Dino 246 ignition advance

I have recently rebuilt my distributor from a '73 246GT.
It is the original from the car and although a UK car, runs the R1 R2 point system with 6 lobe cam.
Over the Christmas break, i will write up exactly how i went about undertanding the distributor advance system and setting the advance through adjustment of the springs and shims. i have plenty of photos and measurments. Remarkably satisfyigly and controllable with a bit of experimentation. However, in the mean time I have a question for the experts.

The curves below are from rpm and advance measured with a snap-on timing light aimed at the crank pulley. I have replaced the advance weigths and springs using the superformance kit, with a bit of smoothing of the damper pistons and bores. By feel, the mechanism feels smooth and good.

1) red curve weights and springs as supplied and no shimming (shims are supplied in the kit). Jitter below 1200 rpm is due to lack of shimming on first pair of pistons, ie. those closest to the pivot.
2) pink curve weights, springs, shimming 1st pair of pistons to set initial gradient. 2nd piston set not in place during test.
3) blue curve weights, springs, shimming with all four pistons in place.
4) yellow curve weigths, springs, further 2nd piston pair shiming, with all four pistons in place.

I am very happy with the advance below 3250 and can understand the leveling off at 4500 rpm even though it appears a little early (weights full out). I should be able to solve this by carefully shaving off a little from the weigths. I know how to calculate the amount to remove.

My question, is what is happening between 3250 and 4500 rpm. The advance appears to accelerate. Has anyone else experienced this and know of a solution. From my school physics days, I don't remember spring strain behaving non linearly unless yeilding takes place, which obviously is not the case.

Kevin
Image Unavailable, Please Login

 

From physics:

(MV**2)/R Anyway, as the weights move outward for a given speed, they exert higher centripetal force. I had the exact phenomenon you see here with the curve changing midstream. The system is a brain teaser full of non-linearities. I'm going through this exercise again right now on a Marelli distrib for another car. BTW. You mentioned trimming the weights to allow for more advance. As I understand the system, the nylon black ring on the shaft that the weights rest against is a specific diameter to control the starting point, and setting the range for total advance. Be patient. Very patient, and avoid modifying unless you have back up parts. One restoration issue I see on these distrib's is the wear point that the spring point create in the drum cause significant headaches. Throws off the shimming, and causes another mechanical feature for the advance system to overcome making the adjusting that much more fun. I've welded up these divets with hard face rod, but should have had the surface ground. If I weld these again, I'd use a softer alloy to machine more easily.

Just my two cents from from working on my distrib on a SUN tester for 3 days! Everytime I get close, I try to perfect it better, and make a mess necessitating my starting over. I get the closest when I think of the mechanism as a clock movement, and deal with it like Harrison in the movie Longitude. Until we become experts, we just need to study the system with great care.


PS: To tackle the runaway curve, I'm going to try an add a small (.1mm) shim under two springs at a a time. you need to figure out which are the "HI SPEED" springs, and act on those. Springs on parallel increase K, In series they decrease K.

 

I was beginning to think i was the only one who had experienced issues with setting the advance up. Thank you for the response. I like the analogy with Harrsion, it would be nice to have some of his skill, dedication and patience. I doubt that I will be going to the Astronomer Royal with my distributor though.

I am convinced that getting the advance right is important. Although I have not yet been on the road after the distributor set up, its already running much smoother and there are no back fires. I will put a report together over the holiday and look forward to the feed back. I am envious of your sun machine, a far more civilised way of working.

You raised one issue that I did come across. The wear by the four piston pins on the drum wall. It was clear by simply pushing the weights out by hand that the pit wear was causing hesitation in the mechanism. I thought long and hard about what to do. Given that a replacement is available in case of disaster (superformance web site), I had a go at grinding the wear out (see attched). The wear was quite deep and I did not feel comfortable going too far. However, the movement is a lot smoother. In hind sight your idea of filling the holes would have been better. I did contemplate using silver solder (the silver not lead variety), but was worried about heat distortion on the relatively thin housing. Perhaps I was being over cautious.

I'll have a think about the plastic ring. As you say, its too easy to modify things and regret them later. Shaving a small amount off the outer part of the weight may be safer and easier.

