Thanks Dan,

That really make sense. I was going to basically compare my datalog MAP and TPS and try and find common spots. Your method seems like it will save me time and get better results.

Rutager

References for dyno shop familiar with Porsche, amd MS-2
 

Great write up Dan, I'm a long way from getting my 74, 2,8, twin plug/10.3 C/R, PMO rebuild running well, much less, optimized.
Can anyone recommend a dyno tuner, relatively close to Humboldt county, CA???

If my wife would just slow down on the Honey Doo List, i might actually be able to bring my dad's 12 year restomod to completion.

Thanks,
Chris

Hi Chris,

The guy I use for dyno tuning is Gary Bains, Bains Tuning - he is in Fremont, CA but he can tune remotely via other dyno shops (he connects to the target ECU via internet, while the technician drives the car on the dyno per Gary's instructions). See Bains Tuning on FB, bainstuning.com, bainstuning_gary on IG. If he can't help you directly, he knows a lot of tuners and may be able to recommend someone near you.

Hi Julian,

I have been adjusting my AE. Would you mind sharing your AE and how much fuel you are adding? I'm still pretty new to this EFI tuning game so knowing what others do is very helpful. Thanks.

http://forums.pelicanparts.com/uploa...1749416116.jpg

Hey Dan,

I'm using Accel Pump-AE (in Accel Enrich >> Accel Enrich Settings) as opposed to time-based. With 100% TPS.

This operates much like a carb setup where there's just one squirt of fuel- no time variables or percent variables and I've found it to work so much better than time-based.

Of course you gave to have a really well dialed in VE map before doing any AE adjustments.


I'm still playing with mine, but it's really snappy and sharp and I think I only need minor adjustments.


http://forums.pelicanparts.com/uploa...1749420314.jpg

Hi Montauk,

I just ordered the Variohm sensor from Digikey as well - I should have it in a week. Then I will be able to see first hand the difference in signal stability between the potentiometer based TPS and the Hall effect based TPS. Thanks for the info! :)

Thanks Dan
 

Thanks Dan, I'll give him a call.
chris

Hi Dan,

What dwell settings do you have setup in Tunerstudio for triggering the MSD Streetfire?

Sean

Hi Sean,

I'm using the default settings in TunerStudio. Honestly, I don't know enough about dwell time with CDI ignition to experiment - I have read 3 ms is pretty typical for a running engine.

http://forums.pelicanparts.com/uploa...1749706611.jpg

Added The IAT (Intake Air Temp) Sensor
 

While I am waiting for the Variohm Hall effect TPS to arrive, I decided to add an intake air temperature sensor (IAT also referred to as MAT - manifold air temperature in MS / TunerStudio documentation) to my ITB EFI set up. The conversion kit from Al at x-faktory.com doesn't use the IAT sensor and my system worked without it, but without the IAT sensor data the MegaSquirt ECU assumes the intake air temp is 70 deg F and does not adjust fueling to compensate for air density changes as temperature rises and falls.

The first order of business was to carefully remove the vinyl electrical wrapping I had applied to the ECU wiring harness back to the point where I separated signal / sensor wires from power / ground wires.

The short orange wire is connected to ECU connector pin 20 - MAT (IAT) sensor. I joined a 3 foot length of orange 16 gauge wire with a crimp connector.

http://forums.pelicanparts.com/uploa...1749788511.jpg

The IAT sensor has an orange signal wire and a black ground wire which is connected to ECU pin 7 sensor ground which is also used for the TPS, so I used a crimp connector with 1 wire in / 2 wires out to split off the sensor ground for both the TPS and the IAT sensor.

http://forums.pelicanparts.com/uploa...1749788511.jpg

This time instead of re-wrapping the wiring bundles with vinyl electrical tape I decided to use 3/8 inch plastic split wire wrap.

http://forums.pelicanparts.com/uploa...1749788511.jpg

http://forums.pelicanparts.com/uploa...1749788511.jpg

Close up of the IAT sensor with connector and pigtail. This is a GM unit and supported with calibration setpoints in TunerStudio. I found this listed on the diyautotune and summit sites.

http://forums.pelicanparts.com/uploa...1749788511.jpg

The IAT sensor mounted on the base plate of the air filter weather hat on the 1,2,3 bank ITB's.

http://forums.pelicanparts.com/uploa...1749788511.jpg

I updated my EFI documentation for the addition of the IAT sensor signal and sensor ground connections.

http://forums.pelicanparts.com/uploa...1749788511.jpg

With my notebook PC running TunerStudio and connected to the powered up MegaSquirt2 ECU, I unlocked the temp sensor calibration tables and selected "Burn". Then I opened "Calibrate Thermistor Tables", selected "Air Temperature Sensor", selected "GM", then selected "Write to Controller". Then I locked the sensor calibrations and selected "Burn". Now the IAT sensor is calibrated and reading the correct air temperature and ready to use.

http://forums.pelicanparts.com/uploa...1749788511.jpg

I also installed one this winter. I haven’t seen any massive benefit yet, but it’s not yet hot or cold enough to really make a difference. Either way, it’s a good piece of data for tuning and understanding what the engine wants/needs.


