The shop manual lists a spark plug gap between 0.020" - 0.025", and I usually set it near the middle. Will using the slightly smaller gap of 0.018" that you suggest help my hot-starting issue, and/or in some other way?
Although that BSA manual does indeed recommend a gap in that range, Triumph manuals recommend 0.020". Despite this, for at least 40 years I've used the classic magneto value of 0.018".
A larger gap is "better" because the bigger the gap the higher the probability a droplet of fuel will be in it when the spark happens. However, it is "worse" because it takes a larger voltage to initiate the spark. Because of emission requirements modern cars can't risk anything less than 100.0% combustion so they use very large gaps. But, they also have the necessary ignition systems to create the spark across those gaps.
Anyway, 0.018" has worked for me for decades. But, I have to wonder if your bike has other symptoms. Typically it's easier to start a bike when it's warmed up. I know that's the case with both my ET-sparked bikes. I wonder if your pilot circuit is too lean.
Yes MM, I do think I'm running a little lean and come Spring I plan to install a 0.107 I bought late last Fall. I did not realize that that might be the cause of my hard starting, as I assumed a full fuel bowl would obviate that issue. I thought the pilot jet mainly controlled idle. Thanks, it's fuel for thought. Adam
#635742 - 01/18/166:55 pmRe: Repairing an ET Ignition System
Joined: Nov 2011 Posts: 3,974Magnetoman
BritBike Forum member
I plan to install a 0.107 I bought late last Fall. I did not realize that that might be the cause of my hard starting, ... I thought the pilot jet mainly controlled idle.
You're mixing together two different jets responsible for two different regions. Your new 0.107 needle jet will make things leaner than a 108 (or a worn 107) for much of the normal riding range. However, a too-small pilot jet will make it harder to start and leaner at idle.
V. The ET Ignition System on a Twin-Cylinder Motorcycle
The setup and troubleshooting procedure for twins is the same as for singles, although with one very important item to note. When the left set of points opens to fire its coil the path the current takes back to earth is through the right set of points (and vice versa) rather than via a direct connection to earth as is the case for singles. Thus, the condition and adjustment of both sets of points determines whether or not a twin will run at all, let alone whether it will run well.
Make sure both sets of points are clean and in good condition, that they are properly gapped, and that the set of points on the right is fully closed when the ones on the left open (and vice versa).
Although you should check the Lucas listings for your own particular machine to be sure, it appears only a few types of points were used on singles and twins, with and without distributors. The following are the relevant Lucas part numbers:
BSA Singles with distributor (e.g. early C15S): 421106 Singles with points in timing cover (e.g. 441 Victor): 54415803 Twins (e.g. Cyclone): 54415803
Triumph Twins with distributor (e.g. TR5A/R): 425219 Twins with points in timing cover (e.g. TT Bonneville): 54415803
VII. Wiring Diagrams
There are other wiring variations but because of obvious similarities the following diagrams with notes should be sufficient to allow troubleshooting:
3-wire stator (late 1950s)
5-wire stator (early 1961 TR5AR)
4-wire stator for single-cylinder motorcycles (early 1960s) Brown - Black/White: 4.3 Ohms Brown - Red: 0.42 Ohms Brown - Brown/Blue: 0.81 Ohms Black/White - Brown/Blue: 5.0 Ohms
One configuration of 5-wire stator with a total of 4 ignition windings and 2 lighting windings inside the stator (mid to late-1960s). --For use in a single, run the Black/Yellow lead directly to earth.-- Ignition windings are ~1.1 Ohms each (i.e. ~4.5 Ohms total for B/W - B/Y) Lighting windings are ~0.4 Ohms each (i.e. ~0.8 Ohms total for Red -Brown/Blue
Two additional configurations of 5-wire stators, each with a total of 3 ignition windings and 3 lighting windings inside the stator (mid to late 1960s). --For use in a single, run the Black/Yellow lead directly to earth.-- Ignition windings are ~1.1 Ohms each (i.e. ~3.5 Ohms total for B/W - B/Y) Lighting windings are ~0.4 Ohms each (i.e. ~0.8 or 1.2 Ohms total for Brown/Blue - Red
I made the notes in this section primarily for my own use, and they aren't needed for troubleshooting an E.T. system. However, they will be useful for anyone who wants to wind their own coils or to pursue further understanding of this type of ignition system.
