Ignition timing and vacuum advance?

[pekka]

Ok. Lets imagine that there is a programmable ignition that has the option to use vacuum advance to control ignition based on engine load. Vacuum advance should increase mileage when "just cruising". Maybe not much but some anyway.

How would you use the engine load signal to change the timing?

Stock timing is somewhere around 10deg below 1200RPM and about 35deg above 3300RPM.

How much advance can the XS safely take when cruising at part throttle? Say going 50MPH in 5th gear (~3200RPM)?

Would it help performance to retard the timing to less than 35deg when going to WOT from part throttle cruise?

Is there any way to safely test the limits of timing other than a knock sensor?

Thanks,

Pekka

 

[peanut]

blimey thats a step backwards isn't it ? reminds me of my MGB and Morris Minor

Why not use one of the dozens of off-the -shelf electronic advance ignition systems

 

[pekka]

Maybe I should have said MAP sensor instead of vacuum advance. Not talking about rotating the distributor...

Why not use an off-the-shelf ignition? Because I can build my own.

 

[scabber]

Interesting concept. Build one.

 

[TwoManyXS1Bs]

Interesting. Been cogitating the same thing for quite some time. Long ago, spent a lot of time fine-tuning the "timed port" driven vacuum advance on my Holley carb. Was able to get 32 mpg on my 327 sbc Vega.

Was thinking of making a kind of throttle position linkage that would do this in reverse, retard the ignition timing gradually once past about 1/3 throttle. It's still in "brain mush" mode...

 

[peanut]

Why not use an off-the-shelf ignition? Because I can build my own.

well if you are that clever why not build a fuel injection system instead

 

[pekka]

well if you are that clever why not build a fuel injection system instead

Because I do not want to. Any comments on ignition timing?

 

[peanut]

sorry I didn't mean to knock your intentions although I didn't understand your motives at first.
Sometimes a project needs following as much for the challenge and satisfaction of doing it as it is for the result. Its all about the journey
good luck

 

[TwoManyXS1Bs]

Because I do not want to. Any comments on ignition timing?

My thinking was no more than 42°-45° at highest manifold vacuum.
And, something like 30°-35° at WOT, when below 4,000 rpm.

There's info out there on appropriate timings based on combustion chamber design.
One chart is in here: http://www.xs650.com/forum/showthread.php?t=45197

I've had this idea of a worm-drive, precision-fit points timing sub-plate, rotatable for this 10°-15° range, with a lever coming out the left points cover. Finger movable, or by cable to left handgrip.

Then, go out and play. Find out what goes zoom, or boom...

 

[pekka]

My thinking was no more than 42°-45° at highest manifold vacuum.
And, something like 30°-35° at WOT, when below 4,000 rpm.

There's info out there on appropriate timings based on combustion chamber design.
One chart is in here: http://www.xs650.com/forum/showthread.php?t=45197

I've had this idea of a worm-drive, precision-fit points timing sub-plate, rotatable for this 10°-15° range, with a lever coming out the left points cover. Finger movable, or by cable to left handgrip.

Then, go out and play. Find out what goes zoom, or boom...

Pretty much the same as I've been thinking. After reading stuff online... I found your thread by accident. Good info on the combustion chambers.

EDIT: The hand adjustable timing should not be too "dangerous". I mean, that's how it was done back in the day. A friend of mine still rides a Shovel with left handgrip turning the points plate. It has not gone BOOM yet.

One "easy" option might be to use a Harley VOES switch and two ignition maps instead of a true MAP sensor. The two maps could be something like:

Full WOT advance at little later RPM and plenty of advance early at part throttle?

peanut: The idea of this thread was more to get a discussion going. The "why" and "is it really worth it" are not the questions here. Rather "how does it work", "what is the benefit" and "when do we do it" But all is good. This thread follows the unwritten laws of any internet discussion. All good.

 

[pekka]

I've been running 4-pin HEI for three seasons now. No problems with it but the advance mechanism is wearing out and WOT timing is all over the place.

The reason why I really asked about the vacuum advance was that after couple of years of thinking about it I ordered a 8-pin HEI module and an Arduino uC board. I've got the software all figured out. Now I just need to build it and actually make it work... The load based timing would give it a nice little touch compared to "the dozens of other electronic ignitions available off-the-shelf" .

8-pin HEI module should be a good base since it can run at by itself with base timing and can be controlled externally to advance or retard as desired. Also it has all the necessary power electronics, protection circuits, signal conditioning etc. built in.

Pekka

 

[TwoManyXS1Bs]

I don't think a "one size fits all" would apply here. Would have to be tuned for 'your' configuration.