Sent from my iPhone using Tapatalk

I haven't driven my car or even started it in particularly hot or cold weather. Once warmed up, the IAT stays around 25°C. Mine is mounted in the old MFI type air filter so I think it's getting an accurate sample of the intake air.

Before I put the MFI filter on, I ran the car with the IAT loose near the front of the engine and no air filters at all. I thought the MFI filter might make a slight difference because all six cylinders would be getting air at the same temperature from the back of the engine. So far, I can't see any difference.

You should have intake air temperature and intake air pressure sensors along with fuel pressure and fuel temperature sensors. Well, I hope the EFI system you are using adjusts for changes in fuel pressure and temperature...

Does/did the 1977 CIS adjust for fuel pressure, fuel temperature and air pressure and air temperature.

Not a sarcastic comment, I really want to know- I can't remember since it's been a while since I had a CIS...

Hi Julian,

My 75 CIS system did not monitor and adjust for air temp, air pressure or fuel temp, but the CIS system is controlled by fuel pressure and there were no O2 sensors, so no Lambda control.

I read Scott's (Winders) comment to be directed toward the EFI ECU control capabilities. I am running a MegaSquirt2 ECU so air pressure (BARO) is measured by the MAP sensor when the engine is key on / engine off, the air temperature is measured by IAT sensor, and both air pressure and air temp are compensated for in the fueling pulse width calculations.

Fuel temp and fuel pressure are not monitored directly, although air temp could be considered a proxy for fuel temp as delivered to the injectors (not accounting for cooling effects due to the latent heat of vaporization when the injector sprays fuel into the intake port).

On the flip side, my system is running closed loop (when it is warmed up and within upper / lower limits of MAP and TPS) as I have the EGO (exhaust gas oxygen) Correction enabled and the ECU monitors the actual AFR against the target AFR and makes corrections to the fuel injection pulse width in real time.

Air temp is not a proxy for fuel temp. Fuel temperature affects fuel density so an EFI system monitoring fuel temp can compensate for changes in density. Changes in fuel pressure need to be compensated for as well.

A closed loop EFI setup can compensate for only so much variance. Changes in air density, air temperature, fuel temperature, and fuel pressure can easily exceed the limits of compensation when running in closed loop.

I'm monitoring fuel pressure, IAT, MAP, cylinder head temperature, oil pressure and oil temperature. I can add fuel temperature since it's built into the same sensor that monitors fuel pressure. Maybe I'll do that and see what the results are. I've got a few unused inputs left.

Scott - how much variation are you seeing in fuel temperatures?

I stand corrected, the MegaSquirt2 ECU has limited capabilities which in turn limit its capabilities to control AFR, see below from a quick AI search of the internet:

Fuel temperature and intake air temperature are often different in an engine's fuel injection system due to several factors:

Heat Transfer: Components within the fuel injection system, such as fuel lines, pumps, and injectors, are located near the engine's heat sources (like the engine block and exhaust system). This heat can be transferred to the fuel, raising its temperature.

Fuel Recirculation: Many fuel injection systems have a return line that sends unused fuel back to the fuel tank. This recirculated fuel has been heated by being in the engine compartment and can increase the temperature of the fuel in the tank and lines.

Intake Air Source: The intake air is drawn from the ambient environment, often through a cold air intake system designed to minimize its temperature, to improve air density.

Different Paths: Fuel and intake air travel through different parts of the engine compartment. The fuel path is near the engine, while the intake air path is typically routed to avoid heat.

Ambient Conditions: The temperature of the fuel in the tank is affected by the ambient air temperature, whereas the intake air temperature is directly influenced by the ambient air temperature at the point of intake.

These factors contribute to a difference between the fuel temperature and the intake air temperature, even though both can be influenced by the external environment. The ECU monitors both temperatures to ensure the correct air-fuel mixture is delivered for optimal performance and emissions control.