Kevin
Image Unavailable, Please Login

 

Hello Kevin,
So glad you know of the Harrison reference. We are the foolhardy ones. These cars run so nice when the distrib's are right. My best distrib repair/adjustment came from a 9,000 mile 308 creampuff. It was truly perfect, but was slow as ever. The distrib advance was completely off. No advance until near redline, then only partial. It wasn't just simply aged grease like I assumed (and hoped). They had other issue that took a lot of time to correct. I got hold of some great used distrib's and cannibalised springs etc. to do the setup. After a ton of work, it ran fantastic! The car felt brand new.

Attached is a pix of my 1948 Sun Distributor Machine. Don't be too jealous. I spend a much time keeping it going. It wasn't really meant to operate at the speeds Ferrari distrib's need. I've replaced the original lawn mower belts inside, and put in precision poly-flex belts with stainless pulleys. I also updated the ancient tube power supply. These parts are much better, but other parts on the machine want to fall off, and the sound is aweful. It does give me a precise flash at a known angle, so I'm good for now.

If you shave material off of the weights to get more movement, you'll change the mass of the flyweights and add another variable. Something I've done is make a rotor mount, and attach a protractor to see if the angular movement is right. (see pix). The Dino distrib has 15 degrees of advance. take out the springs/pins/shims, and mark somehow to know where they came from, then re-assemble. Now fashion a pointer next to the protractor and see if the movement will give you the 15 degree range of motion. If it doesn't, I'd opt to re-make the black plastic ring where the diameter controls the advance starting point.

How did you make the nice graphs? Did you plot points an feed them into EXCEL?

Best regards,
Buce
Image Unavailable, Please Login
Image Unavailable, Please Login
Image Unavailable, Please Login
Image Unavailable, Please Login

 

Advance was measured with timing light with engine running (noisy!). Points plotted in microsoft Excel. Spread sheet is ideal as can add/subtract static advance to compare test curves, as well as calculate the shim thicknesses. I like the idea of calculating max advance angle. I will try it.
Kevin

 

Some people only swear by those
Image Unavailable, Please Login

 

I like this machine. Looks like it takes up less space than the old Sun. Is it smooth / quiet? Give you good info?

 

Not too quite, very precise. An indispensable tool to set a distributor properly.

 

What is the make and model of this machine?

Your Stratos buddy in the US

 

It's a Franceschini my friend, from bella Italia.

 

Two issues still troubling me:

1) The apparent increase in advance gradient at greater than ~3400rpm, ie. non linearity.
2) The maximum advance stopping short.

To test the maximum adavance and to see if I could feel what was happening, I following Buce's helpful comment. I made a simple bench advance indicator. I don't have the desirable distributor testers that you guys have. It is a steel baton with rectangular hole cut out to sit over the cam assembly. The baton has a needle on the end which points to an aluminium protractor plate. Using gentle finger pressure I rotate the baton and cam assembly to sense the intial slack (there should be none of course) and the changes in spring resistance that should correspond to the 19deg and 39deg inflection points shown in the manual.

The advance that I sense is shown in the first figure for two different weight, piston and spring assemblies (green and blue). The manual figures shown in black. The results are fairly close to the manual, but note that the blue set up is 2deg short of the required maximum advance, and the green 1deg too far advance. Both these are in tolerance, assuming my test apparatus is perfectly accurate. However this does not tell me at what rpm the maximum advance is acheived. To do this I need to spin the system.

I ran both set ups on the engine and measured the advance with a strobe. The second figure shows the advance vs rpm.
.....Couple of side comments: I know the blue set up does not fit that well but its a set up that I previously ran to high enough rpm to check the maximum advance (noisy job). The green set up has no shims on the first springs, hence the slightly erratic adavnce at low rpm. Both easy to rectify......
The blue setup reached the same maximum advance as that measured in the static bench test. Great.
The green set up failed to reach the maximum advance measured in the static bench test by 5deg! It is also very short of what is required.

The blue set up is using new polished piston pins, the green is using used pins that do have some minor flattening on the heads. The latter coupled with the remaing piting in the wall of the advance assembly may be restricting the maximum adavnce. However, this supprises me as the manual rotation of both set ups felt smooth.

Setting the distributor advance up, while intriguing is proving to be more involved and perplexing than I had initially thought. Comments welcome.

Kevin
Image Unavailable, Please Login
Image Unavailable, Please Login

 

I wonder, why don't you switch to 3 lobe cams?

