Rotax 912 Ignition Problems.

[rogerCT2K]

Hi Carl and all,

 

I didn't know sending an Email adress was illegal!

 

About the 2000rpm idle of the Rotax 912: I saw a 912UL turning at 750 revs in Germany, equipped with an electronic injection and ignition system of a Check compagny (Silent Hectic or so if my memory serves me well, it has been a long time ago) and it turned smooth like butter. One of the reasons you have to let the Rotax run at 2000 revs is the ignition, 6° advance is way too much at low revs! So electronic weirdo's stop what youre doing and make us a ingintion box with a good curve!

 

Best regards

 

Roger

 

 

[Admin]

Carl, you can just use the PM (Private Message) system to let Roger know your email address

 

 

[bones]

Hi Carl and all,I didn't know sending an Email adress was illegal!

About the 2000rpm idle of the Rotax 912: I saw a 912UL turning at 750 revs in Germany, equipped with an electronic injection and ignition system of a Check compagny (Silent Hectic or so if my memory serves me well, it has been a long time ago) and it turned smooth like butter. One of the reasons you have to let the Rotax run at 2000 revs is the ignition, 6° advance is way too much at low revs! So electronic weirdo's stop what youre doing and make us a ingintion box with a good curve!

 

Best regards

 

Roger

hmmmmmm seems as though my 2 rotax engines are just not right then, purr nicely at 1200-1400 rpm.

 

how the hell do people tune their engines ????

 

No wonder i hear so many horror stories.

 

 

[PapaFox]

Prop RPM perhaps?

 

 

[rogerCT2K]

No prop rpm's, engine rpm's!

 

 

[rogerCT2K]

Hi all,

 

I just checked the net for "Silent hektic rotax 912" and it seems the system still exists.

 

So there is one of the answers to the question "how the hell do people tune their engines ????"

 

There are probably others too.

 

It seems the system takes 10 Amps, I am not sure if I have 10 Apms to spare ??

 

I also do not kow if one can purchase the ignition system seperately.

 

Best regards

 

Roger

 

 

[carlsnilsson]

Roger et al:

 

Well, I thought it was time for me to butt out, but I can't let Roger's brief report on the "Silent Hektic" go without comment. It's the "takes 10 amps" that has stirred me up. As an old "semi-electronic wierdo", I have to comment that if a system takes 10 amps (i.e. 120 VA) , then it is almost certainly supplying the power for the primary ignition. That's not necessary to provide a smooth advance - retard timing system. For the Rotax 912 /Ducati system, you can let the magneto continue to provide the primary power and provide the timing control independently - as the present Ducati system now does - except that that damn part is sealed and buried in the units. In principle, you simply (?) delay the triggering pulses by an appropriate amount dependent on the rpm. According to the manual, the present system does that now in two crude steps - 6 deg BTDC for starting and 26 deg BTDC for rpm > about 600-900. One reason the engine may run very rough below 900 rpm is that the timing will chatter between the two. (I thought "piston slap" was the stated reason behind having to keep rpm > 1400 on the 912 UL - I don't fully understand that - why is that worse at low rpm?). In any case, the triggering of the spark is done by triggering a silicon-controlled rectifier (SCR) and that doesn't take much current. Takes more power at high revs because it has to trigger more often, but I can't see needing anything like 5 amps (!!) per circuit. Gate currents for SCRs may be up to 100 ma (I think) , but that is on for only a small duty cycle, say 2 X 5% for one ignition unit or about 10 ma average. (I could be a long way out, because I have no idea what Ducati actually have in their "timing control circuit" within the ignition module and I've never actually used an SCR). The fact is that SCR's are pulse amplifying devices and the power going into the whole modules for sparking will be less than, say 100W, and so only a few watts at most should be needed for triggering. My guess is that the two triggering circuits per module would need less than 200 ma. (yes, I know I don't actually "know"!).

 

Integrated circuits have progressed since the Rotax/Ducati systems first came on the market. Conceptually, you could now measure the rpm with a cheap microprocessor drawing virtually no current and use that to control a (always positive) delay on the triggering pulse you send to the SCR.