I don't have any data anymore as I sold the race car and the laptop with the data on it.

Also, I changed the fuel cell setup in the race car and did not have enough data to come to any real conclusions.

I went with a Fuel Safe fuel cell with a Radium FCST-X in cell surge tank setup. The return line returned fuel to the surge tank. So, in theory, the heated fuel returning from the fuel rails to the the surge tank might have less time to cool down in the fuel cell than it would it the fuel was returned to the bladder area. The Radium Radium FCST-X setup operated perfectly though and it may have had little actual impact of fuel temps compared to my previous setup.

But, fuel does get warm over the course of a long race so it is a very good idea to have an ECU than can compensate for fuel temperature changes.

Installed the Hall Effect TPS
 

I used the same Variohm sensor from Digikey as Montauk used, see his link in the quote above. This unit is made for mounting holes spaced 32 mm or 38 mm apart. For the RHD ITB's, I used the 32 mm version.

The Hall effect TPS arrived today and I got busy and installed it in the same location as the existing BMW TPS on the cylinder #3 ITB. The throttle valve shaft protrudes 3/4 inch from the end the throttle body, so I fabricated a 1/2 inch spacer out of aluminum bar stock (1 1/2 inch wide x 2 inch long x 1/8 inch thick, sandwiched 4 pieces) with a 1/2 inch hole in the center and 2 x 1/4 inch holes centered about the 1/2 inch hole and spaced 32 mm apart.

I removed the 1,2,3 ITB's to get better access to the TPS.

http://forums.pelicanparts.com/uploa...1749937379.jpg

The existing TPS set up.

http://forums.pelicanparts.com/uploa...1749937379.jpg

The existing TPS removed showing the throttle shaft with D protruding 3/4 inch out of the throttle body.

http://forums.pelicanparts.com/uploa...1749937379.jpg

Test fitting the 1/2 inch spacer block.

http://forums.pelicanparts.com/uploa...1749937379.jpg

The Variohm TPS installed on the 1,2,3 ITB's.

http://forums.pelicanparts.com/uploa...1749937379.jpg

A page from the Variohm Datasheet showing the wiring diagram - for my application I used the CW sensor connections.

http://forums.pelicanparts.com/uploa...1749937379.jpg

The 1,2,3 bank ITB's re-installed with the Variohm TPS and the IAT sensor.

http://forums.pelicanparts.com/uploa...1749937379.jpg

Re-calibrated the TPS on TunerStudio for the new Variohm TPS.

http://forums.pelicanparts.com/uploa...1749939319.jpg

The TunerStudio dashboard while running the engine with the Variohm TPS and the IAT sensor.

http://forums.pelicanparts.com/uploa...1749937379.jpg

Engine running with the Variohm TPS and IAT sensor installed.

http://forums.pelicanparts.com/uploa...1749937379.jpg

Nice work. I hope it gets better resolution for you. BTW, I found the plastic wire sleeves to be too difficult to fit particularly with tight bends. I've been using a combination of shrink wrap, abrasion-resistant wrap from Mcmaster Carr, and fabric tape.
https://www.mcmaster.com/catalog/131/1015/2649K521

Thanks. I am going out for an auto tune session this morning and I'll record a couple of data logs to verify operation.

I agree with your assessment about the plastic wire wrap. I think I'll de-populate the wire connectors and try heat shrink tubing where ever possible. I'll also check the Mcmaster Carr wrap - thank you for the suggestions.

Update: I ordered some of the abrasion resistant sleaving from McMaster-Carr - I'll give that a go next. Thanks.

EFI Wiring Re-wrap
 

I researched various wire wrap and wire sheath options. The heat shrink tubing I found is dominated by 3:1 shrinkage with an inner adhesive and I thought that would be more of a mess trying to remove in the case of a needed repair. I also looked at and ordered some abrasion resistant, length slit, wire sheathing from McMaster-Carr which I may install if the need becomes apparent. In the end, I settled back upon the method used by the factory for my 1975 Porsche: black electrical tape wrapped around the wires in long continuous lengths.

But first I removed and discarded all of the corrugated black plastic length slit wire sheathing - that stuff is bulky and didn't look right on my car. I also dismounted and laid over the EFI electrical panel and disconnected the EFI wiring connectors and I removed the fuel supply and fuel distribution hoses from the fuel pressure regulator to give me ample room to work on the re-wrapping of the EFI wiring bundles and sensor connection lines.

Once the wrapping was completed, I reconnected all of the EFI wiring, remounted the EFI electrical panel and reinstalled the the fuel lines.

