Johny's Electric Vogue

Post up a thread for your EV. Progress pics, description and assorted alliteration
Post Reply
User avatar
acmotor
Senior Member
Posts: 3603
Joined: Thu, 26 Apr 2007, 03:30
Real Name: Tuarn
Location: Perth,Australia

Johny's Electric Vogue

Post by acmotor » Fri, 19 Apr 2013, 19:19

So if preload washers are both on the outsides then the rotor can move with say 100N force applied axially in either direction from the neutral position. Motor torque/VFD torque ripple could be moving this ?

A lever and dial gauge could be used to check this out.
Maybe a go-pro strapped on somewhere to view the action on a real drive ?

Mind you, the movement would probably exceed the bearing internal clearance so there would be polishing on the outer face of the bearings in their housings.

Yes, get the diff sorted as the backlash thump is an issue on its own.... but is it the source of vibration ?

You said you ran the vogue on blocks.
Was that with suspension at normal ride height not body jacked up ?

Just thinking out loud.

Image
iMiEV MY12     110,230km in pure Electric and loving it !

User avatar
acmotor
Senior Member
Posts: 3603
Joined: Thu, 26 Apr 2007, 03:30
Real Name: Tuarn
Location: Perth,Australia

Johny's Electric Vogue

Post by acmotor » Fri, 19 Apr 2013, 19:30

Johny wrote:
Image


Did you try shims at various positions around the UJ to coupler flange to check the alignment of the flange since spline unit will be alignment critical not just in runout but face alignment? A UJ can't correct drive face angle causing unbalance. Gentle bolt up and test.

and.... has the sliding sline been pulled apart and gone back in a different position negating any balance or alighnments done in the past ?

Sorry, still just thoughts and maybe I should have read your posts more closely. Image
iMiEV MY12     110,230km in pure Electric and loving it !

User avatar
Johny
Senior Member
Posts: 3729
Joined: Mon, 23 Jun 2008, 16:26
Real Name: John Wright
Location: Melbourne
Contact:

Johny's Electric Vogue

Post by Johny » Fri, 19 Apr 2013, 20:58

I ran a dial gauge on the flanges and on the tail shaft on the long shaft side of the UJs. The UJs are aligned correctly re the sliding spline.
The Vogue was run on axle stands under the rear solid axle housing, so normal ride height. The spline has about 15mm left in travel (almost all the way in) so it's not limiting travel.

After a trip home I got under the car and felt the UJs - almost cold so not binding (they are also new).

User avatar
acmotor
Senior Member
Posts: 3603
Joined: Thu, 26 Apr 2007, 03:30
Real Name: Tuarn
Location: Perth,Australia

Johny's Electric Vogue

Post by acmotor » Sat, 20 Apr 2013, 00:45

OK, just another thought flow....
Have you rotated the UJ body through 180 deg on the emotor flange at hex head cap screws ?
Have you removed (disengaged) the spline and refit also rotated through 180 deg ? (taking note of original orientation and assuming it doesn't have an alignment key)
In each case RPM and magnitude of vibration being noted.
Are the UJs at the same orientation at each end of the shaft ?
(only asking. I'm sure that you've been through UJs 101 !)
Have you tried reversing the shaft and putting the spline at the rear ?
(assuming here that the UJ flanges are the same both ends)
Still concerned that (a) vibration is in emotor to start with.
Image
iMiEV MY12     110,230km in pure Electric and loving it !

User avatar
Johny
Senior Member
Posts: 3729
Joined: Mon, 23 Jun 2008, 16:26
Real Name: John Wright
Location: Melbourne
Contact:

Johny's Electric Vogue

Post by Johny » Sat, 20 Apr 2013, 04:25

Tailshaft can't be reversed.
I put it back on 180 out at the front after rotor balanced.
Yes UJs are aligned.
Sent from phone.

