Stranger things...Another puzzler...

[bjdecker]

To the Hive Mind -- If anyone can figure this out - I'll buy you a beer at Airventure.

As part of solving any puzzle, it's good to have a view of the forest first - so here's the configuration:

RV-7
IO-360-A1B6 - 9.75" Pulley
Plane Power AL12-EI60B Alternator - 2.75" Pulley
Monkworks MZ30L Standby Generator
Lightspeed Plasma III Ignitions (2)
EarthX ETX680 Primary Battery
TCW IBBS Backup Battery - outputs wired to G3X system(s)
Old fashioned Klixon 7277 breakers, no VPX or similar "systems"
Whelen LED Lighting - "Off"
Avionics, Radios - "Off"
--
No Capacitor filtering on alternator B lead
Standby Generator is "Off"
TCW Backup Battery is "On"
--
Scope probe is connected to the aircraft buss and airframe ground and set to X1 attenuation. AC coupled. DC triggered

----Here we go----

The attached scope shots are from the aircraft electrical buss while running the engine at ~1200RPM.



And another with the Alternator Field turned off; only the battery is supplying the buss:



If you look at the first image, you can see the "humps" in the second image superimposed on the waveform.

Turning off the ignitions causes the "humps" to go away. I think I have an understanding of where this signal is coming from (CDI charge/discharge -- 4cyl, 2 firings per 360 degrees - at 1200RPM (20RPS) you would expect to see 40 firing events per second -=> 25mSec, which is about what I can derive from the humps above.)

If turning the ignitions off causes the little humps to go away, and turning the alternator field off causes the bigger swings to go away, what is the *real* cause of the bigger swings? Instinctually, you might say "alternator ripple" -- but look closely at the frequency/period of the signal, that's not "normal" ripple.

For giggles and grins - here's an image of "normal" ripple from the same alternator (in a different airplane)...note time scale and amplitude difference (ignore the cursors -- just count the dots...)



Enjoy!

 

[Jjackh10]

Years ago while performing assured grounding PM's I found 5 amps coming from the ground wire of a 400hp down hole pump. Broke out the meters and went to testing. It was real, but hitting the ground grid before switchgear so the Multilin wasn't tripping. More testing, McEmax online testing, quizzing the EE's with no real answer. My solution became stop looking at the meters and let her pump.

Put the scope away and go fly buddy.

 

[rmartingt]

I wonder... are the big bumps in the first image actually an aliasing effect from the combination of a much higher frequency alternator ripple with the pulses from the ignition? If the ripple frequency is just slightly off from a multiple of the ignition pulses I wonder if you could get some interesting interference effects. Add in a different sample rate....

Edit:

 

[bjdecker]

I wonder... are the big bumps in the first image actually an aliasing effect from the combination of a much higher frequency alternator ripple with the pulses from the ignition? If the ripple frequency is just slightly off from a multiple of the ignition pulses I wonder if you could get some interesting interference effects. Add in a different sample rate....

Edit:

I love XKCD -- I'll buy you a beer just for posting this

Ok -- back to work

 

[bjdecker]

Years ago while performing assured grounding PM's I found 5 amps coming from the ground wire of a 400hp down hole pump. Broke out the meters and went to testing. It was real, but hitting the ground grid before switchgear so the Multilin wasn't tripping. More testing, McEmax online testing, quizzing the EE's with no real answer. My solution became stop looking at the meters and let her pump.

Put the scope away and go fly buddy.

You know that somewhere out there in the patch there's a big mound of hyper-excited harvester ants -- and it's all because of that 5A ground loop.

 

[rmartingt]

I love XKCD -- I'll buy you a beer just for posting this

Ok -- back to work

What do you think I'm doing instead of working?

 

[ReedC]

To the Hive Mind -- If anyone can figure this out - I'll buy you a beer at Airventure.

As part of solving any puzzle, it's good to have a view of the forest first - so here's the configuration:

RV-7
IO-360-A1B6 - 9.75" Pulley
Plane Power AL12-EI60B Alternator - 2.75" Pulley
Monkworks MZ30L Standby Generator
Lightspeed Plasma III Ignitions (2)
EarthX ETX680 Primary Battery
TCW IBBS Backup Battery - outputs wired to G3X system(s)
Old fashioned Klixon 7277 breakers, no VPX or similar "systems"
Whelen LED Lighting - "Off"
Avionics, Radios - "Off"
--
No Capacitor filtering on alternator B lead
Standby Generator is "Off"
TCW Backup Battery is "On"
--
Scope probe is connected to the aircraft buss and airframe ground and set to X1 attenuation. AC coupled. DC triggered

----Here we go----

The attached scope shots are from the aircraft electrical buss while running the engine at ~1200RPM.

View attachment 84874

And another with the Alternator Field turned off; only the battery is supplying the buss:

View attachment 84871

If you look at the first image, you can see the "humps" in the second image superimposed on the waveform.

Turning off the ignitions causes the "humps" to go away. I think I have an understanding of where this signal is coming from (CDI charge/discharge -- 4cyl, 2 firings per 360 degrees - at 1200RPM (20RPS) you would expect to see 40 firing events per second -=> 25mSec, which is about what I can derive from the humps above.)

If turning the ignitions off causes the little humps to go away, and turning the alternator field off causes the bigger swings to go away, what is the *real* cause of the bigger swings? Instinctually, you might say "alternator ripple" -- but look closely at the frequency/period of the signal, that's not "normal" ripple.

For giggles and grins - here's an image of "normal" ripple from the same alternator (in a different airplane)...note time scale and amplitude difference (ignore the cursors -- just count the dots...)

View attachment 84875

Enjoy!

Do you have any undesirable symptoms you are trying to get rid of? Or just not happy with the 150 mv ripple on the scope?

Are all the interesting artifacts visible on the ac coupled waveform or is there more clues in the dc coupled picture?

I’m not sure this is practical to do, but it would be interesting to see a scope screen grab of the alternator on and the ei off. Do you see normal ripple like picture 3?

If so, I’d guess it’s interaction between the alternator closed loop regulation and the EI. Adding some bulk capacitance close to the EI might clean things up- providing some isolation between the alternator and the EI current demand.

Reed

 

[lr172]

Imo, The humps are likely the ign capacitor charging between firing events. You can see the rise is steeper than the fall. The voltage regulators are not highly precise and it overshoots a bit with the rapid rise in current draw.

I suspect the tall spikes are some leakage past the rectifier diodes in the alternator. These are far from perfect at chopping the sinusoidal ac waves. Without a scope, you would never see the .15 volt swings.

You can’t compare the third image with the first two, as the time scale is wayyy off - ms vs us. If the third image was in ms, it would probably look like a fuzzy0 - 50mv line. Just like the lines in pics 1&2 are fuzzy bands and not clear lines.