Excessive noise on bench test after transmission rebuild

[2017FLHTK]

I'm in the middle of my first transmission rebuild (on a 1979 FL). All internal parts are new Andrews, including the bushings / gears / dogs / mainshaft & countershaft. Obviously new bearings inside the gears, and the mainshaft bearing on the kicker end is brand new as well. The (new) Koyo transmission drive bearing on the case was installed to the proper depth with the correct Jim's Tool.

Here's a list of the parts used:

Gears:
-1st Gear, 2.44 Ratio, Part Number 201105 [correctly paired for 2.44:1 first gear / 1.35:1 third gear set]
-2nd Gear, Stock 1.82 Ratio, Part Number 292160
-3rd Gear, 1.35:1, Part Number 203375 [correctly paired for 2.44:1 first gear / 1.35:1 third gear set]
-4th Gear, stock 26 tooth main drive gear (1977-1986), Part Number 204280

Shift Clutches
1-2 Shift Clutch, Part Number 205120
3-4 Shift Clutch, Part Number 205340

Shafts
-Countershaft (Late 1976-1979), Part Number 207770
-Mainshaft (2970-2986 chain drive), Part Number 208700

With just the mainshaft installed in the case, this is what it sounds like as I rotate the gears:

I have a generous coat of assembly lube applied to the shafts, bearings, and bushings. I have light coat of oil on the teeth of the gears as well. When I rotate the assembly I can hear an alarming amount of mechanical noise.

Is this the normal sound of gear engagement for a bench test? Does it just sound so loud because it's not in the oil bath it will live in once everything is buttoned up and topped off with gear oil? Or is something off?

[kd]

Welcome to the world of straight cut spur gears.  It sounds normal to me.  Fill it the the correct level with fear lube and retry it.  The lube will act as a cushion between the gear teeth and silence it somewhat.  Put the lid on and it will improve it even more. I think you will see a big difference.

BTW, you'll love the Andrews 1st and 3rd gear ratios if you haven't tried them yet.

[JW113]

 

Sounds normal to me too.

What overall final drive ratio are you planning to use? I'm running 3.42 overall final drive, and even with my stock 3.0:1 1st gear, I need to slip the clutch a little. Just wondering how that 2.44:1 is going to launch. But once rolling, I'm sure that closer gear ratio set is gonna be nice!

-JW

[bump]

Sounds about right to me. As long as the gears turn without much effort and shift dogs move smoothly. Button it up and try running it through gears while turning shaft by hand.

[kd]

Sounds normal to me too.

What overall final drive ratio are you planning to use? I'm running 3.42 overall final drive, and even with my stock 3.0:1 1st gear, I need to slip the clutch a little. Just wondering how that 2.44:1 is going to launch. But once rolling, I'm sure that closer gear ratio set is gonna be nice!

-JW

I ran 21 and 22 trans X 51 or 49 final drive and it worked great.  I switched around occasionally but never found it too high with the lower tooth count transmission sprocket.  It did time on the highway quite a bit and had an 80 mph sweet spot.  As JW says, the Andrews first is not the stump pulling ratio you took out so you may want to consider a transmission sprocket change. The 3rd gear splits the difference between 2nd and 4th about prefect.  The OEM 3rd was very close to 4th and not all that great for sport riding.

[2017FLHTK]

Welcome to the world of straight cut spur gears.  It sounds normal to me.  Fill it the the correct level with fear lube and retry it.  The lube will act as a cushion between the gear teeth and silence it somewhat.  Put the lid on and it will improve it even more. I think you will see a big difference.

Great feedback. Thanks!

BTW, you'll love the Andrews 1st and 3rd gear ratios if you haven't tried them yet.

If I ever get this project finished..... 

Sounds normal to me too.

What overall final drive ratio are you planning to use? I'm running 3.42 overall final drive, and even with my stock 3.0:1 1st gear, I need to slip the clutch a little. Just wondering how that 2.44:1 is going to launch. But once rolling, I'm sure that closer gear ratio set is gonna be nice!

