What's new
Van's Air Force

Don't miss anything! Register now for full access to the definitive RV support community.

Dual P-MAG Reliability in 2017

Wiring issue primer ?

I have two Pmags and two CPI . This tread has been extremely stimulating. That said, it appears the bottom line issue affecting CPI reliability is wiring termination failures ( ignoring errors ) , i.e. Corrosion, vibration etc. So, perhaps Ross or someone would start a separate thread on optimizing wiring termination from corrosion, vibration, etc ???
 
I have two Pmags and two CPI . This tread has been extremely stimulating. That said, it appears the bottom line issue affecting CPI reliability is wiring termination failures ( ignoring errors ) , i.e. Corrosion, vibration etc. So, perhaps Ross or someone would start a separate thread on optimizing wiring termination from corrosion, vibration, etc ???

We've used the same type of connectors on the CPI and our ECUs for the last 15 years and MILLIONS of hours collectively in environments much harsher vibration wise than the aviation environment (Baja 1000 races). When the wires are collected and tie wrapped at the controller end and supported as we recommend, I've yet to see or hear of a broken wire at the connector. Since our controllers must be cabin mounted, they don't see moisture, therefore corrosion is a non-issue in aircraft. Everything FF has a strain relieved, waterproof connector.

Best practice on wiring tells us to support bundles from movement or physical strain. Done properly, we simply don't see wiring failures at either end. Our old shop car had SDS on it for 19 years and about 5000 hours. Coarse old engine with fair vibration, heavily salted roads in the winter, no particular care in support or strain relief. No wiring breakages or corrosion in that time.

The Hall sensor is one piece with the cable, no interim connectors.

We just don't see problems here.
 
No, but that's why you have two independent ignitions.

Seeing you are a supplier of ignition products to the consumer, I'm going to assume that what you really meant to say was: "...No, but that's why you choose the proper connectors and follow established wiring practices..."

Mitigation of the hazard is much more productive than relying on a fallback.
 
Last edited:
My impression is that you are more likely to see a public complaint from someone that has an issue than a positive post by someone who has not.

That may be true, but it doesn't undermine the theory that faults in experimental ignition systems will be largely under-reported by users for numerous reasons. As I said previously there is no obligation on users to publicly report faults and quite a bit of incentive to not do so.

Brian, if you personally have a problem with one of your PMAGs, are you going to report that fault on VansAirforce...probably not if it's not in your personal interest to do so. I mean, there's no moral or legal duty of disclosure.

I have a LightSpeed Plasma 11 on one side (magneto on the other....I'm with Vic Syracuse on this). I'd have to think very seriously about publicly whinging about a fault in my EI on VansAirforce and then going cap in hand to Klaus at some time in the future for support. ;)

I mean, when you buy an EI from a very small, virtually backyard company like EMAG or LightSpeed you're not just buying a product....you're hopping into bed with them for good or for bad.

Incidentally I think that the transparency shown by Ross Farnham regarding SDS products in post #88 is just so refreshing. I'm sure his honesty will attract more purchasers to his CPI ignition system. I'm impressed.
 
NASA form

...Brian, if you personally have a problem with one of your PMAGs, are you going to report that fault on VansAirforce...probably not if it's not in your personal interest to do so. I mean, there's no moral or legal duty of disclosure....
Perhaps we need a NASA form for suppliers. :D
 
Brian, if you personally have a problem with one of your PMAGs, are you going to report that fault on VansAirforce...probably not if it's not in your personal interest to do so. I mean, there's no moral or legal duty of disclosure.

Absolutely and I have done so on other products as well. The best manufacturers take the criticism and do something about it.
 
Dan, as always, you bring in good logic; however, once the P-mags are spinning, one or more of its wires can fail and it will keep the sparks flowing. (Unless the failure is in a plug wire.) How many wires can fail in a Distributed system before it stops producing sparks?

Fair point Bill.

Recall wires can fail open, or shorted. I typically make a list with each wire represented and think about both open and grounded failure modes, ignoring the probability of failure, and initially only considering what happens if it does fail. Example here: http://www.vansairforce.com/community/showpost.php?p=1153558&postcount=63

A wire list for the CPI says shorting or opening almost any pin will shut it down. On the other hand, it does not appear any open or short can make it do strange things. Put another way, it works or it is dead.

In the case of the P-mag, the result of an open or short on a few of the wires is obvious. However, I don't know enough about the p-mag to be sure what happens when some of the wires are opened or shorted, notably software driven events.

Pin 1: I think you just stated the system works with this one open. If shorted is normal and open doesn't matter, the logical question becomes "Why is it there?" Back up for case ground to engine block?

Pins 2 and 3: Without an EIC, an open shifts timing aprox 5 degrees advanced. Field experience says that's probably not critical at WOT with 100LL fuel, just higher CHT. I have no idea what the result might be by shorting either 2 or 3 to ground, with or without the EIC. How does the EIC drive a timing shift with these wires?

Pin 4: Obviously a short kills the mag, like any mag.

Pin 5: An open above 800 RPM or so is no issue; that's the whole point of the internal generator. What about a short to ground?

Pin 6: Tach lead. Assumed to have no effect on the P-mag alone, open or shorted. Does the EIC use the tach signal for anything other than an RPM display?

No, but that's why you have two independent ignitions.

A perfectly reasonable answer. It's an extension of the above; list the possible failures, and consider the result of each. Here the result of all failures is the ignition doesn't work, but the flight continues, as we have two of them.
 
SNIP

I have a LightSpeed Plasma 11 on one side (magneto on the other....I'm with Vic Syracuse on this). I'd have to think very seriously about publicly whinging about a fault in my EI on VansAirforce and then going cap in hand to Klaus at some time in the future for support. ;) SNIP

I have recent personal experience on this. I'm helping a budding who has dual Lightspeed ignitions and he ordered (and paid for) a replacement coil from Klaus. Klaus would not mail the coil to me. He mailed it to my buddy and then he had to mail it to me.

