Ignition timing mechanisims, centrifugal and vacuum.

[Juanito84]

**EDITED AND REVISED 2-May-2012**

IGNITION TIMING THEORY

When a sparkplug ignites the fuel in a cylinder it takes time for the fuel to burn. Because of this the fuel is always ignited before the piston reaches the top so that maxumum presure can be made in the cylinder before the piston is on it's way down. The ignition may occur as many as 5, 10, 20, 30, even 40 or more degrees BTDC (before top dead center) in some engines.

If the fuel is ignited precisely at the best moment, the fuel will finish burning at about 10 degrees ATDC (after top dead center).

If the fuel is ignited too late the timing is said to be too retarded, and the fuel will still be burning as the piston starts its way back down causing a loss in pressure and therefore a loss in torque.

If the fuel is ignited too soon the timing is said to be too advanced and the pressure in the combustion chamber will rise to early causing excess back pressure on the piston and the remaining fuel will suddenly combust causing a shock wave that wreaks havoc on the mechanical parts of the engine. This is called detonation, pinging or knocking, depending on the degree. Heavy knocking can even punch holes through the piston tops.

Therefore ignition timing is crucial to engine performance and efficiency.
However timing is not just one set number for any engine. Many factors such as engine temperature, load, RPM, fuel ratio, etc. change the timing needs of the engine. Most of these can be made considerably constant, like fuel ratios with proper jetting or engine temperatures with the thermostat. However RPMs and engine load are allways changing and need timing compensation while driving.

It is possible to make changes in ignition timing automatically because timing is controled by a small dwell cam or trigger that charges and discharges the ignition coil at specific moments for each ignition cycle. The internal parts of the distributor can adjust this triggering mechanism automatically to advance or retard the ignition. The distributor reacts to changes in RPM and engine load and adjusts the timing accordingly. Exactly how and why does it do this?

RPM changes

Speed or RPM cause the piston to move up and down faster. If ignition timing were set to just one single possition then:

As RPMs increase the timing would become too retarded. So with increasing RPMS the timing must advace to keep up.

As RPMs decrease the timing would become too advanced. So with decreasing RPMs the timing must retard.

So how does an distributor automatically advance and retard the ignition timing in sync with RPMs?

On mechanically advanced distributors like SVDA's, DVDA's and 009's this is done with centrifugal advance (aka mechanical advance) The distributor shaft has a pair of weights with springs. As the shaft spins faster and faster the weights move out against the springs and move the dwell cam or trigger mechanism to a more and more advanced possition.

On vacuum only distributors the carburetor (i.e.28 and 20 PICTs carbs) feed a RPM vacuum signal from the venturi to the distributor diaphragm. The venturi creates a highe vacuum at higher speeds and a lower vacuum at lower speeds. In turn the diaphragm turns the points plate and changes the timing.

Load changes

However RPM is not the only factor. As you step on the accelerator pedal you allow more air and fuel into the engine. This is called increasing the load. When more air and fuel are compressed together due to high load they burn faster. If ignition timing were set to just one single possition then:

As the load increases the flame would burn faster and the timing would become too advanced. So with increasing load the timing should be retarded.

As the load decreases the flame would burn slower and the timing would become too retarded. So with decreasing load the timing should be advanced.

So how does the distributor automatically advance or retard the timing in sync with the load?

On all mechanical/vacuum advanced distributors like SVDA's and DVDA's the vacuum diaphragm is controlled by manifold vacuum. As load decreases the vacuum signal increases. As load increases the vacuum signal decreases. This in turn moves the diaphragm with directly moves a plate inside the distributor with the points or triggering mechanism on it. This way it adjusts the timing in relation with load.

On all vacuum only distributors the manifold vacuum signal is combined with the venturi vacuum signal giving a combined vacuum signal that responds to both load and RPM changes.

One note about vacuum advance is that the signal is sometimes cut off (retarded) under certain circumstances. This is called a ported signal. On a ported signal the vacuum is derived from a port that recieves manifold vacuum next to the throttle plate. It is placed there in such a way that the throttle plate covers it over when you let off the accelerator pedal. This is an emissions device to lower emissions durring idling. It also helps with engine braking and can help some with starting, especially when the engine is warm. However some people actually perfer running the vacuum advance strait off the manifold vacuum.

One downfall to vacuum advance is that changes in altitude affect its accuracy. The higher the altitude the less advacement the distributor provides for load changes.