I have a 24 tooth engine sprocket paired to a 37 tooth clutch basket.  Transmission sprocket will be a 23 tooth and the rear wheel sprocket will be a 48 tooth.

According to http://www.nightrider.com/biketech/calc_gearratios.htm that works out to be a 3.217 final drive ratio.  She'll be a slug off the line, but that's ok.  I like to run around 75 mph or so on the highway, so I'll sacrifice my low-end oomph in exchange for smooth highway RPMs.

Sounds about right to me. As long as the gears turn without much effort and shift dogs move smoothly. Button it up and try running it through gears while turning shaft by hand.

Sounds good.  Thank you Sir.

I ran 21 and 22 trans X 51 or 49 final drive and it worked great.  I switched around occasionally but never found it too high with the lower tooth count transmission sprocket.  It did time on the highway quite a bit and had an 80 mph sweet spot.  As JW says, the Andrews first is not the stump pulling ratio you took out so you may want to consider a transmission sprocket change. The 3rd gear splits the difference between 2nd and 4th about prefect.  The OEM 3rd was very close to 4th and not all that great for sport riding.

Just trying to get everything buttoned up / ok to use, but I'll keep your comments in mind if I decide that I need to fine tune the gearing on the sprockets in the future.

[2017FLHTK]

Taking a look at the end play today.  The mainshaft is fully assembled and installed in the case. Since I have a 1979 model I have a roller cage bearing and cannot adjust endplay on the mainshaft. I can insert a 0.014 feeler gauge between the face of the main drive gear and the transmission case wall.

I can move (ever so slightly) first gear on the countershaft gear cluster. It's possible to just barely slip a 0.004" feeler gauge in between the case bushing and the countershaft gear end washer. Technically that's within spec, but I'd ideally like something closer to the middle of the tolerance range when I set my endplay up.

I found a set of NOS / OEM thrust washers (Harley part number 35875-36) that includes all seven of the various thicknesses of washers that Harley uses to set spacing. One of those should get my end countershaft end play correctly set.

[JW113]

Yes, that sounds like a pretty tall first gear. Will be interesting to hear how that works out, so keep us posted.

cheers,
JW

[Ohio HD]

Yeah I always put the 2.60:1 Andrews set in when rebuilding them. Gets the heavy scoot moving quicker.

[76shuvlinoff]

I have a 24 tooth engine sprocket paired to a 37 tooth clutch basket.  Transmission sprocket will be a 23 tooth and the rear wheel sprocket will be a 48 tooth.

lh
 This combo is what I have on my FLH behind a 93" mill and 5 inna 4 transmission (that I really wish had an OD)  I have played with rear sprockets from 51T down to 46T and settled on 48.  It's just my country bomber/bar hopper not an 80 mph long distance highway bike but it's still fairly happy at 70.

[2017FLHTK]

I sent the videos to Andrews tech support and asked them what they thought. This was their response:

Doesn't sound bad although watching a video is not ideal at all/ can't hear much there. You may have a local shop experienced in transmission re-builds that may be able to take a second look/ listen? I would put the transmission in gear spin the shaft with some load on the drive gear applied by hand or hold the drive gear/ spin the shaft if in neutral (this not how QC is performed here, just giving you some ideas). Make sure the gears roll smoothly together as pairs (1st, 2nd 3rd, 4th rolled individually and make sure they don't feel "notchy") otherwise short of sending them back to be QC'd that's as much as I can recommend doing at this time

[2017FLHTK]

Yes, that sounds like a pretty tall first gear. Will be interesting to hear how that works out, so keep us posted.

Guess we'll find out when it's all done.  This is my first time messing around with gearing, so I'm mostly going on reports from other guys who seem to have a similar riding style (happy doing a steady 75mph on the freeway but don't care about taking off quick from a light).