The good news is both ignitions are working.

Carl
 
...Recall wires can fail open, or shorted. I typically make a list with each wire represented and think about both open and grounded failure modes, ignoring the probability of failure, and initially only considering what happens if it does fail...

You bring up a good point that this fault analysis is a great tool for uncovering vulnerabilities in a system. But it seems to me that many people don’t understand that this is but one step in the process. Many people think that vulnerable = unacceptable, and the preferred way to fix that is with redundancy. Many times, the “fix” is actually ensuring that the probability of failure is below the acceptable threshold through a robust installation. One grounded wire can take out a magneto, for example. So is it better to add another ignition or protect that wire from a short?

The point I’m trying to make is that we spend a lot of time on this forum searching for the perfect redundancy scheme, but not nearly enough time executing the basics of the system. Wiring, including connectors, are phenomenally reliable when designed and installed correctly. “Wiring” is significantly more reliable than the components they service, yet we see a lot of electrical issues in the E-AB world. Certainly much more so than in cars or spam cans. The problem, as it turns out, is “us”.

Talking about the latest magic scheme/gadget to ensure airliner levels of dispatch reliability may be sexy, but maybe we should spend some time learning how to terminate and route wires first?
 
We get the redundancy/ reliability question often in our business during initial emails or phone calls with our customers. Some are wary of total electronic dependency, others are ready to ashcan all the legacy fuel/ ignition bits and step into the new world.

In the end, anything can fail and you're still in a single engined aircraft where the engine itself is a single point a failure. People seem to think that engine is infallibly reliable. It isn't. We all know or have heard of someone who's had one fail mechanically. Well designed electronics are many times more reliable than aircraft engines in my experience since there are no moving or wearing parts (remember our test ECU with 145,000 hours on it). I don't know of a single Lycoming engine which has gone even 1/20th that time without being touched.

As someone else posted here, there is always some risk on each flight and everyone has a different level of risk they will accept. Some happily fly single engined at night over the mountains, others would never accept that risk. If you can't accept ANY risk, best to stay home on the ground.

We can mitigate many risks by doing good work on our planes and making good decisions on the ground and in the air.
 
Anyone out there running one Pmag and one SDS CPI EI?

Are these two devices compatible together on one lycoming engine?

If the Pmag failed or misbehaved in flight as described in some posts would it/could it adversely effect the operation of the SDS CPI EI?

Anyone else notice that the CEO of SDS participated in this thread? I did. I appreciate hearing from the MFG of products I'm considering.

Anyone notice the conspicuous absence of representation from Pmag?

Nope, nada, nothing. Just saying.
 
...

Pin 1: I think you just stated the system works with this one open. If shorted is normal and open doesn't matter, the logical question becomes "Why is it there?" Back up for case ground to engine block?
I can't answer with certainty because I didn't design the P-mag but I believe you are correct that it is a backup ground. It is really needed for communication through the serial port.

Pins 2 and 3: Without an EIC, an open shifts timing aprox 5 degrees advanced. Field experience says that's probably not critical at WOT with 100LL fuel, just higher CHT. I have no idea what the result might be by shorting either 2 or 3 to ground, with or without the EIC. How does the EIC drive a timing shift with these wires?
Since 2 and 3 are a serial port, I don't believe it would impact anything in flight. We have shorted them when developing the EICommander with no impact to out test P-mags.

As for what would happen if a break happened in flight, the P-mag only checks for the jumper upon startup, after that it never checks again. So, if you start up on the A memory location, it will continue on that configuration until shutdown.

If you have loaded a custom configuration via Emag's EICAD program or our EICommander, you are running off of the B memory location and when running without the EICommander, there is no jumper and when running with the EICommander, you can disconnect it and you will continue to run off of whatever configuration you have loaded, be it the A, B, or a custom configuration.

Pin 4: Obviously a short kills the mag, like any mag.
Correct.

Pin 5: An open above 800 RPM or so is no issue; that's the whole point of the internal generator. What about a short to ground?
An internal sort to ground would probably kill the generator; however, it would continue to fire, if ship's power is available.

Pin 6: Tach lead. Assumed to have no effect on the P-mag alone, open or shorted. Does the EIC use the tach signal for anything other than an RPM display?
Correct and they list it as a "courtesy (optional) connection".


A perfectly reasonable answer. It's an extension of the above; list the possible failures, and consider the result of each. Here the result of all failures is the ignition doesn't work, but the flight continues, as we have two of them.
This comes down to individual choice. How much control does one want of their ignition timing map and what type of back up system do they want.

As we have seen in this thread, there are many difference of opinion and risk tolerance. All good things.
 
Anyone out there running one Pmag and one SDS CPI EI?

Are these two devices compatible together on one lycoming engine?

If the Pmag failed or misbehaved in flight as described in some posts would it/could it adversely effect the operation of the SDS CPI EI?

Anyone else notice that the CEO of SDS participated in this thread? I did. I appreciate hearing from the MFG of products I'm considering.

Anyone notice the conspicuous absence of representation from Pmag?

Nope, nada, nothing. Just saying.

We have a some people flying with one Pmag and one CPI if I recall. I see no issues with that. If the Pmag went down I don't see any reason it would affect the CPI or vice versa.

CEO? We're a small company, I double as the floor sweeper... :)
 
We have a some people flying with one Pmag and one CPI if I recall. I see no issues with that. If the Pmag went down I don't see any reason it would affect the CPI or vice versa.

