Variable Advance/Retard: How I Found Out
10-08-2005, 11:22 AM
#1
Senior Member
Thread Starter
My Feedback: (10)
Join Date: Feb 2004
Location: Friendswood,
TX
Posts: 453
Likes: 0
Received 0 Likes
on
0 Posts
Variable Advance/Retard: How I Found Out
I designed a circuit to allow me to vary the advance and retard angles at will. I tested the design using a signal generator to generate the timing that would come from a typical Hall Effect sensor on our gas engine. I made calculation using Excel spreadsheet. I verified that things work as designed by watching the signals on my Tektronix digital storage oscilloscope and listening/watching sparks on the spark plug. As I varied the frequency, the spark plug buzz sounded like a normal speed up and down of a gasser being throttled. The signals were verified on the scope to have the expected advance or retard timings.
I then built and mounted the setup on one of my planes to check things out. The setup included a separate transmitter and receiver. The advance/retard board plugged into the throttle channel of the spare transmitter. I used the throttle stick on the spare transmitter to give command to the board to change the advance/retard angles at will: low being full retard, mid being zero, and high being full advance.
This project taught me that there was no advantage in varying the angles in hopes of getting a smoother running engine at low speeds, and no gain to be had at WOT. The design worked as I intended and did not mess up at any point, i.e. causing a misfire or backfire, etc. The reason for doing this was to answer my own curiosity.
I then built and mounted the setup on one of my planes to check things out. The setup included a separate transmitter and receiver. The advance/retard board plugged into the throttle channel of the spare transmitter. I used the throttle stick on the spare transmitter to give command to the board to change the advance/retard angles at will: low being full retard, mid being zero, and high being full advance.
This project taught me that there was no advantage in varying the angles in hopes of getting a smoother running engine at low speeds, and no gain to be had at WOT. The design worked as I intended and did not mess up at any point, i.e. causing a misfire or backfire, etc. The reason for doing this was to answer my own curiosity.
10-08-2005, 12:39 PM
#2
My Feedback: (11)
Join Date: Jan 2002
Location: NWest,
IN
Posts: 962
Likes: 0
Received 0 Likes
on
0 Posts
RE: Variable Advance/Retard: How I Found Out
Pretty interesting: How far advance are you able to go? I have always wondered if ignitions all advance to the same point and if increases could be made by fine tuning each individual engine with an adjustable advance control. I have seen RPM increase by moving the pick-up on 3W engines from the factory install. Its only a couple degrees but on one engine it was enough to show 200 RPM more by fully advancing the pick up to the extreme end of adjustment. I believe some engines/ignitions may hold some power back on account of this. Neat experiment, Walt
10-08-2005, 12:44 PM
#3
Senior Member
Join Date: Mar 2002
Location: Kingston,
ON, CANADA
Posts: 4,925
Likes: 0
Received 3 Likes
on
3 Posts
RE: Variable Advance/Retard: How I Found Out
Bob :
Most engines at full retard are at 5 degrees advance. Still takes time to burn the fuel. Full advance at 28 to 30 degrees. shows no advantage to either fixed or automatic timing. Only advantages are lower smoother idle and easier starting, no kickback. Good stuff though.
Most engines at full retard are at 5 degrees advance. Still takes time to burn the fuel. Full advance at 28 to 30 degrees. shows no advantage to either fixed or automatic timing. Only advantages are lower smoother idle and easier starting, no kickback. Good stuff though.
10-08-2005, 03:32 PM
#5
Senior Member
My Feedback: (2)
Join Date: Mar 2004
Location: Bucyrus,
OH
Posts: 143
Likes: 0
Received 0 Likes
on
0 Posts
RE: Variable Advance/Retard: How I Found Out
That's a very neat experiment. Did you try different props? Reason I ask is because varying the timing may not show any changes if you are only measuring peak rpm vs. one prop. It could be that the engine is limited in other ways, not by the timing being less than ideal. Changing the timing really only changes the torque. It would take a lot of instrumented testing with a steady-state dyno but you could develop a simple ignition map that varies timing vs. rpm vs. throttle, if you could find a way to hook up a throttle position sensor. There would be a bit of artwork involved in choosing a prop... obviously the goal would be to swing a larger prop for a given rpm. Or somehow tie in a variable pitch prop. Hmm!
This wouldn't be of much benefit for a warbird, but it could be for a aerobatic/3D model where the throttle is varied a lot and loads on the engine change constantly.
Just some offhand thoughts.
