Overflow on VE

[Tony Cockshutt]

Just started doing VE learns and I'm getting a few cells changed to 100% and a message "overflow on VE:  2400RPM 80kpa".  I searched this forum and found a thread from 2010 discussing this situation and the suggest was to increase BPC a small amount, reduce the 100% cells and do another VE learn.  Is this the correct change?

My setup is:  455 with new heads and headers.  Affordable EFI with EBL purchased from Applied.  

I've attached the Bin created today.  The corresponding log file is 8mb and seems to be too long to include.  I can create a shorter one tomorrow if that helps.

Thank you

Tony

[Tony Cockshutt]

VE and spk tables

[Bill Van Vlack]

The file that Applied supplies with their EBL package has a 'feature' of too high VE numbers in the high MAP cells..

Here's a BIN that was revised for me by Randy Van Winkle. as a good starting BIN for VE Learns.

https://drive.google.com/file/d/1txJRv6pWe1MmP4wYD6EXuPXAPh1672Nu/view?usp=drive_link

A snapshot of the starting VE table and BPC table  is attached

[Tony Cockshutt]

Thank you.  I'll flash it in tomorrow and give it a try.

--
You received this message because you are subscribed to the Google Groups "GMCMH EFI" group.
To unsubscribe from this group and stop receiving emails from it, send an email to gmcmh-efi+...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/gmcmh-efi/b9884da5-6c68-499b-90ec-c8d30a0ffcd1n%40googlegroups.com.

[Tony Cockshutt]

Went for a 11 mile drive.  VE Learn seems to be heading to the same outcome.  I'm at 99 for 2200 and 80kpa.  Here are the tables:

[MICHAEL ORLANDI]

What TBI unit are you using? What is the fuel pressure? What is the inlet diameter of the feed line going into the TBI?

--
You received this message because you are subscribed to the Google Groups "GMCMH EFI" group.
To unsubscribe from this group and stop receiving emails from it, send an email to gmcmh-efi+...@googlegroups.com.
To view this discussion on the web visit

https://groups.google.com/d/msgid/gmcmh-efi/5bdffb8d-d172-49bd-b4a3-1fc1c06b08f2n%40googlegroups.com

.

[Tony Cockshutt]

Rochester TBI, 13psi verified with gauge this morning, 3/8 mostly steel line 

[MICHAEL ORLANDI]

Take your fuel filter out and try to blow through it. Any restriction replace it.  Tie a fuel gauge into the system at or near the TBI and go drive it up a steep hill. You should not lose any fuel pressure when driving. If you do you have a bad fuel pump or a leaking hose.

https://groups.google.com/d/msgid/gmcmh-efi/85c48077-8a31-4eeb-b33a-e18286f120f4n%40googlegroups.com

.

[Bill Van Vlack]

Tony,

You could try that;  Pulse Width per injector firing, PW = BPC x VE% x BLM%.  If you raise BPC a little, VE's after a VE Learn should come down for the cells you've hit. Also, BLM's should be close to 100%  during a run. This means that the NBO2 crosscounts are balanced lean vs rich and there's no more adjustment needed to VE. 

Take a look at %duty cycle on the log. If it's never close to 100% in closed loop and the O2 readings look reasonable, you're getting enough fuel.

[Tony Cockshutt]

Thanks Michael.  I replaced the filter just 100 miles ago and cut the old one open.  Clean as a could be.  I'll try the hill and see what happens.

To view this discussion on the web visit https://groups.google.com/d/msgid/gmcmh-efi/364823522.1610423.1724965555148%40mail.yahoo.com.

[Tony Cockshutt]

Very good.  I'll try increasing BPC.  Should I do 5% increase across the entire table or just the cells that are over 90?  Duty cycle seldom gets over 50%.  Max appears to be about 80% during WOT acceleration.

I'm not sure what's happening with the indicated fuel pressure.  Today's log file did not have a fuel pressure sensor connected.  I've not been able to get a reliable fuel pressure reading.  Two different sensors, separate 5V power supply.  However, an analogue gauge shows the correct 13 psi.

Thank you

To view this discussion on the web visit https://groups.google.com/d/msgid/gmcmh-efi/CA%2BX-fMeY%3DQG_QzXS%2B8FojixhVQY7cb9LZcoVMijqLZ53sC8zdA%40mail.gmail.com.

[Bill Van Vlack]

Don't touch the VEs; I'd just increase all the BPC table values (BPC vs Kpa). This will initially lengthen the pulse width; then as VE Learns take place the VE values should come down to lean out the mixture and shorten the pulse width back  to where it should be. This will give room for the VEs at/near 99 to adjust back up if needed on subsequent Learns.

When setting the analog point for FP you can also add a filter value. This acts as a damper to the FP readout and may keep your display FP from jumping around.