I have a 24 tooth engine sprocket paired to a 37 tooth clutch basket.  Transmission sprocket will be a 23 tooth and the rear wheel sprocket will be a 48 tooth.

lh
 This combo is what I have on my FLH behind a 93" mill and 5 inna 4 transmission (that I really wish had an OD)  I have played with rear sprockets from 51T down to 46T and settled on 48.  It's just my country bomber/bar hopper not an 80 mph long distance highway bike but it's still fairly happy at 70.

Everything in life is a compromise.  I just hope I picked the right all-around gearing for the way I ride.
 

[JW113]

Since Mark's bike is a 5 speed, it's first gear is probably around 3.25:1 or so. With your four speed and that gearing, your first gear is going to be very close to a HD 5 speed's 2nd gear. So if you have a chance to ride a EVO or TC and launch it from a dead stop in 2nd gear, that's what you're going to be looking at with this close ratio 4 speed set up.

-JW

[76shuvlinoff]

Since Mark's bike is a 5 speed, it's first gear is probably around 3.25:1 or so. With your four speed and that gearing, your first gear is going to be very close to a HD 5 speed's 2nd gear. So if you have a chance to ride a EVO or TC and launch it from a dead stop in 2nd gear, that's what you're going to be looking at with this close ratio 4 speed set up.

-JW

I tried to find the data on my 5/4. All I can find in the paperwork is it's a 2.94:1 1st and of course 1:1 5th. I know I have all the gear ratios written down somewhere, or at least I once did. 

edit to add :
1st 2.94 to 1

2nd 2.21 to 1

3rd 1.57 to 1

4th 1.23 to 1

5th 1 to 1

[2017FLHTK]

I had the countershaft out of the case while trying various endplay washers and and noticed something: the high gear / fourth gear on the mainshaft was INCREDIBLY hard to turn, whereas the third gear spun freely. See video below.

I'm not crazy. Fourth gear should spin easily on the case bearing (similar to third gear), independent of the mainshaft, correct?

[JW113]

Yes, that is correct. What you're calling 4th gear is the main drive gear. Semantics, yes I know. The main drive gear spins in the left side of the case in it's roller bearing, and the mainshaft should be floating in the bushing in the main drive gear. It's not clear to me why the main drive gear would bind if you tighten the mainshaft nut, they should be totally floating independent of one another.

The retaining ring located between the 3rd gear and shifter clutch is snapped in it's groove, yes?

-JW

[2017FLHTK]

Yes, that is correct. What you're calling 4th gear is the main drive gear. Semantics, yes I know. The main drive gear spins in the left side of the case in it's roller bearing, and the mainshaft should be floating in the bushing in the main drive gear. It's not clear to me why the main drive gear would bind if you tighten the mainshaft nut, they should be totally floating independent of one another.

The retaining ring located between the 3rd gear and shifter clutch is snapped in it's groove, yes?

-JW

Yes, snap ring was correctly installed.

[2017FLHTK]

I think I've discovered the source of the problem.  Short version: I had my Andrews fourth gear modified by saddlebagrail by machining the gear to accept oil seals.  The seals on the gear appear to be dragging on the mainshaft and causing an interference issue.

Expanded discussion:

I've had this transmission apart a few times now trying to figure out what was going on.  The last time I marked the transmission case (on the inside) where the final drive butts up against it with red sharpie.  I also marked the face of the final drive gear with black sharpie.  It seems like the black sharpie washed off the face of the gear as oil splashed around on it, but the red sharpie on the case wall stayed in good shape.  That tells me that the main drive gear was NOT touching or dragging on the inside case wall.

I took the mainshaft out of the case and confirmed that the fourth gear bushing would allow the gear to turn on the mainshaft.  Although it would turn, there was still considerable drag (1:15 mark on the video).  Looking at the parts, I noticed that the oil seals installed on the fourth gear seemed to be causing the issue.  If I slid the gear to the top of the shaft such that the bushing was engaged on the mainshaft but the rubber seals were not touching the shaft, it spun freely (2:00 minute mark on the video).  As reference, you can see my original OEM gear turning freely on the shaft (2:30 mark).

