What's new
Van's Air Force

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

Rectifier regulator location help

We had two failing regulators in 300 hours.
Both just quit during cruise flight and not in extreme ambient temperatures or high engine loads.
The effect was the current reading going from -13 to + 1 Amp and back all the time in a period of a few seconds.
On the last one I installed a number of thermo strips and found the all time max (over more than one year) has been 82 degrees C. This is with all options installed and active.

The official spec says max 90 degrees C but more sensible is not to exceed 80 which it did more or less.

I made a full test setup with a car battery ; large AC transformer 22volts @30 Amps max to sumulate the real thing.
I used a car headligth as a load at the battery.

What I find on both broken regulators is that when activated the charge current is about 1 Amp and stable with no battery load. When I load the battery with the lamp the regulator charges about 13 Amps which is correct. When I remove the load the charge current starts hunting between 0 and a few amps which dies not stop until I disconnect the source power (transformer)

Installed the third one now and improved the airflow into the blow tube.
I don't like the idea of the regulator inside and await the results of others who did this.
A better regulator such as the John Deere seems like a good alternative. Anybody using it for some time now? It is easier to get and cheaper.
 
John,

You're reading my mind. I bought a John Deere regulator as a spare and will make up an install plan with standby harness --- just in case. I also have a spare fuel pump. If this keeps up I'll have to enlarge my hangar!😜

Merry Christmas!

Rich
 
cactuspilot said:
This article is interesting - the failure rate is pretty consistent across the industry of airplanes using the 912, so much so it seems prudent to have an extra regulator in your airplane away repair kit.

http://contrails.free.fr/elec_ducati_en.php
Click to expand...

Thanks John, very enlightening indeed. Of course this is research dating back to 2003, 10 years ago. The Rotax/Ducati regulator may have evolved since then but as what we see now is still an excessive failure rate it may not have. In this context, I feel even more that the relocation of the RV-12 regulator may prove futile in the long run. Only time will tell but for those who don't want to wait, the John Deere regulator may be a good option (and one that our fellows S-SLA owners don't have).
 
WingedFrog said:
Thanks John, very enlightening indeed. Of course this is research dating back to 2003, 10 years ago. The Rotax/Ducati regulator may have evolved since then but as what we see now is still an excessive failure rate it may not have. In this context, I feel even more that the relocation of the RV-12 regulator may prove futile in the long run. Only time will tell but for those who don't want to wait, the John Deere regulator may be a good option (and one that our fellows S-SLA owners don't have).
Click to expand...

Which John Deere regulator? p/n AM101406?
 
rgmwa said:
How did you improve the airflow?
Click to expand...

One easy thing you can do is to pull a little of the slack (if you have it) of the conduit and turn the end of it 90 degrees inside the air horn. Turn it into the airflow from the scat hose. Drill 2 tiny holes in the air horn and use a tie wrap to hold it in place. Make sure not to bend the conduit so sharply that it crimps.
 
Dave12 said:
One easy thing you can do is to pull a little of the slack (if you have it) of the conduit and turn the end of it 90 degrees inside the air horn. Turn it into the airflow from the scat hose. Drill 2 tiny holes in the air horn and use a tie wrap to hold it in place. Make sure not to bend the conduit so sharply that it crimps.
Click to expand...

Presumably the whole duct is pressurised in flight by ram air, so the orientation of the end of the conduit shouldn't make much difference to the amount of air flowing to the regulator. On the other hand, if there was some kind of secondary venturi effect generated at the right-angle junction of the small conduit with the much larger main duct, that might inhibit the flow of air into the conduit, so your idea may be worth trying. Can't see that it would do any harm anyway. As Groucho Marx said (more or less), "It's only a theory. If you don't like it, I've got plenty more".
 
Dave12 said:
One easy thing you can do is to pull a little of the slack (if you have it) of the conduit and turn the end of it 90 degrees inside the air horn. Turn it into the airflow from the scat hose. Drill 2 tiny holes in the air horn and use a tie wrap to hold it in place. Make sure not to bend the conduit so sharply that it crimps.
Click to expand...