BillI

To view this discussion on the web visit https://groups.google.com/d/msgid/gmcmh-efi/CAL6hAKeWsm1Qb1q2tcedPxDj4DTBcwrFacbRmZ78h%3DT2Wgx0MQ%40mail.gmail.com.

[Tony Cockshutt]

Very good.  I'll try 5% increase to the entire BPC table and report back.

[Tony Cockshutt]

Short drive.  Learn data is better, but still remaining very close to 100% on several cells.

I'm seeing many spark knocks as well.  I don't hear any pinging, but the log is recording many knock events.  If I somehow screwed up on setting the initial timing could this be part of my problem?

Today's data

[Jon Roche]

Knocks can be tricky to know if you are getting a true ign knock/ping.    Or it is getting false readings.  

How is your knock sensor installed, and what esc module did you install/use?

How did you do initial timing?

On Aug 30, 2024, at 6:49 PM, Tony Cockshutt <to...@cymadesign.ca> wrote:

Short drive.  Learn data is better, but still remaining very close to 100% on several cells.

To view this discussion on the web visit https://groups.google.com/d/msgid/gmcmh-efi/4ff2d7fd-0224-4cab-ade3-d63c35e1f75an%40googlegroups.com.

<Screenshot 2024-08-30 194246.png>

<Screenshot 2024-08-30 194321.png>

<aug 30 2024 revised applied bpc.dat>

<Screenshot 2024-08-30 192947.png>

[Bill Van Vlack]

Did you take a log? I'm pretty sure you can do a playback and take a log.  It would be interesting to see crosscounts and WB02 numbers - if you have one.

I've found the knock counter hard to believe. I get lots of knocks on startup.

It should have done a -1 adjustment at 2000/80, but it's still 99%. As long as it's not 100% you're OK.

To view this discussion on the web visit https://groups.google.com/d/msgid/gmcmh-efi/4ff2d7fd-0224-4cab-ade3-d63c35e1f75an%40googlegroups.com.

[Bill Van Vlack]

PS.. You can do a VE table (or any table) compare using TunerPro to see if the WUD VE changes agree with the BIN files.

[Tony Cockshutt]

Yes, log file was included in my last post.

[Tony Cockshutt]

My knock sensor is threaded into the block.  I think its a standard GM sensor.  I'm not sure which module it is connected to.  The sensor and module were included in the Affordable kit from Applied.  Tomorrow I'll see if there's a part number on the knock module.  Why are you asking about the module?  Are some more reliable than others?

Thank you to all who have been asking questions and suggesting things to try.  Very helpful.

Tony

To view this discussion on the web visit https://groups.google.com/d/msgid/gmcmh-efi/03436A87-77DA-4F7A-8798-3728EE26FB3E%40gmail.com.

[BruceHislop]

The Knock sensor is just a "microphone"  The module is tuned to the Knock sound (frequency) of the engine which changes between engine sizes based mostly on the cylinder diameter. So it's important that the module is matched to your engine.


During startup, there are lots of noises from the starter, ring gear, drained-down lifters and other loose valve train noises. These settle down right after startup.

Bruce Hislop

To view this discussion on the web visit https://groups.google.com/d/msgid/gmcmh-efi/CAL6hAKfFXuwYDSte59ovLjsqaVWZdcJPJcjYZUogpKHPanR_UQ%40mail.gmail.com.

[Bill Van Vlack]

A couple things from the Log;

  WB02 doesn't appear to be working; it stays at 19.5-19.7 AFR.  Even crosscount periods should be stoich.

  Fuel pressure starts at 5psi and rises to 20psi. Should be steady

  IAC at idle should be 10-20 steps.

  AC comes on early and stays on. Might want it off for testing; it messes with fueling.

  The bin is set for 6deg initial advance.

To view this discussion on the web visit https://groups.google.com/d/msgid/gmcmh-efi/CAL6hAKfFXuwYDSte59ovLjsqaVWZdcJPJcjYZUogpKHPanR_UQ%40mail.gmail.com.

[Tony Cockshutt]

Thank you for reviewing the log.  Some additional info:

-I don't have a wideband O2 sensor. 

-Fuel pressure is indeed wacky.  I've tried two different sensors and 2 power supplies.  I even get readings without a sensor connected.  Something is wrong in my connections.  I need to trace every wire to see what's happening.

-I'll reset IAC before next test

-I've been using the AC sometimes. Good to know I should be testing without it.

-I can't remember if I set initial advance to 6 or 8 degrees.  I'll check it and set to 6 if it isn't already there.

Thank you

Thanks also to Bruce for knock module explanation. 

[Cameron]

Tony,

If you resolve your fps issues, please share!  I have had the exact same experience as you are describing.  Hooked up a manual tester and pressure was fine.  But through the ebl I can’t get a reliable reading.