Mechanical only distributors

Mechanical advance only distributors lack one advancing device the vacuum advance distributors have: a way to adjust timing in relation with changes in load. Any mechanically advanced distributor, like a 009, can and should only be adjusted to be most efficient at full throttle. However at any part thottle the timing will be retarded. Although this will not hurt top power and speed it will hurt efficiency at any part throttle possition. This will cause poorer fuel mileage and that famous "flat spot".

Reveiw

Each ignition cycle is set off by the sparkplug at various degrees BTDC due to RPM and Load changes.

Increasing RPMs can tolerate a more advanced ignition timing.

Increasing Load will require a more retarded ignition timing.

RPM changes are normally compensated for by centrifugal weights and springs.

Load changes are normally compensated for by a manifold vacuum signal.

SVDA's and DVDA's use both mechanical and vacuum advance mechanisms.

009's only compensate for changes in RPM.

Vacuum only distributors use a combined vacuum signal from the venturi (changes with RPM) and the manifold (changed with load).

**Did you know?**
Vacuum only distributors timed the #3 cylinder's igntion in a more retarded possition than all the other cylinders. This is because vacuum only distributors were used before the doghouse cooling shroud. These early style shrouds placed the oil cooler in a spot over the #3 and #4 cylinders which caused the #3 cylinder to run hotter than all the others. Because of this the #3 cylinder needed less advance, which was provided by the vacuum only distributor's special timing.

Adjustments

Not all engines are created equal. When a car owner makes changes to his engine many times he is unknowingly changing the ignition timing curves of the engine. This is especially true when the owner changes the cam profile, the deck heighth and compression ratio. Changing the exhaust compression, fuel ratio, rocker ratio, valve size and displacement also affect the engines timing needs and limits.

One of the easiest adjustments on a distributor is the total advance. This is the setting you make when you set the distributor at idle with the aid of a timing light by swiveling to the desired position and then clamping it down.

With this single adjustment you can power time your engine to get better power and efficiency. To do this you must connect a vacuum gauge to any tap on the intake manifold, start then engine, and then open the throttle enough to engage the fast-idle cam to achieve a fast-idle speed of about 2,000 RPM. NOTE: The exact speed is not important. Now turn the distributor body to obtain the highest vacuum reading on the gauge, and then back it off slowly until the needle dips slightly from the maximum reading. The exact amount of backing off needed must be determined by the engine itself. If there is a lot of carbon in the combustion chambers you will have to retard the timing more than for a clean engine. NOTE: The more advanced you can run the ignition timing, the better the gas mileage you will obtain.

Now, road-test the car to determine if the engine pings. If it does, retard the timing a small amount and repeat the road test. Do this until you lose all traces of ping. Now your engine is timed more efficiently. NOTE: Prolonged pinging as well as heavy knocking can damage the engine!

The next step up would be to custom taylor your advance curves. This can be done by changing the weights and springs in the mechanical advance mechanism on your distributor. There also might be tabs that can be bent in or out to change the total advance of the distributor. These adjusments are best done with the aid of a dynanometer, but can be done by use of some very acurate road tests.

Vacuum advance is not normally adjustable. However there are some distributors like some Mallory Unilites that do come with an adjustable vacuum advance diaphragm.

Another extra device is a knock sensor. A knock sensor will send a signal if the engine starts to ping. This signal can be used to automatically retard the timing. This device is handy when the timing is set extremely close to perfect, since it protects the engine from damage.

For the guy that wants to get perfect timing the final step up would be to invest in a digital ignition system. These use a computer to calculate when and where to fire the plugs. RPM sensing is normally triggered by a multi-tooth crankshaft position sensor, although some use camshaft position sensors. Load sensing can be done with an MAP (manifold absolute pressure) or a MAF (mass air flow) sensor.
_________________
If a water cooled engine cools its water with air, isn't it just an overcomplicated air cooled engine?

[krusher]

I think you have manifold pressure MAP and a carb's vacuum port signal mixed up as one and the same thing, there not.

For engine load sensing.

A SDVA need vacuum signal from a vacuum port on a carb

A digital magnaspark distributor can use MAP (manifold pressure)
_________________
(06:31:07) RoachGhia: "i drink dick way too fast"

[Juanito84]

Thanks for the info. Your are right that they are not the same and SVDA distributors work best with a proper ported signal from off the carburetor. Although I may have juxtaposed them, I didn't mean to say that they are the same. A vacuum signal responds to whatever it is hooked up to. In the case of a SVDA it is either atmospheric (idle) or manifold vacuum (above idle). If you don't believe me put it on a dyno and stick a vacuum guage on each. Once the engine is above idle the two will read the same.

