SrTelemaster150

At TDC the bores of the slave rods are in line with the main crank journal hence there is no difference in piston position in relationship with the length of the bore. At 30 crankshaft degrees, or any value between BDC and TDC, the bores will no longer be aligned in a straight line with the crank journal. The effect will be similar to a rod that bends throughout the travel of the piston in the bore as the rod angle changes.

I will find my graph velum and hopefully my highlighters to make a semi full scale, color coded drawing that will illustrate the differences later this evening.

Given that the OEM system was running at at least 15* too much ignition advance, the situation was exacerbated over what 30* would result in.

The only way that 120* magnet locations could result in a variance in crankshaft degree ignition timing is for the module to have some sort of timing variance built into it. Retarding #3 cylinder timing and advancing #2 cylinder timing would be the only way to get the ignition timing to occur at the same piston travel/cylinder pressure value.

By having an exact 30 crankshaft degrees BTDC ignition event, one will have #3 ignition occur when the piston is too far down the bore (ignition too far advanced) for optimal performance and the opposite for #2 cylinder.(not enough advance)

MDavis28

SrTelemaster150

I am not disagreeing with you on that point. In fact it is the twisting, bending if you will, of the rods that causes the variance in linear piston position relative to the head deck.



That is why a true 30, or any value for that matter, crankshaft degree ignition timing setting for all 3 cylinders is not correct for optimum detonation prevention.

You need to get your mind off of crankshaft degrees and look at linear piston position down the bore and the resulting variance in cylinder pressure at 30 crankshaft degrees. With a given octane fuel, cylinder pressure is THE major factor in detonation if A/F ratio is correct. Cylinder pressure will determine the optimum ignition timing advance for maximum power without detonation. If the cylinder pressure varies between cylinders at a given crankshaft degrees, the optimum ignition timing value will also vary.

If you don't believe me, measure the distance from the top of the spark plug hole to the piston crown at TDC for each cylinder and then measure the distance from the top of the spark plug hole at 30 crankshaft degree BTDC for each. Let me know if they are all equal. They won't be if I am correct.

In order for one to see this a cutaway of the 3 cylinders would be necessary. barring that, if you have some patience, I will do a drawing that will illustrate that variance.

MDavis28

JieM

I have measured the degrees BTDC exacly like shown on this video. And as said previoulsly the out of the box values are 45, 53 and 40. The video shows the measures made with a modified magnets position ring.


JM

SrTelemaster150

MATHEMATICS IS AN EXACT SCIENCE. It does not vary regardless of where the slave rods are linked to the master rod. We are talking about geometry here and that is mathematics.

30 crankshaft degrees is 30 crankshaft degrees regardless of which cylinder is coming to TDC. A magnet ring with 120* magnet spacing will trigger the spark at exactly the same moment in crankshaft degrees UNLESS there is some variance built into the module or the Hall Affect sensor arrangement.

There has to be a factor in the module or the Hall Affect sensor arrangement for this to work the way it does.

SrTelemaster150

I am still unsuccessfully searching for an article that cites the differing ignition advance between cylinders of the P&W R-2800 engine.

In the mean time, here is an article that illustrates the differing cylinder pressures between cylinders. This would affect the optimum ignition advance values needed for maximum performance while providing optimum detonation protection.

https://www.researchgate.net/publica..._RADIAL_ENGINE

SrTelemaster150

Set #1 cylinder ignition advance at whatever value would be best. I use 35* BTDC for methanol, (glow fuel) but that's not relevant for what we are discussing here.

Use a piston stop to set that piston distance before TDC.

Using the piston stop, determine the crankshaft degree values that reflect that same combustion chamber volume for #2 and #3 cylinders.

Have a magnet ring fabricated that will trigger the spark at those degree values. While this may be a bit simplistic considering the dynamics involved, it would seem to be more appropriate than having timing fixed at a given crankshaft degree value for all 3 cylinders.

Why this particular ignition system does not trigger the spark at even intervals with even 120* spacing of the magnets is puzzling to me. I will have to look into that.

