ADVrider

Go Back   ADVrider > Bikes > GSpot > Parallel Universe
User Name
Password
Register Inmates Photos Site Rules Search Today's Posts Mark Forums Read

Reply
 
Thread Tools Search this Thread Display Modes
Old 05-28-2013, 07:34 AM   #1
JRWooden OP
Beastly Adventurer
 
Joined: Jul 2008
Location: The great state of confusion
Oddometer: 4,207
Oscilliscope Traces - OEM & Series style Regulator R/R project

I figured I'd start a new thread with the results of my project to document the differences in system voltages between the OEM (shunt-style) R/R and a aftermarket series style R/R.

For the series style unit I choose the Cycle Electrics CE 605 SB which I purchased from Jack here:

http://roadstercycle.com/Roadstercyc...Regulators.htm

Jack sells a true "kit" it has a self-resetting circuit breaker, a variety of crimp connectors, and heat-shrink tubing. For those not familiar with electrical work he has some videos on proper techniques on his website.

To allow side-by-side testing I soldered a donor set of connectors from a dead OEM R/R onto the new R/R (thanks again to imintman for the dead unit)

Ok ... on to the testing....
First we tested the OEM R/R. It took us about 10 minutes to get set up and capture our readings ... and with only 10 minutes running time (most at idle) the OEM unit too hot to touch and we had to let it cool down in order to swap it out.

Note: All of these traces were collected with the probes connected directly to the bike's battery.


Here's what we found with the OEM unit:

Trace #1:



Sorry this is a bit hard to read, but what you see here are random noise spikes that the OEM R/R is tossing into the system they are roughly +0.5v to -1.0V (scope set @ 2V/division & 1ms sweep)

Trace #2



The above is also from the OEM R/R
Note the expanded vertical & faster sweep times -
1V/div and 200ns sweep....
As you can see there are several different things going on here with the OEM unit.... An underdamped "ringing" and then a couple of slower undulating waveforms....

Trace #3




Trace #3 is the best picture I captured of the Cycle Electrics unit.
First ... there is no comparison trace to "Trace #1" above ... from the OEM unit ... the CE unit does not toss off the spikes that are shown in trace #1

It does "ring" in a similar fashion to the OEM unit, but not that the two "undulating" waveforms are missing.

After 10 minutes or so of testing the CE unit it was barely above room temperature at the end of our testing ...

My EE buddy is not surprised by the results, being the hyper-brilliant EE that he is, he pretty much expected it would turn out as it did...

I make an adapter plate to get the thing permanently bolted to the bike ... this is a picture of it in process:
Fasteners are all M6 x 1.0 x 20mm - four cap screws to mount R/R to plate and then two FH machine screws to mount the plate to the OEM bracket. I could have made a new bracket instead of using the adapter, but I had a chunk of aluminum sitting there and figured this way the project would be 100% reversible if needed.



While the CE unit is somewhat larger than the OEM unit I was able to mount it in the stock position without much drama.

Open items:

1) I still need to to do some ABS testing (at various levels of battery state-of-charge as Joel suggested) ...

2) I stupidly forgot to capture EMI readings with the oscilliscope probes in "free air" close to the R/R. I may try to go back and do that ... The heatsink on the CE unit is massive, and it is grounded. With my "almost" 100% coverage adapter plate I would like to think that I've managed to capture most of any EMI that would be generated, but then I'm not an expert at EMI type stuff....

3) dbh had suggested capturing some readings from the stator side of the circuit with a current probe... alas... we don't have one....

So anyway... that's what I've got so far...

JRWooden screwed with this post 05-28-2013 at 07:40 AM
JRWooden is offline   Reply With Quote
Old 05-28-2013, 08:24 AM   #2
ebrabaek
Beastly Adventurer
 
Joined: Jan 2010
Location: Grand Valley, Colorado
Oddometer: 4,677
Bravo....... I'l be your first groupie.....
__________________


Erling
ebrabaek is offline   Reply With Quote
Old 05-28-2013, 08:26 AM   #3
dbh
d00d
 
dbh's Avatar
 
Joined: Jun 2009
Location: Northern Alberta, Canada
Oddometer: 48
Very interesting, thanks for this.
Nice captures of the detail in those pulses!

The ringing pulses from the CE unit have a freq of about 7.5MHz and the OEM unit has some even higher frequency components plus the oscillation you mentioned which has a freq about 625kHz i.e. much more complex a/c information. Were these at idle, about 1500rpm or so?

Good measurements answer questions. Good science raises more questions.
Now I'm wondering about the origin of these signals.
So good job.

The amount of a/c in either case is not going to put significant current through the battery so no heating issues there. If one of the phases should have a winding burn-out/short-circuit/break I wonder how much of the increased ripple would be passed on by each r/r - that could shorten a battery's life.