A MAP sensor reads from the same point of view as a manifold vacuum port except it senses absolute pressure and doesn't compared it to atmospheric pressure. A diphragm can only compare the pressures on both sides of it, in this case manifold against atmospheric. Fuel injection systems and digital ignition systems can calculate the presice load on an engine with either a MAP or a MAF sensor although the equations are considerably different.
_________________
If a water cooled engine cools its water with air, isn't it just an overcomplicated air cooled engine?

[Juanito84]

Here are some diagrams I drew. They are not to scale but serve the purpose of depicting how these systems work.

First is a 34 PICT although the 31 PICT and fuel injected modles work the same. Notice how the vacuum advance does not respond directly to engine speed. It's sole purpose is to provide vacuum advance in compensation for low load. The centrifugal mechanisms in the SVDA and DVDA do all the compensation for engine speed.

When idling and starting the vacuum advance mechanism does not advance. In fact with the DVDA it even retards it a bit more. This is mainly for emissions but also helps the vehicle start. All gasoline engines start better when the ignition is close to 0 degrees TDC since there is less back pressure during the end of the compression stroke that could stop the engine. For this reason the port (or ports) are located just above the closed throttle plate. This provides atmospheric presure that can't advance the distributor.

Next is the 34 PICT while you are just lightly cruising along. The low load on the motor can withstand much more ignition advancement, perhaps as much as 45 degrees, to produce more power with less fuel. For this reason the design of the carburetor causes the port to pass onto the manifold side of the throttle valve once it is open. Now the distributor can "tell" how much load is being put on the engine and compensate so that it advances with the lower load. The restriction of the throttle plate causes the vacuum that lets less fuel and air enter the engine and therefore lowers the load on the engine.

Now let's see what happens when you floor it. The engine would ping or even knock if it were as advanced as the part throttle position above. But the open throttle reduces the vacuum being fed to the diaphragm. Remember that the mechanical advance in the SVDA and DVDA is still advancing the ignition at the same time, but only in relation with engine speed. The vacuum advance is strickly for load compensation.

Now as you can see this has been the part that those mechanical advance only distributors, like the 009, don't accomplish. They are like an SVDA that is stuck in the high throttle or idle possition. Both 009's and SVDA's have mechanical advance. But only the SVDA has the vacuum part of it too. Sure they run, but they are less efficient unless you have to drive around at full throttle anyway.

Now let us look at the pre 1971 carburetors and distributors. They are different for the fact that they don't have mechanical advance. The whole system works on vacuum, both for speed and load compensation.

These carburetors when at idle are much like the SVDA and DVDA in one respect, they are not advanced.

Now when you step a little on that accelerator the same thing happens as with the SVDA and DVDA's, the port is now on the manifold vacuum side and so advances the distributor. But wait. The engine is not at high RPM's at the moment. This causes a low vacuum in the venturi. The combination of high and low vacuum cause the distributor to find its happy spot.

Now what happens as you pick up speed. The combination of a low load and higher RPM's will allow for even more avancement. Since there is no mechanical part of the distributor to do this it must do this by adding more vacuum. Since the speed is even higher the venturi is going to be causing a lot more suction. The high vacuum with high vacuum causes... well... high vacuum.

Now what would happen if you floored it at these high engine speeds. Although it would need some advancement for the higher RPM's it wouldn't need so much for the quicker flame caused by the higher load, that is the open throttle. Now we have a combination of low and high vacuum so now the timing finds itself somewhere in between again.

Now last but not least what happens if you lug the engine? Your engine speed is low and your load is high causing a high flame speed. You don't need very much advancement at all. But these two conditions conveniently cause two low vacuum signals. Walah! very little, if any, advancement.

Please note that the vacuum advance curves vary from engine to engine. You can't allways slap on a SVDA and expect better performance than a 009. The curves of the distributor have to match the curves of your engine. The best way to find out is to put the engine on a dyno and swivle that distributor back and forth under many different combinations of load and speed. Checking for best torque and where pinging starts to set in you can find the correct advance curves for both speed and load. RPM curves are changed by changing the weights and springs in the centrifugal (mechanical) part of the distributor. Adjustments for the load curve are possible on some distributors like Mallory Unilites by adjusting the tention of the diaphram spring.
_________________
If a water cooled engine cools its water with air, isn't it just an overcomplicated air cooled engine?

[krusher]

why dont you host this as a mini website somewhere?
_________________
(06:31:07) RoachGhia: "i drink dick way too fast"

[stevehenderson]

cb megaspark
http://www.cbperformance.com/catalog.asp?ProductID=1884

Megagolt Lite Jr -- like a megasquirt so just get it.
http://www.autosportlabs.com/
Vw parts to go with it.
http://www.thedubshop.net/shopping.php

Some graphs for existing dizzys

[Juanito84]

Thanks Steve! That's really helpful!
_________________
If a water cooled engine cools its water with air, isn't it just an overcomplicated air cooled engine?

