Caution: Innovation at Work

We are pleased and gratified by the level of interest and enthusiasm the E-MAG ignition system is receiving.  Even so, it's important to not loose sight of our place in the Experimental environment.  

E-MAG is brand-new ignition technology.  That's good in the sense that it represents fresh ideas and innovation.  It also means the E-MAG system is less proven.  The population of systems in service is relatively small.  More and more systems are being installed, but the "real world" has only begun to judge the merits of our work.  The real world is not like a test bench, and not well represented by one or two test planes.  What matters is how they perform in the field, in a wide array of aircraft configurations, and over a long period of time. 

Several changes have been made, based on what we've learned from the first small production runs, and customer feedback.  We've provided follow-up service for some units, and made refinements to future systems.  This will be an ongoing process, and we want it to be as transparent as possible.  

As such, we've reserved this section of our web site to report on the status of E-MAG's integration in the field.  Here we can also report on our enterprise, and changes we make to improve production, quality, and service.  We'll try to keep these summaries short, factual, and with less promotional fluff.  They are presented here in familiar FAQ format. 

Note:  Much of this activity is in-process.  As such, we make no representation as to the accuracy or completeness of the information presented.  

Q: What kind of service issues have you run into, and what have you done to address them?

Q: Can you promise you won't have a "new and improved" unit come out right after I get mine?

Q: I've heard about an occurrence of excessive drive gear wear, and your gears not being hardened.  Can you elaborate?

Q: What kind of service issues have you run into, and what have you done to address them?

A1:  We identified several early units that had oil seeping past the oil seal.  The electronics section is protected, but oil was able to interfere with the position encoder.  We addressed the problem by repositioning and/or changing the style of oil seal. 

A2:  A few installations experienced start-up stumble and/or starting difficulties.  We are not (yet) clear on why this manifests with just a few engines, and not others.  Even so, we modified the firmware that manages the transition from start mode to run mode, which seems to have resolved the few problem situations.  This firmware update is applicable to all engines, and early production ignitions were all brought current.   

A3:  To further improve the low speed operating range of P-MAG units (well below 1000 rpm), our second series of P-MAGs (thirty three units) were made with improved (closer) spacing between the rotor and stator sections. Power output improves (or worsens) with the square of this distance. This change did improve the low speed performance, but also increased the high speed output of the alternator. Enough so that after extended periods at high speed, the power regulating components were operating close to their limit.  Built-in safeguards then cause these components to reduce output to avoid damaging themselves.  This produced symptoms such as a high-speed miss and variable tach output in some of units.  These P-MAGs are being serviced with components that will reduce component load without sacrificing low end alternator performance.

A4: We found instances where one of the wires from the P-MAG alternator was positioned directly underneath a thru-hole component on the electronics board.  The tip of the component was able to pierce the insulation of the lead which sent alternator current into the board.  Separation between the board and alternator has been increased, and the mechanism for attaching the nose section to the main case has been changed to permit greater control over the alternator wire orientation relative to the board. 

A5: Position encoder disks, if mounted too high, too low, or off axis, can cause an occasional high speed miss.  A fixture for positioning the heads during assembly will provide much tighter control.  

A6:  A percentage of customer reported an occasional high speed miss.  This was eventually traced to voltage shifts on the 12 volt supply buss.  It’ was not so much the amount of voltage change as the rate of change that caused a momentary advance or retard of ignition timing before E-MAG’s automatic dwell compensation was able to react.  The symptom is buffered to a large extent by the ignition’s onboard capacitors.  But a combination of a) longer wire runs, b) lighter gauge power wire, c) load from the ignition (rpm), and d) the size and rate of voltage shifts all work against this buffer.  These are conditions unique to each installation.  The pattern had not presented on our test bench, our test engine, or on any of the local customer installations, so it was tough to trace down.  But once we identified the problem and could replicate it in the lab, the cure was relatively easy.  Going to 18 gauge wire for power and ground (alone) will fix the problem for many.  If not, we modified the firmware routine (standard now in all units built since 7/5/05) to eliminate the symptom.  This firmware modification is available for all E-MAG models at no charge.  So far, installing this modification has cured all (but one) of the instances reporting the symptom.  This instance is still under study. 

