Charging system help

[tps8903]

I think I screwed up. I'll start there. Any Electrical geniuses around to help me out?

In an effort to get my Earth X charging better I sourced a couple of Lithium Specific Regulators. I bought one that was 20Amp and designed for permanent magnet systems up to20 Amp. Score Right? Unfortunately, it was full wave, not half wave. That does make a difference apparently. So that regulator is like fried ($160)

Shortly after take off my RPM went to zero, and my Charge never really went above 14.4V, which was a minor improvement. Then it started decreasing. To top it off I had a top Magnetron that was acting up in flight, which appears to have sorted itself out for now.

Not satisfied with the new product I put the stock Aerovee VR back on, and went to do a circuit around the airport. No RPM, no charge. So I swapped the AC Leads to make sure RPM was on the correct AC pin. Same results. While doing a run up my Earth X went I to BMS safe mode and turned off.

So now I have a dead battery, an unknown status for my Aerovee VR, and an unknown status for my Alternator, and or I wrecked the RPM input on my RDAC XF.

I'm going to check the Stator for Continuity tonight. It's my understanding the should be zero ohms between AC outputs and no Continuity to ground. Also it should put out about 30VAC? at 3000rpm?

Any other advice appreciated.

[tps8903]

So I have NO Continuity on the AC Leads. They also have NO Continuity to Ground. I suspect the solder let go. Or I melted the stator

[BRS]

Chris,

Sounds like the exact mistake I made about a year ago. Got the nice B&C full wave regulator built for the Rotax 9xx engines. Quickly let the smoke out of my stator (at idle!). Mine is a Revmaster so the alternator is a bit different, I'm not sure how Aerovee does their stator.

I'm using the EarthX ETX680c on two planes. One of which has a 60amp alternator. It's common to see 50+ charging amps at idle but in this plane that works. It will try to do the same thing on the Aerovee, for a second or two. I've found, that on both planes, the key is to let the battery charge at idle/low-idle until the current goes down and voltage comes up. Usually takes about 4 minutes which is a non issue as the oil is still warming up and I'm going through checklists.

[tps8903]

Chris,

Sounds like the exact mistake I made about a year ago. Got the nice B&C full wave regulator built for the Rotax 9xx engines. Quickly let the smoke out of my stator (at idle!). Mine is a Revmaster so the alternator is a bit different, I'm not sure how Aerovee does their stator.

I'm using the EarthX ETX680c on two planes. One of which has a 60amp alternator. It's common to see 50+ charging amps at idle but in this plane that works. It will try to do the same thing on the Aerovee, for a second or two. I've found, that on both planes, the key is to let the battery charge at idle/low-idle until the current goes down and voltage comes up. Usually takes about 4 minutes which is a non issue as the oil is still warming up and I'm going through checklists.

I'll take a picture of mine once it's out. I ordered a new stator, and a new Sonex regulator. I'll have a back up regulator, and if the stator only broke a lead wire I'll have a back up stator. At $320 a pop, they aren't cheap. This is my lesson learned for thinking I could do it better than the factory.

In hindsight my factory regulator with the Earth X was probably fine. I always landed with 12.2-12.3 which the folks at Earth X says is 90% charge....

[tps8903]

I have the orange epoxy?? coated stator. The casing around the stator is completely cracks and small chunks have been ground into dust inside the rotor. I cant see exactly where it failed as the entire stator is covered in the orange casing.

Based on the Sonex Store site the new stator seems to omit the plastic casing and has the copper insulated wire visible.

[Bryan Cotton]

If it is of interest, at the bottom of this page is my mechanical master switch. A little lighter than the solenoid and it draws no current.
viewtopic.php?f=6&t=2009&start=20

Edit:

[WesRagle]

HI Guys,

I'm having charging system problems with my GPAS 2180 VW. I was working on the VeeCU trying to figure out why my Alternator RPM reading was not registering. I was looking at one phase of the alternator and noticed that it was "flat topping" at 20 volts. So, I shut the engine down. Then I went and touched the battery and it was almost too hot to handle.

Without much thought I decided that the voltage regulator had to be on the fritz. I looked at the old regulator and saw it had five wires. One was power, one was ground, and three that fed directly from the alternator. So, I determined I needed a three phase Harley voltage regulator. I found this one and bit. Ref. (https://www.amazon.com/dp/B0081SD6VI?ref=ppx_yo2ov_dt_b_fed_asin_title). I installed the new regulator and resumed my search for the missing alternator signal.

Next I probed the VR inputs, but they didn't look like I expected (referenced to ground). I expected to see three flat top sine wave inputs. Instead I saw a rectangular wave form of varying duty cycle and, on top of that, missing pulses.

Next I unplugged the regulator and wired the three phases of the alternator to the bench so I could have a look with the Oscope. Each phase had a nice sine(ish) wave when measured with reference to ground. Pleased with that I wired one of the phases to the VeeCU. Nothing. Still no reading.

Next I hooked a probe to a phase and watched as I hooked up the VeeCU input. The signal disappeared when I hooked up the VeeCU.
I know full well that the VeeCU alternator input impedance is 10 K Ohm to ground. Plus, a short would throw some sparks.

Next I checked the impedance of each phase to ground. They were open. So, it appears that I have three isolated outputs.

Next I measured the resistance of each phase to phase. They were all the same, very low.

Next I got a 1 Ohm resistor (to use as a fuse) and connected one of the phases to ground through the resistor. No problem, no current.

Next I hooked up the VeeCU to one of the other phases and (angles singing) the correct reading appeared on the display.

I also remember that the signal of the two phases not connected to ground increased in amplitude when I connected the one phase to ground. It looks like I have three phases wired delta.

Next I got tired and came in.

I am on a quest to make the VeeCU as immune as possible to sensor failures. To make that happen I need a reliable alternator signal to back up the crank position sensor (secondary ignition module).

So, I need some electronics guru to unscramble my brain and offer some advice on my next move.

Thanks for Your Time,

Wes

[WesRagle]

Hey Guys,

No thoughts? I know some of you have studied the charging systems on VWs.

Here is a pic of the two un-grounded phases with the third phase grounded. Yellow and blue are the two phases. Purple is the difference between yellow and blue calculated by the scope. 20 Volts Peak at a relatively low RPM.



EDIT: I found a regulator with the same plug as the old one. I'll give that one a try.

Wes

[Dave Wolfe]

Re the missing pulses that could be a switching regulator control strategy see here for example:

https://m.youtube.com/watch?v=YpvTz8fif ... Rpaw%3D%3D

The "flat topping" sounds like a shunt type linear regulation scheme where voltage is maintained by dumping extra current to the regulator.

[WesRagle]

Hey Dave,

Thanks for the reply.

Yea, I don't know what's in that new regulator I bought, but I don't like it :-)

I think I'll just cool my heels and see if the problem goes away with the new regulator. Thankfully the stator checks good so probably no damage there.

I really don't want to run down a rabbit hole and lose focus on the task at hand. Having said that, I do need to know if I need to modify the alternator input circuit to accept most/all alternators. I only have the GPAS and the Hummel in my Onex as examples. The GPAS has three wires (phases) coming out ot the alternator and the Hummel has two.

Question: Isn't it true that most folks use the alternator for a tach input to avoid losing their RPM signal while doing mag checks during runup?

Wes