F800GS R/R Mosfet replacement....How to...

Discussion in 'Parallel Universe' started by ebrabaek, Apr 13, 2012.

  1. 14TLC

    14TLC Adventurer

    Joined:
    Jun 19, 2009
    Oddometer:
    49
    Location:
    Out Of Space
    [FONT=&quot]Hi everyone. I’m a new proud owner of 2012 F800GS and I’m closely watching all threads regarding F800GS issues in order to be well prepared. But this stator failure makes me very concerned. I do not ride as much as many of you and for sure mine will be out of warranty when it reaches the threshold mileage after which a stator failure is expected.[/FONT]
    [FONT=&quot]
    [/FONT]
    [FONT=&quot]While searching for solution I came across one small company in Croatia ([/FONT]www.sper.hr[FONT=&quot]) which makes all sorts of electronic devices and among all other things they make R/Rs for motorcycles and boats. I‘ve contacted them and described to them the problem we have with the stator and concerns about using a serial type of R/R. I will translate in English what their engineer answered and what is stated on their web page. Unfortunately the web page is only in Croatian.[/FONT]
    [FONT=&quot]
    [/FONT]
    [FONT=&quot]“Using serial type of R/R should not cause an issue with other electronic devices on the motorcycle. What about the ignition system and ignition coils? They produce by far higher impulses and noise in the electrical system. However the serial type of R/R suffers from another issue. All serial types of R/Rs which have thyristors and diodes in the design do not operate well on higher revs. The thing is that the inductive component of the generator’s impedance gets higher, and higher amount of energy is produced in shorter time intervals. In this condition the serial R/R cannot properly discharge the energy and then constant DC current will occur from two phases towards the third phase of the generator. This usually occurs above 8000 rpm. The solution is hybrid type of R/R which has both serial and parallel type of regulation. Most of the time it operates in serial mode and when it senses the critical voltage it will switch to parallel mode. It also has voltage protection in order to limit the voltage at 14.6 V”[/FONT]

    [FONT=&quot]So, they produce parallel, serial and also hybrid type of R/R. As I can see on their web page the hybrid one is their best selling R/R and so far they never had any issues or faulty R/Rs of this type. They state that the components which they use are high quality components and are rated for 40-60% higher current than the rated current of the R/R which is 30A. The thing is that they use large heat sinks, usually larger than most R/Rs on the market, especially on their hybrid one. But they offer option for the hybrid one to be built in two housings, one for the serial and other for the parallel one. You can also buy the serial one and if you notice charging issues later you can add the parallel one to complete the hybrid R/R. This is the model they recommended for the F800GS and the list of motorcycles on which they fitted this R/R[/FONT]
    [FONT=&quot]
    [/FONT]
    http://www.regler.sper.hr/sp_trofazni_regler.htm

    [FONT=&quot]Unfortunately they don’t sell outside Croatia.[/FONT]
    [FONT=&quot]
    [/FONT]
    [FONT=&quot]On their web page they have a lot of practical advices and information regarding R/Rs, charging systems, generators and examples how to test the R/R, generator and also measuring insulation of the generator. I will translate the most interesting topics and I will post them here.[/FONT]
    [FONT=&quot]
    [/FONT]
    [FONT=&quot]P.S. Here is one interesting battery monitoring gadget:[/FONT]
    [FONT=&quot]
    [/FONT]
    http://www.argusanalyzers.com/batte...tery-bug-battery-monitor-argus-analyzers.html
  2. Indy Unlimited

    Indy Unlimited Long timer

    Joined:
    Jan 21, 2007
    Oddometer:
    1,604
    Location:
    Parker, CO
    So my head is now splitting with all this great information and I am back to thinking we need to cool the stator or reduce the wattage with a rewind?:tough
  3. Mike.C

    Mike.C Stelvio Dreamer!

    Joined:
    Dec 31, 2007
    Oddometer:
    594
    Location:
    Brisbane-Australia
    Yeah me too, but my head split ages ago trying to understand this topic....