This way of scooping the air from the conduit would likely improve the air flow to the regulator. I am concerned that you would intercept a significant part of the air cooling the cylinder fins. My preferred solution is to do the same thing after relocating the blast tube pick-up to the larger cooling conduit going to the coolant radiator. From the air stream I receive through the heater vent, there seems to be plenty of air flow available there. This would require minor surgery on the fiberglass conduit and also an easy way to disconnect the tube when removing the lower cowl. I think it is the best option at this point to keep cooling the regulator with forced air as the research studies listed earlier have recommended.
 
WingedFrog said:
This way of scooping the air from the conduit would likely improve the air flow to the regulator. I am concerned that you would intercept a significant part of the air cooling the cylinder fins. My preferred solution is to do the same thing after relocating the blast tube pick-up to the larger cooling conduit going to the coolant radiator. From the air stream I receive through the heater vent, there seems to be plenty of air flow available there. This would require minor surgery on the fiberglass conduit and also an easy way to disconnect the tube when removing the lower cowl. I think it is the best option at this point to keep cooling the regulator with forced air as the research studies listed earlier have recommended.
Click to expand...

I wondered about that option too. However the air in the duct flows past the oil cooler and runs up against the coolant radiator, so I imagine that the air in at least parts of the duct could get quite warm, which is not what you want for air going to the regulator. At least the air coming from the engine shroud inlet should be cool. Perhaps the best option would be to graft a fibreglass stub onto the side of the shroud inlet at an angle that would direct air smoothly into the conduit. Shouldn't be too difficult.
 
Last edited:
rgmwa said:
I wondered about that option too. However the air in the duct flows past the oil cooler and runs up against the coolant radiator, so I imagine that the air in at least parts of the duct could get quite warm, which is not what you want for air going to the regulator. At least the air coming from the engine shroud inlet should be cool. Perhaps the best option would be to graft a fibreglass stub onto the side of the shroud inlet at an angle that would direct air smoothly into the conduit. Shouldn't be too difficult.
Click to expand...

I doubt that the air upstream from the coolant radiator is warmed by either radiators. The air passes partially in front of the oil radiator, a small amount is sucked/pushed through it, cooling the oil and venting against the exhaust but I doubt there is any significant warming of the air flowing downstream to the coolant radiator. My plan is to pass the blast tube through an elongated hole into the main conduit and fasten it inside the conduit against the wall with its opening oriented forward. The fastening can be done with tie-wraps passing through slits in the fiberglass wall and probably a helping of the same silicon caulk I used for the initial installation of the blast tube. A better solution would be to create a scoop in fiberglass with an adequate stub to connect the blast tube but I prefer to test the results of this minimally invasive set-up before doing something more permanent.
 
This thread, it seems to me, begs the question: Are there better, heavier duty, regulators out there that would solve this problem? A John Deere regulator has been mentioned, so it looks like others are thinking about this as well.

Maybe a better regulator, combined with the new location. Of course now that dual Skyviews are on the table that might tax the system even more.

Jim
 
Maybe I'm whistling past the graveyard, but I have three Phoenix summers on my 12 with no issues. I plan to maintain the original configuration until failure (hopefully not in my dwindling lifespan) and then try a John Deere regulator in the same spot. After all -- nothing runs like a Deere! :D
 
RFSchaller said:
Maybe I'm whistling past the graveyard, but I have three Phoenix summers on my 12 with no issues. I plan to maintain the original configuration until failure (hopefully not in my dwindling lifespan) and then try a John Deere regulator in the same spot. After all -- nothing runs like a Deere! :D
Click to expand...

You are right, Rich, the issue is far from being settled. What seems to be clear from the research paper from Grenoble is that there is a need for cooling that goes beyond just convection. The initial solution by Vans with the blast tube made sense and I would like to understand under what rationale they rushed to this relocation solution. Some basic tests need to be made about the volume of air flowing through this tube and what could affect it. Although the temperature under the cowl raises after engine shut down, I don't believe that this is an issue because after shut down the regulator is off, not generating any heat. Even the need to use a conductive paste between the regulator and the firewall shelf is not clear as the regulator is filled with foam on the bottom side which means that heat dissipation from the bottom is very limited (another good initial call by VAN's instructions).
As the saying goes: before acting on this issue, it is urgent to wait and I am with you.
 
