How does this target AFR table look?, VE table results after autotune Options

[rwilner]

One is that both vacuum advance and retard need to be taken into account

This is an interesting point. The MS takes into account only Manifold Pressure "position", i.e. the instantaneous MP, to calculate advance. However, the factory arrangement takes into account MP"change in position", i.e. whether it's increasing or decreasing, to calculate advance. In this way, the factory setup is actually more sophisticated than MS! (IMG:style_emoticons/default/blink.gif)

two is that the factory chart is listing the vacuum at the distributor ports which is not the actual manifold vacuum as these ports go to various locations on the throttle body so they dont translate directly into a table based on manifold vacuum.

This is also a great point. For the manifold vacuum sense pickup, I've connected the port on the plenum just downstream of the throttle body to the MPS. Is this the best place to measure manifold pressure? Would it be better to use one of the ports further down on the plenum (ex. where the vaccum elbow connects), or better yet to install a nipple someplace special?

[ChrisFoley]

This is an interesting point. The MS takes into account only Manifold Pressure "position", i.e. the instantaneous MP, to calculate advance. However, the factory arrangement takes into account MP"change in position", i.e. whether it's increasing or decreasing, to calculate advance. In this way, the factory setup is actually more sophisticated than MS! (IMG:style_emoticons/default/blink.gif)

Analog vs digital. The sampling rate of MS should be fast enough that instantaneous mp measurements are sufficient to at least be the equal of an electro-mechanical device connected to a pcb filled with resistors and transistors.
Ie., the MPS "sophistication" was needed to overcome the shortcomings of rudimentary electronic circuitry.

[rwilner]

Analog vs digital. The sampling rate of MS should be fast enough that instantaneous mp measurements are sufficient to at least be the equal of an electro-mechanical device connected to a pcb filled with resistors and transistors. Ie., the MPS "sophistication" was needed to overcome the shortcomings of rudimentary electronic circuitry.

Chris,
Let me see if I understand your point:

For the MS, each value of KPa and RPM will result in a single commanded advance. This advance is commanded instantly because it's done digitally in software (at least, it is with my crank-fired setup).

For the factory setup, each value of KPa and RPM will result in TWO possible commanded advance values, based on whether manifold pressure is increasing or decreasing. The retard was needed because the elecro-mechanical limits of the system did not enable the "advance only" numbers to be achieved fast enough under decreasing MP conditions.

[KaptKaos]

(IMG:style_emoticons/default/popcorn[1].gif)

(IMG:style_emoticons/default/bye1.gif)

Subbed.

[McMark]

The retard connection is manifold vacuum and the advance port is at the throttle plate. At zero throttle the retard side of the vac canister has full engine vac and the advance side has no vac, so you get full retard. As the throttle opens you get decreasing manifold vac and increasing advance vac. At some point the venturi effect produces more vacuum on the advance side of the vac can. Did I get that right?

It's an interesting and complicated setup. Another side benefit I see is that on overrun/deceleration the timing is retarded which is could be a good idea when leaning out on overrun.

[Richard Casto]

Hey Richard! Make any progress on your resto lately? I checked your thread...been over a year since you updated it...

Been crazy busy with other stuff. So much so that even my wife said I should take some time to step way and actually work on the 914 to help me decompress! The weather is supposed to be decent this weekend, so I will probably work on it for the first time in a number of months!

Richard

[rwilner]

The retard connection is manifold vacuum and the advance port is at the throttle plate. At zero throttle the retard side of the vac canister has full engine vac and the advance side has no vac, so you get full retard. As the throttle opens you get decreasing manifold vac and increasing advance vac. At some point the venturi effect produces more vacuum on the advance side of the vac can. Did I get that right?

It's an interesting and complicated setup. Another side benefit I see is that on overrun/deceleration the timing is retarded which is could be a good idea when leaning out on overrun.

In control systems land we call this "hysteresis," which means that the system output is a function not only of the instantaneous input but also some finite number of past inputs.

The basic concept is illustrated via the graphs below. The second one shows the hysteresis, which is how I believe the factory setup works.

