = 0.0277778 ms/degree; .027778 x 576* = 16ms. There is even less time once you approach 7000 RPM, 13.7ms. Of course if your injector DC is over 80% you have a longer injector pulse which would carry the injection into the intake stroke.
So if you have a 13ms injection pulsewidth @6000 RPM, the engine travels 13/0.027778 = 468*. If injection starts @TDC of the compression stroke the end pulse would finish at 468* BTDC well before the intake valve opens, so if you have the injection timing set at 400* it would get all the fuel in no problem. But, I just read the tool tip in TS that says this is based on the spark event so I assume injection start for the next cylinder combustion event would start at the spark advance point of the current cylinder? If true that would add 10* - 3x* more for injection to occur.
At any rate, if your injectors are big enough there is no issue getting all the fuel in. My injectors are little too small. At 6800 RPM full boost my pulsewidth is 18ms (94%DC). So I have my pulse ending with the intake valve open in that region (180* BTDC).

As for small pulsewidths, you need that information about your injectors, which is not provided by most manufacturer's. If you have ID1000's they provide that info. I don't have any experience with this.

Good info, thanks. I have small pulse data for my ID1000s. Need to find my specsheet for the cams but I hope I can put some calcs together for some initial timing numbers with the help of your data above.

One datapoint from yesterday: 5800rpm, 0.9bar, DC80%, PW 17ms. And that was too lean as it was the first WOT high boost run of the day.

Could not find info on elgrin 964 but this was here:
964 for Turbo Use (1.25 @ overlap)
intake is .470 lift and 240 duration
exhaust is .430 lift and 230 duration

How do I come up when it starts to open intake valve?

http://forums.pelicanparts.com/uploa...1504814431.gif

From what i've seen, injector timing only makes a significant difference in the idle and low rpm areas. I just used trial and error to see what gave me the smoothest idle and also ended up in the 400 range at idle. At higher rpms the injectors are going to be open most of the time anyways, so its not quite as critical. I'm sure there is some optimization that can be achieved with it, but I wouldnt worry about it too much.

So to determine the optimum value should I look for richest VE cell or some other measurable unit?

This morning I tried from 400 to 450. Nothing tha tI could feel different but I still have massive lean spots of idle light accell. I tried richening those cells a lot but still lean stumble. Yesterday I tried to lean those cells (if the lean stumble was too rich --> no ignition --> read lean) but that was not it.

My TPS still gives too low numbers low revs so I cant dial AE to help small trhottle inputs. After the stumble it goes very rich on low kpa cells. The same happens when WOT. First lean surge and after AE rich for some time. I have tried to adjust but still work to do and I even added MAP AE to the mix to counter TPS problems.

My ms3 reads 12.4-12.2 Volts and my Voltage gauge reads 13.1? I also measured only 3-4Volts from Vref. This could make TPS wonky? One time I saw both CLT and MAT jump up at the same time the voltage droped from 12.4 to 12.2V. I think my CLT sensor tip touches cylinder head once in a while as I had to dremel it thinner to be able to squeese it in between the fins on cyl4.

I have a new high Amp alternator to install next winter but I guess I can check the wiring once more.

EDIT: After some battery voltage calibration searches I think my main problem is the vref being too low. Might even be bad wiring as vref should be at least 4.7 at any sensor and mine shows about 3-4Volts? All sensors seem to work so hopefully not ground to vref issue but maybe short somewhere. BUT, the low vref should in theory rise the ts vbat not make it lower (maybe this behavior is different from ms2 to ms3pro?)

EDIT2: I might have had one sensor ground momentarily on the vref when I measured those voltages above ( I remember switching some wires around) but now they were really good at 4.95V. Alternator is having hard time as battery was 12.95 not running and 13.1 running and injector 4 had 12.5V running. I need to check what ts batV shows and adjust the difference for the 12.5 V the injector sees. Injectors are from relay for fuel pump and Injectors. Should I hunt where the 0.5-0.6V goes between battery and injector?

Smurfbus I would suggest you turn AE off and go out and do some Autotune with the injector timing set to where you want it. I followed Dennis's settings and it made a huge difference in my VE table. When auto tuning, richened up the cells that are causing the lean stumble, then lock them out so they can't be altered.
I had the same issue, but implemented this approach and car now runs amazing.
Mind you I am not running E85.

It seems its after all too rich on those low rev 80-100kpa cells. Autotune and WEAL both just added fuel log after log even on 'very hard' and ended up to 250 there. Still lean = misfire or no spark due to too much fuel so now I have leaned the cells even more than the last leaner-test and it helped but still lean. so I just try to find the correct low numbers. Now I'm at about 100 and my WOT max is 250 upper right corner. AE does not switch on at off idle.

Outch. My pro took a dump. shut down off light cruise close to garage so managed to roll it home. pro does not ground the pump/injector relay anymore? pump works from separate ground for relay.

pro gets 13V off ecu relay but I cant connect to it anymore. member requests pending on FB addicts and owners

The fuses look ok inside the ecu. Maybe I need to do some continuity tests on them.

Found this: Matt Cramer wrote:
The MAF and TPS are both powered from the VREF. Usually a short on the VREF would be the most likely suspect, although it is possible for a short on the IAC valve to blow the fuse too.

I just installed a bosch ICV ! Drove maybe 5miles before it shut down.

Any ideas?

EDIT: yup, the main fuse was blown. I dont have spare fuses and the closest shop with them is about 50 miles away. First I thought about dirty alufoil trick but realised its too hazardous for the ecu if its not the ICV that I have now disconnected. Then I figured I could feed the sensors 5Volt off a separate circuit and use the vref fuse for main fuse. That seems to work and the car started and all sensors read fine. I'm still not confident enough to take the 50mile drive with this patent as if there is some intermittend short somewhere it would blow the only fuse and I would be forced to wait for bedtruck. I hope to get a loaner for tomorrow as the nice people at autotune.fi were willing to sell some fuses to me even though they are closed for weekends.