Bendix Mag SF-4 issue - 400 rpm drop on right mag/sparkplug.

[skyking3286]

Started three flights ago.

I had perfect mags until three flights ago. 25 rpm drop at both mags at 1500 rpm mag checks.

Three flights ago...Right mag dropped 75 rpm... seemed to clear a bit, so I flew around bit.
Two flights ago...dropped 200 rpm, taxied around a bit and did a few full throttle run ups. Dropped to 100 rpm, so I flew around to
see if I could clear it with some running. Runs just fine.
Last time... actually no flight. It dropped 500 rpm on the right mag. Left mag is 25 rpm just like normal. And at full throttle on the right
mag all I can do is 1500 rpm with a backfire once in a while. Rough running, almost flooding feeling. Go to both and its' like nothing is wrong.
OK, now we have to act. We know it's not the fuel because it runs just fine on the left mag.

Here's what we did today.

On my own, I ScotchBrited the coil contacts, the spark plug wire ends and the contact ends. Started it up, still had 400 rpm drop.
Checked the top spark plugs, which the R mag fired, looked OK.
Checked with infrared temperature reader on the exhaust. Looked pretty much the same all around after running it on the rough
mag for two minutes and shutting down.

Now, we have to run to the shop and let them have at it.

With a voltmeter, we checked the switch. Fine and dandy. Right is left and left is right as it should be when you ground one out.
Checked the plugs. Nice spark... Shoot. There goes the cheap solution.
Now the mag. Took the right mag off the engine.
Looked good on top when we popped the hood.
Now on the bench we found that the contact breaker on the back end of the mag wasn't opening at all.
The contact was moved by the cam, but never opened.
So we adjusted it and retimed the mags, tried to start it. Right mag now dead.
Hmmmm....
Found out that the longer wire actually goes the shorter distance to the mag, switched the ground and the P-lead.

PERFECT.

Well, except for the bill that's coming.

Just for your information anyone else gets a drop of 400 rpm on one mag. Most of the stuff on the internet says sparkplugs, wires and then mag.
In my case, it was the contract breaker being out of adjustment and not opening.

[skyking3286]

UPDATE:

Well, I went out on Tuesday to fly it for the first time after the mag issue was seemingly resolved.

I taxied out after warming the plane up for ten minutes and the mags were equal, dropping the usual 25 rpm at 1500 rpm! Hooray.

Flew south and back for 1.7 hours. Wonderful flight.
Turned off the grass runway and just for fun, hecked the mags.
SHOOT...
Right mag is now dropping 150 to 200 rpm, almost stopping if I let it run for any time on the right mag alone.
Can't get over 1500 rpm on it. Just like before, almost worse, in that it really didn't want to run.

Options anyone?

From what I can gather:
1. Usually, the condenser goes first, then the coil.
2. The coil can go bad when it's hot, but check fine on the mag check early in the flight.
3. No one around here wants to work on the SF-4 mags.
4. Toyota is making the condenser that people sell now, it's the same specs as the old Bendix one.

Does anyone have a place to check a mag with the old oven temperature method?

[skyking3286]

Looks like its fine again.... on run up it was normal. Something to keep watching.

[Paul Agaliotis]

Mark,
The coils are pretty good on these little mags, It could be an issue but would be low on my list. It sounds more like a resistor in a plug giving you problems. The next time you fly and have this problem leave it on the bad mag for five minutes, shut it down and find the cold cylinder or wet plug. I have had a bunch of plug failures in the last 10 years that I never saw in the previous 35 years.
If you still want to send it out give Al a call at Savage Magnetos. I don't know if he has kept up his Repair Station certification but you'll find nobody better with these mags.
Paul

[skyking3286]

Thanks Paul, I'm beginning to think that too. The mag should be fine. Tonight I went up and the mag check after 5 min. warm up was less than 25 rpm at 1500 rpm for both mags. Flew for an hour and then landing I checked again, the left was as smooth as butter, but the right one was acting up again. 250 rpm drops and rough at 1500 rpm.

So, if the plugs get warm, then the resistance goes? I'm thinking of getting two plugs and putting them in front and then checking and then moving them to the back and checking. Would that be a plan?

