Re-Torque Engine Mount Bolts

The engine is, of course, the primary source of vibration on the bike. Over time or due to factory variations the engine can get slightly out of alignment on the mounts and generate additional vibrations. Sometimes re-torquing the bolts that secure the engine to the mounts can bring it back into alignment and reduce vibrations. The procedure is to loosen the bolts, run the engine briefly, then re-torque the bolts. There are four main mount bolts, left and right front, upper rear, and lower rear.

Here are some tools that you will need. Not shown are sockets or allen wrenches to remove the frame attachment bolts if slider brackets are in the way. If you do not have slider brackets then loosening those four frame mount bolts is optional and may or may not provide any additional benefit. Also not shown is the torque wrench.

1600 x 1200

Place Bike on Centerstand.

Loosen the front mount bolts first. I do have sliders on brackets that impede access to the front mount bolts. The bolts that attach the slider brackets also attach the front engine mount bracket to the frame. I loosened the top bolt, removed the bottom bolt, swiveled the slider bracket out of the way, then replaced the bottom bolt temporarily. Do this procedure completely on one side then the other.

1600 x 1200 1600 x 1200 1600 x 1200 1600 x 1200

This shows the two front mount bolts. Loosen them so that they are free to move just slightly in place.

1600 x 1200 1600 x 1200

Next, loosen the rear top mount bolt. This is a long bolt that runs the width of the engine. The head is on the left and the retaining nut is on the right. You will need to hold the nut while you apply force to loosen the bolt. It is on tight and creaks as it is loosened. A deep 14 mm socket would have been useful here to hold the nut on the right but I was able to cheat a bit and not plug in the 6" extension all the way into the socket and gained just enough clearance to hold the nut securely without fear of rounding it.

1600 x 1200 1600 x 1200 1600 x 1200 1600 x 1200

Next, loosen the rear bottom mount bolt. This is also a long bolt that runs the width of the engine. The head is on the left and the retaining nut is on the right. You will again need to hold the nut while you apply force to loosen the bolt. It is also on tight and creaks as it is loosened as well. The nut on the right is close enough to the end of the bolt here so the 14 mm socket fits well over the nut here.

1600 x 1200 1600 x 1200 1600 x 1200 1600 x 1200

Next, with the all the mount bolts free to move just slightly in place, start the engine (transmission in neutral) and rev it a few times to 3000 rpm then shut it down.

If you have slider brackets, swivel them back in position removing and replacing the bottom M8 bolt. If the bottom bolt doesn't easily pull out it can be slowly unscrewed out.

Re-torque the mount bolts in this order. You will need to hold the nuts again as you torque the bolts.

  1. Lower Rear (40 lb-ft, 55 Nm)
  2. Upper Rear (40 lb-ft, 55 Nm)
  3. Right Front (or Left) (40 lb-ft, 55 Nm)
  4. Left Front (or Right) (40 lb-ft, 55 Nm)
  5. Frame Mount/Slider bracket bolts (if necessary) Right Side (or Left) (40 lb-ft, 55 Nm for top M10 bolt, 24 lb-ft, 33 Nm for bottom M8 bolt)
  6. Frame Mount/Slider bracket bolts (if necessary) Left Side (or Right) (40 lb-ft, 55 Nm for top M10 bolt, 24 lb-ft, 33 Nm for bottom M8 bolt)

The service manual lists the Frame Attachment bolts before the Right/Left Front Mount bolts but this is not possible because of the slider brackets.

For the front M10 bolts, a regular 14 mm flat wrench can hold the nut as you torque the bolt. The bottom frame mount M8 bolts take a 13 mm wrench to hold the nut.

My main issue with vibration is an annoying resonance I feel in the seat at 6500 rpm. This is due to the solid rubber feet on the Corbin Seat which press directly onto the frame and do a miserable job of damping any vibrations. A couple of things I have done, installing the Carburetor Equalization Manifold and doing Ivan's Dynamic TPS Adjustment in addition to installing Ivan's jet kit have helped alot but I was hoping to dial this out completely. On my first ride after this re-torquing I was initially disappointed as that 6500 rpm vibe was still there but as the engine warmed it seemed to subside and became nearly undetectable aside from a little roughness. We'll see as time goes on whether or not this re-torquing has any lasting effect. You may indeed wish to give this a try if your vibe conditions are excessive or intrusive.