User avatar
Richo
Senior Member
Posts: 3737
Joined: Mon, 16 Jun 2008, 00:19
Real Name: Richard
Location: Perth, WA

Johny's Electric Vogue

Post by Richo » Sat, 20 Apr 2013, 05:16

Since you seem to see the same vibration without the shaft connected I would look more at how the motor mounts to the car.
Perhaps a small vibration in the motor is becoming resonant with your motor frame.
The only test I can think of would be to put the car on stands then put the jack on the cross beams of the motor mount and apply some force there to see if this changes the vibration frequency.
So the short answer is NO but the long answer is YES.
Help prevent road rage - get outta my way!

User avatar
Johny
Senior Member
Posts: 3729
Joined: Mon, 23 Jun 2008, 16:26
Real Name: John Wright
Location: Melbourne
Contact:

Johny's Electric Vogue

Post by Johny » Sun, 21 Apr 2013, 02:09

Hmmm. That's a new take.on it. I'll definitely try that when its back up on stands. I won't get time until the 1st weekend in May but that'll be on the list. Thanks Richo.

User avatar
Johny
Senior Member
Posts: 3729
Joined: Mon, 23 Jun 2008, 16:26
Real Name: John Wright
Location: Melbourne
Contact:

Johny's Electric Vogue

Post by Johny » Mon, 22 Apr 2013, 17:57

acmotor wrote: So if preload washers are both on the outsides then the rotor can move with say 100N force applied axially in either direction from the neutral position. Motor torque/VFD torque ripple could be moving this ?
Just an update on motor torque ripple possibilities. A couple of times I ran the car down a hill with drive disabled and the vibration was just as bad. It's a totally passive condition.

User avatar
Johny
Senior Member
Posts: 3729
Joined: Mon, 23 Jun 2008, 16:26
Real Name: John Wright
Location: Melbourne
Contact:

Johny's Electric Vogue

Post by Johny » Mon, 22 Apr 2013, 17:59

BTW I'm really dying to test Richo's suggestion re motor mounting cradle resonance. I'm going to do the axle stand stuff this Thursday - I'll report back.

marty11
Groupie
Posts: 191
Joined: Sat, 28 Mar 2009, 03:26
Real Name: Marty
Location: Various Locatio

Johny's Electric Vogue

Post by marty11 » Mon, 22 Apr 2013, 18:28

re    " ran downhill with drive disabled"   

Could it be a buckled wheel? slightly bent front axle?

to test: car up on jacks, hold a bit of chalk against a rim. Spin rim and check if chalk is even, or there is a "bump" against the chalk whilst spinning,

If you have one wheel that is out of balance a lot, can also be a prob.

Apologies if these two points are already covered in your posts.. havent had time to read the whole thread. .am at work :)

User avatar
Richo
Senior Member
Posts: 3737
Joined: Mon, 16 Jun 2008, 00:19
Real Name: Richard
Location: Perth, WA

Johny's Electric Vogue

Post by Richo » Mon, 22 Apr 2013, 20:33

He did the test before with the wheels off and it was still there.
So is unlikely to be the wheels.
So the short answer is NO but the long answer is YES.
Help prevent road rage - get outta my way!

User avatar
Johny
Senior Member
Posts: 3729
Joined: Mon, 23 Jun 2008, 16:26
Real Name: John Wright
Location: Melbourne
Contact:

Johny's Electric Vogue

Post by Johny » Wed, 24 Apr 2013, 18:17

Regen braking. I did some calculations on last Thursday's log of the trip home (https://dl.dropboxusercontent.com/u/463 ... EV-TWO.TXT) to see what part regan played. Regen contribution was close to 14%.

           Avg        Drive    Regen
AH        4.55       5.279    -0.7263
Watt/Hour 2796       3242     -446

This was a a 36 minute trip. The trip had about 15 minutes of slow traffic in it (starting around line 1740) where I didn't touch the brakes until stopped (accident ahead).

Edit:
Shocking power economy at 165 Wh/km. Without regen it would have been 191 Wh/km.
The 20 minutes logged of the trip in was 143 Wh/km.
https://dl.dropboxusercontent.com/u/463 ... EV-0NE.TXT
A bad day overall.
BTW The sum of motor revolutions divided by 291180 is distance travelled in kilometers (don't ask why).
Last edited by Johny on Wed, 24 Apr 2013, 08:31, edited 1 time in total.