[ChrisFoley]

To me, hysteresis is when my wife says its cold and gets up to adjust the thermostat, then a little while later she says shes too hot and gets up to lower the setting on the t'stat - and so on ad infinitum. If the temperature changed instantaneously to match the t-stat setting there would be no need for her to ever get hysterical. (IMG:style_emoticons/default/rolleyes.gif)

[rwilner]

To me, hysteresis is when my wife says its cold and gets up to adjust the thermostat, then a little while later she says shes too hot and gets up to lower the setting on the t'stat - and so on ad infinitum. If the temperature changed instantaneously to match the t-stat setting there would be no need for her to ever get hysterical. (IMG:style_emoticons/default/rolleyes.gif)

actually...thermostats have a built-in deadband...this is a special case of intentional hysteresis.

(IMG:style_emoticons/default/biggrin.gif)

[914_teener]

To me, hysteresis is when my wife says its cold and gets up to adjust the thermostat, then a little while later she says shes too hot and gets up to lower the setting on the t'stat - and so on ad infinitum. If the temperature changed instantaneously to match the t-stat setting there would be no need for her to ever get hysterical. (IMG:style_emoticons/default/rolleyes.gif)

actually...thermostats have a built-in deadband...this is a special case of intentional hysteresis.

(IMG:style_emoticons/default/biggrin.gif)

(IMG:style_emoticons/default/popcorn[1].gif) (IMG:style_emoticons/default/popcorn[1].gif)

Interesting comment by Chris....particulary with the wife thing...
I am assuming still discussing timing here with MS:

Are you suggesting Rich that perhaps the stock setup does a better job and retarding or advancing spark than than the circuitry interpolation on the table?

[rwilner]

Interesting comment by Chris....particulary with the wife thing...
I am assuming still discussing timing here with MS:

Are you suggesting Rich that perhaps the stock setup does a better job and retarding or advancing spark than than the circuitry interpolation on the table?

Not exactly.
I'm suggesting the factory setup has an additional retard feature which compensated for a built-in limitation of a mechanically-advanced system. The retard feature is not needed with a fully electronic system because it is not subject to the same mechanical constraint (a distributor).

I guess I am also suggesting that motor performance does not benefit from factoring in change in manifold pressure. Stated differently, instantaneous manifold pressure is enough information to compute optimal advance, assuming you can achieve that advance instantly.

[aircooledtechguy]

I think some among us are WAY over thinking this stuff. (IMG:style_emoticons/default/av-943.gif)

What is manifold pressure?? Vacuum! Those kpa#s are all just various levels of vacuum until you reach 100kpa. That is atmospheric.

To even suggest that a traditional vacuum signal going into a vacuum can on the side of a mechanical dizzy can out perform in ANY way a completely digital and infinitely tunable modern MAP system is ludicrous.

MS is instantaneous advance. There is always a slight lag with anything mechanical or even electro/mechanical. Once you get that motor dialed in with either an EDIS or COP ignition system, you will look at ANY distributor as a HUGE compromise. Even a tunable one like a Mallory does not compare.

Just as carbs are a compromise compared to tunable EFI system, so is a dizzy to tunable fully digital ignition.

[rwilner]

I think some among us are WAY over thinking this stuff. (IMG:style_emoticons/default/av-943.gif)

What is manifold pressure?? Vacuum! Those kpa#s are all just various levels of vacuum until you reach 100kpa. That is atmospheric.

To even suggest that a traditional vacuum signal going into a vacuum can on the side of a mechanical dizzy can out perform in ANY way a completely digital and infinitely tunable modern MAP system is ludicrous.

MS is instantaneous advance. There is always a slight lag with anything mechanical or even electro/mechanical. Once you get that motor dialed in with either an EDIS or COP ignition system, you will look at ANY distributor as a HUGE compromise. Even a tunable one like a Mallory does not compare.

Just as carbs are a compromise compared to tunable EFI system, so is a dizzy to tunable fully digital ignition.

Nate
I agree with you 100%. My point was that the retard is the factory's "fix" to overcome the (mechanical) limitations of the distributor.

[914_teener]

I think some among us are WAY over thinking this stuff. (IMG:style_emoticons/default/av-943.gif)

What is manifold pressure?? Vacuum! Those kpa#s are all just various levels of vacuum until you reach 100kpa. That is atmospheric.