Planning any more tests?
dbh is offline   Reply With Quote
Old 05-28-2013, 08:32 AM   #4
WayneC1
Beastly Adventurer
 
Joined: Oct 2008
Location: Sydney, Australia
Oddometer: 2,354
Thanks for the info, it is appreciated, re the EMI on the CE unit, with the heat sink earthed it should be virtually nil
__________________
650GS Info - www.f650gs.crossroadz.com.au
WayneC1 is offline   Reply With Quote
Old 05-28-2013, 09:57 AM   #5
JRWooden OP
Beastly Adventurer
 
Joined: Jul 2008
Location: The great state of confusion
Oddometer: 4,207
Thanks for the kind words guys....

dbh:
Yes, all of the pictures were taken at idle (~1200 rpm), we did rev up the bike with both R/R installed and while the frequency changed, the magnitude of the noise/spikes did not change appreciably so we dis most of our dinking around at idle speed...

Wayne:
Thanks! I was hoping that between the heatsink covering 5-sides of the unit and my adapter plate covering 90% of the 6th side and all tied to earth that even if the series R/R did generate more EMI it would be confined.

ABS testing this afternoon!

Oh... and one other plus.... I think the baseline load on the charging system is ~100W so the series regulator "frees up" 300W or so of horsepower that would otherwise just get wasted and turned into heat in the stator & the R/R so I get about 0.4 HP extra to use for other things ...

Given my harness setup it's not hard to swap R/R so I'm open to suggestions on extra measurements that could be made, although I may not get around to setting them up until fall ... lots of riding to do this summer!

I wish I didn't have 30K miles of "cooking time" on the OEM stator
JRWooden is offline   Reply With Quote
Old 05-28-2013, 11:36 AM   #6
TowPro
Getting there
 
Joined: Aug 2009
Location: Compass PA
Oddometer: 3,018
What is the difference in voltage measured in the input of the New vrs Old regulator? This would be the voltage seen on the windings.

Interested to see the differance on a fully charged battery with all ACC's turned off.
__________________
2014 R1200GS - My Naked Street Touring Bike
2000 KTM LC4 640 E - For everything else
TowPro is offline   Reply With Quote
Old 05-28-2013, 02:55 PM   #7
JRWooden OP
Beastly Adventurer
 
Joined: Jul 2008
Location: The great state of confusion
Oddometer: 4,207
That's a good question ....
I don't know but I'll measure it ...

I should have thought a bit harder about my "test protocol" before riding over to my buddy's house but I was too damned excited about being almost done with this project.

I have a Fluke 77 multi-meter handy here at my house... but not an Oscope ....

If I go across two of the stator wires or from one of them to ground,
let the battery discharge a bit so that the R/R will allow "full output" for a while before starting to actually "regulate" I might be able to capture the (lower) voltages during use and then the higher voltages that would occur at "interrupt" ???

Is the Fluke 77 "fast enough" for this type work or is it time to bring my EE buddy a beer?
JRWooden is offline   Reply With Quote
Old 05-29-2013, 05:48 AM   #8
gstreff
Adventurer
 
gstreff's Avatar
 
Joined: May 2010
Location: Colorado Springs Colorado
Oddometer: 13
ElectroSport on stator and regulator

Here is an interesting reply to an email inquiry to ElectroSport about their redesigned stator for the F8 and the regulator to go with it.

Hello,The part does have some design changes. The output is on par with the OEM unit, but it does run cooler for a couple of reasons. The main reason is the quality of the copper wire used on this one stator is extremely pure copper. Most of our stators we sell for around $150, this one is a bit more expensive at about $200 because of the extra material cost of the copper. You really can not find more pure copper for a winding application like this. The result is a lower internal resistance and less heat produced. Also, the winding configuration is tweaked a bit to produce more voltage and a bit less amperage, which helps reduce heat as well. Along with some higher temp insulation, the part ends up dealing with the heat it does produce more effectively than the OEM version.



The OEM reg/rect would be perfectly suited for this bike. No need to try the goofy series/switching reg/rect, our stator will fix the overheating issue at its source.


Kyle Wood - Sales Manager
-------------------------------------
Procom Engineering / ElectroSport

21 Brookline, Aliso Viejo, CA 92656
PH: (949) 305-4200
www.electrosport.com
__________________
09 F800GS
gstreff is offline   Reply With Quote
Old 05-29-2013, 06:12 AM   #9
WayneC1
Beastly Adventurer
 
Joined: Oct 2008
Location: Sydney, Australia
Oddometer: 2,354
Well the first part is well written and appealing, the second part unfortunately is a real shot to the foot on their part as there is nothing goofy about the move to either more efficient FET regulators or series pass regulators to reduce engine loads it is the way of the future

Even the cars are moving to reduce electrical generation loads with improved systems and BMW is claiming some impressive power and mpg results
__________________
650GS Info - www.f650gs.crossroadz.com.au
WayneC1 is offline   Reply With Quote
Old 05-29-2013, 07:06 AM   #10
JRWooden OP
Beastly Adventurer
 
Joined: Jul 2008
Location: The great state of confusion
Oddometer: 4,207
I've had a long running e-mail exchange with Kyle ... which in part prompted his company to come up with the improved design ... the first part of his e-mail to you....

