Page 1 of 3

Interesting VR failure

PostPosted: Thu Sep 21, 2023 1:11 am
by Bryan Cotton
We had a new VR from Sonex. Voltage was edging up and we had seen 13.6 early in the flight. We were doing a night flight to an airport 26nm away. About the time we started descending, my hard wired red LED headlamp started going bonkers. Checked the voltage and it was over 15. Adam thinks he saw 18 before the display changed. The battery light fault came on. So we headed home. Voltage was variable, between 13.4 and 14.1 or so. Battery fault light went away on the ground, but voltage was only 13.2.

I'm going to try my Amazon VR and see what it does.

Re: Interesting VR failure

PostPosted: Thu Sep 21, 2023 10:15 am
by daleandee
What you are seeing my not be a failure of the regulator. I had a similar condition and when replacing the regulator I found a poor connection that I had missed earlier.

My Corvair uses a PM alternator. The same type regulator used on the Corvair & Jabiru engines may also work for the VW conversions:

Image

Wires are as follows:

blue wires (2) are AC voltage from the PMA
the red wire (1) is charging voltage to the battery
black wire (1) is ground
yellow wire (1) is the sense wire & turns on the regulator
green wire (1) idiot light - generally not used

Pretty sure I'm not telling you anything you don't already know. I've bought cheap reagulators and didn't get what I paid for so spend the money for a good one. My plane is electricity dependant so I take a lot of care to be certain my electrical system works well all the time.

These kinda things can be frustrating ...

Dale
3.0 Corvair/Tailwheel

Re: Interesting VR failure

PostPosted: Thu Sep 21, 2023 11:11 am
by Bryan Cotton
Dale,
What regulator is that? Do you have a part number and source?

Edit: one thing I'm considering is adding a dedicated ground. Currently the VR gets its ground through the firewall.

Information about PMG static winding failure

PostPosted: Thu Sep 21, 2023 11:50 am
by GordonTurner
I wrote to Bryan in a different thread and suggested that per the BandC Technical Manual for their AVC-1 rectifier/regulator, it mentions that the voltage may have to be increased to maintain wet lead acid or LITHIUM type batteries. It comes pre-set for Odyssey.

Also interesting was that BandC do not recommend their AVC-1 for use with Jabiru motors. This sparked my curiosity, as their statement was “may cause overheating of stator windings”. I did a little further digging, on the Jabiru website, and found this:

https://jabiru.net.au/wp-content/upload ... edures.pdf

I believe that actually any installation using a PMG generator is susceptible to this problem. In summary, if the charge on the battery is very low, the PMG will burn itself up trying to match the current inrush to the battery when full power is applied. PMG’s do NOT govern current, current is a function of energy going into the PMG (engine rpm is a simplified way to look at it) and DEMAND.

So. Example. Generator rating is 20amps. Try to keep it at 80% or less most of the time for long life. So 16amps. Controlled by what electrical loads the pilot selects, and available based on rpm. Depending on the installation, the PMG may be able to provide significantly more at takeoff/climb power should the load be very high. In the case of a depleted battery, any battery but specifically a LITHIUM battery, its initial charging current demand could be much higher than 20amps.

Newer Jabirus have OVER-CURRENT protection specifically to guard against this case.

Presumably Rotax does too, hence they are the prime market for AVC-1.

Everything else leaves it UP TO THE PILOT to control the load. The only control is don’t take off with a deeply discharged battery.

The AVC-1 is not alone in this limitation btw. All of the VR’s out there that I am aware of do not provide OVER-CURRENT protection. So operate accordingly.

Good luck out there. Gordon

Re: Interesting VR failure

PostPosted: Thu Sep 21, 2023 11:53 am
by jaflint
The VR Dale is referring to is a Yanmar 129150-77712. Available at seafarerstore.com.

Re: Interesting VR failure

PostPosted: Thu Sep 21, 2023 12:18 pm
by daleandee
jaflint wrote:The VR Dale is referring to is a Yanmar 129150-77712. Available at seafarerstore.com.


That appears to be an even better unit! The one I found was on ebay under the John Deere name:

https://www.ebay.com/itm/354602415740

Image

Buy a genuine JD or Yanmar regulator made in Japan (not China) as these work well. I've bought cheaper and was disappointed ...

Dale
3.0 Corvair/Tailwheel

Re: Interesting VR failure

PostPosted: Thu Sep 21, 2023 11:26 pm
by Bryan Cotton
I find it curious that I had two voltage regulator failures in a row. One that has been good for 145 hours, and one that was brand new from Sonex. So, I'm going to investigate the potential ground issue. I put the old "failed" VR back in, and added a 10ga ground wire.
VR ground wire.png
VR ground wire.png (446.03 KiB) Viewed 10474 times


The other end goes to my grounding point on the engine.
VR ground on block.png
VR ground on block.png (546.42 KiB) Viewed 10474 times


My neighbors were spared a ground run at 10:00 tonight because it's pouring out. I'll report back tomorrow.

Re: Interesting VR failure

PostPosted: Fri Sep 22, 2023 3:31 am
by Skippydiesel
I have always been a big promoter of dedicated earth returns for all 12V systems.
This does not mean that the airframe can not be part /parallel in the circuit, but does mean that you don't rely on the airframe alone.
Simple reasoning: an airframe earth return much like car chassi return, is dependent on the electrical conductivity of multiple joins. These joins usually work well, when the aircraft is new, but add years of vibration, may be a bit of salty air and you may experience elevated resistance through to complete failure- run an earth wire and you will never have an earthing problem. Simple low cost solution, that seems to get a lot of experts, LAIM's and such, a bit "hot under the collar".

Re: Interesting VR failure

PostPosted: Fri Sep 22, 2023 10:50 am
by Bryan Cotton
I am also a fan of dedicated returns. This was one place I skipped it for some reason.

I did a test run with the original VR. It is still failed. Will try the Amazon VR next. I've ordered a John Deere VR from the ebay link.

Re: Interesting VR failure

PostPosted: Fri Sep 22, 2023 2:19 pm
by Bryan Cotton
On a thread here somewhere, a link was posted to a Stens 435-081 VR. Here it is again:
https://www.amazon.com/dp/B000G65X4Q

Looks like a chinesium copy of the transpo VR that came in my Aerovee kit. Bolt in replacement.
Transpo vs Stens VR.png
Transpo vs Stens VR.png (604.99 KiB) Viewed 10389 times


Ground run looks promising. Here is a pic of my MFD showing 14.5V in the bottom left corner at just under 3000 RPM. No electrical load except for recharging the battery.
Alpha screen VR ground run.png
Alpha screen VR ground run.png (515.28 KiB) Viewed 10389 times


I turned on everything and it went down to 13.9V, which still seems good.

Now the question is - do I trust it to fly 630nm to Elmira, NY next weekend?

Edit: for $27, I have a spare on order.