Timing Experiments
obsolete info, past experiments
When testing my v3 setup on the bench with a 60-2 trigger simulator i noticed a thing out of the ordinairy:
above a certain rpm (which seemed to be dependened on the generated advance value, the dwell start shifted towards missing tooth.
a normal cycle (20 degrees advance, 3.2 mS dwelltime, 5020 rpm):
strange behaviour (20 degree advance, ~9mS dwelltime, 5345 rpm):
Where is the missing tooth from this?
These tests were performed with this [config], [tables] and [my_make].
The ign_tdcdelay of 110 degree just doesn't make sense.
For a multitooth trigger you have big freedom for moving it (in 6 deg steps in your 60 toothwheel). If your max ignadv is 48 degree, move trigger tooth to a later tooth so tdcdelay becomes 62 degree.
| mda value | 05 | 10 | 15 | 20 | 25 | 30 | 35 | 50 | 60 |
| prob. rpm | 9250 | 9000 | 8920 | 8780 | 8560 | 8320 | 8259 | 7630 | 7270 |
| mda value | 70 | 80 | 90 | a0 | b0 | c0 | d0 |
| prob. rpm | 6938 | 6584 | 6291 | 5971 | 5605 | 5252 | 4914 |
I played with spreadsheet (the trick is to calculate in usec, not degrees): the result is clear:
Dave has the trigger 48 deg after the missing tooth, and the tdc 62 deg after trigger:
(62+48)/62 * tdcdelay - ignadv is constant (1960 usec): the dwell value minus a small few usec constant
When the dwellstart reaches the missing tooth, the other channel was be set. It only effected certain setup of multitooth.
The above problem was corrected in dispatcher.
Back to: MembersPage/DaveBrul