Last night it occurred to me that I had never asked him his opinion of how using a series style R/R might impact stator life ... it is possible that, depending on the design, higher voltages could be induced into the stator when the coils are "disconnected" ....
My EE buddy thought that unlikely but alas we collected no data on that...

Anyway, I sent Kyle an e-mail this morning asking his opinion of series style R/R ....
JRWooden is offline   Reply With Quote
Old 05-29-2013, 07:22 AM   #11
WayneC1
Beastly Adventurer
 
Joined: Oct 2008
Location: Sydney, Australia
Oddometer: 2,354
Yes in any open circuit stator winding the unloaded voltage generated will be higher as there is no circuit to "pull" the voltage down, this is normal in any electrical circuit, there may be some design considerations for them to take into account but there is no problem in terms of stator heat or damage. There are normally figures in the BMW maintenance documentation to guide what the open circuit voltages should be.

His use of the term goofy is totally unprofessional to say the least but we all have foot in mouth at some point and at least he did not put in writing, as a UK BMW Service Manager did to a UK customer, that I rode a damaged machine and overloaded my machine to cause the fork lug failure, still waiting for an official apology on that one but I am not holding my breath
__________________
650GS Info - www.f650gs.crossroadz.com.au

WayneC1 screwed with this post 05-29-2013 at 07:27 AM
WayneC1 is offline   Reply With Quote
Old 05-29-2013, 08:59 AM   #12
TowPro
Getting there
 
Joined: Aug 2009
Location: Compass PA
Oddometer: 3,018
Quote:
Originally Posted by WayneC1 View Post
Yes in any open circuit stator winding the unloaded voltage generated will be higher as there is no circuit to "pull" the voltage down, this is normal in any electrical circuit, there may be some design considerations for them to take into account but there is no problem in terms of stator heat or damage. There are normally figures in the BMW maintenance documentation to guide what the open circuit voltages should be.
I would expect to see higher voltages on the stater running the FET VR when the system is fully charged, but there still is a couple amp load all the time with the lights, ECM, fuel pump running.. As long as the stater (and the input to the FET VR) can handle those higher voltages everything should be ok.
__________________
2014 R1200GS - My Naked Street Touring Bike
2000 KTM LC4 640 E - For everything else
TowPro is offline   Reply With Quote
Old 05-29-2013, 11:04 PM   #13
JRWooden OP
Beastly Adventurer
 
Joined: Jul 2008
Location: The great state of confusion
Oddometer: 4,207
I do wonder about the dielectric value of the stator windings ....

My EE buddy speculates that the R/R "could" be designed in such a way as to unload near the zero-crossing and thus not have much impact ....

Have not had a chance to do ABS testing, but will do that soon.

Will then ride "as-is"....

If stator cooks I will use an Electrosport replacement, as I do believe Kyle on the low-resistivity wire
JRWooden is offline   Reply With Quote
Old 06-02-2013, 07:52 AM   #14
WayneC1
Beastly Adventurer
 
Joined: Oct 2008
Location: Sydney, Australia
Oddometer: 2,354
JR, can you post full dimensions for the CE605 VR, A nice pic of it off the bike would be good as well if possible
__________________
650GS Info - www.f650gs.crossroadz.com.au
WayneC1 is offline   Reply With Quote
Old 06-02-2013, 08:16 AM   #15
JRWooden OP
Beastly Adventurer
 
Joined: Jul 2008
Location: The great state of confusion
Oddometer: 4,207
Here it is installed:




For comparison here's one of the stock R/R (taken right after the hose blew off my radiator :-/)



and here's a side-by-side ...
The CE unit is larger but roughly proportionally larger in both dimensions,
while I think the compufire is more rectangular (less square?),
have not had a chance to hold the CompuFire in my hand ...
It might be easier or harder to mount ... not sure.




It does stick out farther than the stock unit as the heatsink its self is taller ...
and then I have a 1/4" adapter plate behind the R/R mounted to the OEM bracket...
but even riding around in 85 degree weather urban driving the CE unit is only slightly warm to touch,
while the old OEM unit was to hot to hold on to....

Maybe some year when the dust settles I'll go back and do a more "professional" mounting job .............
Wait... WTF am I saying ... that's never happening......

JRWooden screwed with this post 06-02-2013 at 08:43 AM
JRWooden is offline   Reply With Quote
Reply

Share

Thread Tools Search this Thread
Search this Thread:

.
Display Modes

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is On

Forum Jump


Times are GMT -7.   It's 01:03 AM.


Powered by vBulletin® Version 3.8.5
Copyright ©2000 - 2014, Jelsoft Enterprises Ltd.
Copyright ADVrider 2011-2014