[Paul Agaliotis]

Mark,
The resistors are ceramic and may have opened with the heat. Find the plugs that the right mag fires and check them. When you have mag drop just sit and let it run on the right. The cylinder not running will cool off pretty quick. Check for it as soon as you can, before the under cowl temps warm it up. If you're still unsure pull the four plugs being fired by the right mag, one should be wet from not firing.
Make sure the switch is wired correctly and the selected mag is actually the mag in operation. Don't laugh, it happens, and it makes you look like a monkey. Don't ask me how I know this.....back to my banana.
Paul

[skyking3286]

Found this at http://www.ultralightnews.ca/articles/r ... dplugs.htm. Interesting how resistence works in sparkplugs...

Spark plug safety tip

From: Rainier Lamers <Rainier@Questek.co.za>
Subject: Was: Spark plug safety tip
Date: Fri, 26 Feb 1999 16:35:56 +0200

Recently an interesting thread on this newsgroup discussed the relative merits of resistive plugs as used on many Rotax engines.

In order to get to the bottom of this Spark plug vs. resistance story I decided to do some experiments. I am a electronics engineer (when I am not flying , so I have the tools for the job.

First a coil was rigged for the job to create the high voltage required for the plugs. I used a standard car coil combined with an electronic interrupter/generator. Spark plugs with and without resistors as well as caps with and without resistors where obtained. The plugs where all equally gapped to 0.4mm. I used a scope to monitor the voltage at the plug tip as well as at the coil (via high impedance probe - scopes are expensive !). The result was quite interesting.

As expected, the voltage at the cable (before any resistor) rises at the same rate regardless of resistance until the point of firing. Thereafter however the picture changes.

The coil generates a certain amount of energy. This energy wants to go somewhere. At a voltage of about 7KV the plug fires (irrespective of resistance). Until that happens NO CURRENT FLOWS. Whether you have resistance or not does not matter. A paltry 5KOhms does not do anything when compared to the nearly infinite resistance of the gap itself (until the plug fires that is).

Once the plug fires the resistance comes into being. The coil cannot get rid of its energy in the shortest possible time due to the resistance.

This reduces current flow in the spark and it takes quite a bit longer until the energy in the coil has expelled itself via resistor and spark gap. This results in a longer spark. However the spark is weaker due to energy loss in the resistor.

The scope shows this very nicely and it also gives reason why the resistor helps to suppress RF. With the resistor changes in the rate of current flow are much less resulting in "flatter" edges on the voltage vs. time curve that the scope shows you. This implies less high frequency components of the signal in the cable feeding the spark plugs.

Conclusion: It is quite safe to insert a resistor. It will not stop the plug from firing at all. High values will however lead to a weaker spark -but it is going to spark, no matter what.

However, contaminate the plug with fuel (even just a little) and a high value resistor will cause the plug not to fire. 10K plugs as are used on some motorcycles now are about maximum I would guess.

I do not recommend you use resistance plugs or caps at all on inverted Rotax installations due to the possibility of plug contamination by oil and fuel.

As further experiment I increased the resistance value to see what effect this is going to have. I tried values 10K, 47K, 100K, 220K, 470K, 820K and 1M.

Even with 1M the plug (dry, not contaminated) still fired but noticeably weaker and longer (you can actually see it firing longer !). But contaminate this plug even slightly (a little moisture by exhaling onto it) and you get no spark.

Finally, a disclaimer: All of the above has been found out by a little experimenting. Use the results at your own peril. Don't get back to be with a heap of bent metal that used to be your plane claiming it is because I said you should put 1M resistors into your spark plug leads. I did NOT.

Rainier Lamers.

Editors Note:
A brief recap of the two types of caps that are currently being used on our engines. Originally Rotax engines were equipped with breaker point ignitions and Bosch spark plug caps. A breaker type ignition is not the hottest spark producer at low cranking speeds and at low RPM’s.
Because of this a spark plug cap with a 1000 ohm resistor built inside was used along with a sheet metal type shield on the outside of the cap. The intention of these two items was to reduce EMI, “electro-mechanical interference” or RFI, “Radio frequency interference”.

