Ignition point of interest

[zeeprogrammer]

Hi George. In an earlier post you had mentioned "Now from the nut to the Hall I'm at about .125 or less, not the .250 from the top of the rotor. Hmm I wonder if that could be my problem?"

The 'could be my problem' implies the distance was .125 and a larger distance of .250.
Was the solution to increase the distance?
To minimize arcing or the chance of current finding a way to the Hall?

[Roger B]

Mike,

I am not sure if 9000 rpm is the present limit for Rcexel units, the instructions on the just engines site give a maximum current draw at 8000 rpm.

https://www.justengines.co.uk/acatalog/ignitioninstructions.pdf

As a little background: I bought an Rcexel unit for my 3cc vertical engine on the (mistaken) assumption that it would eliminate any ignition problems. I mounted the timing magnet on the camshaft drive gear which runs at half engine speed. Discussions on this Forum came up with two points 1) there was a built in advance curve and 2) the ignition sensor was expected to be fitted to the crankshaft and the curve was designed for this. The result was double the expected advance  . With the help of another poster we were able to work out the advance curve (attached). The nasty bit is that up to 400 rpm (800 rpm if driven from the camshaft) there is full advance then this drops back by 24°. No problem for a high speed engine but was a problem for Jo’s R&B.

I then made up a simple ‘points assist’ amplifier from a kit and used this with the hall sensor and a 6V motorbike coil. This ran but wouldn’t rev due to the lack of dwell angle to build up the magnetism in the coil. In this case the coil was energised when the magnet was in front of the sensor and fired as the magnet moved away. This is only a problem with a conventional coil system CDI type systems are not affected.

Finally I moved the hall sensor to the crankshaft and had good results with the Rcexel unit.

[gbritnell]

Hi Zee,
That's what I'm assuming happened. I really don't have any other answer for it. By increasing the distance/insulation factor the errant spark can't find it's way to the Hall transistor. In the old days of working on cars a distributor cap or rotor would cause misfires which could be traced to carbon tracking caused by a repeated spark path going somewhere other than to a terminal post. I had mentioned that when I took the cap off it was a little dirty inside probably from the contact strip on the rotor running against the center post of the distributor. (I usually put a dab of high temp grease there) In a distributor cap the center post is carbon for lubrication.
When you build 6-7 engines with similar distributors and don't have a problem with them then you have to start looking for the culprit.
gbritnell

[Admiral_dk]

Rogers dwell time is down to how long the magnetic field is exposed to the Hall-Sensor if it is a "amplified system".
This can be solved with both the Hall-Sensor and the magnet being stationary and moving a slotted iron/steel shield between them - I seem to remember George did this in one of his other builds.

I believe that the Rcexel system uses the Hall-Sensor as a "Trigger system" and these do (almost) not care about the dwell time of the signal as long as both the "On" and "Off" part of the signal are within minimum length and sometime a maximum too. Here the Trigger fires the spark in simple systems - or does so after a delay that follows an advance curve programmed into the MCU in the "Intelligent Systems".

[Vixen]

Thank you Per

A very precise and concise explanation of the differences between and amplified (TIM-6) system and the RCXEL triggered systems. That's just what I was looking for.

Do you have any additional information regarding the separation distance between the magnet/ Hall and the thickness required for  the moving metal shield?

Mike

[gbritnell]

Hi Mike,
Here's some pictures of the distributor I made for my inline 6 cylinder engine.
I use a single magnet (.125 dia. x .062 thick) pressed into an aluminum angle plate. This is mounted to the floor of the distributor. The bracket has a slot so it can be moved back and forth to adjust it's position relative to the timing disc.

[gbritnell]

The cap is made from Delrin. I turn the I.D. and O.D. then hold it with a tiny 3 jaw chuck in the mill (vertically) to drill the terminal holes and cut the outside of the posts. I make an annular cutter with a tiny amount of taper in it then feed it down on the center of each terminal to form the posts. 

[gbritnell]

The rotor is also made from Delrin. I turn it then cut a slot to accept the phosphor bronze strip with the brass terminal on the end. The is glued in place with CA adhesive. (The larger distributors have the brass button threaded 0-80 and screwed into the rotor to hold the strip in place. I drill and tap for a 1-72 set screw in the side of the rotor so that I can adjust it's position on the shaft.

