Which way to turn 3.2 air flow disk?

[Discseven]

Originally Posted by 76FJ55

If you look at the manual it actually specifies to disconnect the return hose hose and to connect the hose to you r catch bucket on the return side. this will test the flow rate at the 36psi load not at open rail (0psi). your pump will likely still pas sthe test as you are seeing higher than specified pressures when running, so it doesn't seem like it is a pump supply issue.

Interesting. Thank you 76. In this case, measured flow is what makes it past regulator.

Am looking at Porsche Technical Specifications booklet. Page you show is not in there. Is that the large format service manual you have? Is there a volume indicated for the 30 second flow?
.

[76FJ55]

Yes, this is from the Porsche Work Shop Manual. This page (20-2) covers the setup for the test. Page 24-3 shows the required specifications.

Do you not have a copy of the WSM? what is your email address?

[proporsche]

Karl: you can also get all the info from Porsche specs booklet..
link in your message

Ivan

[Discseven]

Originally Posted by 76FJ55

Yes, this is from the Porsche Work Shop Manual. This page (20-2) covers the setup for the test. Page 24-3 shows the required specifications.

Do you not have a copy of the WSM? what is your email address?

Thank you 76. Do not have that. Only have the small spec booklet. PM'd you my email.

Originally Posted by proporsche

Karl: you can also get all the info from Porsche specs booklet..
link in your message

Ivan

Just saw your PM. Thanks Ivan.
.

[wazzz]

Originally Posted by Discseven

Interesting. Thank you 76. In this case, measured flow is what makes it past regulator.

Am looking at Porsche Technical Specifications booklet. Page you show is not in there. Is that the large format service manual you have? Is there a volume indicated for the 30 second flow?
.

What this test is measuring here is the pump max flow, and it appears that the pump is good. No regulator is involved, since engine is not running. Pressure regulator and damper work only with vacuum produced by running engine.
Also, injectors are not involved in this test. Again, because engine is not running, injectors are bypassed and not actuated. They are closed and fuel running inside the rails is skipping the injectors.

I can send you a copy of the spec book pdf if you want. I have it OCRed, which allows for text searches. PM me your email if you want it.

[mysocal911]

Originally Posted by Discseven

Pressures measured at stock rail port:



Pump ON, engine OFF. Air is bled. Calling this 38.5 psi. Spec = 36.26 psi +/- 2.9 psi Within spec. "Spec book" am referring to here is page provided by Ivan --- TY.




Engine now at idle. See this as 34 psi. Spec = 29 psi. Over spec by 5 psi.

Assuming test gauge just used at rail is correct, installed Radium gauge in fuel line is off (high) 3 psi.






Flow test. Hose is connected to rail port. Pump ON for 60 seconds. From spec book, 60 second flow is 1,700 cc / 1.79 quarts. Being at at 3.5+ quarts, flow is measurably over spec.

I have original Bosch injectors that were removed from engine in 2020. Given uneven look of plugs and down pipe temps, am leaning towards Lucas injectors being questionable.

Yes!

Originally Posted by Discseven

Monday, am sending Bosch injectors to Mr. Bill to test & refurb. Provided they come back in good form, idea is to have them on hand to eventually test them against Lucas (from FiveO) that are currently installed and now suspect (without proven guilty as it stands. I know better that to think one thing is wrong here. Could be a combo and so testing this & that to rule things in or out.) Sal suggested swapping injectors side-to-side... running engine... then having a look at plugs again.

Still have to sort out opening mixture screw as Ivan suggested. 3mm Allen has temporarily escaped me.
.

You mean just cleaned. They can't be "refurb", a misconception by many.

[76FJ55]

Originally Posted by wazzz

What this test is measuring here is the pump max flow, and it appears that the pump is good. No regulator is involved, since engine is not running. Pressure regulator and damper work only with vacuum produced by running engine.
Also, injectors are not involved in this test. Again, because engine is not running, injectors are bypassed and not actuated. They are closed and fuel running inside the rails is skipping the injectors.

I can send you a copy of the spec book pdf if you want. I have it OCRed, which allows for text searches. PM me your email if you want it.

This is not entirely correct. the test is actually testing the fuel flow at engine off rail pressure which is ~36 psi. the fuel flow created in this scenario is the amount of fuel the pump can push at that rated pressure. This specification is drastically different than than asking for fuel flow at 0psi.

For instance if you have a fuel pump for a carded engine (~5psi) and you test it as described earlier in this post with the fuel rail open you may get a result around 2 g/m, however if you use this same pump and test it on the outlet side of the regulator which requires 36psi to open, you will have 0 fuel flow since the 5 psi generated by the carburetor pump is insufficient to force the fuel through the regulator.

the idea of this test is to test the minimum flow produced by the pump at the maximum fuel pressure (no vacuum at the regulator).

Originally Posted by wazzz

Pressure regulator and damper work only with vacuum produced by running engine.

This statement is wrong. the vacuum supplied by the engine to the regulator specifically lowers the fuel pressure. The highest fuel pressure is when there is no vacuum supplied to the regulator Either engine off, vacuum source disconnected or effectively at WOT.

[wazzz]

Yes, sorry for the misconception and wrong information. I overlooked the test setup and thought the fuel flow was measured right at the feed line.
Thanks!

[Discseven]

...Sensor gets very hot very quick.