 

In practice, I would not want to replace my Ducati modules and so, like the "SoftStart" module, I would only want maybe an additional delay to the triggering to make for a "softer" start. You don't need a microprocessor for that - just some simple fixed delay. You can't generate a true negative delay to advance the spark for high revs; however, with a microprocessor, come to think of it, as the whole thing is cyclic, if you measured the rpm, you could delay the triggering by a whole 360 deg (crankshaft) LESS however much you wanted to advance the spark! Hmm, that would work. Then you could produce a nice timing curve as a smooth function of rpm. That would be a cool add-on! You saw that here, dear readers! (If I've thought of it, it is probably very old hat and used in cars already).

 

Rergards

 

Carl

 

 

[rogerCT2K]

Hi Carl,

 

I think you undertstood me wrong, the 10amps is for the ignition "and" the injection system, mainly to boost the fuel pressure (pump) I think.

 

And for the strong vibrations at low revs: -6° is just way too much, it could even be that the best is + a little something!

 

That can be tried out with some electronic or mechanical hocus-pocus.

 

Best regards

 

Roger

 

 

[rogerCT2K]

It is strange that Rotax sticks with a system that dates back to the seventies. The ignition timing should be a curve instead of a two level system. To take an example somebody said here mine turns perfectly at 1200 revs, but don't forget that at that moment the pre-ingition is 26°, that is probably the ideal value at max rpm.but certainly not at 1200 rpm.! In this digital era it is very simple to let the timing be at the ideal moment at any number of revs, one could even change it with the load and the atmospheric conditions. But all this has le

 

 

[rogerCT2K]

But all this has lead us far away from a quite simple problem, how to solve a bad mag drop.

 

Best regards

 

Roger

 

 

[carlsnilsson]

Oh dear! There's me rattling on about SCR's based on a misunderstanding! Comes of being late at night and those funny pills! I agree, the 26 deg may be fine for full power, although Ducati didn't have much room to play around. (You have to be able to start - would not want any more than 6 deg BTDC or the poor old sprag clutch would suffer even more - and the 26 deg would be set by a trigger pulse width apparently of 20 deg). If you only need positive delays - i.e. no more than 26 deg BTDC, then that makes it simpler to design a "front-end add-on". I'm a bit curious as to how they switch from trailing edge of the trigger pulse (6 deg) to leading edge (26 deg). Have to look at the pulse shape sometime. Anyway, all (!) one really needs to do is adequately simulate that pulse shape and provide a variable delay based on rpm. As I said before, I have an ambition to be able to hand start the 912 (in the case of a flat battery in some corner of a foreign field) and it would be beneficial to time the spark a bit later for that. The SoftStart module must do that. An agent in CA for that module told me that it was no use on these older ignition units - which doesn't make sense to me. Has anyone who might be still be following this thread actually looked at the timing on these older 912UL engines with the modules inside an Interference Supression Box? If all the SoftStart does is simply delay the two timing pulses in one circuit, then there is no obvious reason why it should not work.

 

Yes, I know we have drifted away from solving a slightly excessive mag drop problem, but then that's the beauty of a forum like this and maybe why they are called "threads". All hail and thanks to Ian the webmaster.

 

Carl

 

 

[rogerCT2K]

Thanks for thinking along on the path of the ignition of a Rotax 912 with me for a while Carl!

 

Best regards

 

Roger

 

 

[Guest willie7]

I pushed my Rans S6 out of the shed this morning, to go flying but she wouldn't start. Few a couple of weeks ago, without issue.

 

After dropping cowls,checking fuel delivery etc I eventually ended up at the plugs with no spark on either side.

 

Unbelievable.

 

I pulled everything to bits, metered the stator, checked connections and wires, still nothing.

 

I read this thread and thought I'd need to send the engine away as I was sure it was a module problem.