I have in hand one of the early prototype systems for 3 cylinder radials developed by C&H Ignitions. I will examine the Hall Affect sensor arrangement and call Adrian for his input.

I am engineering a maximum compression FA-180 FA-450R3 hybrid that will be a metanol version of the FG-84R3. I will cal it FA-512HC/CDI. Since static CR will be in the neighborhood of 13:1, (much higher that what gasoline would allow) I want maximum performance while providing optimum detonation prevention.

I will discuss this phenomenon with Adrian at C&H. Perhaps we can engineer the best magnet ring spacing for that particular application.

JieM

Thanks for this document.
Extract of this document: "The other situation occurs in the angle value of the top dead center position of the pistons. The maximum difference occurs between the cylinders 4 and 7, and is 7.6 CAD causing a significant shift in ignition timing in most cylinders according to TDC."

Cylinder 4 and 7 on a 9 cylinder radial are exactly at the same place (120 deg) as cylinder 2 and 3 on our Saito.

SrTelemaster150

Yes you are correct.

One must be cautioned however to not take those values too literally. Rod length/crankshaft stroke ratios will skew those values a bit.

Interesting though is the relationship of the nearly 7.6* variance in the Saito magnet spacing that results in the close to +7.6* and -7.6* shift in ignition timing.

45* BTDC #1 tming +8* for 53* BTDC #2 timung - 5* for 40* BTDC #3 timing.

The problem isn't the variance in timing, the problem is too much advance for all cylinders.

If a 10* shift was employed by cutting a new slot in the prop hub where it engages the prop shaft pin, net advance would then be 35* BTDC, 43* BTDC and 30* BTDC for #1, #2 and #3 respectively. An addition 5* retard in timing advance would result in a slight decrease in power but even more detonation protection.

By employing the Morris magnet ring that has 120* spacing in the magnets for 30* BTDC timing for all cylinders, you are about 5* too far retarded for #1, 13* too far retarded for #2 just right for #3. Yes you are preventing detonation over what the Saito timing does, but you are also leaving a lot of potential power on the table.

I spent almost an hour on the phone this morning with the man that conceived the radial engine CDI concept. He worked with RCEXL on the prototype systems. I have one of those early prototype systems for the FA-450R3 conversion. Originally the magnets were all spaced evenly for 120 degree firing order.

Once rod agulatity was considered, it was discovered that an uneven firing order would result in more power by allowing the optimal ignition timing while still providing maximum detonation prevention. My magnet ring is spaced at 120*. I will fabricate one with the proper uneven firing order once I determine the correct offset to take into account the proper rod angularity factor for the FA-450R3 geometry.

Those of you that have employed the Morris even firing magnet ring are now using the 1st prototype ignition timing concept. You are going backwards.

Just in case you doubt my findings, measure the distance between the 3 Hall Affect tripping magnets on your Morris rings. DO NOT CONFUSE HALL AFFECT TRIPPING MAGNETS WITH THE REVERSE POLARITY SEQUENCING MAGNET. If you have an even 120* firing order, you have the same 120* spacing of the Hall Affect tripping magnets.

Morris Mini Motors

The Morris' Mini Motors magnet ring conversion kit for all Saito FG- R3 engines or #1 to #2 = 126, #2 to #3 = 108, #3 to #1 = 126

SrTelemaster150

How does that result in 120* firing order?

Morris Mini Motors

cathurga

SrTelemaster.... it doesnt....

As far as I am aware, and Adrien, and Morris (as well as some dude at RCxel) the ignition systems do NOT have any timing variance, the timing is triggered solely by the magnets relation to the hall sensor (in other words, the magnet approaches the sensor, and fires as soon as it touches) so the position of the magnets is what we are referring to.
The cylinders are placed 120Deg apart, we know that. The angles on the slave rods mean that TDC on those cylinders is NOT at 120Deg, its around 126Deg. All of us have ASSUMED (right or wrong) that we need to have THOSE two cylinders fire 30Deg before those cylinders reach TDC, which has influenced the placement of those two relative magnets accordingly.
I think I understand what you are trying to say, but it seems to indicate that those two cylinders would have a different stroke. If we did your test with the toothpick, in other words, measure where Cyl 1 TDC is, then go back 30Deg on the crank and measure the depth of the piston relative to the plug hole and then take the same toothpick, put it in cyl 2, go back 30deg from TDC, that the mark on the toothpick would not be in the same place? I know there is a lot of geometry involved here, but if those lower cyl's have the same stroke as cyl 1, then the timing must be the same distance in degrees, surely.