CEO? We're a small company, I double as the floor sweeper... :)

Ross, you sure know how to jump in and take all the credit......:D
 
Ross;
In Post #115, you referred to single point reliability. Those failures come in 2 kinds: Instant total failure, and progressive failure.
Although modern electronics have a reputation for reliability, they are also perceived to fail utterly and instantly.
A Kettering (points) ignition is less reliable, but has a reputation for slowly failing, giving fair warning so it can be repaired before dire consequences.
I don't need to heap additional cliche stories, and I have a personal experience where my 1988 Harley electronic ignition module failed softly, I changed it before being stranded.
I would be comforted to know that an electronic flight control would have a progressive failure mode.
 
Ross

Anyone out there running one Pmag and one SDS CPI EI?

Are these two devices compatible together on one lycoming engine?

If the Pmag failed or misbehaved in flight as described in some posts would it/could it adversely effect the operation of the SDS CPI EI?

Anyone else notice that the CEO of SDS participated in this thread? I did. I appreciate hearing from the MFG of products I'm considering.

Anyone notice the conspicuous absence of representation from Pmag?

Nope, nada, nothing. Just saying.

Charlie,

Did you get the chance to meet the aforementioned CEO at Reno?

Skylor
 
Ross;
In Post #115, you referred to single point reliability. Those failures come in 2 kinds: Instant total failure, and progressive failure.
Although modern electronics have a reputation for reliability, they are also perceived to fail utterly and instantly.
A Kettering (points) ignition is less reliable, but has a reputation for slowly failing, giving fair warning so it can be repaired before dire consequences.
I don't need to heap additional cliche stories, and I have a personal experience where my 1988 Harley electronic ignition module failed softly, I changed it before being stranded.
I would be comforted to know that an electronic flight control would have a progressive failure mode.

I think it's best not to group our electronics in with others like those in old Harleys. Our goal is to NEVER have a failure which could stop the engine and we've got over a half million flight hours showing that's possible. Also, remember the bench test ECU with 145,000 hours on it. Nobody here is ever going to fly an RV even 1/10th of that time, even Vlad. ;) The components of today coupled with good design should yield something with a MTBF many times greater than the Lycoming engine they are controlling.

We'll have some new people coming in to the market which have no previous track record in the field trying to convince you their new whiz bang XXX is the best thing since sliced bread. Only time will tell but usually a perfect controller design does not come on the first iteration. We've seen lots of smart people try and fail. Initial lessons are hard won and only with actual experience doing it. Lab testing is important but the real world is the actual proving ground which defines your success and only time will tell if you did well or not.

As a competing manufacturer, we are in a unique position to hear from people using those other products when they fail and come looking for a solution. We hear the good and bad about customer service going along with the failures or problems. We probably never would have developed the CPI if all the other EIs were really good because we'd never get any market share. I could tell you of a conversation with someone using a competing product who had 5 failures in not too many hours and others who've had multiple failures as well. Those people have lost total confidence in those brands obviously.

Some stuff we see on other brands, we shake our heads at after seeing the failures. They were very predictable, at least from our experience.

We've heard of several high end (expensive) ECUs fail almost regularly in the heat and pounding of the BAJA/ SCORE off road races where SDS just keeps running year after year. One of our clients has over a dozen class wins with the same old SDS ECU. Several competitors running the other brands finally switched over to us and have had no more failures. We were very happy to have one of our clients win the SCCA GT3 championship for the 4th time last month. The auto market gives us a chance to test reliability that doesn't present itself in the aviation environment but there have been spinoffs both ways to improve the products overall.

Our reputation for reliability did not come overnight and it did not come without some hard lessons in the earlier days but I'm proud of what our small team has accomplished over 23 years. Hard work but very satisfying. I enjoy going to work every day. We're working to innovate, design new parts to make installation easier on a wider variety of engines, bring customer requested features and ideas into our products and improve them across the board. I've started on a series of videos to help people with installation and use of SDS products-sorry they took so long. I'd like to thank many of our loyal customers who've directly helped us to improve, test and uncover problems during development. We can't think of all these good ideas ourselves!
 
Last edited:
...I would be comforted to know that an electronic flight control would have a progressive failure mode.

Edit - Ross beat me to the punch again...

Anyway, I'll just add that as long as the failure mode is benign to the ship (no fire, explosion, leaking, etc), then the demonstrated SDS product reliability well exceeds accepted range for aircraft use. Also keep in mind that electronics usually fail on power up, not in use. If I do a runup some day and find a coil has "suddenly" died since the last flight - might be inconvenient, but it's safe. One notable exception to this is P-mag, of all things. The history of "lost timing events" often happened in flight.

Besides, I think that's better than having dying magneto right on the edge and tempting me to make "one more flight"
 
Last edited:
Charlie,

Did you get the chance to meet the aforementioned CEO at Reno?

Skylor

I tried but the SDS CEO was in high demand and the line was long. Didn't want to be pushy. Besides, he was multi tasking what with all that floor sweeping and such.

I did speak with other SDS users and many who are considering SDS. Got many questions answered.

It was great watching you race at Reno. You had quite the cheering section and it was good meeting your very dedicated support team.

Are you considering EI? I think I remember you're running mags. Is that correct?
 
Returning to P-mag reliability....



I can't answer with certainty because I didn't design the P-mag but I believe you are correct that it is a backup ground. It is really needed for communication through the serial port.

(discussing Pin 1 ground) So what doesn't work if it is not connected...or connected somewhere other than a case bolt? You've been very specific in the past:

http://www.vansairforce.com/community/showpost.php?p=1080846&postcount=10

http://www.vansairforce.com/community/showpost.php?p=1072337&postcount=8

http://www.vansairforce.com/community/showpost.php?p=978738&postcount=3

http://www.vansairforce.com/community/showpost.php?p=979945&postcount=4

Since 2 and 3 are a serial port, I don't believe it would impact anything in flight. We have shorted them when developing the EICommander with no impact to out test P-mags.

Jumper in = ignore B
Jumper out = use B info.
EIC connected to jumper terminals = use signals from EIC to modify B info, in flight or on the ground.

Recall the mystery below? Since both the EIC and the P-mags were found to be A-OK, the conclusion was it had to be something about Nigel's serial wiring...