WingedFrog said:
You are right, Rich, the issue is far from being settled. What seems to be clear from the research paper from Grenoble is that there is a need for cooling that goes beyond just convection. The initial solution by Vans with the blast tube made sense and I would like to understand under what rationale they rushed to this relocation solution. Some basic tests need to be made about the volume of air flowing through this tube and what could affect it. Although the temperature under the cowl raises after engine shut down, I don't believe that this is an issue because after shut down the regulator is off, not generating any heat. Even the need to use a conductive paste between the regulator and the firewall shelf is not clear as the regulator is filled with foam on the bottom side which means that heat dissipation from the bottom is very limited (another good initial call by VAN's instructions).
As the saying goes: before acting on this issue, it is urgent to wait and I am with you.
Click to expand...

When mine goes into paint, I plan to relocate mine and redo the panel with that carbon fiber overlay....I was wondering if in addition to ensuring the ground is proper and the proper use of heat dissipation silicone....perhaps the use of a CPU style cooler and lower amperage fan could be used on top of the the reg when mounted up side down using the wiring for the avionics fans?

I have noticed that there is a lot of heat coming off that firewall and indeed my center lower console (where the red cube and rudder cables go) is extremely hot - to the degree that I am considering some kind of additional insulation blanket.

I heard from Mike S that the demo/trainer did blow a lot of these regs as well - so I expect they did do some testing with the newer location.
 
Last edited:
WingedFrog said:
I would like to understand under what rationale they rushed to this relocation solution.

Some basic tests need to be made about the volume of air flowing through this tube and what could affect it.
Click to expand...

Rushed?
The design that was abandon was in place since the introduction of the RV-12 (4 + years?)

You assume tests were not made.....? ;)
 
Jim T said:
This thread, it seems to me, begs the question: Are there better, heavier duty, regulators out there that would solve this problem? A John Deere regulator has been mentioned, so it looks like others are thinking about this as well.

Maybe a better regulator, combined with the new location. Of course now that dual Skyviews are on the table that might tax the system even more.

Jim
Click to expand...

I wonder if the regulator that B&C sells (designed for use with their alternators) would work? The B&C external regulators are rock solid but probably larger than the regulator that is supplied by Vans. They are designed to be mounted on the cold side of firewall
 
RFSchaller said:
John,

How did you notice the hot firewall? It's tough to reach in the wheel well.
Click to expand...

lol - and I am not nearly that flexible either!

No - I got one of those cheap laser thermometer pointer things when I noticed how hot the center got while running it up in shorts and sandals - no carpet.

It's frequently 90+ when I point it at the firewall or center console even after a while at higher altitudes and at cruise with all the vents open - point it at the vent and get 55, point it at the firewall near the center console - 91.

With the carpet in place and at altitude it doesn't seem too bad and of course in the desert with that greenhouse canopy I thought it was just typical. There are a few threads here talking about it, and I think I saw someone put a vent/fan in the red cube well......

I wonder if anyone has thought about or done cowl flaps for the 12?
 
Last edited:
Couple of comments from long time Rotax 914 owner / operator in my Europa.

If you look at the photo of the regulator / rectifier on the Contrails website, the two diodes (MR2510) that make up half the bridge rectifier are soldered to a copper plate located at the bottom left corner of the photo. The plate is at +12 so it cannot be attached to the Ducatti enclosure which is at ground potential. What happens over time is that the solder connection fails due to thermal stress and so the bridge rectifier no longer functions. Often times the solder will reattach when the regulator cools and will function properly until it is heat stressed again, hence the intermittent nature of the failure. Interesting to note that the thermal photos do not show a hot spot near the diode location which affirms that they do not have a good thermal path.

My understanding is that a certified installation has no choice but to replace the regulator. In an experimental installation, you can add a pair of external diodes in parallel to the internal ones (anodes connect to GG and cathodes to +B/R). Its actually easier to use a bridge rectifier. I have done this and have over 500 hours on my Ducatti.

As shown on the Contrails website, the Ducatti is heat stressed especially when operating near its specified limits of 240 watts (20 amps). It really needs cooling air. I have temperature indicating strips on my Ducatti, ignition modules and external bridge rectifier heat sink and inspect them during preflight.

Aside to Rick, the Rotax engine has a dynamo which requires a regulator / rectifier. The B&C LRC3 regulator does not provide rectifier function. The PMR1 seems to, but B&C does not recommend it. There are shunt rectifier / regulators used on motorcycles but these also will not work with Rotax. The german Schenke regulator rectifier used by Contrails seems to be more robust but is not readily available in the US.