    I am going to do something soon as the launch date for my next big adventure is July 28th and count down has commenced, just not sure what exactly. IIRC Joel was talking to some R/R manufacturers about a potential solution.
  4. ebrabaek

    ebrabaek Long timer

    Joined:
    Jan 12, 2010
    Oddometer:
    5,426
    Location:
    Grand Valley, Colorado
    So far....this solution is working great. I have been monitoring voltage....battery starting....etc....and all is well. I did speak with the importer of this R/R....and he'd mention that these were on back order...... But that perhaps another source from a well know US bike supplier could become an option. If there is enough interest....I could ask...If there is enough interest, as there seems to be little if none of these R/R's available. I do feel though that regarding Stator lonlegivity ...or should we say...prolonging it.... Only time will tell. Unless you mount a probe inside the case...on the Stator itself. I have no doubt the charging issue is resolved....and the more I look into the series R/R.... I am not willing to take the chance, as there appears to be issues. My aluminum bracket will be here on monday....but as cool as it runs now.... it will only get better......:thumb:thumb

    Erling
  5. Mike.C

    Mike.C Stelvio Dreamer!

    Joined:
    Dec 31, 2007
    Oddometer:
    594
    Location:
    Brisbane-Australia
    Hey Mr Beak, your too popular! Your pic sharing site has blocked em because so many users like em... go figure:eek1
  6. WayneC

    WayneC Long timer

    Joined:
    Oct 3, 2008
    Oddometer:
    3,523
    Location:
    Sydney, Australia
    Thanks for posting the info re the Croatian Co, very good info, really appreciated, we have some F650 single owners in that part of the world so have relayed the info over to F650.com as there are VR probs with the single & will see if they are
    able to go further with Croatia

    On the F800 it is clear from the brilliant work done by Joel, ebrabaek and others that the primary problem is lack of cooling to the stator & that wont be resolved without a design change

    The work around if you like is to try to reduce stator temps by a VR change, Increasing the VR regulation voltage using the Shindengen FH12 shunt regulator provides a limited but useful reduction in load on the stator. Going to the series pass VR further reduces the load & hence temperature of the stator

    All I can say is I am glad I am only looking for a replacement VR for the single to overcome our limited problems
  7. ebrabaek

    ebrabaek Long timer

    Joined:
    Jan 12, 2010
    Oddometer:
    5,426
    Location:
    Grand Valley, Colorado
    I know... I'm sorry..., It will reset tomorrow....and all will be well...... I have had some issues with Photo bucket last time they revamped their site... I lost hundreds of links.... so I stopped paying....and sometimes when projects fly off the wall.....I bust their buffer.....:D:lol3... All will be visible...tomorrow.....:thumb:thumb

    Erling
  8. cathulu

    cathulu Been here awhile

    Joined:
    Jan 4, 2010
    Oddometer:
    179
    I suspect your heat sink mount is fine, the mosfet has a low Rds so the i^2r losses are lower than the stock regulator... Your anecdotal observations support that. Joel did not make a distinction between the two.

    This is a very interesting thread along with all the others on the stator issue.

    Sent from my GT-I9100M using Tapatalk 2
  9. ebrabaek

    ebrabaek Long timer

    Joined:
    Jan 12, 2010
    Oddometer:
    5,426
    Location:
    Grand Valley, Colorado
    Yeppers...... What will be interesting is the temperature before and after the Aluminum bracket.... My gut feeling is that this mosfet R/R unit runs so cool to start with... It does not need the frame cooling.... I was stumpted as of how hot the stock R/R runs.... Really hot...:thumb:thumb

    Erling
  10. JRWooden

    JRWooden Long timer

    Joined:
    Jul 5, 2008
    Oddometer:
    4,643
    Location:
    The great state of confusion
    If the i^2 * R losses are less in the R/R ... might that be actually be a bad thing?

    This MOSFET R/R is still a shunt design right?
    So if the losses are less in the R/R and the resistance of the stator is unchanged,
    then total circuit resistance is lower and the max. current would be higher ... thus creating MORE i^2 * R heating
    in the stator due to the higher current....