_________________________Primary___________________High Tension Coil____________Resistance______Inductance______Resistance______Inductance Lucas ET2_______0.640 Ohm______15.88 mH__________?_______________? Lucas ET3_______0.638 Ohm_______6.44 mH______5.2-5.6 kOhm________? Honda XL125____1.00 Ohm_________6.90 mH_______5.39 kOhm_______2.785 H
According to a 1965 BSA manual: ET3 Coil primary_____________secondary 160-165 turns_______12,000-12,500 turns (i.e. the physical turns ratio is 73-78:1; ~75:1)
By applying a known AC voltage to the primary and measuring the AV voltage output of the secondary the electrical turns ratio can be determined. This will be lower than the physical turns ratio because of losses in the coils. My measurements found:
3ET (measured) 50:1 Honda coil (measured) 67:1 i.e. the Honda coil is ~35% higher, usefully reducing the necessary input voltage by the same 30%.
A 1966 Triumph 500 uses stator 47197B and rotor 54215824 The four ignition windings in series (black/yellow - black/white) measure 4.8 Ohm The two lighting windings in series (red - brown/blue) measure ~0.5 Ohm
According to BSA for a C15 RM13 ET alternator the two ignition windings in series (black/white - red) should be 3.9 Ohm, and the two lighting windings (brown/blue - red) should be 0.3 Ohm.
According to a Triumph 500 manual, the windings on the stator are:
Ignition windings (2) 250 turns each of #25 SWG (equiv. to #24 AWG) (2) 98 turns each of #20 (#19 AWG) Lighting windings (1) 98 turns of #20 (#19 AWG) (1) 98 turns of #21 (#19 AWG)
According to a BSA A65 manual, the windings on the stator are:
Ignition (2) of 250 turns each (2) of 88 turns each
The following list is of stator numbers taken from Lucas spare parts lists (bold where a given stator is used for both singles and twins) --Singles-- '60-'61________47176________BSA (w/o stop light) '60-'61_______ 47177A ______BSA (w/ stop light) '61-'62_______ 47177_______Triumph '62-'66________ 47173 ______BSA; Triumph '67___________ 47173B ____Triumph '67-___________47197A ____BSA '71-'72_________47226______BSA. Stator cut-down by 2/3 to save weight, having only the two ignition coils. Used on the B50MX. --Twins-- '61__________47149________Triumph '61__________47165A_______Triumph '62-'63_______47173 _______Triumph '61-'62_______47177 _______Triumph '63-'67________47188A______Triumph '65-66________47188________BSA, Triumph '66___________47197________BSA '67___________47197A _____BSA; Triumph
Although it doesn't seem to be an issue in practice, a "feature" of the ET system is it doesn't have any voltage regulation so it could be possible to burn out bulbs at high enough rpm. That said, a light bulb itself provides crude regulation since the resistance of its tungsten filament increases with temperature so the hotter it gets the more resistant to increased current it becomes. Anyway, there is an easy fix for this for anyone who wants to minimize the chance of blowing a bulb on high speed runs at midnight. I implemented this some years ago on my C15S but only now made the measurements needed to post what follows.
For voltage regulation of their DC electrical systems BSA and Triumph added a single "14 Volt" Zener diode between the supply line and earth to keep the voltage at what was needed to charge the battery. When the voltage from the rectifier in such a circuit is less than the Zener voltage the diode has an extremely large resistance to earth so it's as if it isn't present at all. But, if the Zener voltage begins to be exceeded the diode becomes almost a dead short to earth so the voltage can never rise higher.
The same principle can be applied to the AC voltage in an E.T. system, although in this case two Zeners need to be connected in series between the stator's two wires that feed the headlamp and tail lamp. It doesn't matter if the diodes are wired "head-to-head" or "tail-to-tail" (but they can't be head-to-tail). This is shown in the next photograph where I have connected two low-power Zeners of nominal 6.5 Volts and attached the pair to an oscillator and an oscilloscope.
The next photograph shows the waveform at increasing levels from the oscillator, starting at 6.9 V peak (13.8 V peak-to-peak, as is shown in one of the boxes to the right of the trace) at the top and increasing to 14 V peak at the bottom. While this oscilloscope automatically computes peak-to-peak, it's half that value (i.e. peak voltage) that matters to a bulb. Note that the diodes act to keep the peak voltage in either direction from exceeding the nominal ~6.5 V of the Zeners. Hence, if high power versions of these diodes were added to the wiring of a motorcycle, the voltage across the bulbs never could exceed this value no matter how fast the engine was reving. Thus, the Zeners would allow the bulbs to lead a longer, happier life than they otherwise might have.