The latest engines are equipped with stronger dual electronic ignition systems and use a 5000 ohm NGK resistor cap, these may be either black or magenta in color, mechanically these are identical. Along with these resistor caps the newer engines also are equipped with resistor spark plugs. All of this just to reduce that annoying hum or whine you may hear in your headset while your engine is running.

[Richard Murray]

Mark,

That post says it all. Tempest has an interesting blog on testing plugs http://www.flyplatinum.com/blog/?p=690

I don't remember the type of Bendix mags I had in an old Taylorcraft several years back, but if you tried to prop it after letting the engine sit for a half an hour, you could spend the next hour propping your arm off trying to get it to fire. If you opened the cowl and let things cool down after shutting down, it would always start on the second blade. The OFs around all said its 'those Bendix mags'. A good friend who is an IA and has rebuilt many mags said the coils seem to be influenced by heat and perhaps this was the case.

What Paul suggested is right on and for $30 at Harbor Freight you can get a infrared digital thermometer and get an more accurate temperature of the cylinder head than the ole hand will tell.

Richard

[skyking3286]

I have two M41Es on order from ACS and there is another clue. When we pulled the plugs, one of the right mag fired ones looked a little bit shiney, but not quite wet.
That has my interest now. Update later.... and a reminder for folks to check the mags on shut down too. That's where my problem is showing up, not on preflight mag check. Not to mention the closed breaker contacts we discovered when the adventure began.

While we are at it, there is precious little on the SF4 mags on the internet. I was expecting more but there's not much.

[MikeB]

Mark,
I've been following the thread and may have missed something, but have you considered the possibility may be in the mag switch itself? They tend to get a little 'flaky' after years of service and turning the switch on and off several hundred times. I've taken the switch apart and cleaned the contacts on them although it's more likely that it would affect grounding rather than open side.
Just a thought.. .

Mike

[clipperfixer]

Mark,
If you get to the point that you want the mag rebuilt I have a friend that will overhaul them. He has one of mine now. I can get you contact info if you need it.
Ray

[skyking3286]

Thanks for the tips all!

Mike - I cleaned the contacts on the switch about five years ago and we wiggled the key in the ignition a bit during the first check and no change. So at the moment, I'm not thinking of the switch as being a part of the issue.

Ray - I'll certainly keep that in mind. The Savage Mag folks in LA are recommended but pricey according the rumors I hear. The parts from Fresno aren't that expensive
and a whole rebuilt mag is in the $600 range (?). But the coil and condenser will run well under $200.

[clipperfixer]

I hear you on the parts prices Mark, I told my friend that rebuilds them the same story about not finding anybody that wants to touch them. He said he does not know why part are available. Anyway he built mags for a living before he started with us. His old customers still hunt him down for rebuilds. I don't know what his pricing is but I am sure it is more than reasonable.
Ray

[skyking3286]

Ray, I'll keep that in mind... looks like a good lead.

Now, here is the latest. Yesterday, the 22nd of August, I went flying. Mags were perfect on check, in fact, only a 25 rpm drop on both. Went flying for an hour and then checked on landing. R mag is rough, only 1500 to 1700 rpm static at full throttle. Ran it for a three minutes on the right mag, with just a little bit on both when the engine seemed about ready to quit.

Got out and used my pocket infrared Harbor Freight at the valve covers. At first the back cylinders on both sides were suspect. Let it cool a bit, figuring that the hot
cylinders would be more hot soaked than the poor firing ones.

#1 126
#2 163
#3 115
#4 152

In other words, the two back cylinders were the coldest. Which goes against what I would think... that the front cylinders being in the airflow first, would be the coolest
and the back ones the hotter ones. Unfortunately I had the wrong sized wrench to get the sparkplugs out so next time...

I have the mag checks on YouTube, HD. The run up is about 1 min into the tape, the right mag running is at 1:01:07, after I land. Sounds like an old tractor. At least I don't see a lot of smoke coming from the pipes. Enjoy!

http://www.youtube.com/embed/TpJW8Ukzx_ ... 3c4gZZp2Qw

[Paul Agaliotis]

Mark,
Don't run it on both when checking it . It cleanes the wet plug. Bring it back, find the rough mag and let it run on it for about 5 min. Shut it down and check for a wet plug.
Paul