[evilphoton]

as others have, thank you for this info!

I have a '71 with a DVDA and a 30/31. In testing with the venturi providing advance vacuum, it wouldn't advance much past maybe 10 or 15 degrees (before actually reading any of this info).

When I get the engine back together, I'm going to try your diagram.. modify the vacuum lines to combine the two of them and see if I can get more (proper) advance.

From what i've read, the 30/31 isn't made to natively work with SVDA or DVDA because of the weak venturi signal, so now I have a new path

[Juanito84]

Here is another interesting article: http://www.gofastforless.com/ignition/advance.htm There are some good true facts (not just some guy's personal opinion, but actual proovable facts) about advance systems.

One thing that I forgot to mention about manifold vs ported vacuum advance is that many professionals actually perfer manifold vacuum over ported. Perhaps the main reason is that since your engine can idle at a higher speed with less fuel it helps with cooling at idle. This is why some cars in the 70's had a thermo activated valve that would switch vacuum advance from the ported signal over to direct manifold vacuum when the engine began to overheat.

Personally here in the mountains, unless you have a way to shut off the fuel when needed (i.e. programable fuel injection), the retarding that both the ported SVDA and even more so the DVDA both create do make a noticeable difference with engine braking down steep mountainous grades and even in stop and go traffic. But then again that is just a personal preference.

_________________
If a water cooled engine cools its water with air, isn't it just an overcomplicated air cooled engine?

[jtwaller]

I would love to hear your opinions on the aircooled.net SVDA. I own one and absolutly love it! I am running it on a weber progressive on a 1904cc. My real question is the timing of it. ACN tells me to time it to 28* with the vacuum unplugged 38*-41* with it plugged in. Everyone I have talked to says that 38*-41* is too advanced for an aircooled engine and I will overheat it. I have a recent incident where my local shop knowing how my engine was set up, "recurved" the distributor retarding the distributor (bending in the tabs) and timing the distributor to 30* total advanced with the vacuum plugged in. My engine ran 20* hotter and my oil light came on at every stop light. I reversed everything and now I'm back to perfect. Why is this? I would love to shut up all the nay-sayers who tell me I'm wrong for running so much timing. John, you have an amazing product. Can you explain why it does what it does without giving away the secret behind the ACN SVDA?
_________________
1967 Daily Driver
..[-_-_-_-]..
*/l_____l\*
( O \ ! / O )
1973 Thing
1969 Squareback
1967 Bug
1968 Bug
1971 Bug

[74 Thing]

First of all timing should be set like ACN suggested-28-32 total advance with the vacuum hose disconnected. With the hose connected the vacuum advance will allow another 10-12 degrees of advance under light load when there is vacuum.

I would not use that shop again since they have no clue. Tell them to look up the stock timing of an SVDA carb engine which I believe is 7.5 degrees BTDC with the hose disconnected which should yield 28-32BTDC total advance.

What they did is adjust the centrifigal stops and you were only running 30 total advance with the vacuum hose connected so when the vacuum was not present as in full throttle conditions you were only running a total advance of 18-20 degrees which is 10 degrees less than you need!

[Juanito84]

[[email protected]]

I've said this many times, but I guess it needs saying again.

The spark plug is fired before the time that max pressure is needed on the piston, because it takes time for the flame to propagate, and for pressure to build in the cylinder. TIME.

Given the propagation time is a constant (it's not, just follow me a minute), the faster the engine is running the more advance you need to get the pressure at the proper point. This is why ignition curves have CENTRIFUGAL ADVANCE. The faster the engine is going, the less time the combustion has to get the pressure when it needs to. So the ADVANCE gives the combustion more time to do this. Most "get" this concept.

Where they fall short are in the other aspects of combustion WRT ignition timing.

1) a hot engine will have faster combustion than a slow one. This is seen with an engine that only pings when it's hot. The timing is fine for a cooler engine, but is excessive for hot engines.

2) combustion speed also depends on LOAD. Load is "how full the cylinder is with air and fuel). WOT (full load) requires less timing, since a full cylinder has more pressure (HEAT) when compressed.

For example, at idle the throttle is closed. The only air being admitted into the air is via the idle circuit, a puny passage in the carb, and what is bleeding around the throttle plate. While the compression RATIO is unchanged, the actual amount of air being compressed is very small. So "Pressure" is low (relatively), and combustion speed is slow.