 

Kevin - I have been following this thread with interest, but am beginning to question the initial premise. Your original graph revealed a remarkably accurate tracking of the manufacturer's THEORETICAL advance curve. I do not believe that the manufacturer of the distributor every intended (believed) that they could produce a mechanical device to stay within the tolerance limits throughout the RPM range. You could spend a great deal of time trying to "tune" your weights and springs and never produce an advance curve mimicking ideal. Nor do I believe that the engine will know the difference.

I assume that your pursuit represents an interesting exercise, but if you really wanted to tune the distributor to performance beyond original, then why not introduce a modern electronic approach with magnetic pickup on the pulley or flywheel and microcontroller programmed advance. Then you could, indeed, design a perfect advance system.

Jim S.

 

Hello again,
Remember what I said about Harrison. Patience. Your work is not in vain. A few points to keep motivated: First off, you can see from your graphs that having the pins polished produces a smoother curve. Also, electronic setups get the timing within 1/10 of a degree or better. Modern engines, and race engines use crank triggered devices etc etc to get the spark dead on. When it comes to tuning, having the distributor confidently set is a great starting point. I've setup my 308 with the distrib setup not-to-spec, and I've been back in there to re-do it, and it was worth the effort. Another point I'll throw out there is it won't work better on the engine magically. What I mean is: If it's not set correctly on the bench, it won't be fixed bolting it to the engine.

Some other tips: Spin the distrib up and down in revs, maybe with an electric drill to "settle in" the parts before you test. Bosch grease for distribs is thick. Use care when filling the spring cups. Do not fill them solid. It will restrict the advance. With your pointer setup, did you get the 15 degrees range of motion without the springs? I can't recall. If your advance is shy by a few degrees with springs in place, you could have too many shims and coil bind. You can't force it to work. Take a break and stratagize your next attempt. Try to cling to how it was setup in Italy. Pins with flats should be replaced. Smooth out divets as best as you can. Keep the parts clean from grit, and grease carefully. I found my Dino distrib sheet. I'll try and post it. It took many attempts to get it as good as it is, and it's not perfect. It's about 1/4 degree too high between 4800, and 5000. Don't let it make you crazy. The difference isn't going to be between your engine making 63 HP, and 500 HP, so if you don't get it in spec after some endpoint you set for yourself, just get it as close as you can so you can enjoy the car, then try again later. Obviously you don't want to be too far off to the point you hurt the engine. ie. too retarded at high end will make headers glow, too advanced at the low end makes it tough to start.

BTW: I spent a few hours today polishing out divets from the spring pins. Tomorrow I'll be back on the machine.

Hang in there.

 

Jim, a few reasons for the approach:
Before I started the distributor work, the car regularly spluttered and back fired on the over run, more so than most other Dino's have listend to. I have progressively worked through potential causes, including air leaks and carburetor adjustments. The distributor was next on the list. Plotting the advance as it was, showed too much progressive and erratic advance. In order to correct the advance I needed to know how the system worked. Progressively I have been through simple cleaning, logical spring and shim replacment, each time plotting the advance to see if the changes make the anticipated changes. Its taken a few attempts (currently 10), but I have iterated and converge on a reasonable advance profile. Furthermore, the engine is running much smoother and without backfire as well as minimal carburetor splutter at idle. I will shortly be summerising what I have done and the result, so others can get there quicker.

I too doubt that the manufacturer expected a real distributor to follow the profile shown in the manual. It does appear odd that the advance follows four discrete linear profiles. I would have expected an engine to respond better to smoother inflection curves. However, the manufacturers profile and tolerance is all I have to go on.

The reason for my most recent questions, has been to see if anyone has come across the non linear advance above 3400rpm and the maximum advance falling short, both out of tolerance. I think its likely that the maximum advance is quite important for high speed running. I would expect a 4deg error is significant.

I am sure there are modern and more accuarte ways of acheiving the advance, however, I would like to stay as original as reasonable. The coughs and splutters are part of the character of the car, even if its not the optimum set up.

Kevin

 

Yes i have considered this, but have been holding off for a few reasons:

1) I don't like replacing anything until I have got it working the way it originally was supposed to. (just stubbon and probably irrational)
2) changing to a 3 lobe system does require buying a 3 lobe assembly (not cheap), replacement base plate for the R2 points, jig drilling the distributor casing for the correct phasing of the replacement points, and a small wiring change.
3) I am considering replacing the points for an optical or magnetic system. Adrian has put a very useful list together under "dino ignition alternatives". After some follow up discussion, it looks like with the Pertronix system I will not have to buy a 3 lobe assembly replacement.