A7: Composite Oil Sumps:  In a wasted spark system, plugs are fired in pairs at A) the charged cylinder that’s ready to ignite, as well as B) the opposing cylinder during the exhaust stroke.  This second spark is not intended to ignite, hence the name “wasted spark”.  However, after engine shutdown fuel vapors can accumulate in the intake manifold.  A hot start can then ignite what is normally the “wasted” side of the spark during valve overlap.  [Valve overlap is when intake and exhaust valves are momentarily open at the same time.]  If this happens, the event itself is rather unremarkable and is likely to be heard only as a hard “puff” prior to engine start.  However, the intake manifold channels this “puff” to the sump.  In one test cell case, this 15 to 20 psi pulse was enough to crack a light weight composite sump.  When replaced with a standard aluminum sump, the problem did not recur. To reduce the likelihood of "wasted ignition", a Run Mode Starting Delay (”RMSD”) has been added (a firmware change only) that prevents plug firing until the ignition sees two (2) passes of the TDC index.  At start up, this will help ventilate the intake chamber of accumulated vapor.  To date, we have no other reports of sump related issues.  So it’s tempting (and maybe appropriate) to narrowly define the one test cell example as the only configuration (fuel injection with side entry composite sump) at risk, but we don’t know that with certainty.  Either way, we’d rather approach such issues with solutions, rather than require customers avoid innovative and exciting new products.  The RMSD is standard in units after 9/1/05, and is available as a free upgrade to everyone.  The RMSD update is not mandatory unless you plan to use a composite sump.   NOTE: RMSD is a new feature, and we cannot guarantee that RMSD will, in all cases, prevent "wasted ignition" if conditions are just right.  In which case a composite sump, if used, could be damaged.  

A8:  Customers receiving the first of the 113 series ignitions reported seeing excessive cylinder head temperatures [the 113 production series introduced a number of changes from prior runs].  A review confirmed the system was executing command code perfectly, but the advance tables were calling for more advance than was needed.   Not all owners reported temperature issues, but we issued a Service Update covering all series 113 units shipped prior to that time to alert owners of the condition, and to arrange for reprogramming.  

A9:  We received occasional reports of some series 113 ignitions that would stop firing on one or both banks of cylinders.  After being powered OFF and back ON, the units resumed normal operation.  The frequency of occurrence varies, and some units don't seem to exhibit the problem at all.  We traced the cause to a latent firmware bug that was has been corrected with a firmware update.  

A10:  Some 113 series ignitions have failed to initialize correctly on power up.  When this happens, the stored index value is not picked up and the system defaults to the native index, which will cause the timing to be way off.  Relatively few units have reported the problem, and a firmware upgrade (version 20) guards against this condition.  If it does occur, proper timing can be restored by following the steps found in "Troubleshoot Timing". 

A11:  E-MAGs have defenses against 12 volt power brown-outs and black-outs that commonly occur during startup.  However, we found that if voltage dips into a very specific (narrow) voltage band and then rebounds, the position sensor chip may not reset.  This condition would appear as a non-operative ignition during the R/L ignition check.  When the unit is powered OFF and then ON again, it works fine, because the second start was not re-visiting the sensitive voltage.  We have altered the code sequence for power-up initialization and the problem has not recurred.  

 A12:  Quick-Set has been trouble-free for most customers, but difference in field elevation, buss conditions at start-up, operator over/under "puffing", etc. can complicate this seemingly simple task.  We have successfully addressed such issues as they come up by offering refinements to the operating code (firmware).   One such modification (in firmware v21 thru v23) assists with Quick-Set sensitivity when it's being used at lower field elevations.  This set of revisions inadvertently made another change that was NOT intended.  It allows the Quick-Set pressure triggers to be set by negative pressures as well as positive pressures.  This has the potential to defeat two of the three safety boundaries that were created to guard Quick-Set against accidental activation.  See Service Bulletin for full description and remedies.