From yamahablue:

"I just removed the engine from my FZ and found a potential problem. The top rear engine mount bolt was corroded and almost seized up in the engine case. The reason, dissimilar metal corrosion. ie: steel against aluminum. This is a big no no unless the bolt is properly plated with a resistant coating. This bolt obviously was not plated very well and corroded terribly in the bike's short 1 year life span. There is no doubt that, if gone unchecked, the bolt would have failed in a couple of years. I will have to replace the bolt and clean the engine case well to prevent any further deterioration. The bottom bolt was properly treated and showed no signs of corrosion. Here is a pic of both top and bottom bolts."

Addendum II:

From Bill Jinks, an improved procedure:

It is important to completely remove the "Rear Upper Through Bolt" as both Wyn and Yamahablue have proclaimed that theirs were seriously corroded...and upon removal and inspection I can confirm their claims of corrosion as...well....almost an understatement!!!

Pretty ugly eh??? was fire up the compressor and wire brush on a handgrinder time...cleaned her right back up...dressed her out with some copper base anti-sieze...and back in she went.

And's.... the things I did differently...How I did'em...& Y...

Thing #1.

After removing and dressing out that ugly corroded upper rear mounting bolt I proceeded to loosen the lower rear mounting bolt and then I loosened..."Just The Two Through Frame Bolts" up front..but...not "BOLT #3" in the pic below..

The reason I didn't loosen "Bolt #3" is a simple one has no "Floating Value" simply allows a vibration dampening (through the rubber bush) attachment of the forward engine case mounting points to the actual forward engine mounting bracket assy...which in and of itself is a "Hard Given Dimension"..with.."Zero Floating Value" other'd simply be loosening and re-tightening the bolts as..from what I can tell...there is no moving/floating feature/value involved there...just a bolt going through a rubber lined bushing.

Thing #2.

After completely loosening the the "rear upper and lower through bolts" and....just the "through the frame/downtubes" engine mounting bolts (bolts #1 & #2 in the pic above) on both sides...I didn't just start my engine....let it idle a while and..rev it a few times to 3K rpms....I also...

Pulled in the clutch lever and gently snicked it into 1st and then "VERY GENTLY" permitted the clutch to engage allowing the rear wheel to spin and the chain to "Find & Settle Into It's Most Linear Tracking Path" possible....while at idle...never reving the engine...while...all the engine mounting bolts were dead loose and...I could almost hear the chain find it's straightest path of travel as it aligned the front and rear sprockets and quieted down the chain tracking noise considerably.

Afterwards I was amazed to find that...after I gently pulled in the clutch lever..while the engine was still running...with the tranny still in 1st gear...the rear wheel kept right on spinning (until I reached over the bike and actuated the rear brake with my right hand rendering the rear wheel motionless even after releasing the brake) but prior to was as though I had never disengaged the clutch!!!....and this told me front and rear spockets were previously not in perfect alignment with the chain they were..effectively eliminating a very noticable amount of chain drag/friction.

Thing #3.

I didn't re-tighten the mounting bolts per Mr. Glenn's recomended sequence for the following reason....

Reason: It's the forward mounting bolts #1 and #2 (noted in the pic above) that have "The Floating Feature" incorporated into their design...which tells me that for the most's the "front of the engine & mounts" that biases engine postion...where...the rear mounting through bolts engage metal tabs welded to the frame...and... if you tighten them will most likely mechanically influence/shift the engine right back to the position you started from as...the forward mounts (designed to float) are...well..still loose and.."FLOATING". So to me the thing to do was...

a. Leave the rear through bolts generously loosened and...Re-Tighten The Floating Forward Mounting Bolts First!!!...starting with "BOLT #1" (the lower forward through frame bolt with the smaller 12mm head in the pic above)...first snug each one...then torque each one...on each side..sequentially so as not to disturb the engines new found proper position. Then....

