R50 advance curve

[laurent31]

Hello,
Since the advance curve of our twin seems to be lost in Bosch archives, I undertook to measure it.
I have a Souriau testbench for car dizzy. So I hade to make some adaptations.

testbench

detail

Here is the curve of my advance unit

advance curve

We can clearly see the two slopes. The low RPM one driven by the coil springs and the high RPM one driven by the U-shape spring (plus the coils). This U-shape spring is often called the stop spring but actually it is not. The maximum advance is determined by the peg of the bob weight with moves into the slot of the plate.

The advance unit is highly adjustable with the metal plates. Metal plates can be rotated but also shifted. Their hole is bigger than the screw diameter. Herse is my initial tuning :

advance unit

Has anybody done the same measurement on his advance unit ?
Any advice to adjust the curve ?

Laurent.

[schrader7032]

Wow...pretty cool! I thought the degrees of advance would be capped at around 3000 RPM...nominally when viewing the timing marks in the window, once the engine gets to a high RPM, the F-mark stops advancing such that the ignition will fire at the same degrees before TDC. Does your curve reflect that?

[Daves79x]

With that later style 'stop' spring, it must take lots of centrifugal force to compress it, hence the relatively high rpm needed for full advance. Love to see curves for the old style stop spring. My timing light says the old style spring gets to full advance far before 5000 rpm. It appears that the 'shelf' at 3000 rpm is the pin hitting the stop spring, then relatively linear from there to full advance as it compresses the spring.

Dave

[schrader7032]

So what is being plotted? "Advance crank" and "RPM Crank". What are the units for "Advance crank"? Seems clear that the "RPM crank" is just the RPM.

[laurent31]

Cranckshaft angle Vs Crankshaft RPM is plotted. For sure, what was measured on the bench is Camshaft angle Vs Camshaft RPM. But refering to camshaft is not meaninfgul. I prefer engine RPM.

Surprised that old style and new style advance unit would not give the same advance curve. What did change in the engine that needed a such a different advance curve ?
Note that the units are all adjustable.

Laurent.

[schrader7032]

I guess I'm not familiar with what crankshaft angle means. I'm still trying to wrap my head around what you're showing and how the point of ignition firing reaches a maximum advance degrees before TDC and then doesn't change no matter how much the RPM goes up from there.

When you indicated that the two slopes were due to the small spring and then the u-shaped spring, were you able to physically see that happening with your test setup? I suppose a timing light or strobe hooked to the RPM signal would provide a stop-action view of what was happening.

I notice that the post on each weight in the advance doesn't move linearly...first it moves radially, and then the post looks like it gets into the angled slot. Would this change in direction of the post on the weight contribute to a change in the curve? It's been a while since I manually played with an advance unit to see how the weight is constrained as it moves out from the rest position...maybe I'm reading too much into that.

[malmac]

A very interesting post.

I am not even close to being ready to factor in this information however this is high quality information.


Thank you for posting.

Mal

[jwonder]

Laurent,

I did some similar testing as you, although not as exact and I was not able to produce a chart. I mounted my full setup on my lathe and used a light to watch what was happening. There are really three steps, two stiffnesses to the springs, and then the third U-shaped spring. You can somewhat see it in the curve you show although the two springs come in fairly close to one another I was able to control that in the lathe and watch it. The U-Shape spring, as you said, is not a stop, but another spring. That is why several people came up with the adjustable screw to control the maximum advance as seen at the bottom of this page: https://www.flickr.com/photos/roundel/sets/394821/

I would love to see curves for more advance units and see how sliding the metal plates in three settings (all the way out, even with the sides of the unit, in X amount) would affect the curve. That would be quite interesting!

Keep up the good engineering work!!!

[laurent31]

Kurt, I'm only measuring the dynamic advance. Remember that static advance is tuned at 9 °BDTC on the bike.
So 0 in my graph means 9° on the bike. And 30° in the graph means 39° on the bike.
I could observe the movement of the weights only on the bike with a timing light. And yes, things are moving the way I say. But measurement is not easy. Need rev counter installation, need a hand on the twist grip. And running the bike at 5000 RPM at stop is not recommended for an air cooled engine. But this is only physics whith spring, centrifugational strength.
Another way of testing that is to maitain the weights on a certain position and statically observing the points break out. That way, you can measure the advance angle for different position of the weights. But it is not relating to RPM.
Pegs are not guided by the slots. They are part of the weight and moving also on their axis. The springs are sorta guided by the slots, introducing probably a non linearity. More a drawback than a feature I think.

Yes I'll try some measurements with other positions of metal plates. But we can infer that the metal plates are adjusting horizontally the low RPM part of the curve.
The two springs are acting in parallel. On my advance unit, the two springs have same length. No free play on one of them. Hence only a single low RPM segment.
But one can imagine a three segment curve with one spring having a free play thanks to a long loop.
First segment : only one spring
Second segment : both springs acting
Third segment : both coil springs + Ushape spring.

Laurent.

[schrader7032]

So you're just spinning the advance unit which is connected to the camshaft on the bike. Doesn't the camshaft turn twice as fast as the crank? If you change the numbers on the bottom scale by half and get out to 2500 RPM and have reached 30 degrees (or 39 as you suggest), then that all makes sense. My guess is if you went up to 6000 RPM on your test which is 3000 RPM at the crank, maybe you'd begin to see the curve flatten. Just wondering...