This wouldn't be of much benefit for a warbird, but it could be for a aerobatic/3D model where the throttle is varied a lot and loads on the engine change constantly.
Just some offhand thoughts.
10-08-2005, 08:32 PM
#6
Senior Member
Thread Starter
My Feedback: (10)
Join Date: Feb 2004
Location: Friendswood,
TX
Posts: 453
Likes: 0
Received 0 Likes
on
0 Posts
RE: Variable Advance/Retard: How I Found Out
It has been many months since I did this. I have forgotten half of the details. However, I can recall the basic principle as follows. We know that a full advance around 28 degrees will give us ease of starting, and zero deg (firing at TDC) at speeds above 5krpm will give us max rpm at WOT. The question became whether changing the angle in between idle and 5krpm would produce lesser engine shakes.
The circuit worked with two points: the 'bottom' point was determined by the mechanical position of the Hall sensor and the 'top' was preset at 5krpm. The 'bottom' point represented the largest advance, which I set at about 28 deg for ease of starting. At 5krpm, a fixed retard was set by a trim pot at zero deg to as much as 5 deg after TDC. Any speed between idle and 5krpm was linearized by the circuit to arrive at a proper advance angle. Anything above 5krpm, the angle was held fixed at zero deg. Note that the Hall sensor served as rpm sensor as well.
By using the spare transmitter, I could alter the linear curve between idle and 5krpm at will by counting how many clicks I placed the throttle stick. First, I placed the stick at its lowest position which meant that the timing was a straight line. This could be thought of as having a mechanical coupling between the Hall sensor and the throttle arm like the Brison's setup. I then started the engine and ran it at a fixed speed say 3krpm. I picked up the spare transmitter and started moving its throttle stick one click at a time.
Here was the subjective part as I listened to the engine and felt the airplane for more or less vibration level as I counted the number of clicks. The number of clicks commanded a corresponding amount of retard to be added to the ignition by the circuit. In effect, this allowed me to 'see' if an other than straight line timing would buy me anything. The circuit allowed me to easily alter the timing as the engine ticked away. It would be like being able to mechanically move the Hall sensor more advance or retard as the engine ran at a fixed 3krpm, etc. The experiment showed that tuning the timing produced a slight improvement that was not worth the added complexity.
Sorry I was not clear in my original post: 28 deg advance at idle and 0 deg at and above 5krpm IS ideal for our gasser. With a fixed 28 deg advance, we would lose say 200rpm at WOT. It does make starting the engine easier and the simplest setup.
The circuit worked with two points: the 'bottom' point was determined by the mechanical position of the Hall sensor and the 'top' was preset at 5krpm. The 'bottom' point represented the largest advance, which I set at about 28 deg for ease of starting. At 5krpm, a fixed retard was set by a trim pot at zero deg to as much as 5 deg after TDC. Any speed between idle and 5krpm was linearized by the circuit to arrive at a proper advance angle. Anything above 5krpm, the angle was held fixed at zero deg. Note that the Hall sensor served as rpm sensor as well.
By using the spare transmitter, I could alter the linear curve between idle and 5krpm at will by counting how many clicks I placed the throttle stick. First, I placed the stick at its lowest position which meant that the timing was a straight line. This could be thought of as having a mechanical coupling between the Hall sensor and the throttle arm like the Brison's setup. I then started the engine and ran it at a fixed speed say 3krpm. I picked up the spare transmitter and started moving its throttle stick one click at a time.
Here was the subjective part as I listened to the engine and felt the airplane for more or less vibration level as I counted the number of clicks. The number of clicks commanded a corresponding amount of retard to be added to the ignition by the circuit. In effect, this allowed me to 'see' if an other than straight line timing would buy me anything. The circuit allowed me to easily alter the timing as the engine ticked away. It would be like being able to mechanically move the Hall sensor more advance or retard as the engine ran at a fixed 3krpm, etc. The experiment showed that tuning the timing produced a slight improvement that was not worth the added complexity.
Sorry I was not clear in my original post: 28 deg advance at idle and 0 deg at and above 5krpm IS ideal for our gasser. With a fixed 28 deg advance, we would lose say 200rpm at WOT. It does make starting the engine easier and the simplest setup.
10-08-2005, 09:49 PM
#7
Senior Member
RE: Variable Advance/Retard: How I Found Out
We know that a full advance around 28 degrees will give us ease of starting, and zero deg (firing at TDC) at speeds above 5krpm will give us max rpm at WOT.
Have I totally misunderstood?