First bench of 4.9 wideband. Put a lit match at tip of sensor and data dropped. Still needs much attention but given things are ON and data is moving in the right direction... is one step closer to sorting this MPG puzzle out.

Power supply here = Arduino 5v / connected to laptop. Lambda shield 12v / connected to car's battery.
Display = 0.96" OLED. Plan is to mount this where Fasten Seatbelt plaque currently is. (TY Joe.) Bezel designs in Onshape then 3D prints.
Data = Lambda on top line. AFR on bottom line. (Screen is focused. iPhone pic blurred it.)
.

[Discseven]

First in-exhaust test of 4.9 - engine at standing idle / 800 rpm

Plugs: W6DPO - cleaned since prior photo
Injectors: Lucas low impedance 24lb. from FiveO
Mixture screw: 3 ¾ turns out now - was just recently adjusted from being 2 ¼ turns out.
Wideband: ArduinoUNO w/Bylund Shield & sketch + u8g2lib.h + SPI.h —- no mods yet
Sensor: Bosch Lambda 4.9
Power: Arduino via laptop. Shield via car battery

OLED Display = AFR top line / Lambda bottom
Data = 1-2-3 side (SSIs)


Only the Bosch sensor is installed here. All else is sitting on the ground next to the car. Is a progressive bench test to see what numbers are displayed, how the display is working, see if any glitches exist to be sorted out. If there are glitches, what scale are they? What I’m after here is to decide if the system is worth duplicating for 4-5-6 side of engine. Or, does it go in the bin? If it works well and I can sort problems out, I’ll pull all the cables and do all the in-car wiring once.

(Why bother? All this is for the purpose of displaying data on a 0.96” display. ZADA TECH offers a miniature display system using 4.9 but one has to choose between seeing Lambda or AFR data. The USA rep for ZADA is a Porsche guy, Joe, aka “Joesmallwood” on Pelican. Excellent to deal with IMHO. Would certainly be simpler to go the ZADA route. I fancy torturing myself and so the design/build route.)


Start engine… then display is shot at noted minutes from engine start. While these are static images of the display, in real time the numbers are in constant motion. I did not stay by the display to study the range as the setup was too near the exhaust. Only got near when the timer went off to shoot the display.


2 min.


5 min.


10 min.


15 min.


Temp at 15 min.


Not saying AFR/Lambda figures are correct here. Rather, they are what they are for the time being, notably at a standing idle only.

Plan now is to swap out installed plugs for Iridiums, and installed Lucas injectors for original Bosch injectors---provided Bosch injectors test & clean well by Mr. Injector. Then repeat this interval test with no adjustment made to mixture or idle screw, or anything else other than as noted. When replacing plugs, will do compression test to rule out plugs---as seen in last photo---suffering from any excess oil in cylinders. Will be a few days before I hear on the Bosch injector status from Mr. Bill in Idaho. Iridiunms arrived all gapped to 0.8mm.

Going through this testing, replacements, testing, then adjusting whatever is to be adjusted according to the numbers seen be the numbers right or wrong... THEN checking the city mpg... those mpg results will be revealing.

As I finish these notes, realize it would be possible to test AFR/Lambda with the engine under load provided a 4.9 extension cable long enough to reach from the wheel well into the cabin. Simple enough to Blue Tape cables to car's exterior. Would be interesting to see "loaded" figures with current plugs & injectors... Other fish to fry so will pass on this.
.

[Discseven]

Update...

5 of the 6 Bosch 0280 150 158 injectors I sent to Mr. Bill tested good. 1 failed. Found another and sent that to Mr. Bill to see if that works.

Arduino based wideband system is in progress. Diagram completed. Waiting on various electronic parts to arrive.
.

[Discseven]

Wideband controller for both exhaust sides is done. 4-5-6 side in SSI needs a hole put in and bung installed. Wanted to position 4-5-6 sensor exactly the same as on the 1-2-3 side but is questionable whether it's possible to drill a hole on the top side of the exhaust tube with the SSI installed on the engine. Nothing appealing about removing that exhaust section. Need a low profile 90 degree drill bit fixture. Also on menu is finding a stainless welder willing to get in under the car. And all the cabling needs to be run.




My original Bosch 0280 150 158 injectors serviced by Bill. As always, Bill is a pleasure to deal with.
.

[Discseven]

I've been flogging a dead horse for the past weeks. Doing so, I've found Bylund Automotive's shield and other products of his being useless. The Bylund forum is an empty shell. All Bylund YouTube vid's have comments shut down---a sure tell that I missed because I wanted this gear to work. And all of what was invested was because I wanted wideband data displayed in a tiny 0.96" OLED that fit into the Fasten Seatbelt plaque space. I'd flog myself but "win some, loose some" comes to mind and that is more appealing space.

Have been holding off on swapping plugs and injectors wanting to see what the O2 status was under various engine loads with the existing W6DPO plugs and Lucas injectors. A baselining of sorts. I figured to be past that stage by now but the wideband build... I don't give up easily, sometimes to a fault.

What the course is now depends on a response from Zada. I will finish this project.

[Discseven]

Am continuing this project in new thread. Dual Innovate wideband is now installed...

https://forums.pelicanparts.com/porsche-911-technical-forum/1170007-wideband-data-3-2-getting-10-mpg-city-what-do-you-think.html
.