User avatar
Johny
Senior Member
Posts: 3729
Joined: Mon, 23 Jun 2008, 16:26
Real Name: John Wright
Location: Melbourne
Contact:

Johny's Electric Vogue

Post by Johny » Fri, 26 Apr 2013, 15:05

Richo wrote: Since you seem to see the same vibration without the shaft connected I would look more at how the motor mounts to the car.
Perhaps a small vibration in the motor is becoming resonant with your motor frame.
The only test I can think of would be to put the car on stands then put the jack on the cross beams of the motor mount and apply some force there to see if this changes the vibration frequency.
I had my heart set on this being the cause - alas no. I'd even measured up for motor-cycle steering dampers.
Placing a jack with a block of wood under the centre of the motor frame very marginally decreased the vibration is accord with what you would expect in that it offered a bit of dampening to the real cause.
The vibration is DEFINITELY coming from the front. It's exactly like the tailshaft wasn't balanced at all. I recorded the audio at around 60 km/h and I think the phone can record down to 40 Hz so I'm about to analyse that - I'll report back...
Last edited by Johny on Mon, 20 May 2013, 10:55, edited 1 time in total.

User avatar
Johny
Senior Member
Posts: 3729
Joined: Mon, 23 Jun 2008, 16:26
Real Name: John Wright
Location: Melbourne
Contact:

Johny's Electric Vogue

Post by Johny » Fri, 26 Apr 2013, 15:31

Here is a spectrum plot of the recorded audio with speed somewhere between 55 and 60 km/h.
Image
The lowest peak occurs at 37 Hz which is rotation frequency at 57 km/h - that fits. The next peak is at twice rotation frequency which suggests the universal joint as it rotates over the "top" on each side. The third and most prominent peak is at 118 Hz - which I can't explain - three times rotation frequency?
It still looks like interplay between the UJs and the motor/mounting.

Crawling around under the car travelling at 55-60 km/h (on stands) the vibration was almost non-existant at the diff end. The motor itself was the worst point. I could not visually see any runout consistant with this behaviour. I've had some really bad UJs in tailshafts in my life that didn't vibrate as badly as this apparently OK one.

Precision Balancing have a portable setup for analysing vibration that they use in the field - it may soon come to that.

I'm also thinking about CVs....

Edit: Changed to 37 Hz
Last edited by Johny on Fri, 26 Apr 2013, 18:26, edited 1 time in total.

User avatar
Richo
Senior Member
Posts: 3737
Joined: Mon, 16 Jun 2008, 00:19
Real Name: Richard
Location: Perth, WA

Johny's Electric Vogue

Post by Richo » Fri, 26 Apr 2013, 20:55

Yeah the window and phone roll off probably suggests the first is ~38Hz.

Another possible test:
Disconnect the motor mount from the chassis but just lower it a few mm away from the chassis.
Have the car on stands and run the car see if the problem is still there.
Just be careful not to give it lots of torque since the motor is loose.
Perhaps a slight change in angle will show the problem.

What about the coupler itself is that secured well?
So the short answer is NO but the long answer is YES.
Help prevent road rage - get outta my way!

User avatar
acmotor
Senior Member
Posts: 3603
Joined: Thu, 26 Apr 2007, 03:30
Real Name: Tuarn
Location: Perth,Australia

Johny's Electric Vogue

Post by acmotor » Fri, 26 Apr 2013, 21:16

Just trying to read the messages in your FFT graph....

BTW, I find it helps to use linear frequency rather than log frequency on the x-axis when chasing vibrations. Harmonics are easier to read then.

Help me here... the second peak is at 76Hz -33.8dB, there is lower one at half that frequency.
Wouldn't your emotor be doing 2280RPM at 57kmph ? or have I got that wrong ? That means ~38Hz

The third peak is at ~118Hz which is the sum of the first two frequencies and is the 3rd harmonic of the first peak. These being the typical vibrations of a shaft with fixed nodes at each end. Even the next peaks 4th (~154Hz) and larger 5th (~195Hz) follow.