Next will be to perform a fuel pressure / leak check, then start the engine and after warm up to perform an ITB flow balance check / adjustment.

http://forums.pelicanparts.com/uploa...1750173073.jpg

Wire wrapping in process. I used small 4 inch cable ties to hold the wire bundles together, then removed them, one by one, as I wrapped along the length of each wire bundle and sensor connection line.

http://forums.pelicanparts.com/uploa...1750173073.jpg

http://forums.pelicanparts.com/uploa...1750173073.jpg

http://forums.pelicanparts.com/uploa...1750173073.jpg

http://forums.pelicanparts.com/uploa...1750173073.jpg

Nice job!!!

The best stuff:

Raychem DR-25

https://www.milspecwiring.com/DR-25-Shrink_c_325.html

Just be aware to if possible separate power and ground wires from signal wires and again, if possible concentric wrap your harness as it’ll make it more durable, flexible and so much neater.

https://uploads.tapatalk-cdn.com/202...da9b3be013.jpg
https://uploads.tapatalk-cdn.com/202...fde7009531.jpg


Sent from my iPhone using Tapatalk

The Haltech harness uses shielded cables for the crank, cam and knock sensors. There's also separate ground sensor wires that go to the ECU but are not bonded to the chassis ground. 5VDC sensor power wires are included too. So far, I haven't had any wiring noise problems.

Agreed. The Raychem DR-25 heat shrink tubing is very good and available in many diameter sizes.

https://www.te.com/en/product-9908044002.html

http://forums.pelicanparts.com/uploa...1750261927.jpg

Eurosensor uses Raychem DR-25 on the wire lead of their Variohm Hall effect rotary position sensor that I installed as a TPS.

http://forums.pelicanparts.com/uploa...1750261976.jpg

Stuff like that looks amazing and pretty but also looks like a pain to repair.

Thank you again Dan for all your documentation in this thread. I finally got my CIS -> EFI conversion started for the first time today. Needs a lot of tuning and it’s not an ITB setup yet but it’s a start.

The goal is to plan sufficiently such that you don’t ever have to cut into your harness to make modifications. If you have to alter the wiring after install, the prep was insufficient.

Tidy wiring is much easier to repair if ever that’s needed than rats nests.


Sent from my iPhone using Tapatalk

I am close to dumping the CIS on my 1975 911S, but will go to Weber triples.

I like simple

Awesome write up! Dan what fuel pressure regulator are you using? Can’t make it out from the pictures and also at what PSI do you currently have it set at? I’m chasing down some fuel issues and just trying to verify few numbers. Thanks

The fuel pressure regulator is from K Motor Performance:

https://kmotorperformance.com/shop/universal-adjustable-fuel-pressure-regulator-kit-an6-6an-fittings/

I set the fuel pressure to 40 psi static and it drops to 38 psi when the engine is running.

Dan,
I went with the Bosch 0280155831 injectors—same ones you used—mostly because you did. I’ve set them as close as I can to the spec’d 43.5 PSI (3 Bar), but my data logs show 45 PSI. I noticed you ran yours at 38 PSI (2.6 Bar), which is over 10% lower. Any particular reason for choosing that pressure?

One odd thing I ran into: accurate injector data is surprisingly tough to find. I couldn't track down the dead time versus voltage specs online. Then it hit me to check the Maxxecu software (which had made my shortlist of ECU options), and sure enough, the data was right there.

Here’s what I’m using:
Dead Time (ms): 1.96, 1.43, 1.11, 0.94, 0.82, 0.71
Voltage (V): 8, 10, 12, 14, 16, 18

This is a fantastic FPR with a damper built in. This worked flawlessly on my race car:

https://www.radiumauto.com/FPRD-RA-Fuel-Pressure-Regulator-Damper-P2751.aspx

P.S. You want a damper....

Hi montauk,

I set the fuel pressure based on my observation of the fuel pressure running on other cars I have owned or worked on and not on any engineering calculations. When I initially installed the ITB EFI system on my stock CIS motor and checked the fuel pressure, the fuel pressure regulator adjustment was screwed all the way loose so I saw the max pressure of the fuel pump at 90 psi and turned it down to 40 psi and left it there.

Since the stock CIS motor with ITB EFI was making 205 hp (crank) on the dyno before I rebuilt the motor, I was way below the estimated max hp for the given injector size (I remember reading the 30 lb/hr injectors would be good for 288 hp), so if I was running the injectors with 38 psi versus 43.5 psi I wasn't in danger of running out of injector capacity. But now with rebuilt motor making 274 hp (crank) I am much closer to the edge.