[JW113]

So the old sandbagrail o-ring, eh? I guess we left that little piece of into out, didn't we?
 

I have one of those also, but not on the main drive gear. It's in the mainshaft spacer. Unless this is something else? As in, an o-ring to keep it from leaking out the main drive gear bushing when it or the mainshaft gets worn?

-JW

[2017FLHTK]

So the old sandbagrail o-ring, eh? I guess we left that little piece of into out, didn't we?
 

Yeah, truth is I forgot about the modification (work was done in March).  This ongoing project has been apart for way longer than I would have initially guessed.    I'll haze myself later......

I have one of those also, but not on the main drive gear. It's in the mainshaft spacer. Unless this is something else? As in, an o-ring to keep it from leaking out the main drive gear bushing when it or the mainshaft gets worn?

-JW

I have the spacer from him as well (haven't installed it yet), but the oil seal on the main gear is another machining service that he does.  I'm waiting to hear back from Ray / Saddlebagrail on what his thoughts are.  I'll make sure to post an update when I have more info.

[JW113]

Of course, this is just a wild ass guess, based on little fact, but from the description... not enough clearance in the groove for the o-ring, and it's binding...

-JW

[2017FLHTK]

Further observations:

There is a good machined fit on the main gear assembly, and the bushing / gear slides nicely on the lubricated mainshaft. Ignoring the drag once the end with the oil seals engage the shaft, there are no "tight spots" along the mainshaft that the bushing is binding on. With the main drive gear mounted on the mainshaft I tried to slip feeler gauges in between the main drive gear bushing and the mainshaft. Because the feeler gauges were so thin, I had a difficult time getting them to both conform to the curvature required to check clearance and slide forward / into the gap. My 0.0015 and 0.002 feeler gauges looked like they would have fit. But, because they were so floppy I couldn't actually get them to slide in place.

There is no indication that the final drive gear is making contact with splined portion of the mainshaft. The splined shaft tops easily clears the ID of the main drive gear with PLENTY of space.


I see no marks indicating that the main drive gear is physically touching the end face of the splines on the mainshaft.

The bushing inside the main gear is not interfering with the fit of the shifter clutch / dog.




I reassembled the transmission with all the Andrews parts, but reused my OEM / factory final drive gear. The gearbox turned easier and was much quieter. See video below:

Saddlebagrail (the machinist who modified the main drive gear for the quad seal) wrote "The drag cause by the 2 small seals is «NORMAL»,instead of having one small seal that drag when new,now you have 2 small seals that double the drag,but let me tell you from my experience,it will not drag long when in service,put some quality grease in between seals before installing,

The sound of the teeth meshing have nothing to do with the update on the main drive gear,and it is hard to say from here if you really have problem or not,but as mentionned ,take it one step at a time ,until your have done the complete steps by steps"

l'll swing by the auto parts store and pick up some Mobile 1 synthetic grease (saddlebagrail's recommended grease). I had previously put a thin film of disc brake grease on the quad seals prior to installing on the mainshaft, so I don't expect a large change in status by swapping over to a different brand of grease on install, but I'm willing to give it a shot.

So assuming no change of status, short of buying another Andrews final gear and seeing if the problem recreates itself, any input on next steps?

[JW113]

After thinking this over a little, I'm not so sure adding an O-ring between the main drive gear and the mainshaft is such a good idea. Think about it for a second. O-rings are typically used in places where the is no relative motion between the surfaces. When there IS motion, a lip seal is typically used. In fact, can you think of any instance where an O-ring was used to seal a rotating shaft? At least, in a high relative speed application? When the bike is in neutral and the motor running, the relative speed differential between the mainshaft and main drive gear can be over 4000 rpms if the motor is rev'd to redline. You think that's a good idea to have an O-ring in that situation? Especially when it's being squeezed at tight as what it appears in your videos above.

Personally, I'd leave it out and move on...