I did think of 3 phase torque ripple just only to try to come up with a x3 ratio.Image forget that.

Given that these are UJs, this comes back to the plane of the coupling faces at each end of the shaft not lining up. i.e. runout is ok, you've checked that, but are the faces parallel ?
Image

CVs may overcome this problem and a sliding CV replaces the need for the sliding spline you have at present. Need to reseach that further.

"Crawling around under the car travelling at 55-60 km/h" had me respecting your persistance until I read the content of the brackets.   Image
iMiEV MY12     110,230km in pure Electric and loving it !

User avatar
acmotor
Senior Member
Posts: 3603
Joined: Thu, 26 Apr 2007, 03:30
Real Name: Tuarn
Location: Perth,Australia

Johny's Electric Vogue

Post by acmotor » Fri, 26 Apr 2013, 22:41

Another thought is that given the speed ripple of the UJ system and the zero backlash of the emotor end to the rotor inertia, is there an issue with using a UJ in this arrangement ?

Image
from
http://commons.wikimedia.org/wiki/File: ... _speed.png

CV joints would be required at both ends of the tail shaft ?
iMiEV MY12     110,230km in pure Electric and loving it !

User avatar
Johny
Senior Member
Posts: 3729
Joined: Mon, 23 Jun 2008, 16:26
Real Name: John Wright
Location: Melbourne
Contact:

Johny's Electric Vogue

Post by Johny » Sat, 27 Apr 2013, 04:28

Richo wrote: Yeah the window and phone roll off probably suggests the first is ~38Hz.
My mistake - true. I was thinking about the 78 Hz peak when I wrote it.
Another possible test:
Disconnect the motor mount from the chassis but just lower it a few mm away from the chassis.
Have the car on stands and run the car see if the problem is still there.
Just be careful not to give it lots of torque since the motor is loose.
Perhaps a slight change in angle will show the problem.

What about the coupler itself is that secured well?
Yes, coupler is held on with M12 in the end of the motor shaft tightened to about 70 NM and two grub screws holding key.
Last edited by Johny on Fri, 26 Apr 2013, 18:28, edited 1 time in total.

User avatar
Johny
Senior Member
Posts: 3729
Joined: Mon, 23 Jun 2008, 16:26
Real Name: John Wright
Location: Melbourne
Contact:

Johny's Electric Vogue

Post by Johny » Sat, 27 Apr 2013, 04:34

Thanks for all the responses guys. I have been looking at CVs to replace the UJs. The other thing is something called 1/2 critical rotation speed. What it boils down to is that even if a tailshaft has a critical speed of say 5000 RPM, half that speed can be excited by UJs causing excessive vibration.
The guys who made the tailshaft are looking into the tube size etc. now.
I had them make it out of 2.5 inch when the original was 3 inch (trying to keep weight down). Perhaps a mistake in retrospect but even so there would have been a problem.

I'm pretty convinced that the UJs are responsible. Now it's how to fix it without outlaying a mint.

acmotor, I have looked at so many websites on driveshaft vibration that the diagram you put up is in my aftervision now.
Last edited by Johny on Wed, 08 May 2013, 06:15, edited 1 time in total.

User avatar
Johny
Senior Member
Posts: 3729
Joined: Mon, 23 Jun 2008, 16:26
Real Name: John Wright
Location: Melbourne
Contact:

Johny's Electric Vogue

Post by Johny » Sat, 27 Apr 2013, 04:38

acmotor wrote:Help me here... the second peak is at 76Hz -33.8dB, there is lower one at half that frequency.
Wouldn't your emotor be doing 2280RPM at 57kmph ? or have I got that wrong ? That means ~38Hz
Yes. Speed in km/hr times 40 = RPM.
The third peak is at ~118Hz which is the sum of the first two frequencies and is the 3rd harmonic of the first peak. These being the typical vibrations of a shaft with fixed nodes at each end. Even the next peaks 4th (~154Hz) and larger 5th (~195Hz) follow.

I did think of 3 phase torque ripple just only to try to come up with a x3 ratio.Image forget that.