Originally Posted by nigelspeedy
Approximately 1 in 20 times I have experienced a fault after sending a new ignition setting. I cant say where the root cause of the problem lies but the symptoms are as follows. After you press the send button the engine begins to run very roughly and lose power. Like a very LOP mag check. Unlike the brief firing suspension that is normal this roughness does not go away. On the EICommander one of the PMags will constantly display the correct new timing value you have just sent (this is normal). The other PMag display will vary between the new value and 19.6 deg about every 5 seconds (this is not normal). The CHT will rise at a rate of about 20F per minute and the engine oil temperature will do the same. I have found two ways of fixing this fault. First, you can reload a new ignition setting, (just repeating the one you just sent has worked for me) but this takes a few button pushes and about 30 seconds all the while you will be distracted by your engine which feels like it is under great duress. Second, you can take the P Lead for the misbehaving ignition and turn it off then back on. This is quicker but be very careful to properly identify the correct ignition. I normally associate top with 1 or left and bottom with 2 or right. On the EICommander display the upper timing advance number is associated with the right ignition. It is labeled as such but the display is quite small and you could easily make a mistake in haste. Although the faulty ignition 'display' is alternating between the correct value and 19.6 deg I don't think this is when the ignition is actually firing as the level of roughness and CHT rise are counter to this. After seeing this fault I checked the PMag timing and they were both at TDC so I don't think timing was lost, as the odds of it being subsequently found perfectly are pretty low. I also sent the PMags back for an overhaul and they were both within spec. They have always run the latest firmware version. I have seen the error on both the left and right ignition, not simultaneously though. My EICommander was also returned and it had one of the comm boards replaced but I have seen this fault a couple of times since. I have never experienced this fault except immediately after sending a change. I never had this fault with just the PMags before I installed the EICommander. I have never used the PMag proprietary software to make changes to the settings.

An internal sort to ground would probably kill the generator; however, it would continue to fire, if ship's power is available.

(discussing a short to ground on the Pin 5 wire) Ship's power would not be available; it's a grounded power lead. Right after it grounded, the breaker opened. So, the system is on internal generator power, with the external power lead shorted to ground. Does it provide a path to ground for the internal generator?

Bill, I know these are hard questions, but one can't claim superiority based on immunity to external wire faults without clearly defining exactly what happens given wire faults in the competing system.
 
fault testing

Returning to P-mag reliability....

...

Bill, I know these are hard questions, but one can't claim superiority based on immunity to external wire faults without clearly defining exactly what happens given wire faults in the competing system.
Seems like some bench testing would not take too much time and effort. I'd do it on one of mine but I'm in "get 'er done" mode, and I don't want to fry one of my p-mags. Anyone have any pull with the emagair guys?
 
I am new to Experimentals. Have many hours flying behind regular magnetos. I have read this complete thread and others and still cannot find an explanation of what P-Mags are, how they work and what the advantages/disadvantages are. Or at least I cannot decipher this information. Can someone please explain in layman terms how P-mags differ from a regular Bendix mag, how they work and what the advantages/disadvantages are? I would just like to have an understanding of P-Mags. I am talking to a gentleman that has an airplane I have an interest in with dual P-Mags on an IO360 Superior Engine with Cold Air Induction.
Thank you
 
I am new to Experimentals. Have many hours flying behind regular magnetos. I have read this complete thread and others and still cannot find an explanation of what P-Mags are, how they work and what the advantages/disadvantages are. Or at least I cannot decipher this information. Can someone please explain in layman terms how P-mags differ from a regular Bendix mag, how they work and what the advantages/disadvantages are? I would just like to have an understanding of P-Mags. I am talking to a gentleman that has an airplane I have an interest in with dual P-Mags on an IO360 Superior Engine with Cold Air Induction.
Thank you

http://www.emagair.com/
 

The current Pmag manual (v LC114.28) has this to say regarding the pin 1 ground wire:

"Note: You cannot rely on the ignition’s mechanical attachment to the engine to provide ground. Aluminum anodizing acts as an electrical insulator, so the clamp connection to the anodized flange will not be a reliable ground."

Additionally, the Troubleshooting Tips version 04 on the website indicates "not having a proper ground connection" to be one of the most frequent installation errors.

This language seems to indicate the ground wire is pretty darned important.
 
Returning to P-mag reliability....





(discussing Pin 1 ground) So what doesn't work if it is not connected...or connected somewhere other than a case bolt? You've been very specific in the past:

http://www.vansairforce.com/community/showpost.php?p=1080846&postcount=10

http://www.vansairforce.com/community/showpost.php?p=1072337&postcount=8

http://www.vansairforce.com/community/showpost.php?p=978738&postcount=3

http://www.vansairforce.com/community/showpost.php?p=979945&postcount=4



Jumper in = ignore B
Jumper out = use B info.
EIC connected to jumper terminals = use signals from EIC to modify B info, in flight or on the ground.

Recall the mystery below? Since both the EIC and the P-mags were found to be A-OK, the conclusion was it had to be something about Nigel's serial wiring...