Since there a six magnets in the rotor the wave form goes through three oscillations for every revolution of the engine. Hence the 100 Hz I used corresponds to 2000 rpm (x3 / 60 sec/min. = 100 Hz). Referring to the bottom trace, it can be seen that even when the output of the oscillator/stator is high (i.e. 14 V peak) the voltage is clamped at +6.5 volts for half the cycle, then "instantly" (i.e. 1 ms for this example) jumps to -6.5 V. Since the power supplied to the bulb goes as V^2 the minus sign doesn't matter, and since the time the voltage is at a reduced value is less than that experienced by a light bulb operated from the mains at 50/60 Hz, the bulb reacts as if it were supplied with a constant DC voltage of this magnitude.
The next diagram shows in magenta where a pair of Zener diodes of appropriate power should be added.
When placed after the lighting switch the diodes only have to deal with the excess output of the stator since the bulbs will be dissipating most of the 35 W when the switch is 'on'. If the bulbs haven't failed the Zeners might have to dissipate an excess of maybe 10 W (5 W each) at most. However, if the headlamp bulb blows out the Zeners would be subjected to nearly the full output of the stator. This argues for using a pair of 20 W Zeners (40 W total) located outside the headlamp shell, for cooling, to be absolutely safe.
It's been a decade since I bought the ones I installed in my C15 so I don't remember where I got them nor their specifications. However, 10 W and 50 W seem to be the easiest to find these days, with more power handling costing more.
Update: In rearranging some drawers I came across a spare clipper diode array I made some years ago, shown in the next photograph:
As can be seen I have the diode pair connected as in the circuit diagram above and held in an insulating "bracket" that I made from teflon or delrin. I fully tightened the nuts and then soldered the terminals to avoid mechanical strain.
The part number on the diodes (NTE5180A) shows that they are 6.2 V and 10 Watts each so they can handle an excess of up to 20 Watts. This is more than sufficient to keep the voltage from exceeding 6.2 V, although if the headlamp burned out all ~35 Watts from the stator would be dissipated in the diodes. Unfortunately, unless someone finds a source of 20 Watt diodes the next step up is to the much more expensive 50 Watt. Another note is this array can't be located inside the headlamp shell because it needs free flowing air to cool it.
A final note on this is that a nominal "6 volt" DC circuit has the voltage regulator set at ~8 volts. Since "6 V" bulbs survive that voltage it means use of an 8 V Zener is appropriate, and also will result in a significantly brighter light than if operated at 6 V.
[The End (unless I think of something additional...)]
Last edited by Magnetoman; 01/24/1711:33 am. Reason: Update
can i use the 6ac points plate instead of the 4ac type with the ET IGNITION
Also ,can i eliminate the condensers that are mounted on the points plate, still using only the long ones mounted on the coils
what is the best timing for just riding around. S,37 or M, 39 degrees
Yes, the 6ac is a direct replacement. Ideally, you will have condensers at the coils and across the points. Electrically, it might look like having them at the coils is the same as having them at the points but the inductance of the wire between the locations add reactance, i.e. somewhat reduces the effectiveness of the condenser as providing a low resistance alternative path for the current when the points open. The bike should run but the lifetime of the points will be reduced to some degree due to increased arcing.
Is the 6ac points plate better than the 4ac plate for the ET system?
The 6AC incorporates an additional adjustment allowing the gaps and the opening positions of the two cylinders to be separately adjustable. With the 4AC production tolerances almost always result in having to have different gaps in order to have both cylinders fire at the correct position BTDC. A 4AC will work, but a 6AC definitely is better.
Thanks Magnetoman for schooling me on the points plate.
The wiring and coils on my í67 TR6C were in real bad shape. But the stator passed all the continuity tests described in the Triumph Service Bulletin 67/14. Also, I sent the rotor to Joe Hunt to get remagnetized.
After reading your posts on restoring an ET system I bought a new wiring harness and connectors, two Honda 30500-950-405 coils to replace the crumbling Lucas coils, and 4 new re-pop condensers. Now I'll get a 6CA points plate and new points.
Iíll use those new parts along with the original ET 5 degree AAU. I'll connect the points to the condensers that are on the Honda coils and wire those in parallel with the separate condensers. Does that sound like the right approach?
I'll connect the points to the condensers that are on the Honda coils and wire those in parallel with the separate condensers.
Your wording is a little ambiguous, but wire the Honda coils just as the Lucas. The Honda coils have capacitors riveted to them so those take the place of the separate capacitors Lucas mounted at the coils.