The 6 lobe cam R1, R2 system is a little clumsy and somewhat an after thought to satisfy US emissions. I don't have too many reservations about changing it to an earlier dino set up.

Kevin

 

Kevn - makes perfect sense. I would do as you are doing. Trying to keep the car original works for me.

BTW, I suspect that most of the trailing-throttle popping is related to carburettor tuning.

Jim S.

 

3 lobe cam is more original.

Agreed.

 

Kevin,
Wrote a series of "Sagas" about my distributor a couple of years ago. A sample:
http://ferrarichat.com/forum/showthread.php?t=98213&highlight=springs+weights
You really don't need a machine if you have some model making tools, a plug-in electric drill and a way to find how fast the drill is spinning. Putting the 0-80 tapped hole in the rotor doesn't hurt the rotor. Cheap 3/8" drills usually go up to 2500 rpm, it will say on the label. The LED is not special, anything Radio Shack has is fine. Small is good. The 200 Ohm resistor is not critical, just figure on about 20 mils current to the LED from whatever power source you use. Any power supply over 5 volts would be fine, just adjust the resistor to suit. Wire so the points short out the LED and the light comes on when the points open.
I triggered a 'scope from the points, six pulses per rev. Crank speed is twice cam/distributor speed so 3 pulses are 1 rev. 1,000 RPM / 60 = 16.6 crank revs/second. *3 = 50 pulses/second. 5,000 rpm = 250 pulses/sec. Easy to measure on a scope or counter.

I held the distributor in a vise and hand held the drill. I coupled the drill to the distributor using rubber tubing over a stud clamped in the drill and the nut on the end of the distributor shaft. Took 2 sizes of tubing.

Started out with long springs touching and weights against inner plastic stop. Short springs just a little loose. Spun and saw break as inner springs came in, too late. Hit 15 deg too soon. Total travel was fine. Added shims to long and short until 2,000 and 5,000 were right. Ballanced sides, used some of my washers. Did not try for midpoints, just looked for smooth operation up and down. Car runs great. 1,000 idle works well. Looks nice and smooth when checked with timing light in car. Does not seem to be a really critical process.
John

 

John, yes I did read your post on this. In hind sight I should have followed with your approach. I guess I was hoping to achieve a good curve with even less effort, but got involved anyway. Part of the issue is just getting to know how it works and what the variables are. I am there now, just need address the maximum advance issue and I will leave it alone and get on with the more important job of getting the heads off to sort the valve out.
thanks to all for the input.
Kevin

 

I love that approach. Very clever. I was starting to rig something up using an electric drill when a friend got a distrib machine. I finally got my distributor set tonight. It took literally days to do. I guess someone who is really talented could do it way faster.

 

Not a great deal to report, but thought you may be interested in the wear development on the distributor advance components. Its only 4000 miles since the drum and pins were polished. As expected the wear is minimal.
Kevin
Image Unavailable, Please Login
Image Unavailable, Please Login
Image Unavailable, Please Login

 

After installing a Bosch ignition and Lumenition trigger, ref:
www.ferrarichat.com/forum/showthread.php?t=276251&highlight=bosch+ignition
I thought it would be interesting to check the dynamic timing. The mechanical advance mechanism has remained unchanged from the previous measurements in 2008, 4000 miles ago when the distributor was cleaned and rebuilt. Replacing the electrical ignition should have no effect on the advance, however, I took the opportunity to do some checks. The ignition was previously running a Super 4 module with standard R1 and R2 points. The static timing in each case has been adjusted to match.

There are some small differences, most notably from 2400rpm where the latest set up is advancing approx 2deg less than before. I am not sure why this would be the case, the same pins, shims and springs and order have been re-installed. may be some bedding down of the pins and spring bases. Fortunately, the advance curve now fits very closely to that of the manual.

The maximum advance at 4500rpm is still falling short of the workshop manual 5500rpm. Nothing has changed to the advance system so would not expect it to do better. To address this I need to re-work the advance mechanism so that the weights have more spreading distance.

Another project on the list of to do.

Kevin
Image Unavailable, Please Login