Given that these are UJs, this comes back to the plane of the coupling faces at each end of the shaft not lining up. i.e. runout is ok, you've checked that, but are the faces parallel ?
Not quite. The diff flange faces down by 2.5 to 3 degrees.
CVs may overcome this problem and a sliding CV replaces the need for the sliding spline you have at present. Need to reseach that further.
Agreed. Centre CVs are common in split drive shaft at the centre but I haven't seen one at each end. I may be the first.

User avatar
Johny
Senior Member
Posts: 3729
Joined: Mon, 23 Jun 2008, 16:26
Real Name: John Wright
Location: Melbourne
Contact:

Johny's Electric Vogue

Post by Johny » Sat, 27 Apr 2013, 04:49

Extra info - may be nothing but the UJs at each end look like they are different sizes (I think). I'll have to check but the front one looks larger than the rebuilt original at the rear.

Wellsey
Groupie
Posts: 43
Joined: Wed, 30 Jul 2008, 07:29
Real Name: Brad Wells
Location: Canberra

Johny's Electric Vogue

Post by Wellsey » Sat, 27 Apr 2013, 08:10

The graph posted above indicates the differing speed of the output shaft given a constant speed of the input shaft. But why? Here is my crude way of explaining it AND why the faces of the universal joints MUST be parallel.

As we know, time x distance = speed. and because we can't change time, and the distances are geometrically fixed for a given angle, speed must change.

The drawing shows the flat face of the input shaft as a circle (path of the fixed uni joint) and the ellipse represents the path taken by the output shaft.

All this is ONE universal joint....When you have a second universal joint at the other end of your tail shaft, the varied speeds transmitted through an EQUAL ellipse provides a final (and constant) speed as the first input shaft.

Having the second universal joint at a different angle (not parallel) changes the final output speeds slightly and causes vibrations.

v=xgQgm3GwaFsImage

The answer to you problem (if you need to keep the universal face planes out of parallel) may be the Thompson Coupling. (Or adjust the geometry.)

I would have thought that the guys that did your tail shaft would have know of this type of issue.

http://www.youtube.com/watch?

Interesting video. The three minute mark explains (to some degree) the different speed caused by the geometry.

Wellsey
Groupie
Posts: 43
Joined: Wed, 30 Jul 2008, 07:29
Real Name: Brad Wells
Location: Canberra

Johny's Electric Vogue

Post by Wellsey » Sat, 27 Apr 2013, 08:14

Oops, sorry, Here is the Youtube clip:

http://www.youtube.com/watch?v=xgQgm3GwaFs

User avatar
Johny
Senior Member
Posts: 3729
Joined: Mon, 23 Jun 2008, 16:26
Real Name: John Wright
Location: Melbourne
Contact:

Johny's Electric Vogue

Post by Johny » Sat, 27 Apr 2013, 14:06

Thanks Wesley.
Yes they do know about it (as do I as you can see in earlier posts) and considered up to 3 degrees as fine (they race speedway so of course could have a perverted idea of what is "fine").
One of the early checks I did was to adjust the diff pinion angle to align the faces and it looks like I should revisit this.
Thomson coupling are not available for mere mortals. Also not really needed as CVs would suffice.

Even when faces align the shaft accelerates and deaccelerates which may be the issue.
Last edited by Johny on Sat, 27 Apr 2013, 04:59, edited 1 time in total.

User avatar
Johny
Senior Member
Posts: 3729
Joined: Mon, 23 Jun 2008, 16:26
Real Name: John Wright
Location: Melbourne
Contact:

Johny's Electric Vogue

Post by Johny » Sat, 27 Apr 2013, 14:56

Link to differential pinion angle fiddle:
viewtopic.php?title=johnys-electric-vog ... 707#p41829

I was convinced that this was the issue since the original car (as ICE powered) exibited the same kind of problem. At the time I was surprised that it made no difference so, since it's an easy one to try again, I will. Since then I have read that over 1 degree face difference can lead to 1/2 critical speed tailshaft problems.
Last edited by Johny on Sat, 27 Apr 2013, 05:02, edited 1 time in total.

Post Reply