To even suggest that a traditional vacuum signal going into a vacuum can on the side of a mechanical dizzy can out perform in ANY way a completely digital and infinitely tunable modern MAP system is ludicrous.

MS is instantaneous advance. There is always a slight lag with anything mechanical or even electro/mechanical. Once you get that motor dialed in with either an EDIS or COP ignition system, you will look at ANY distributor as a HUGE compromise. Even a tunable one like a Mallory does not compare.

Just as carbs are a compromise compared to tunable EFI system, so is a dizzy to tunable fully digital ignition.

Nate
I agree with you 100%. My point was that the retard is the factory's "fix" to overcome the (mechanical) limitations of the distributor.

No.....I think the discussion was going on with the initial mapping for the advance curve and the effect of the retard on the AFR...and why the stock setup used a retard can.

The concern with MS was the relationship of timing and head temps during diferrent loads on the engine...I believe?

Some good discussion I thought.

Good luck with your inital setup Rich. Keep us posted. I am assuming this inital setup is for your 1.7 first?

[rwilner]

The concern with MS was the relationship of timing and head temps during diferrent loads on the engine...I believe?

correct!

Some good discussion I thought.

Good luck with your inital setup Rich. Keep us posted. I am assuming this inital setup is for your 1.7 first?

actually I have a 2.0. I ran autotune tonight and it worked great!!

will post results tomorrow.

[rwilner]

ok...my new AEM setup is working great. The dash gauge is functioning and the feed to MS matches the gauge.

I ran autotune using the updated AFR target table.

VE table before autotune:


VE table after autotune:


In some areas there were major changes, in others, small changes. No changes over 4k rpm because I didn't rev the engine that high.

I tuned this by taking a 45 minute drive through back roads where I could floor it in several gears and also on some hilly highway at varying loads.

Car revs much more smoothly and cyl head temps never got above 300, that's on premium gas though. I'm going to try midgrade and see what happens.

Overall I'm very pleased with the AEM setup and autotune.

[falconfp2001]

ok...my new AEM setup is working great. The dash gauge is functioning and the feed to MS matches the gauge.

I ran autotune using the updated AFR target table.

VE table before autotune:


VE table after autotune:


In some areas there were major changes, in others, small changes. No changes over 4k rpm because I didn't rev the engine that high.

I tuned this by taking a 45 minute drive through back roads where I could floor it in several gears and also on some hilly highway at varying loads.

Car revs much more smoothly and cyl head temps never got above 300, that's on premium gas though. I'm going to try midgrade and see what happens.

Overall I'm very pleased with the AEM setup and autotune.

Where did you ground the gauge? Also, how fast does it start to read? How long is the cabling?

[rwilner]

It starts to read within 30 seconds of starting up. The cabling is maybe 2 runs that are 7 feet long -- the sensor connects to the gauge, and the gauge has an output that feeds the microsquirt which is mounted in the stock ecu location.

The signal output ground is grounded at the signal ground of the microsquirt through the seven foot run. The gauge power ground is connected to a bolt that I installed on the center tunnel.

[falconfp2001]

It starts to read within 30 seconds of starting up. The cabling is maybe 2 runs that are 7 feet long -- the sensor connects to the gauge, and the gauge has an output that feeds the microsquirt which is mounted in the stock ecu location.

The signal output ground is grounded at the signal ground of the microsquirt through the seven foot run. The gauge power ground is connected to a bolt that I installed on the center tunnel.

What is the total cable length that comes with the AEM? My Megasquirt is under the dash and the gauge is off the ash tray. I need it to go from the muffler to the dash. Sorry if I'm being a pest.

[rwilner]

What is the total cable length that comes with the AEM? My Megasquirt is under the dash and the gauge is off the ash tray. I need it to go from the muffler to the dash. Sorry if I'm being a pest.

no problem.

I think you need only the O2 sensor cable to reach that far, and it will easily reach that far. Even if for some reason it didnt, I believe there is an extension available.

The power and signal wires are short but they might reach since your gauge and ECU are so close. If not, they're easy to extend.

The AEM website should have specs on the cables.