Cameron

To view this discussion on the web visit https://groups.google.com/d/msgid/gmcmh-efi/b499255a-23d4-4754-a306-3356bc350968n%40googlegroups.com.

[Tony Cockshutt]

Cameron, I have been following your fuel pressure troubles hoping to hear of a solution.  Somehow, we'll figure it out.

Tony

To view this discussion on the web visit https://groups.google.com/d/msgid/gmcmh-efi/CAMM5UngZjoVGB152OMB5K6c96LyLBF-67to-46cQE-U%2Bkpc%3D5Q%40mail.gmail.com.

[Bill Van Vlack]

For the FPS.

 If you're getting readings with the sensor disconnected, either it's connected to the wrong ADC point, or the ADC point is bad. Try connecting 5v DC to the ADC point and change the 'Preferences' to a 0-5v sensor. It should display the sensor range; i.e with a 0-30psi gauge it should read 30psi.  Or try a >5v battery; it should read less than full range. Assign a few other ADC points the same way and try them.

On Sat, Aug 31, 2024 at 4:38 AM Tony Cockshutt <to...@cymadesign.ca> wrote:

To view this discussion on the web visit https://groups.google.com/d/msgid/gmcmh-efi/b499255a-23d4-4754-a306-3356bc350968n%40googlegroups.com.

[Tony Cockshutt]

Bill, that's a great idea to test with a battery!

[Bruce Hislop]

Tony 

Don't exceed 5Vdc on the ADC inputs or you may damage it.  I'm sure RobR put some protection against higher voltages, but best not test their limits.

If you have a potentiometer, you can connect that between 5v and ground, then connect the wiper to the ADC input and watch the results as you adjust the potentiometer.

Bruce Hislop

To view this discussion on the web visit https://groups.google.com/d/msgid/gmcmh-efi/e85b15b3-3baf-40a8-856b-15a49c755fafn%40googlegroups.com.

[Bill Van Vlack]

Whoops...  I meant <(less than) 5v battery.

To view this discussion on the web visit https://groups.google.com/d/msgid/gmcmh-efi/CAFziMfNL-j3xLYw8M_7KU9Z5yr3PUJiNjZJuHvDGeeKjy0WnuA%40mail.gmail.com.

[Tony Cockshutt]

I didn't get out for a drive today, but did some troubleshooting on the fuel pressure.  Peculiar results:

1.  the sensor I assumed was wrong, was indeed wrong.  0-30 psi sensor produced 4V at 13psi while it should have been about 1.8V

2.  then I removed the wires at the ADC board and configured every channel as an Autometer 0-30 psi gauge.  With no wiring attached the WUD reported fuel pressure readings between 6 and 18 psi depending on the channel.

3.  With no wiring attached the incorrect readings slowly increased the longer I watched.  e.g. 6psi moved to 6.4 psi over a 3 minute range

4.  with 3v battery connected to the ADC I got readings on 4 of the channels of 18.8psi which is about right for a 0.5-4.5V range.  I didn't have enough time to check the remaining channels, however when I removed the 3V the channel went back to reporting 6psi, or 7 or 8 or some other number in between.

[Bill Van Vlack]

It sounds like a bad ground somewhere. But since the 3v battery proved the ACU, try connecting the 3v battery at the harness side of the sensor connector (i.e., with the sensor disconnected).  For ground, connect a wire from the 3v battery ground to the ground point of the ECM. The WUD should read 18psi.

Put a meter between the ECM ground and the 5v supply. If not already, connect the 5v supply to the ECM ground directly.

Good Luck!

To view this discussion on the web visit https://groups.google.com/d/msgid/gmcmh-efi/9c348d74-9957-4ba4-88f5-6efae77bfb36n%40googlegroups.com.

[Keith V]

Unconnected ADC values will float and are undefined.

Connect to ground or 3v as you did

---

From: gmcm...@googlegroups.com <gmcm...@googlegroups.com> on behalf of Tony Cockshutt <to...@cymadesign.ca>
Sent: Saturday, August 31, 2024 7:34 PM
To: GMCMH EFI <gmcm...@googlegroups.com>
Subject: Re: [GMCMH-EFI] Re: Overflow on VE

 

To view this discussion on the web visit https://groups.google.com/d/msgid/gmcmh-efi/9c348d74-9957-4ba4-88f5-6efae77bfb36n%40googlegroups.com.

[Tony Cockshutt]

Well that mystery is now solved.  Thank you Keith.

In the meantime I've rearranged a few ground points.  Tested the FP wiring as Bill suggested and it works properly.  I also rechecked the initial timing.  I was at about 8 degrees.  Reset it to 6 degrees.  That sure is a convoluted job.  Trying to see the timing marks while keeping the timing light, wires and my mirror out of the fan blades is a real challenge.  Will go for a longer drive tomorrow and see how it works.  