Jim Butcher
Europa XS, Rotax 914, GRT Sport EFIS
 
John,

I think that smoothing capacitor is in the instrument panel module from Vans. I noticed one when I put my module in and assumed that's what it was. I'd have to check the schematic to be sure.

Rich
 
This is another piece of information very relevant to this thread. I took it from a Blog belonging to Klaus Truemper, the author of the link provided above by Cactuspilot.

http://pointsforpilots.blogspot.com/2014/08/cooling-rotax-912-voltage.html

This thread becomes a wealth of information about the 912 Rotax/Ducati regulator/rectifier

I had a discussion with an A&P specialist of Remos at my airport about our issues. As soon as he heard the word Rotax/Ducati regulator he started smiling. Remos have the same problems as we have and that are well explained in the above link. He figures that the MTBF for the Remos is about 400 hours.
 
Last edited:
I had heard about the failure rate of the Ducati regulator in the RV-12 configuration early on and researched this quite a while back. It seems that it is worse in the RV-12 than in other Rotax installation but still not very encouraging when looking at the MTBF in those other installations.
I decided to look into motorcycle regulators, particularly the ones that feature modern MOSFET designs which are emitting less heat and therefore are less likely to contribute to the temperature problem we are seeing with the Ducati regulator.
After a lot of searching I settled on a Shindengen FH012, a 3-phase regulator with MOSFETs, 30Amp continuous rating and said to be very tolerant to heat. They are used in Honda and Yamaha motorcycles and ATVs and are readily available.
There are reports that some Ducati drivers that were haunted by failing regulators have successfully solved their issues by installing this regulator instead of the original one. Yes, it seems they also fail on motorcycles. Who knew?
My reg is still doing fine at 70 hours but I have the FH012 sitting on a shelf in case I see problems. The installation will require some changes to the plug and certainly some slight changes to the mounting point but the size is comparable and it should fit on the shelf fine.
I won't be working on this replacement before need arises but as this thread is becoming very popular I thought I let anyone know about what I found so far, in case they like to pick up the idea and go ahead working on this.

BTW, the 3-phase input is not a problem. It is just a feature that we cannot use as the Rotax generator only has one phase. Just connect the Rotax output to one of the 3 phase inputs on the reg and leave the other 2 leads open.

Why didn't I decide to use the Deere regulator that was mentioned here earlier? It's the same architecture as the Rotax and the power rating marginal for our use (considering it will be running in a hot environment too). So it's likely that it might just be cheaper to replace but might still fail - although not so frequently, maybe. YMMV.
 
fyi, just did a search for the Shindengen FH012 regulator. They seem to have replaced that model with the FH020AA. According to one source: "Shindengen has replaced the former FH012AA regulator with a updated version, the new part number is FH020AA . The updated version is physically the same size only minor changes have been made to the heat sink casting."
 
tomkk said:
fyi, just did a search for the Shindengen FH012 regulator. They seem to have replaced that model with the FH020AA. According to one source: "Shindengen has replaced the former FH012AA regulator with a updated version, the new part number is FH020AA . The updated version is physically the same size only minor changes have been made to the heat sink casting."
Click to expand...

I decided against using the FH020 for now because among the "minor changes" they were quoting was a temperature control. This control is supposed to avoid overheating and will shut the regulator off if it feels it is becoming too hot. I think in our environment I would rather not have to worry if the non-charging is coming from a regulator failure or an overly cautious temperature control circuit.
Now once I would have verified that the FH012 is working great, I would probably consider checking the FH020. It is a drop-in replacement as the connectors are identical. It might be that the temperature control would only kick in after engine shut off in our application or for a short time after starting the engine when finished with a refueling stop on a Southern Arizona airport in the summer. Either way, I did not want to have too many parameters in the equation that could have a "not charging" result when trying something new.
 
MacPara said:
I decided against using the FH020 for now because among the "minor changes" they were quoting was a temperature control. This control is supposed to avoid overheating and will shut the regulator off if it feels it is becoming too hot. I think in our environment I would rather not have to worry if the non-charging is coming from a regulator failure or an overly cautious temperature control circuit.
Now once I would have verified that the FH012 is working great, I would probably consider checking the FH020. It is a drop-in replacement as the connectors are identical. It might be that the temperature control would only kick in after engine shut off in our application or for a short time after starting the engine when finished with a refueling stop on a Southern Arizona airport in the summer. Either way, I did not want to have too many parameters in the equation that could have a "not charging" result when trying something new.
Click to expand...
Good to know. The site I was looking at didn't mention the auto shutoff ...
 