-JW

[Ohio HD]

After thinking this over a little, I'm not so sure adding an O-ring between the main drive gear and the mainshaft is such a good idea. Think about it for a second. O-rings are typically used in places where the is no relative motion between the surfaces. When there IS motion, a lip seal is typically used. In fact, can you think of any instance where an O-ring was used to seal a rotating shaft? At least, in a high relative speed application? When the bike is in neutral and the motor running, the relative speed differential between the mainshaft and main drive gear can be over 4000 rpms if the motor is rev'd to redline. You think that's a good idea to have an O-ring in that situation? Especially when it's being squeezed at tight as what it appears in your videos above.

Personally, I'd leave it out and move on...

-JW

 

[2017FLHTK]

After thinking this over a little, I'm not so sure adding an O-ring between the main drive gear and the mainshaft is such a good idea. Think about it for a second. O-rings are typically used in places where the is no relative motion between the surfaces. When there IS motion, a lip seal is typically used. In fact, can you think of any instance where an O-ring was used to seal a rotating shaft? At least, in a high relative speed application? When the bike is in neutral and the motor running, the relative speed differential between the mainshaft and main drive gear can be over 4000 rpms if the motor is rev'd to redline. You think that's a good idea to have an O-ring in that situation? Especially when it's being squeezed at tight as what it appears in your videos above.

Personally, I'd leave it out and move on...

-JW

JW, poor choice of words on my point.  Better terminology would be a quad seal, similar to what was used on the evo era five-speed transmission mainshafts.  On another forum that I'm on several people have spoken very highly about this modification.

At this point my concern isn't the gear noise in the box as it rotates, but instead it's the ridiculous force required to turn the main drive gear. If you haven't done so already, please watch this short video (less than 60 seconds) and tell me if this is what you see when you're rebuilding four speed transmissions.

Every video I've been able to find on four speed transmissions shows the main drive gear spinning freely & independently on the mainshaft with relative ease, as compared to me having to push with all my might to get the damn thing to turn on the mainshaft even with no countershaft installed. As contrast, look at this video at the 9:20 mark where the main drive is being spun on the mainshaft (without the mainshaft turning) with only thumb pressure:

I find it hard to believe that the quad seal is causing anywhere near that amount of drag.

[JW113]

Yes, I did watch your video. And no, it shouldn't take any force at all to turn the main drive gear with the mainshaft stationary. So it's not clear to me exactly where this O-ring or quad seal is being installed, but from the symptoms, it appears to be getting pinched between the inside edge of the main drive gear, and the mainshaft where the splines are. Just for grins, how about removing the quad seal and see how it spins?

-JW

[2017FLHTK]

Yes, I did watch your video. And no, it shouldn't take any force at all to turn the main drive gear with the mainshaft stationary. So it's not clear to me exactly where this O-ring or quad seal is being installed, but from the symptoms, it appears to be getting pinched between the inside edge of the main drive gear, and the mainshaft where the splines are. Just for grins, how about removing the quad seal and see how it spins?

-JW

Nah, the quad seals aren't getting anywhere close to the splined shaft (they're on the other end of the part and normally stick out of the case).

I heard back from Ray / saddlebagrail (the machinist who did the modification).  He wrote:

"you can take one small seal off ,but it would be useless to pay for a work and not getting benefit out of it" and told me that I could "take the small seals out and the gear will spin easily and freely....the 2 small seals are made are what make the gear hard to spin on the shaft"

[kd]

Like you> I can't imagine being that tight is good.  There will be a lot of heat and probably damage because of it. The second part of the video in post #14 where the shaft is tapped back 1/4" and it spins with about the right amount of seal drag makes me think there is something wrong with the stack. Have you measured it with the old gear set installed?  Is the shaft fully into the kicker side bearing? Is the main shaft bearing the correct width? Are all of the parts seated or is a burr holding something back. It has to be something simple that is being taken for granted.  Did you share the video with Ray / saddlebagrail so he could see how tight it is and what happens when you backed it of 1/4"?