  • Go on to "Bolt #2" (the upper forward through frame bolt with the larger 14mm head in the pic above)...and again...first snug each one...then torque each one...on each side..sequentially so as not to disturb the engines new found proper position. And now....
  • Go to the rear and re-snug, then fully re-torque the "UPPER" through bolt first as it has the weaker mounting tabs (designed to support uni-directional up & down) welded to the frame and they will easily pinch inwards as the bolt is tightened meeting the engine case at it's new found home without disturbing the newly established position of the already re-torqued forward floating mounts....and finally...go onto...
  • Re-torqueing the lower rear mounting bolt...with the much beefier mounting tabs welded to the frames lower cross-tube support.

And now...I do know this...even if the shop manual says best mechanical instincts and the reduction of chain/spocket noise and drag/friction tell me that...I just did everything right.

Addendum III:

From Wally Hartung, an aluminum industry metallurgist, warning about using a copper-based anti-sieze.

In a galvanic corrosion situation involving aluminum and steel, we're not worried about the steel. The aluminum is the sacrificial metal and will corrode preferentially, protecting the steel. The presence of water is absolutely necessary for the reaction to occur.

Plated steel bolts (plated with either cadmium or zinc) will afford some protection to the aluminum, but only as long as the plating lasts. It's being used up as it sacrifices itself to protect the aluminum. As soon as the plating is used up, the sacrifice of the aluminum parts to the steel will begin.

The white powdery coating on one of the bolts looks a lot like aluminum oxide. Guess what part is really getting eaten up. That's right, the aluminum case.

Now if you want to really see aluminum go to pieces, get copper involved. Copper eats up aluminum in a situation of dissimilar metals with water involved. Copper based anti-seize should NEVER be used in contact with, or where it can get in contact with aluminum. Use the aluminum powder anti-seize, always.

My knee-jerk reaction recommendation to fix this corrosion problem would be to:

  1. Clean up the existing corrosion. If the steel bolt hasn't been significantly corroded (just light surface corrosion) clean it up and reuse it. If there is any question as to a reduction of the cross area of the bolt due to corrosion, replace it with the proper factory bolt.
  2. Spray the bolt with WD-40 or other anti-corrosive that is compatible with the rubber pieces. (I went out to the garage to look at the mounts and noticed there is a rubber piece in the joint. Like I said, I just got the bike. Vibration isolation part, or seal to exclude water, or both?)
  3. If you've got copper anti-seize where it can get in contact with the aluminum cases, clean it out of there completely.
  4. Use a little RTV to seal the assembly. The basic idea is to get rid of the water incursions into the bolted joint. Without the water, no corrosion.
  5. Blue Loctite on the threads, not copper anything.

This is why I never use a pressure washer on a motorcycle, they'll force water into all kinds of places it's not supposed to be. Also, never use a plain steel or brass(copper) wire brush on aluminum during the corrosion cleanup. Always a stainless steel wire brush. Steel or copper based wire brushes will leave small particles and lead to pitting of the aluminum.

304 stainless fasteners are fine with in contact with aluminum. There is a surface passivation involving the chrome oxides that form on the stainless that precludes the formation of a galvanic cell under most conditions.

Wally advises that by just looking at a couple of photos of one component of the problem he really can't offer a professional opinion from such limited information. There may be other factors involved, but the basic premises of keeping the water out and not using copper are sound.

Back to Main Page

Last Updated: 06-25-2004

Copyright © 2001-03, Patrick Glenn, All Rights Reserved.
Yamaha® and FZ1® are registered trademarks of the Yamaha Motor Corporation.
This site is not affiliated in any way with the Yamaha Motor Corporation.

The information presented here reflects solely my personal experience with my motorcycle and is presented for entertainment purposes only. No information presented here is to be relied upon for issues of rider safety nor to replace the services of a qualified service technician. Any attempts to follow or duplicate any of these procedures are done so completely at your own risk. By reading the information on this site, you agree to assume complete responsibility for any and all actual or consequential damages that may arise from any information presented herein.