Yes, you have to have a capacitor on the output of the regulator / rectifier. It smooths the DC output. It is called out in the Rotax Install Manual and I believe supplied as part of the RV12 electrical system.

Shindengen regulators / rectifiers are shunt regulators. I'm not an expert, but my understanding is that a shunt regulator / rectifier is not very efficient. They dissipate energy not needed to satisfy power demand as heat. The advantage of the Shindengen units over other shunt regulators / rectifiers is that they use MOSFET technology rather than SCR technology, MOSFET being more efficient. So it is a more efficient way to dissipate energy not needed.

The Ducatti regulator / rectifier is a bridge rectifier which only rectifies (converts from AC to DC) the AC from the dynmo when needed to satisfy power demand. If your power requirements are modest, much of the time the SCRs are turned off, generating no heat. When needed, they turn on and heat is generated. It also provides more pure DC so it is easier to filter and smooth.

The John Deere unit mentioned looks pretty good too. Here's a link to more information on it: http://www.n56ml.com/corvair/dynamo.html

Jim Butcher
Europa XS Rotax 914, GRT Sport EFIS
 
rvbuilder2002 said:
Rushed?
The design that was abandon was in place since the introduction of the RV-12 (4 + years?)

You assume tests were not made.....? ;)
Click to expand...

What was I thinking when I wrote this? (I suspect I was in the mind of an S-LSA owner) ;). You are right Scott, as an E-LSA builder I took possession of a sound platform developed by VANs based on decades of experience (and tests!:rolleyes:) and I should be thankful for it (and I am!). As far as tests go, if I disagree with Vans solutions it is my problem to prove them wrong by doing my own tests as the E-LSA regulations allow.
Some people thanked you in another thread for the great help you provide through your participation to this Forum and I want to join the chorus on this thread. Merry Christmas!
 
h&jeuropa said:
Shindengen regulators / rectifiers are shunt regulators. I'm not an expert, but my understanding is that a shunt regulator / rectifier is not very efficient. They dissipate energy not needed to satisfy power demand as heat. The advantage of the Shindengen units over other shunt regulators / rectifiers is that they use MOSFET technology rather than SCR technology, MOSFET being more efficient. So it is a more efficient way to dissipate energy not needed.

The Ducatti regulator / rectifier is a bridge rectifier which only rectifies (converts from AC to DC) the AC from the dynmo when needed to satisfy power demand. If your power requirements are modest, much of the time the SCRs are turned off, generating no heat. When needed, they turn on and heat is generated. It also provides more pure DC so it is easier to filter and smooth.

The John Deere unit mentioned looks pretty good too. Here's a link to more information on it: http://www.n56ml.com/corvair/dynamo.html
Click to expand...
The FH series from Shindengen are shunts. They have just come out with their first series regulator which is the SH775 and they have two more (more powerful) versions in development. The SH775 is good for 14Amps continuous without any cooling and claims 27Amps with light cooling, 35max.
I'd rather wait for the slightly more powerful version as I like to have some tolerance in respect to the currents we are running and as it is in a rather warm environment. I'm just thinking of summer-in-Southern-Arizona kind of warm environment. Anywhere else, I wouldn't be so concerned.

I've tried to find a reliable source for the JD regulators and could not find one that didn't have reports online that the unit appeared to be of bad quality and did not stand the current for a loinger period of time. Lots of burnouts and such on all the models from AM101....6, over MIA8..... up to the latest AKH6004. Also I was not able to locate any kind of spec sheet that could tell me for what current rating these JD regs where designed. It clearly worked for N56ML but the pictures show an old AM101...6 model that used to be sold by JD and was apparently a much higher quality part than what you can get today. As I cannot find a supplier or a spec sheet I'll be staying away from this one for now.

There is also a US made option with series configuration, which I am currently investigating. Compu-fire is building, among other electrical things, Harley-Davidson replacement R/R units. The 55120 @ 32Amp looks very good to me.
I am trying to get wiring information from them for non-Harley installations. The unit is readily available and looks very, very promising to me.
 
Last edited:
You must log in or register to reply here.
Back
Top