For what it's worth, I snip the wires to the Honda capacitors and install spade connectors to reconnect them. That way if I ever suspect the condenser is causing a problem can easily isolate it from the circuit to test, if it's in my garage, or to temporarily connect a spare from my tool kit if I'm on the road.
I am making a 66 ,tt and tr6sc . I would use a different ignition set up but the bikes are restored to factory spec.I Set the timing at the S location on the rotor book, says 37 degrees for s I tried the ac6 points plate and mounted a condenser in line up at the coil area grounded and out of sight. No spark I went back to the ac4 and we are sparking Like to use the ac6 ,so much better for set up. Any advice
The fault is due to either a short or an open connection somewhere. The only way to find the problem is to troubleshoot the circuit with ohmmeter in hand.
Reminder:I'm sorry, but I can't answer individual PMs about issues with ET systems. If you have a question that hasn't been addressed yet please post it to this thread and I'll try to answer it if it's of general enough interest.
Last edited by Magnetoman; 04/07/165:20 pm. Reason: added 'Reminder'
The illustrations that were included with your original ET system posts were very useful, but they have disappeared.
All the figures show up when I go through the posts of this thread. Since they're all hosted externally, on Photobucket, I suspect your computer is blocking access to that site for some reason. Have you recently upgraded your security software? Check your settings.
I know this a old post , but I thought I lost a rotor key and bought one and for the life of me I couldn't understand why I couldn't get every thing lined up with the peg and key ! Now I know why ! Thank you !!
Hello MM. I have a 1966 A65 BSA Hornet that I received as a non-runner. As can be seen from the photo, when the crank is at TDC the locating pin for the rotor is at approx the 2 o'clock position. To get the bike running* I must re-orient the pin back to the about 9 o'clock position... is that correct? *And of course locate the rotor on the "S" position for street use regards
when the crank is at TDC ... I must re-orient the pin back to the about 9 o'clock position... is that correct?
Sorry for the delay in responding. Thanks to satellite wifi technology I read your post yesterday while at 30,000 ft. over Greenland but, thanks to having departed the hotel at 3:30am, I decided it best to sleep for a night before responding.
Yes, with the pistons at TDC the "Timing Disc" (BSA's terminology for the piece on the crankshaft with the peg that locates the stator) should be at approximately 9:00.
The plugs require the least voltage to ionize the air when the center electrode is negative, and correct orientation of the stator on the crankshaft is needed to ensure that is the case. However, the ET system should work with the Timing Disc in any of the six positions. It's just that in three of them the voltage will be rising when the points open and falling in the other three positions. I suppose for simplicity's sake BSA identified just one of the three 'correct' positions as 'Correct'. Note that since 9:00 is the 'Correct' position, the other two 'correct' ones will be two slots (120o) before and after that, i.e. at ~1:00 and ~5:00). Hence, although where yours is now isn't 'Correct', it's 'correct'.
p.s. if the previous paragraph is confusing, just ignore it. Set your Timing Disc at 9:00 with the pistons at TDC and all will be well.
The world traveler! Yes, makes sense. Someone had replaced the ET with battery and coil, but looks like the primary chain was replaced too and that's just how the "disc" ended up (photo). Thanks again magneto... Hope you had a good flight!
Sorry to be a pain MM but I forgot to ask about the auto advance, and hope you will like this question as it touches on the theoretical.
As mentioned, someone converted my Hornet to battery and coil and I am now reverting back to ET. I obtained a proclaimed ET auto advance from a vendor (roll of the dice) and he said he is not sure if the auto advance is from a single or twin, but is definitely ET. Pics:
Thus my question... is the ET auto advance unit the same between the singles and the twins? There are a dizzying number of Lucas part numbers for these depending on model, but in photos I have seen -- at least wrt the cam profile -- they all look very similar. If there are differences is it possible to tell whether the advance in the photo is correct for a twin?
is the ET auto advance unit the same between the singles and the twins?
Disclaimer: I haven't done precision measurements of the cam profiles, spring constants, location of pivot points, or weight of the bobs so my answer is best-guess based on the ones I've examined over the years.
I can't imagine why the shape of the cam or any other aspect of the assembly -- other than possibly the springs -- would be different between those for singles and twins, and as far as I can tell from the ones I've examined there is no difference. Of course, a twin needs two sets of points, but the fundamental auto-advance assemblies seem to be the same. However, since the springs adjust the advance curve, i.e. the rpm at which full advance is reached, I wouldn't be surprised if differences were to be found there between different models.