I did a quick AI internet search:

Calculating Maximum Power for a Given Fuel Injector Size
To determine the maximum power a given fuel injector can support, you'll need to use a variation of the fuel injector sizing formula. The key components are:

• Injector Size (lbs/hr or cc/min): The flow rate of your specific injectors.
• Number of Injectors: The number of injectors in your engine (e.g., 8 for a V8).
• Brake Specific Fuel Consumption (BSFC):This represents the amount of fuel an engine uses to produce one horsepower per hour. BSFC values vary depending on the engine type (naturally aspirated vs. boosted) and fuel type (gasoline, E85, methanol). (0.5 for naturally aspirated gasoline engines)
• Maximum Injector Duty Cycle: This is the percentage of time the injector is open and flowing fuel. It's generally recommended to use a maximum duty cycle of 80-85% for safety and consistent fuel delivery.

Formula: Estimated Crank HP = [(Injector Size in #/hr x Number of Injectors) ÷ BSFC] x Maximum Injector Duty Cycle

For 30 lbs/hr: Est HP = [((30 x 6) / 0.5) x 0.80] = 288 hp

For 31.4 lbs/hr (43.5 psi): Est HP = [((31.4 x 6) / 0.5) x 0.80] = 301 hp

Now assume those injectors at 38 psi flow only 27 lbs/hr (down 10%),

For 27 lbs/hr: Est HP = [((27 x 6) / 0.5) x 0.80] = 259 hp (80% injector duty cycle)

For 27 lbs/hr: Est HP = [((27 x 6) / 0.5) x 0.85] = 275 hp (85% injector duty cycle)

Now, with more experience and considerably more thought, I see that I should turn the fuel pressure up so that when the engine is running the fuel pressure will be at 43.5 psi to take full advantage of the Bosch 62672 (0280155831) injectors that are rated at 31.4 lbs/hour at 3 bar (43.5 psi).

But unlike your Haltech ECU, the MegaSquirt2 ECU doesn't monitor fuel pressure nor fuel temperature directly, so if I adjust fuel pressure, then I will need to re-run auto tune to adjust the values in the VE table since for a given injector pulse width the injectors will deliver more fuel per shot at the higher fuel pressure.

Another thought: My motor makes max hp at 6600 rpm and that carries flat out to 7200 rpm and I have the rev limiter set to 7500 rpm. The ECU fires each bank of injectors once every 2 revolutions, so a given injector fires 3750 times per minute which works out to firing every 16 ms. Per your dead time data and given my Anti-Gravity battery is at 13.2 V (13.8 V while running), my injector dead time is about 1 ms which leaves 15 ms for the injector to deliver fuel at 7500 rpm. At 85% duty cycle, that translates to 12.75 ms maximum pulse width available at 7500 rpm. I need to go look at a recent data log with the a 100% TPS pull to 6000 rpm and see what injector pulse width the ECU was calling for - I wonder if I'm already at the wall?

Thank you for asking the question.

Hi Dan and the rest of the thread,

Just to add on to the chat about injectors. Today I tested my injectors for dead time in preparation for tuning. The setup I have is 986 Boxster Fuel Rails with built in Fuel Pressure Regulator(FPR) that is calibrated for 3.8 Bar(55 PSI). I am using stock Bosch 0280156053 injectors rated at 260cc but with the stock 3.8 bar fpr they are theoretically rated at 294cc.

I did some testing today for dead times and with the stock FPR and stock replacement CIS Bosch Fuel pump they output 310cc with the FPR verified at 55 psi. When checked at different voltages and pulse widths in TunerStudio I ended up with a .81ms dead time. This is on a stock but tired old 77 2.7 CIS engine.

Sean

Dan,

The Bosch injectors will have an improved spray pattern if you bump the pressure up to 60 psi and will also decrease the duty cycle which would be a good thing in your case.

If you don't have a fuel damper, I would recommend installing one. When I installed the FPR I listed above, my MoTec M150 traces for fuel pressure smoothed out.

Dan,

You have invested a lot more time in figuring this stuff out than I have. Great work. I was thinking about the dead times. Those are at 3 bar. If you're running at a different pressure, you'll need to determine what the dead times are. Or doesn't Megasquirt allow for dead times?

As far as running at 60PSI, per Scott's suggestion, I did find one chart that lists flow at 55PSI, 334 cc/min versus 297 cc/min at 43.5PSI.

Stan Weiss' - Electronic Fuel Injector (EFI) Flow Data Table

Dave