 

Before walking up for lunch, some 5 hours later, I put the meter on the earth between the modules and it didn't buzz ! I dropped off the three earths, cleaned up the terminals, washers and bolts, hooked them up again and the 912 fired up first turn !

 

Bloody electrics !!! Thanks for the useful info contained here.

 

 

[Davyxt912]

The ongoing story.....Sent the ignition modules to B Flood, checked OK on their test bench.

 

Got them back and installed one of them on another engine - NO SPARK there either......

 

I've now heard of another engine that had no spark on both sides, modules tested OK by B Flood, but engine still wouldn't start until modules were replaced by new ones.....

 

So I've just ordered a new module from a friend in USA who used to service Rotaxes and has it in surplus stock. ONLY A$850.........

Hi mate iv just seen your post. My ignition modules have failed can you recommen a suplier. ?

 

 

[JG3]

Hi mate iv just seen your post. My ignition modules have failed can you recommen a suplier. ?

Get in touch with Roger at [email protected]

 

JG

 

 

[Davyxt912]

Get in touch with Roger at [email protected]JG

Cheers JG3 thanks so much for that much appreciated.

 

 

[dan tonner]

Hello JG;

     I realize I am replying to a very old thread you started but which was hijacked about 2/3 the way along....

To come to the point:  my very well working 912 UL just stopped while I was back-tracking for take-off. If I didn't know better, I'd have said someone flipped BOTH mag switches to ground.  The engine was included in the Rotax 912-026 SB of 1999, and the logbooks do not reflect a stator change.  The SB was 21 years ago and the engine has been a gem for about 1000 hours - about 300 of them, mine. 

      Here's what I have tried without so much as a *cough* from my engine since the catastrophic failure:

1. The mag switches were tested and found to be working.  (I even tried a start with the kill-switch wires disconnected from the modules.)  Nada
2. Continuity tests were done on the charging coils and the trigger coils - I checked resistance between engine ground and each of the charging coil connectors at the modules; each checked out at 3.6 ohms - exactly on spec. I also checked the resistance of the four ignition trigger coils and all were within spec around 230 ohms. 
3. I followed the fuel supply from the tanks right up to the carburetors - disconnecting, checking and re-connecting at every stage;  the final test was disconnecting the fuel lines from the carbs and opening the valve in the cockpit. "Gravity only" fuel flow was steady through all lines, the filter, BOTH fuel pumps, the "spider junction" and into a cup.   I also carefully removed each float bowl and found the bowls full, the floats "floating" and the float needle adjusted properly.
4. I checked the throttle and choke connections and found all to be as they should.
5. I tried restarting after removing my BushCaddy soft-start module.  Not a cough.
6. I removed both modules and the BushCaddy and installed a single known-to-be-working module wired to one set of plugs through their respective sparkplug coils. When that produced nothing, I switched to the other set of coils. Nothing.  I tried the other combinations of coil-to-modules connections. (I didn't expect this to work since these combinations only involved only two cylinders - it did not.)  I tried these combinations with and without the kill switch connected. Still nothing.    

       All testing was done over a period of a week and every day's testing involved a freshly charged battery.  Cranking speed was normal and on cooler days, I warmed the engine with an external heat source.   

      One thing I did not check was the trigger coil gaps - but it would seem rather odd to have all four suddenly be unable to trigger a spark.  I used the same logic to excuse the four dual sparkplug coils from possible failure.

 

     Ultimately, what resolved your issue back then?    Fortunately (I suppose) my problem does not seem to be intermittent; I just can't find it.

Any suggestions?

 

CanadaDan

[skippydiesel]

New to this conversation -

Have you actually determined you have a spark ? - inside hanger with low light, remove spark plugs from engine, connect plugs to HT leads, rest or clamp plugs to engine,  rotate engine on battery have observer look for spark - repeat for each plug or pair of plugs. If no spark - investigate modules , fly wheel sensors

If you have spark, is it at he correct time in the engine rotation?- use multi meter or similar, plugged into HT lead to determine when spark occurs relative to piston position. If spark timing wrong, fly wheel or sensors may have moved/slipped.