    Or am I AFU here?
  11. Gangplank

    Gangplank Advenchaintourer

    Joined:
    Jul 20, 2003
    Oddometer:
    2,316
    Location:
    Las Vegas
    Isn't stator only one way? So higher Max output of RR Means less draw on stator = less heat.
  12. JRWooden

    JRWooden Long timer

    Joined:
    Jul 5, 2008
    Oddometer:
    4,643
    Location:
    The great state of confusion
    I don't think it's one way but I wouldn't bet more than a 12-pack on it ... :lol3

    I though the stator was basically just coils of wire tied together in a "Delta" configuration.

    Schematically like this picture: http://www.teal.com/products/App%20note%20AN-15.htm
  13. ebrabaek

    ebrabaek Long timer

    Joined:
    Jan 12, 2010
    Oddometer:
    5,426
    Location:
    Grand Valley, Colorado
    That's what I understand...... If the R/R has to regulate the voltage down to 13.8 volts by the way of a fancy short....there is more heat dissipated....than if it only regulates down to 14.4......:thumb:thumb

    Erling
  14. ebrabaek

    ebrabaek Long timer

    Joined:
    Jan 12, 2010
    Oddometer:
    5,426
    Location:
    Grand Valley, Colorado
    Your correct....:thumb:thumb

    Erling
  15. JRWooden

    JRWooden Long timer

    Joined:
    Jul 5, 2008
    Oddometer:
    4,643
    Location:
    The great state of confusion
    Maybe I don't understand what you mean when you say "one-way" ...

    I do agree that regulating at a higher voltage is GOOD
    (more voltage means less current ... so good),

    but beyond that it seems like in a shunt design system, with our stators especially,
    INefficient R/R would be better ...
    i.e. a higher resistance path to ground in the R/R would be an advantage in that it would help limit the current flow thru the stator coils when less than full output is required.
  16. ebrabaek

    ebrabaek Long timer

    Joined:
    Jan 12, 2010
    Oddometer:
    5,426
    Location:
    Grand Valley, Colorado
    Perhaps " one way" is not a good Phrase..... as it really isn't. As you mentioned it is still a shunt....but effectively does a better job shunting the excess power, than the stock R/R. The red lettering is what I'd like to focus on, and it appears that heat dissipated within the mosfet R/R....is less than the stock unit. That does not necessarily mean that more heat is now present in the stator. But someone with more knowledge than me...could perhaps chime in as to the scientific explanation here....:thumb:thumb

    Erling
  17. JRWooden

    JRWooden Long timer

    Joined:
    Jul 5, 2008
    Oddometer:
    4,643
    Location:
    The great state of confusion
    I am certain the higher voltage is a good thing all around ... the battery is happier, and the stator is happier... win-win.

    Without a real set of test cases being run, in which thermocouples are installed in the stator and the various configurations tested... I don't know ........... but thinking about it ... it almost seems like the -SC version of the R/R might be the "local optimum" ............
  18. 14TLC

    14TLC Adventurer

    Joined:
    Jun 19, 2009
    Oddometer:
    49
    Location:
    Out Of Space
    Glad I could help. For the F650 single they recommend their serial type R/R, model 14V-3F-15A-A. It costs about 45 eur. I don't know for which production year of F650 is applicable but one can always consult with them when baying one.
  19. ebrabaek

    ebrabaek Long timer

    Joined:
    Jan 12, 2010
    Oddometer:
    5,426
    Location:
    Grand Valley, Colorado
    I don't think so..... As the higher voltage should ease the load on the stator.....And...just because the R/R dont get as hot...... does not mean that more heat is going to the stator. But as you duely noticed....No body will know until a lead is placed on the stator. We know the old on eventually will eventually kill the stator..... we also know is kills the battery....fairly fast as well..... This one will/should not kill the battery....so at least one achievement here.....Perhaps two...:thumb:thumb

    Erling
  20. ebrabaek

    ebrabaek Long timer

    Joined:
    Jan 12, 2010
    Oddometer:
    5,426
    Location:
    Grand Valley, Colorado
    Let me know if there is any interest in a group buy....... I could ask.......as I keep testing the unit.....:thumb:thumb

    Erling