Originally Posted by nigelspeedy
Approximately 1 in 20 times I have experienced a fault after sending a new ignition setting. I cant say where the root cause of the problem lies but the symptoms are as follows. After you press the send button the engine begins to run very roughly and lose power. Like a very LOP mag check. Unlike the brief firing suspension that is normal this roughness does not go away. On the EICommander one of the PMags will constantly display the correct new timing value you have just sent (this is normal). The other PMag display will vary between the new value and 19.6 deg about every 5 seconds (this is not normal). The CHT will rise at a rate of about 20F per minute and the engine oil temperature will do the same. I have found two ways of fixing this fault. First, you can reload a new ignition setting, (just repeating the one you just sent has worked for me) but this takes a few button pushes and about 30 seconds all the while you will be distracted by your engine which feels like it is under great duress. Second, you can take the P Lead for the misbehaving ignition and turn it off then back on. This is quicker but be very careful to properly identify the correct ignition. I normally associate top with 1 or left and bottom with 2 or right. On the EICommander display the upper timing advance number is associated with the right ignition. It is labeled as such but the display is quite small and you could easily make a mistake in haste. Although the faulty ignition 'display' is alternating between the correct value and 19.6 deg I don't think this is when the ignition is actually firing as the level of roughness and CHT rise are counter to this. After seeing this fault I checked the PMag timing and they were both at TDC so I don't think timing was lost, as the odds of it being subsequently found perfectly are pretty low. I also sent the PMags back for an overhaul and they were both within spec. They have always run the latest firmware version. I have seen the error on both the left and right ignition, not simultaneously though. My EICommander was also returned and it had one of the comm boards replaced but I have seen this fault a couple of times since. I have never experienced this fault except immediately after sending a change. I never had this fault with just the PMags before I installed the EICommander. I have never used the PMag proprietary software to make changes to the settings.

We spoke with Nigel after he experienced this and struggled to reproduce it in our lab. In fact, we couldn't reproduce it at all. This was the first time we had been informed of such an issue.

Not many people ever change their timing in flight, for good reason. While the p-mag did not lose its TDC mark, it appears the shift angle (where it should fire in relation to TDC) was not accepted by the P-mag.

Based on what he experienced, we do not recommend changing the timing in flight. If someone insists on doing it, there are two different ways. One is the On the Fly option, which issues the timing command to the P-mags and is less disruptive to the firing sequence, or do it on the ground when the engine is not running.

We were not aware what Nigel was testing or we would have suggested he use the On the Fly option. This option does change the timing but does not try to write to permeant memory. The idea being that once a pilot finds a setting they like, they can land, and save it permanently.

This really doesn't have anything to do with the P-mag but does have everything to do with how the EIC communicates with them.

The display Nigel mentions is just that, a display. It displays whatever firing angle the P-mag reports to it.

(One other thing, the reason there are so many button presses to send a new timing configuration to the P-mags is so that someone doesn't send them a configuration by accident. That was by design, not a bad programming choice.)


(discussing a short to ground on the Pin 5 wire) Ship's power would not be available; it's a grounded power lead. Right after it grounded, the breaker opened. So, the system is on internal generator power, with the external power lead shorted to ground. Does it provide a path to ground for the internal generator?

Bill, I know these are hard questions, but one can't claim superiority based on immunity to external wire faults without clearly defining exactly what happens given wire faults in the competing system.
Next time you are at SnF or OSH, stop by the Emag display. Their 114 P-mag is wired to four sparkplugs and no ground wire connects the plugs to the P-mag.

When I asked Brad about this, he said they use the second plug as the primary return path. Remember they fire two plugs at the same time (wasted spark), that was a convenient way to always have an adequate electrical return path. So, they have three return paths; the P-mag body, the ground wire, and the other sparkplug.

One other design feature of the P-mag is that once they are spinning and producing power, there is no way to shut them down. All the P-lead does is stop the plugs from firing but the internal electronics are still alive.

Remember, I am not the manufacture of the P-mag ignitions, nor are our companies connected. Everything I have written on this thread and others is based on what we have learned while working with the ignitions over the past 10 years. For answers as to why they are designed and function the way they do, you really need to go to Emag.
 
Last edited:
I tried but the SDS CEO was in high demand and the line was long. Didn't want to be pushy. Besides, he was multi tasking what with all that floor sweeping and such.

I did speak with other SDS users and many who are considering SDS. Got many questions answered.

It was great watching you race at Reno. You had quite the cheering section and it was good meeting your very dedicated support team.

Are you considering EI? I think I remember you're running mags. Is that correct?

I think the cheering section belonged to the guy with the Lancair 320 (Mark)! My team of two consisted of my Dad and my significant other...and they are very dedicated to supporting my Reno hobby!

I don?t currently have plans to swap out the mags, but if that time comes, SDS will definitely be at the top of my list. I really appreciate Ross?s openness and honesty here when it comes to discussing his products!

Skylor
 
We spoke with Nigel after he experienced this and struggled to reproduce it in our lab. In fact, we couldn't reproduce it at all. This was the first time we had been informed of such an issue.

Ok, discard Nigel's report, and consider timing loss with a known cause, the arcing alternator B-lead.

http://www.vansairforce.com/community/showthread.php?t=144336

At first, it was claimed that V40 would have prevented the issue, but that was obvious nonsense, as noted by this gentleman...

http://www.vansairforce.com/community/showpost.php?p=1130130&postcount=31

...and recognized by the manufacturer:

http://www.vansairforce.com/community/showpost.php?p=1131165&postcount=49

So why is it relevant to this particular exercise? The P-mag is unique, in that it's the only ignition which allows the user to insert the body and mesh the drive gear at random, then write a TDC position reference electronically. It's a wonderful feature as it relates to installation and inspection, as it makes timing a P-mag dirt simple. However, all design is the art of compromise. The same feature is an Achilles heel; the easy-to-set TDC position can also be reset. V-40 (and I suspect other patches too) is a example of an attempt to prevent unintended reset given a specific condition. The case of the arcing B-lead is an example of timing reset caused by a less predictable input, so it's not likely to be fixed with a software patch.

All other ignitions physically reference crankshaft TDC to the ignition. As such, installation requires physical alignment of trigger and crankshaft position. Losing that reference requires outright mechanical failure. It is slightly more difficult to install, but more reliable in service.

I've attempted to run though some basic "what if" analysis, which as Mike pointed out, would be followed by an estimate of probability. It does not appear we're going to be able to complete that examination due to lack of information. Frankly, it makes no personal difference to me. I'm just trying to lead the conversation back toward a logical estimation of reliability.