To find out if you have a fuel delivery problem - try spraying a very small amount of fuel into the inlet (s) - if she fires at all, the problem is likely to be fuel delivery.

Blocked exhaust - collapsed muffler baffles or similar, will not alowe gas to escape, engine will not run - check exhaust system free flowing

If spark good & correct time, fuel being delivered problem likely to be more serious mechanical issue.

[dan tonner]

Thanks for the reply Skippydiesel;

     I only did the "spark test" on a couple of plugs - there was no visible spark.     I will use a spray bottle to send atomized fuel into the carb intakes (air filter removed) while cranking the engine.  If there is any spark  -  properly timed or not - there will be some form of ignition - possibly an insurance claim.   = )

     A squirrel trapped in the muffler is a possibility I haven't fully investigated.

     The "suddenness" of the  engine shutdown on the runway after months of perfect performance is, I think, key here.    There was no hesitation, roughness, slowing of the engine (taxiing at about 3000 RPM), and no "kickback".  The engine just stopped.   And no restart.  A visual engine examination afterwards revealed nothing out of order.

 

     We're on the same page with your analysis.

Stay safe. Fly safely.

CanadaDan

      

[skippydiesel]

I have been "caught" quite a few times by blocked exhausts. There is an expectation that if you have fuel & spark the engine will run - This is general true, if spark at correct time & gases are able to " flow" .

Admittedly the engines that have had the blocked exhaust, have all been small  4 & 2/ engines. The cause, mainly mud wasp (not too prevalent in Canada)  or in the 2/  the  gauze spark arrester has gummed up over time . No matter the principal still exists - the gas must flow from inlet to exhaust for any engine to run. Easy test - get someone to rotate engine while a hand is held over exhaust - no (or minimal) air pulse you have a blocked exhaust. In the Rotax 912 this would likely be collapsed internal baffles or unlikely & only if engine burning prodigious quantities of oil, large deposits of partly burnt "glug"

 

Going back to your persistent nil spark - as highlighted in an erlier comment in this conversation - Negative earth (ground in Canada) must be in good condition for ignition system to work properly & safely. Part of your investigation should involve a thorough check of electrical continuity. Do not discount a wire being  broken internally. All earth points clean & tight. Give all wires a gentle tug - look for unusual movement, even elongation of insulation, as a "tell tale" for a bad contact

[dan tonner]

Thanks again SkippyDiesel;

     I wasn't joking about the squirrel in the exhaust...my hangar is rural and I have trapped a squirrel attempting to set up a home in the hangar before.... and,  "bin there; done that with blocked chainsaw and lawn mower exhausts   (carbon though, not rodents.).

     I have removed, sanded, applied dielectric grease and reassembled the ground wires for the coils and modules. Also double checked the ground wire between the engine, battery and frame  (I use a copper busbar grounded directly to my battery for the engine, all instrumentation and switch grounds.)

     The SB on the early 912 stators was due to an insulation issue that included the two charging wires that run to the modules. In these early stators, (and mine was one, but I don't know if it was ever swapped out) both insulated wires were bundled together inside a "fabric-type" cover that was then contained in the grounded wiremesh sheath. It seems the insulation over the copper wire tended to break down, exposing the copper wire and allowing for corrosion, and possible breakage and/or cross-connection with the other lead. The "fabric-type" layer would have to deteriorate or break to cause a short to ground.     My 23 year old motor will have had lots of time for this insulation to fail. The "improved" stator has each insulated wire inside a black plastic insulation tube, bundled inside the "fabric" sheath and further bundled inside the wire mesh sheath which is grounded. I believe the wire itself also had an improved type of plastic insulation.

     HERE'S A QUESTION:

     I am trying to understand how it is that I am still getting spec-level 3.7 ohms of resistance between each charging coil lead and ground if the charging coil leads are cross-connected somewhere inside the wire sheath....it seems clear the leads are not "grounded" before reaching the coils or I would see zero ohms.