[gbritnell]

The timing disc is made from mild steel. I bore out the inside leaving a boss for the through shaft. The metal wall is about .025 on this one. The bigger distributors have a little thicker wall due to the larger diameter. While the timing disc is still part of the stock that it was tuned from I put it in my dividing head and cut the windows. For dwell I don't have much choice with these small distributors. I leave only enough metal between the windows to turn the magnetic field off. I have never played around with making the windows bigger or smaller to see what affect they have on the ignition. Most of the time the engines will rev up fine and don't seem to overheat the ignition modules so they must be pretty good.
I also drill and tap for a 1-72 thread so the timing disc can be adjusted like the rotor.

[gbritnell]

The cap is held in place with two tiny strips of stainless steel cut from the strip found in some wiper blades. When my wiper blades go old I would salvage the strips for future use.

[gbritnell]

These pictures show the timing disc and rotor mounted to the shaft.

[gbritnell]

For my Hall transistor mounting I machine a bracket that mounts to the side of the distributor. It is held in place with 0-80 socket head screws.
In the first couple of pictures of the distributor body you will notice the window and the screw holes. The bracket has 2 flanges cut at an angle so that where they touch the body (tangent point) the screw holes are drilled. The bracket has 2 side rails which locate in the window on the distributor body and locate the Hall transistor. For this small distributor I glued the Hall in the slot with CA. I have read where this is not a good idea because as the adhesive sets it puts stress on the Hall transistor but I have never had a problem. On my larger distributors I use a piece of styrene plastic the same width as the Hall and drill 3-.032 holes for the leads to pass through. When mounting the Hall the only thing that gets glued is the styrene so if what I read is correct it doesn't put any stress on the transistor. Like I said I have never had an issue either way.
The clearance from my magnet to the Hall varies with the diameter of the distributor body. The magnet bracket is moved as close as possible to the timing disc but being as the bracket is wider than the magnet that moves it back a little. From the magnet to the timing disc I would say I have about .020-.025. When I make the Hall bracket I make the side rails only high enough to fit in the window on the body. The Hall is generally flush with these side rails. The timing disc has about .005 clearance from the body so when you stack everything up the magnet is about .040-.050 away from the Hall. You could probably get away with .060-.080 and still have a good magnetic signal to the Hall.
I hope this helps you out some.
gbritnell

[Vixen]

Hello George,

Once again I must thank you for taking the time to write such a comprehensive reply. The design, the build and the photos are all expertly presented. A talent shared, indeed.

It will take a little while for all that information to sink in. The slotted rotor design certainly looks to be a very neat and elegant way to trigger the sparks. I believe I will be able to incorporate your slotted rotor directly into the design of my Bristol Jupiter's two dummy magnetos. I will try to keep things simple, by using an external battery powered coil in place of the magneto coil. The dummy magnetos will act like normal distributors. You do not have to fight every battle.

A scale 9 cylinder radial engine magneto has enough design quirks of it's own before considering the Hall sensor stuff.  For a start. the Magneto's input shaft rotates at 1.25 crankshaft speed and has four contact breaker lobes/ magnets/ slots/ whatever. The distributor rotor arm has a further 9:4 gear reduction from the input shaft and the cap has nine plug leads which fire in the order 1,3,5,7,9,2,4,6,8.  All that gearing provides 4.5 sparks per revolution, though it's easier if you think of it as 9 sparks in every two revolutions of the crankshaft. All cylinders fire once in two revolutions of the crankshaft.

See what I mean about being quirky, At least you have pointed the way with the spark trigger stuff, so all I have to do is to incorporate your ideas.

I guess we don't choose to do this stuff because it's easy

Thanks again

Mike

[nj111]

Thank you George for the information. Your distributor  is superb. Don't want to push this thread off it's topic but have you covered machining the distributor's helical input elsewhere on this forum? I'd  love to know a little more about that end of it! Nick

[gbritnell]

Hi Nick,
Rather than get too far off topic I'll start another thread about the helical gear cutting. It will be under machine tools and fixtures.
gbritnell