The configuration variables would include:
Compression ratio.
Induction volumes at various rpms. Influenced by exhaust, cam, headwork, carburation.
Fuel quality. Ambient temps.
Elevation, especially if above 8,000ft MSL.

Rather than a knock sensor, I'd love to have a gadget that would display chamber pressure graphs, like those used in SAAMI firearms cartridge/propellant development. Strain gauge stuff...

 

[pekka]

Yeah. Higher compression. Less advance needed and so on. Ambient temp would be an easy variable to add. I wonder how Harley did the VOES? Does it account for the temp or elevation changes?

According to the internet the car people are doing this by the ear. Like: "If it pings under load you should retard the timing"

How about running a piece of fuel hose from the engine to inside of your helmet? That should work almost like an engine stethoscope?

One option would be to add for example 5degrees of advance over the stock at light load and see if it brings any noticeable benefit.

 

[TwoManyXS1Bs]

Well, whattyaknow, there's stuff already out there. A Google search of "combustion pressure sensor" yields plenty of info. Sparkplugs with sensor:

Reading the pressure curve (blue line below), the simple goal would be to have:

(area under the combustion curve) minus (area under the compression curve) = Max you can get.

Then, you could ignore all the other stuff, engine/induction/rpm variables.
But, you'll need a crank position sensor...

 

[TwoManyXS1Bs]

A simple code synopsis would be like this:

Init:
Squeeze = 0
Bang = 0

Loop:
Read(CrankPos)
Read(Pressure)

If (-90 < CrankPos < 0) then
Squeeze = Squeeze + Pressure
Else
Bang = Bang + Pressure
Endif

If (CrankPos = 90) then
Do the (Bang - Squeeze) comparison, adjust timing as needed
Goto Init
Endif
Goto Loop

 

[pekka]

Ok. Simple enough... I think I'll stick to measuring vacuum.

Turns out that actual MAP sensors are cheaper than HD VOES switches. Just need to do some experimentation with the intake vacuum vs. MAP output signal.

 

[howardsmed]

Ok. Lets imagine that there is a programmable ignition that has the option to use vacuum advance to control ignition based on engine load. Vacuum advance should increase mileage when "just cruising". Maybe not much but some anyway.

How would you use the engine load signal to change the timing?

Stock timing is somewhere around 10deg below 1200RPM and about 35deg above 3300RPM.

How much advance can the XS safely take when cruising at part throttle? Say going 50MPH in 5th gear (~3200RPM)?

Would it help performance to retard the timing to less than 35deg when going to WOT from part throttle cruise?

Is there any way to safely test the limits of timing other than a knock sensor?

Thanks,


Hi Pekka

if you are using a 360 engine format, then the ignition pickups available from Pamco and Jerry Heiden which are Hall effect, can be used with an Ignitech box which is programmable. the Ignitech box allows for a MAP sensor, so you can run an XYZ ignition instead of just advance/rpm. I have done exactly the same thing for an old ironhead HD. the Ignitech boxes are ever so cheap around E150
I am going the same route as you suggest on an 880cc XS650 I have built his summer

Howard

Pekka

 

[peanut]

Hi Howard welcome to the forum .

...I've been referring folk to your website for a while now ,hoping that one day I'd see you posting here . Looking forward to reading more from you hopefully , work commitments permitting of course

 

[mrriggs]

I know a car with quenchy-wedge chambers like 50+ degrees advance when cruising so I wouldn't be surprised if our archaic hemi with dome pistons didn't want 60 degrees or more under light load.

I started developing a digital ignition for the XS with the primary goal of incorporating a load based ignition curve to improve mileage. Life got in the way of it's completion.

I'm currently building one for my '84 LTD LX. It has two curves; a high-load curve that is used when manifold vacuum is below 6"Hg and a low-load curve that is used when vacuum is greater than 20"Hg. It interpolates the advance based on both curves when manifold vacuum is in between. I've also incorporated a digital EGR control to [hopefully] squeak a few more MPG out of it. The coolest part about the system is that all adjustments can be made on the fly with a hand-held programmer. I dug up an old "Tripminder" gauge for the dash that displays instantaneous MPG so I can adjust the advance and EGR in real time.

 

[mrriggs]

Don't get your hopes up about using cylinder pressure feedback to adjust ignition timing. When you look at a snapshot of a single pressure cycle it looks like it would be fairly simple. See if you can find a chart that overlays ten consecutive cycles. It'll be all over the map and chances are you will even see a misfire or two. If there are three or four well timed cycles in the ten then you are looking at a strong running engine.

Harley used ion sensing coils to detect misfire. You can hook a scope to the sensor output and get a pretty good picture of cylinder pressures. Much cheaper than a pressure transducer setup.