Bill
10-08-2005, 10:29 PM
#9
RE: Variable Advance/Retard: How I Found Out
Thre are a number of new ignitions on the market which are programmable. Basically the engines are run with different profiles -then the "best" profile for a particular engine , is programmed in by the factory.
The same modules also record highest rpm for any given run and hold this info for down loading. A more recent addition is to make the circuit compatable with any ignition power source from 4.8 to 9V. I do know that the manufacturers claim they can produce curves which in actual practice , improve midrange response, etc..
All of this is really of interest to flying aerobatic stuff where the engine is constantly being juggled at mid to higher power settings. In actual practice the setups seem to do this quite well . I have tried a number of em but what the tailoring changes really are - I don't know. I do know that the engine/ignition setups are really user friendly-very easy starting -smooth -really strong response etc..
The same modules also record highest rpm for any given run and hold this info for down loading. A more recent addition is to make the circuit compatable with any ignition power source from 4.8 to 9V. I do know that the manufacturers claim they can produce curves which in actual practice , improve midrange response, etc..
All of this is really of interest to flying aerobatic stuff where the engine is constantly being juggled at mid to higher power settings. In actual practice the setups seem to do this quite well . I have tried a number of em but what the tailoring changes really are - I don't know. I do know that the engine/ignition setups are really user friendly-very easy starting -smooth -really strong response etc..
10-09-2005, 12:03 AM
#11
Senior Member
Thread Starter
My Feedback: (10)
Join Date: Feb 2004
Location: Friendswood,
TX
Posts: 453
Likes: 0
Received 0 Likes
on
0 Posts
RE: Variable Advance/Retard: How I Found Out
As part of this same experiment, I wanted to find out if the capacitive discharge ignition (CDI) would generate interference. On the bench, I found that there was absolutely no problem whatsoever.
I wrapped the receiver antenna around the spark plug cable, had the spark plug buzzing at all kinds of speed in open air near the receiver/battery/servo/switch, etc. I stopped short at sticking the antenna wire in the path of the spark. That could fry the receiver. Yet my scope recorded zero glitch. The nice thing with a digital persistence scope is that if there is ONE stray signal, it would be captured on the screen. Yet by simply scratching sharp metal edges like rubbing the tip of a screwdriver up and down the threads of a bolt, the scope would capture glitches and the servos jumped.
The SCR (Silicon Controlled Rectifier) used in the CDI circuit is known as a self-commutating device. It does a very nice job of depleting the energy stored in a very small capacitor (0.22uF or thereabout) without creating sharp rising and trailing edges. Most of the energy is converted into a nice spark every time. There is no ringing that would cause multiples of the fundamental frequency to glitch our transmitted frequency in the 72-75MHz range.
The glitches we pick up when flying gassers are due to two sources. The first source being metal rubbing against metal especially if you have sharp metal edges rubbing somewhere in your plane. The second source would come from a defective ignition circuit in the form of cracked or unintentionally gapped ferrite cores. These ferrite cores are used to make step up transformers that convert 4.8Vdc (as low as 3V will also work) to about 300V to charge the small cap, which dumps through the SCR into another step up transformer to generate around 15,000V.
I wrapped the receiver antenna around the spark plug cable, had the spark plug buzzing at all kinds of speed in open air near the receiver/battery/servo/switch, etc. I stopped short at sticking the antenna wire in the path of the spark. That could fry the receiver. Yet my scope recorded zero glitch. The nice thing with a digital persistence scope is that if there is ONE stray signal, it would be captured on the screen. Yet by simply scratching sharp metal edges like rubbing the tip of a screwdriver up and down the threads of a bolt, the scope would capture glitches and the servos jumped.
The SCR (Silicon Controlled Rectifier) used in the CDI circuit is known as a self-commutating device. It does a very nice job of depleting the energy stored in a very small capacitor (0.22uF or thereabout) without creating sharp rising and trailing edges. Most of the energy is converted into a nice spark every time. There is no ringing that would cause multiples of the fundamental frequency to glitch our transmitted frequency in the 72-75MHz range.
The glitches we pick up when flying gassers are due to two sources. The first source being metal rubbing against metal especially if you have sharp metal edges rubbing somewhere in your plane. The second source would come from a defective ignition circuit in the form of cracked or unintentionally gapped ferrite cores. These ferrite cores are used to make step up transformers that convert 4.8Vdc (as low as 3V will also work) to about 300V to charge the small cap, which dumps through the SCR into another step up transformer to generate around 15,000V.