But as the throttle is opened, more air (and fuel) is entering the cylinder. This is being compressed the same amount, and temps rise. The more air the more pressure, which means more temp (pre combustion, NOT engine temperature). The more temp, the faster the combustion process (it's actually an oxydizing process for you chemistry geeks). Basic chemistry teaches that every 20C increase in temp (of reactants), doubles the speed of the reaction.. The faster the reaction, THE LESS IGNITION TIMING THAT IS NEEDED. Looked at another way, MORE IGNITION TIMING IS NEEDED UNDER LIGHT LOADS.

Centrifugal Only distributors only change timing based on RPM, they have no load sensing capability. Vac advance senses load, and load timing changes.

Right here you should understand that cruising at light throttle will require more timing than driving WOT. So where does this extra timing come from? If you have a 009/010/019 or other centrifugal only distributor, you are SOL! So the best distributor will change timing on RPM AND LOADS!!!!!

Where people get further confused, is because IF you are running a centrifugal only distributor, you time it for "worst case", which is WOT. If you go more than this is may run great at part throttle, and maybe even WOT when not hot, but at some point the engine will have too much timing for how much air is being let in, and how hot the engine is, and it will ping. Eventually it will be damaged. So the rule of thumb is to time these for "worst case", which is full throttle, which is the LEAST TIMING. That's where the 28-32 "rule" comes in. That rule always applies, even if you have vacuum advance. But the rule does NOT apply to part throttle, which is where the additional timing of vacuum advance will actually kick in. This is because at Full Throttle there is no manifold vacuum to pull any advance in anyways! So vacuum advance has the additional benefit of "load sensed timing adjustment" that is not possible with centrifugal only distributors.

if some asshat is timing, or telling you to time, a vac advance distributor to 28-32 total timing, they have no clue what they are doing and you need to find another "mechanic", because we have no idea what other false information they are working on your car with!!!!!

PS: Combustion speed is also why engines that are "rich" need more timing than engines that are leaner (leanER, not LEAN). Rich A/F's are very cool (pre combustion), and flame front propagation is relatively slow, so the engine needs more timing to get out of it's own way).
_________________
It's just advice, do whatever you want with it!

Please do NOT send me Private Messages through the Samba PM System (I will not see them). Send me an e-mail to john at aircooled dot net

"Like" our Facebook page at
http://www.facebook.com/vwpartsaircoolednet
and get a 5% off code for use on one order for VW Parts ON OUR PARTS STORE WEBSITE, vwparts.aircooled.net

[Juanito84]

Exactly [email protected]! Finaly someone understands how it works!
_________________
If a water cooled engine cools its water with air, isn't it just an overcomplicated air cooled engine?

[[email protected]]

I covered this 15 years ago.
_________________
It's just advice, do whatever you want with it!

Please do NOT send me Private Messages through the Samba PM System (I will not see them). Send me an e-mail to john at aircooled dot net

"Like" our Facebook page at
http://www.facebook.com/vwpartsaircoolednet
and get a 5% off code for use on one order for VW Parts ON OUR PARTS STORE WEBSITE, vwparts.aircooled.net

[jtwaller]

John, your the man! I'm sorry for making you repeat yourself as the samba is huge and I couldn't find a thead that explains it. I am learning everyday how to do everything myself and only rely on my local shop to do jobs that I just don't have the time to do. This was something they took apon themselves and it has been fixed. I have a friend who is running your distributor in his 67' convertable that was just in the latest Hot VW's and he too is running 30* WOT with the vacuum plugged and I need help explaining my case to him as well. Thanks for all the input guys!
_________________
1967 Daily Driver
..[-_-_-_-]..
*/l_____l\*
( O \ ! / O )
1973 Thing
1969 Squareback
1967 Bug
1968 Bug
1971 Bug

[Juanito84]

Ok. Now I got this a bit more organized.
_________________
If a water cooled engine cools its water with air, isn't it just an overcomplicated air cooled engine?

[[email protected]]

I explained it before the Samba existed; back in the RAMVA days (USENET). Then again in our dist article, and some other places. Obscure to find for sure.
_________________
It's just advice, do whatever you want with it!

Please do NOT send me Private Messages through the Samba PM System (I will not see them). Send me an e-mail to john at aircooled dot net

"Like" our Facebook page at
http://www.facebook.com/vwpartsaircoolednet
and get a 5% off code for use on one order for VW Parts ON OUR PARTS STORE WEBSITE, vwparts.aircooled.net

[stevehenderson]

Go electronic its 2012 use a computer for fuel and timing help save the planet from your dirty old engines. Tune it to run clean as possible.

[Stuntmanus]