     Assuming those leads are cross-connected, what would you expect to see on a multimeter checking resistance from the connectors at the module end and ground?     Perhaps more significant, what would happen to the current if the leads "joined" inside the sheath while the engine were running......a surge?.......a reduction of curent?......no current?

 

Are we having fun yet?

"best,

CanadaDan

[Kyle Communications]

Dan if there is a close gap..I mean extremely close from one of the coils to ground you wont read any resistance until the engine is rotated then the higher voltage induced into the coil could track over to the freame..this will stop the CDI from working...did you disconnect the main power wires from the CDI 2 generator coils and test with a multimeter on AC while spinning the engine over?..this is important because it will tell you if is what is happening

 

[skippydiesel]

You have just about exhausted this old cocky's (farmer) knowledge base.

 

A sudden engine out & no spark situation just screams vibration/movement damage to a common,  to both ignition systems, electrical point but the scream may of course be a diversion.

 

If I have read your investigatory process correctly, you have no spark :

 

Being produced by either of the two ignition systems

Isolating the ignition switch did not remedy the no spark

A, known to be working, donor module, did not produce a spark - ego the module(s) are probably OK

So the problem may be in the existing wiring that supports/feeds the modules

I suggest going over the ignition system wiring harness millimetre at a time, with a particular emphasis on any common wiring where + positive power supply is being delivered to the system and - negative output from the system. As you clearely are aware this might include any wiring being contained in a common duct or sheathing. Your scrutiny should also focus on any "tight" pinch spots that wires may have to pass through eg firewall.

 

Sorry I am close to not having much else to suggest.

[M61A1]

Do a resistance check on the stator, the two wires that go to the ignition units. I had a similar issue a while back. The flywheel/rotor had bee removed and the stator inspected IAW the SB with no visible damage noted some time prior.

I did all the things you have done and found nothing....until the resistance check, which was very low. I removed the flywheel/rotor and there was no obvious damage, but them I removed the sheath from the two wires to find that both of them were extremely munted up. Once separated the resistance check was normal.

After being quoted a ridiculous price for a new stator, I removed the old one and de-soldered the two damaged wires and soldered in nice new ones with high temp silicone insulation complete with good quality heatshrink and fresh wire braid.

Haven't had a problem with it since.

[dan tonner]

Thank you (all three of you for taking the time to offer suggestions...)

Mark: 

     Correct me if I've misunderstood you but, as long as there is no contact between the two CDI coils "hot" leads and ground, no break in the wire or windings  and a good ground at the other end, I will see "specification" resistance through these coils of 3.2 - 4.5 ohms. (not "zero" reststance)      (I get readings of  3.6 and 3.7 ohms on the static engine.)   

     I do understand you to say that the current could jump across a tiny "gap" to ground before reaching the modules when the engine is spinning if the wire insulation is deteriorated and being compressed at the hold-downs.    That is one scenario.  BUT; WHAT WOULD HAPPEN IF CURRENT CROSSED FROM ONE POSITIVE LEAD TO THE OTHER  POSITIVE LEAD DUE TO INSULATION FAILURE INSIDE THE WIRING SHEATH?     (I'm trying to understand why all resistance checks are "good", the module is good and trigger coil gaps are good but the engine won't make spark.    I haven't measured the AC from the CDI coils (disconnected modules) with the engine turning on the starter.   I don't know what voltage or amperage to expect and my multimeter has limited current capability; this test is not described in the Heavy Maintenance Manual and I don 't know how to properly set it up.

 

SkippyDiesel: 

      You have summarized where I am headed quite nicely. 

 

M61A1:

     Can you recall what the "very low" resistance readings of your failed stator were?  I am still getting specification readings from each (which happen to be "very low" at a little over 3.5 ohms).   It is possible that there is a crossover between the "hot" leads either at standstill or when the engine is turning.   I don't know what effect this would have on the resistance reading(s) across the CDI coils since they are quite low anyway at 3.3 - 4.5 ohms.

 

Thanks again foir the inpou. Stay safe. Fly safely.

CanadaDan