A13:  You can set E-MAG timing using either our Quick-Set method, or our earlier (mechanical - rotate and clamp) method.   Both routines are described in the current manual.   

• Quick-Set stores a numerical value that serves as an offset reference from the position sensor's "Native" (built-in) index position.  When you set timing with Quick-Set, you overwrite (replace) the previously "Stored" value with a new one.  

• The mechanical method lets you set timing using the "current" (Native or Stored) index position.  

Some have mistakenly assumed the mechanical timing method automatically uses the Native index - which is NOT the case.  To set timing with the Native index, you need to first "zero" the Stored value in order to make the Native index visible/usable.  Until you do, the LED will only signal (by lighting GREEN) the location of the most recent Stored position.  

More recent firmware versions (v21 and after) give the operator a means of "zeroing" the Stored value.  Earlier versions do not have this capability.   Accordingly, earlier versions cannot set timing (with either timing method) to the native index (with the single exception referenced in A10 above). 

A14:  Occasional reports of engine roughness have been traced to faulty spark plug wire fabrication.  The installation manual instruction for crimping terminals now reads [in part] as follows:

....  Use a razor blade to trim 1” of shielding from the end of the wire.  The wire has a black outer jacket, a white insulating layer, and a center spiral wound (wire) core.  Gently cut the outer (black) shield and avoid cutting anywhere near the center core.   The white layer separates easily so you can twist the outer jacket in order to complete the separation. 

Note 1: The center core can easily be nicked by the crimp tool (if used) or a blade.   When you are finished crimping the terminals, a simple ohm test will help you spot a nicked wire.  Measured resistance should be approximately 180 ohms per foot of wire.  A nicked center wire might read several times that level. ...

A15:  Late series 113 ignitions (and after) have a feature that records the maximum temperature seen at the ignition circuit board.  The large majority of ignitions we’ve serviced in our shop have a recorded “Max Temp” in the range of 175 to 185 degrees (F).  This is within our recommended limit of 190 (the same limit specified for some, if not all, magnetos).  This shop reading will not tell us when the Max Temp was recorded (heat soak after shut-down, early break-in procedures, etc.), so we don’t know if the reading reflects the current operating environment.  Also please note:  Our temperature signal has NOT been calibrated, and serves as a general temperature indicator ONLY.   

Ignitions operating at exceptionally high temperatures can experience thermal shut down.  Such instances can be aided by the addition of (or improvements to) blast tube cooling, as is required by the current manual.   Experience has also shown the mere presence of blast-tubes does not guarantee they are operable and/or effective.  So we encourage builders (soon after installation) to verify the ignitions are operating within acceptable temperature limits.  E-MAG’s built-in temperature signal (later series 113 and after) can be displayed by one or more of the free configuration and control programs that are available -see Downloads page.  Other, more accurate, readings can be made with thermal reactive labels or with a thermal probe secured in a manner to measure ignition case temperature.   

The most frequent cause of inhibited blast tube cooling (we suspect) relates to exit air restrictions.  Blast tube air “blasts” because of the pressure differential across the baffle partition.  If accessory side exit-air is restricted, pressure on the back side can rise, and the pressure differential is reduced.  We don’t claim to be authorities in this area, but we have enough reports from customers who were able to significantly reduce ignition temperatures (as well as oil temperatures and accessory case temps overall) by improving exit-air flow, that we are confident this is, at a minimum, one of the primary causes.

Q: Can you promise you won't have a "new and improved" unit come out right after I get mine?

A: Definitely not.  We plan to be as innovative as we possibly can . . . right up to the time we go for FAA certification.  After that, nothing is likely to  change - for ever.  We need to do everything we possibly can before that time.     

Q: I read about an occurrence of excessive drive gear wear and your gears not being hardened.  Can you elaborate?

A:  We try to replicate certified gears in terms of dimension.  They are hardened (as appropriate) for use on our ignitions only.  

[ Home ] [ Introducing E-MAG ]

E-MAG Ignitions (TM)

2014 Greg Street. Azle, Texas 76020 

Copyright 2006.  All products, features, and prices are subject to change without notice.