I really dont see how compression is key here, if the cyls all have the same bore/stroke....whic I assume (right or wrong), they do.

cathurga

I got my parts from Morris (Thanks Morris!) ~~wave~~ and started assembling the engine today, I see that the piston they supplied is the new one, with the head machined out to lower the compression. Surprise, surprise...this piston would not fit in the new cylinder I bought to do the repair....it just won't, to the point where I could barely pull the assembly apart after trying to get it squeezed in there. I have done these a few times, and there is no way that piston was going in there.... very disappointed, and VERY disillusioned with these engines. I am going to be looking for alternatives now.

cathurga

Off topic.....does anyone have any of the original pipes from this engine that they are not using? One of mine (the medium length one) has snapped (my bad) and I am not spending US$140 on a set of 3 new ones....I dont care what condition it/they are in, and will buy a set from someone who is not using them, or swap for a used original Saito ring for this engine. Let me know please.
Also, if anyone knows of an alternative, I would also like to know.

SrTelemaster150

Delete inadvertent post.

SrTelemaster150

I am well aware of how the 4 magnets work. Before the 1st prototype of the 3-cylinder system was even developed, I discussed the concept at length with Adrian. Although I did not grasp the concept at that time, once the prototype was developed, it all became clear to me.

If you magnet rings have 126-108-126 spacing, how does the system fire ar 30* BDC for each cylinder on an engine that is built on 120* cylinder spacing?

Morris Mini Motors

SrTelemaster150

You are avoiding the question. How does 126-108-126 magnet spacing result in 120-120-120 spark timing?

Is the FG-60R3 NOT built on 120* cylinder spacing?

JieM

It does not result in 120deg firing order, it result in 30deg BTDC on each cylinder. And it works fine. Engine runs smoother, and iddle can go to 900 rpm. Just impossible with a 53deg BTDC.

JM

slither

OK gentlemen, I have been in touch with Ray a few times as the timing issue unfolded, and he has created a new master rod that solves the slave-rod geometry problems. He said that all of the R3 engines suffer from the same timing issue, as the rod geometry and rod ratio is identical, just scaled up in the larger engines and down in the smaller ones. As Sr. suggests, the geometry issue never reared its ugly head in the glow engines, since the ignition timing in glow is based on the glow plug/compression and is, therefore, independent of each other cylinder - which is not the case with the fixed electronic ignition timing. To exacerbate the situation, the valve timing on 2 and 3 is also thrown off by this error... Saito really made a colossal error on the radials! They never picked up the error in their petrol conversion process, as they did not know to look for it. However, the valve-timing issue remains in the glow versions.

Since the rod angularity is different, depending on the direction of crank rotation, the piston-stop method will produce some error in determining TDC on the slave cylinders. Ray has some gizmo he uses to take care of that, and he came up with around 4.5*, give or take... I cannot quire remember at this hour. So the whole notion of even firing on 120* spacing by moving magnets is moot... not gonna happen.

I only have preliminary test info on the 90R3, but Ray said his idle went down to like 700rpm and it smoothed out... that is a reliable 700rpm idle with super transition! His top end went up significantly, too - he had to go up 2 inches in diameter and 2 on prop pitch to keep it in the safe rpm range! He said the 90R3 turned into a beast that purrs like a kitten at idle... here's the issue: that rod would cost about a grand to make by hand in his shop.

He is doing some research into mass-producing parts to get the cost down to something reasonable... one remaining concern is the amount of work required to make the swap. More to come as it develops...

Didier

SrTelemaster150

Editing for corrections.

SrTelemaster150

Does Morris Mini Motors have magnet rings with the same uneven spacing for the FG-84? I only need the ring since I am doing this an an FA-512R3 prototype conversion.