I will say this, based on reported field experience. The major P-mag risk is random timing reset. The major risk with other systems is simply no spark. Random timing can take out an engine. No spark merely means soldiering on with the other ignition.

The architecture which allows random timing reset due to outside factors has nothing to do with the P-mag's significant benefit, its internal power supply. A system with a hard TDC reference and internal power generation would be a heck of an ignition...but that's not what we have here.

Next time you are at SnF or OSH, stop by the Emag display. Their 114 P-mag is wired to four sparkplugs and no ground wire connects the plugs to the P-mag.

When I asked Brad about this, he said they use the second plug as the primary return path. Remember they fire two plugs at the same time (wasted spark), that was a convenient way to always have an adequate electrical return path. So, they have three return paths; the P-mag body, the ground wire, and the other sparkplug.

One other design feature of the P-mag is that once they are spinning and producing power, there is no way to shut them down. All the P-lead does is stop the plugs from firing but the internal electronics are still alive.

Two things, and I'm outta here. One, it is easy to make sparks. The issue is the timing of those sparks. Two, all waste spark systems use the other plug in the paired coil as the return path for the secondary; that's why there is a wasted spark. It is very unlikely that the high voltage secondary is part of the ground path for control electronics.
 
....So why is it relevant to this particular exercise? The P-mag is unique, in that it's the only ignition which allows the user to insert the body and mesh the drive gear at random, then write a TDC position reference electronically...

...The architecture which allows random timing reset due to outside factors has nothing to do with the P-mag's significant benefit, its internal power supply. A system with a hard TDC reference and internal power generation would be a heck of an ignition...but that's not what we have here...

Thanks Dan for clinically articulating the engineering challenge P-mag faces.

It seems clear after all these years that the P-mag has battled 3 main issues:

1. Physical alignment and security of the timing magnet.
2. Software driven magnet position (easy to set, vulnerable to random reset)
3. Electronics vulnerable to the operating environment (heat and chemical intrusion)

All of the above could easily be fixed by using an external crank trigger and moving the brains into the cockpit. Add complete adjustability of the curve and you'd have a real winner.

OTOH, add an SD-8 to CPI and you are already there.
 
All other ignitions physically reference crankshaft TDC to the ignition. As such, installation requires physical alignment of trigger and crankshaft position. Losing that reference requires outright mechanical failure. It is slightly more difficult to install, but more reliable in service.

I was applauding your approach of trying to bring logic to this until you made above unsubstantiated claim.

What data do you have to back this up in any way?

You seem to imply that physical TDC alignment in general (not specific to P-Mags) is more reliable then a software/electronics based approach. Having written highly reliable software myself which has run for many years at this point (in other areas) I totally disagree with this broad statement. An electronic/software based TDC setting can be made at least as reliable as a mechanical one if done properly (mechanical once do fail too as you stated yourself).

Now I have no knowledge if P-Mag did it properly or if they fixed it properly after they had those faults. So the one advantage of a mechanical approach is that all of us can look at it and see if we think it's a proper approach whereas with software and most integrated electronics we can not. Only thing we can do on software/electronics is trust the manufacturer and look at overall statistics. If the manufacturer provides no data we are stuck speculating. Openness generates more trust which some manufactures seem to have learned whereas others haven't.

Also just because TDC is set properly doesn't mean the ignition will not fire at random times. There are other possible electronic/software faults which could cause random ignition under a byzantine fault scenario (you mentioned at least one in your thread). So again as soon as you go to ANY electronic ignition you trust the manufacturer to have done the proper fault analysis and use the proper technique. If you actually want to "see" what's going on you really need to stick with the old mags.

Maybe we should have an open source electronic mag project as other communities have for similar things (e.g. drone software/control). That way everybody could look at the code and help the manufacturers to find bugs and fix them. I would doubt that anybody is that open in our community yet though ... .

Oliver
 
"Lost Timing" events would certainly worry me.

You can do all the analysis you want but in the end, the system either works properly all the time or it doesn't which means it's either well designed and tested or it isn't.

Nigel's independent testing uncovered several things about the product that people may not have known before. Information is good.

We don't worry about lost timing events or programming changes which don't take hold with SDS because they never happen. Change the timing on CPI and it's instant and accurate to within 1 degree. Shut off the unit or lose power while the engine is turning and it re-initializes and verifies crank position before starting the spark train again when power is restored. Anything else is not acceptable in our view.
 
Thanks Paul

You know folks, there is a huge amount of useful (and some not so useful - you can sort that out...) data in this thread, but very little of it has to do with "PMAG Reliability in 2017". So most of the really good info will be lost because people aren't going to be able to search for it (let's all admit that the search function is pretty basic here), and they won't look under this title to find info on SDS equipment (for instance).

And there is absolutely no way a Moderator can (or will) sort this mess out.

Think about that folks - if you have presented good, new information here - think about going and starting a new thread. Or it is all lost for posterity.

Paul... this thread has been maddening to read. At this present time I think there over 135 responses.... very few of them actually answer the OP question. I honestly didn't believe there was a moderator reading this thread at all. This thread is so far off the rails it provides very little tangible information regarding the original topic. Paul...you are very respected in this community and I thank you for your comments regarding the thread creep.
 
It would actually be easier for a moderator to remove the posts that *are* on topic, and put them in a new thread... Then rename this one to "SDS Reliability in 2017."
 
They were reliable for us today; dual Pmags since August 2017, one for Easton, one for me:D

2i0dunk.jpg
 
Dual pmags but problems

I have an RV7 with an 0360 A1A which has been converted to Precision Airmotive FI and dual Pmags with firmware 36. At. 98.2 hours the R mag failed and L mag had lateral an axial play in the shaft. Brad at Pmag rebuilt the bearings in L mag and added firmware? 41 in both. I just reinstalled and they appear to operate fine for first few hours.

The engine was overhauled at engine shop in Florida to zero time.