[2017FLHTK]

Like you> I can't imagine being that tight is good.  There will be a lot of heat and probably damage because of it.

I'm a shade-tree mechanic at best, but it seemed a little alarming to me.   

The second part of the video in post #14 where the shaft is tapped back 1/4" and it spins with about the right amount of seal drag makes me think there is something wrong with the stack. Have you measured it with the old gear set installed?

I'm not understanding.  What would I measure on the old gear set? 

Is the shaft fully into the kicker side bearing? Is the main shaft bearing the correct width? Are all of the parts seated or is a burr holding something back. It has to be something simple that is being taken for granted.

Agreed on the part in red.  I'm 99% sure that that main drive is properly assembled. When I installed the cluster gear on to the main shaft, I placed the first gear into an arbor, slid the mainshaft in to the other end, screwed on a thread protector on the end of of the main shaft, and then gave the end of the mainshaft some hefty lovetaps until the gear cluster was seated. I'd say its very unlikely that the 1st gear didn't make it's way all the way fully seated, but is within the realm of things that are possible. Unfortunately I don't have access to a hydraulic press.

The main drive bearing is the correct part number, and I have no reason to believe that it's an incorrect width.  It turns freely in the case with no binding or restrictions.

Did you share the video with Ray / saddlebagrail so he could see how tight it is and what happens when you backed it of 1/4"?

He's commenting on the issue in a thread on another forum, but I don't think he's watched the video. 

[kd]

I'm not understanding.  What would I measure on the old gear set?

If it is OK with the old gears then there is something wrong with the fit of the new ones.  Is's the only thing you changed other than bearings.  You seem confident the bearings are correct and seated properly.

I would consider removing the cluster set and looking for witness marks to see if it did seat. It may be sharp teeth digging into a radius on the shaft? Measure it as it is now and then do the math when apart to see if it was all the way. When you reassemble it, freeze the shaft and heat the gear in the oven.   It should change tone when it is seated. If it was me, I would need to know if that gear set is on all the way.   

[2017FLHTK]

If it was me, I would need to know if that gear set is on all the way.   

On another forum, a guy posted about ensuring that the gear was fully set on the shaft: "A quick easy way to know if it is, how much end play does the mainshaft third gear have once the washer and snap ring is in place."

In my case the snap ring clicks in place in front of the washer retaining third gear.  There is effectively no slop / end-play, which tells me the mainshaft gears are properly emplaced.

However, the big news is that now everything seems to be working!

Content that everything was properly assembled on the mainshaft, I decided to reassemble the mainshaft. However, I first put a healthy film of Mobil 1 synthetic red grease on the main drive gear bushing, and a light film of grease on the mainshaft. Everything turned MUCH better than before. I was able to spin the main drive gear around the mainshaft with only moderate pressure with my thumb. When I inserted the countershaft cluster in the gearbox, everything continued to rotate nicely. I can still hear a small amount of gear mesh noise, but that should go away once it's in a proper oil bath.

Thanks to all who provided suggestions / mentorship / guidance while I stumbled through my first transmission rebuild.

[FSG]

did you wash that bearing before lubing and installing?

[2017FLHTK]

did you wash that bearing before lubing and installing?

Yes. Shipping protective film was stripped off with kerosene, and then assembly lube was applied to the needle bearings. Needle bearings were all rolled / rotated to ensure assembly lube was generously applied throughout the bearing.

[dbdevkc]

Found the discussion on gear ratios interesting - I recently tore down my 4-speed and found the following ratio Andrews gears (probably about 30 years old):
1st: 2.24; 2nd: 1.65; 3rd: 1.35

It's a relatively light FXE, and I'm a light guy, but yikes!

[JSD]

I ran the 2.44 as above and same 3rd with 25 trans to 51 rear but it had a primo 3" Kit and 4 1/2 stroke with A grind cam sat on 80mph all day and no problem from start off 2 up. 16 years 2 rear tyres a year. Should never sold that bike. Got a 68 to play with now as i love a good set up shovelhead.