Many thanks to everyone for their suggestions and encouragement.

Tony

[Dick Paterson]

Tony      once you set the timing  where you want it ….take a small slat of wood  (I like paint stir sticks)  place  the slat  in between the spark plug boots on the dist cap….slide it  back to touch the firewall….   Then  mark around the slats contact point   with a marker pen  …     anytime   you want to verify- or --reset  the timing  -place  the slat  in the cap  ….and  slide  it into the  marked square on the firewall.   You might want to I D  the  plug boots  you use  …with a dot of  office  whiteout.    

To view this discussion on the web visit https://groups.google.com/d/msgid/gmcmh-efi/37c2f077-95f2-4b4f-bd38-569c00552f1bn%40googlegroups.com.

[Tony Cockshutt]

Great idea.  Thank you Dick.

To view this discussion on the web visit https://groups.google.com/d/msgid/gmcmh-efi/002701dafeb9%24677d7110%2436785330%24%40paterson-gmc.com.

[Jon Roche]

Hopefully your not messing with timing often.    

I like tools.   After timing a gmc with an adjustable timing light….  You will never want a standard one again.      The adjustable one you set the light

To the timing you want(in your case 6).  Point it down and just match the pully mark with zero mark on the gauge.  You can mark both with a paint marker or something    And just line up the 2 lines.      You can usually see the zero timing mark straight down front of engine but  Trying to read further up that timing guide as you found out is tough.   The adjustable timing light makes simpler.  

On Sep 4, 2024, at 6:02 AM, Tony Cockshutt <to...@cymadesign.ca> wrote:



To view this discussion on the web visit https://groups.google.com/d/msgid/gmcmh-efi/CAL6hAKcttY4vogOONVLuc3SULksgEVgQDrmGCJ5BmWkDhkAK8A%40mail.gmail.com.

[Cameron]

Tony,

if you solved the fps, pls do share! I was considering going with
just a gauge, since having fps in the wud is not 'critical'.

thanks!
Cameron

> To view this discussion on the web visit https://groups.google.com/d/msgid/gmcmh-efi/37c2f077-95f2-4b4f-bd38-569c00552f1bn%40googlegroups.com.

[BruceHislop]

I would not use a gauge as a permanent fixture for monitoring fuel pressure unless there is a snubber in series with it.  The fuel pressure is not smooth but rather pulses with the injectors opening and closing.  This causes fatigue to the gauges internal tubing used to measure the pressure and it will eventually fail.  Also the gauge will son not return to zero due to the fatigue.

Even fluid damped gauges do not stand up to the pulses.  I believe you should be able to do a search for this info on gmcmotorhome.org in the GMCNet archives.

Bruce Hislop

[Cameron]

Thanks Bruce,

I was thinking of something like this;

https://www.maxtow.com/black-green-maxtow-15-psi-fuel-pressure-gauge/?srsltid=AfmBOor0Z9pS3-xqfuHDYX0_dxBfmPRh1hALCoA4pZ_RFh7Fhj8-OCUQ

essentially the same sort of setup for sensor from the fuel line, but
to a gauge, instead of the ebl. still a bad idea?

Cameron

> To view this discussion on the web visit https://groups.google.com/d/msgid/gmcmh-efi/78af507e-4e5d-4e4a-844c-4bbe25849133%40gmail.com.

[Tony Cockshutt]

Ok, I'm going to try to close out this thread.  To summarize, using the Revised Applied bin I had a few cells in the VE table topping out at 100%.  I increased the BPC table by 5% and reset the initial timing to 6 degrees from about 8.  Then did several short drives with VE learning turned on.  I now have one cell (2200rpm, 80 map) that is at 98, all the rest are lower.  My  reported fuel mileage has increased from 6.2 mpg to 7.8 mpg.  The cells are still changing a bit with each VE learn so probably there will be more improvement to be had.  It is very flat around here, I need to find some hills and do a few more learns.

I reconnected all the grounds.  Then confirmed the fuel pressure sensor wiring with a 3V battery.  And confirmed my cheap 0-30psi fuel pressure sensor was returning a constant 4ish volts.  I now have a steady 26.8-27 psi fuel pressure reading.  Before redoing the grounds the fuel pressure would randomly increase over a drive.  I've ordered a new FuelTech sensor and will post the results in Cameron's fuel pressure thread.

Many thanks to all of you who read my posts and made suggestions.  All the tips were helpful in some way.

Thank you

Tony

[Bill Van Vlack]

Great to hear!

Bill

To view this discussion on the web visit https://groups.google.com/d/msgid/gmcmh-efi/ab897b71-d0dc-48a5-8853-be22c31b145bn%40googlegroups.com.