I understand that others with 0360 A1A engines have had similar issues especially with shaft and bearing problems. If so, please let me know. [email protected]
 
I have an RV7 with an 0360 A1A which has been converted to Precision Airmotive FI and dual Pmags with firmware 36. At. 98.2 hours the R mag failed and L mag had lateral an axial play in the shaft. Brad at Pmag rebuilt the bearings in L mag and added firmware? 41 in both. I just reinstalled and they appear to operate fine for first few hours.

The engine was overhauled at engine shop in Florida to zero time.

I understand that others with 0360 A1A engines have had similar issues especially with shaft and bearing problems. If so, please let me know. [email protected]

Neil,You indicate, "The engine was overhauled at engine shop in Florida to zero time."

Was your engine "overhaul" a result of the problems you described, specifically, "At. 98.2 hours the R mag failed and L mag had lateral an axial play in the shaft. Brad at Pmag rebuilt the bearings in L mag and added firmware? 41 in both."

Thanks for sharing your experience.
 
Thanks for letting us know.

I have an RV7 with an 0360 A1A which has been converted to Precision Airmotive FI and dual Pmags with firmware 36. At. 98.2 hours the R mag failed and L mag had lateral an axial play in the shaft. Brad at Pmag rebuilt the bearings in L mag and added firmware? 41 in both. I just reinstalled and they appear to operate fine for first few hours.

The engine was overhauled at engine shop in Florida to zero time.

I understand that others with 0360 A1A engines have had similar issues especially with shaft and bearing problems. If so, please let me know. [email protected]

Thanks for letting us know about your experience.

For those of you reporting failure modes, please include make and model of your propeller... in addition to your make / model and accessory version of your engine.

Thanks
 
Ok, discard Nigel's report, and consider timing loss with a known cause, the arcing alternator B-lead.

http://www.vansairforce.com/community/showthread.php?t=144336

At first, it was claimed that V40 would have prevented the issue, but that was obvious nonsense, as noted by this gentleman...

http://www.vansairforce.com/community/showpost.php?p=1130130&postcount=31

...and recognized by the manufacturer:

http://www.vansairforce.com/community/showpost.php?p=1131165&postcount=49

So why is it relevant to this particular exercise? The P-mag is unique, in that it's the only ignition which allows the user to insert the body and mesh the drive gear at random, then write a TDC position reference electronically. It's a wonderful feature as it relates to installation and inspection, as it makes timing a P-mag dirt simple. However, all design is the art of compromise. The same feature is an Achilles heel; the easy-to-set TDC position can also be reset. V-40 (and I suspect other patches too) is a example of an attempt to prevent unintended reset given a specific condition. The case of the arcing B-lead is an example of timing reset caused by a less predictable input, so it's not likely to be fixed with a software patch.

All other ignitions physically reference crankshaft TDC to the ignition. As such, installation requires physical alignment of trigger and crankshaft position. Losing that reference requires outright mechanical failure. It is slightly more difficult to install, but more reliable in service.

I've attempted to run though some basic "what if" analysis, which as Mike pointed out, would be followed by an estimate of probability. It does not appear we're going to be able to complete that examination due to lack of information. Frankly, it makes no personal difference to me. I'm just trying to lead the conversation back toward a logical estimation of reliability.

I will say this, based on reported field experience. The major P-mag risk is random timing reset. The major risk with other systems is simply no spark. Random timing can take out an engine. No spark merely means soldiering on with the other ignition.

The architecture which allows random timing reset due to outside factors has nothing to do with the P-mag's significant benefit, its internal power supply. A system with a hard TDC reference and internal power generation would be a heck of an ignition...but that's not what we have here.



Two things, and I'm outta here. One, it is easy to make sparks. The issue is the timing of those sparks. Two, all waste spark systems use the other plug in the paired coil as the return path for the secondary; that's why there is a wasted spark. It is very unlikely that the high voltage secondary is part of the ground path for control electronics.

(I believe they call that a magneto, Dan :D)
 
Resolution

Well, here's the conclusion of this thread from my end.

1.) I ended up installing two P-114 P-Mag's in my airplane. Both have v40 software and cooling fins on the case. They were wired with the existing LASAR wiring and manifold pressure lines, as well as blast tubes.

2.) I timed the P-MAG's 2 degrees after TC, so they should fire at 23 degrees at high power, and limit their advance to 32 degrees. Probably limiting the advantage I gain from them, but I like the increase in detonation margins.

3.) During test flight my performance was identical to the LASAR ignition it replaced WRT fuel flows and airspeeds. EGT's were identical, and CHT's were right around 330 degrees on an 80 degree OAT day.

4.) Cruise at 10,500' density altitude 20" / 2400 RPM (60% power) was 8.0 gph yielding around 165kts true.

5.) Idle is smoother now on the P-MAG's than it was on the LASAR, and engine starts are similar to a car. First blade and the engine fires, hot or cold. It's impressive.

6.) I am not 100% sold on the P-MAG's yet. If I have any issues with the timing, or reliability of the unit they're done and will be replaced with a different system, but for the ease of install I felt they showed the most promise at this time. Hopefully I'm not responding to this thread in a few hundred hours saying that I'm installing CPI and a Magneto...
 
Last edited:
Pmag reliability

My engine was rebuilt at JB Aircraft in Sebring FL to zero time. I am not a mechanic but I was informed that they have a great history and it has performed great to this point. I have a new constant speed Hartzell blended airfoil prop from Vans.

My engine came with two new Slick mags and harnesses which I removed after reading much about Pmags here. I removed these before installation on the RV.

I have a G3X system so I have been able to monitor the engine functions since first test flight. I had difficulty with oil temps (up to 225-230) a few times in the heat of summer in Bham Al mainly after refueling, touch and goes or during ground times after flights. Also some issues with CHT on climb if not careful. These experiences were not much different than experiences I have followed here by many others.

After I switched from AeroShell mineral break end oil, to 15-50, re-checked and modified engine baffling and changed to Stewart Warner oil cooler, oil temps OK and CHT ok but I am careful on initial climb out.

The Pmags have operated as advertised until the issues previously described. I am a little shaken at the final flight as my passenger was my wife who is a white knuckle flier.

I have certainly enjoyed and learned a ton from this building experience and from this forum. I have the pleasure of flying several different planes and I flew in the Army both RW and FW for 20 years. The RV is one of the most fun and versatile planes I have had the pleasure to fly.

All of the other planes I have flown (except turbines) have magnetos and I have never experienced a failure. The simplicity of the Pmags were a draw for me and I hope my issue was isolated and will not be repeated. I have decided to reinstall a non impulse Slick on the L mag which is the side of the bearing failure.

One of the sources of others with failures is listed in this thread. I spoke with Tim Andres and he experienced 5 different events from bearing failure to general failure of the mag just quitting operations. Both of us have 0360 A1A lycomings. Tim has a fixed pitch prop and I added FI and constant speed prop. After I questioned Brad about the cause of the failure, especially bearing failure with less than 100 hours, he stated that certain engine and prop combos may create an issue with Pmags. I hope that is not the case with my engine but I share my experience with you not as a proponent for any ignition system but as a matter of safety. The failure may be an isolated event which I hope but I will be a great deal more alert with the rebuilt Pmags.

Still running two until the Slick and harnesses arrive. Thanks for this forum. It has been an invaluable help to me during and after construction. Be glad to speak to anyone who wishes more info.

Neil Clay 205-253-9595
 
I have an RV6/ io-360/hartztell cs and replaced the slicks with 2 Pmags. 57 hrs flawless. Smooth , awesome hot starts, and I think more power but not really sure. Reading this thread scares me a bit as I originally thought I should keep one mag. I was meticulous with blast tube positioning and wire routing/support but this random timing reset/engine shutdown talk is concerning. I installed and promptly flew to the east coast and all the way around Cape Breton Island at about 200' inspecting the coastline and whale watching. (with a passenger) Probably shouldn't admit that on here but I'd hate to see the fan stop. It's what keeps the pilot cool.
 
I understand that others with 0360 A1A engines have had similar issues especially with shaft and bearing problems. If so, please let me know. [email protected]

What is the source of this information?

Thanks,
Erich

We have had two customers who have picked up bearing failures with the EICommander, before they became more than just a maintenance an issue.

One was an early 113 P-mag on an O-320 and the other on a FP O-360.

Statistically, probably a rounding error.

They show up as a timing divergence issue on the EIC and are a challenge to debug. After we go through all the "usual" questions, we ask the clients to remove and inspect the P-mags. Sure enough, they found some wiggle in the shaft.

As a control point, I have 800 hours on my P-mags, 200 hours in a low compressions O-290-D2 and 600 hours on my O-360 180HP (ECi Equivalent) and have not had an issue with the bearings. However, I did have the internal generator grenade right at 500.? hours. These were the P-mags I had in the plane when I had prop strike with the O-290-D2, so it is possible that this failure is related to that. It is also one reason I always perform a mag check by pulling ship's power from them prior to every flight. The ignitions never missed a beat and had I not checked the internal generator, I would never have known there was an issue. The good news is, they operated off of ship's power just fine and I completed my flight home as it was firing and the other P-mag's internal generator was functioning fine.
 
What i am curious about is what would cause the bearing to wear out or fail? The only thing I can think of is a side loading caused by improper gear clearance or something out of balance on the end of the shaft. As an example, when I had my slicks rebuilt the impulse mag was sloppy and looked in terrible shape but the non impulse one was in good shape. The mechanic said it was due to the big chunk of unbalanced iron spinning around inducing loads on the shaft. It made sense to me but P mags don't have that issue.
 
What i am curious about is what would cause the bearing to wear out or fail? The only thing I can think of is a side loading caused by improper gear clearance or something out of balance on the end of the shaft. As an example, when I had my slicks rebuilt the impulse mag was sloppy and looked in terrible shape but the non impulse one was in good shape. The mechanic said it was due to the big chunk of unbalanced iron spinning around inducing loads on the shaft. It made sense to me but P mags don't have that issue.

Doug, what I am curious about is what happened to XD3 15w40?

Thanks up front for your response....... now back to P mags.
 
Mid 1990s. A big mfgr made a 2 blade composite prototype prop and I believe a friend had the first if not only on a pumped up 4 Cyl Pitts S1T, angle valve, 9:1 or greater compression.

It repeated interacted vibrations to the Slick mags that caused failures of the mags. I believe twice in under 50 hours.

I assume this interaction is still a challenge.
 
Mid 1990s. A big mfgr made a 2 blade composite prototype prop and I believe a friend had the first if not only on a pumped up 4 Cyl Pitts S1T, angle valve, 9:1 or greater compression.

It repeated interacted vibrations to the Slick mags that caused failures of the mags. I believe twice in under 50 hours.

I assume this interaction is still a challenge.

I had a friend with a Continental O-200D and ground adjustable Sensenich composite prop on the front of his Cub Crafter's Cub. His Slicks would be lucky to make 35 hours. He switched to a Catto prop and the problem went away.

It is all about harmonics.
 
Last edited:
Hello gasman. I made a couple follow up comments on the 15w40 on the the"any P-mag pireps?" thread. See the last comment on page 6 and the 1st comment on page 7 over there. Basically a few of us around my parts use this oil. I worked in R&D at Imperial oil for the first 7 years of my 37 year there. We had a large engine test facility which I spent lots of time frequenting as one of my close friends tore down and prepare engines everyday for testing. Also have seen high time lycomings taken apart that used conventional av oil vs XD3 15w40. Lets just say I am VFR orientated. Seeing is believing.
 
Back
Top