Best Of *Science* of solving N scale diesel mechanism drivetrain vibration noise?

[atsf_arizona]

An 'original' Kato SD40-2 chassis...  I've also pulled the same stunt (by putting it under a  Intermountain SD45-2 chassis - same Kato chassis - so it MU's flawlessly with my much larger fleet of Kato six-axle units.   I get a much better performing unit.

Randgust, so you found that the Intermountain SD45-2 shell will fit nicely on a Kato SD40-2 mechanism?   Any shell or mechanism shimming, trimming needed?   

I am enjoying this thread.

[jagged ben]

Randgust, so you found that the Intermountain SD45-2 shell will fit nicely on a Kato SD40-2 mechanism?   Any shell or mechanism shimming, trimming needed?   

I think you misread.  He clarified he was talking about a Lima FP45.    Kato mechs are generally too wide for other mfrs hoods.

[atsf_arizona]

Do they get quieter with the shell on after the (Ron Bearden "spring in the universal joint" modifications ?

.

The Atlas B40-8 / C628 trucks mechanism, with it's RSD-7 shell, IMHO no change in the sound:

/>For comparison, here is that mechanism without the shell:

/>
=========

However, the Intermountain SD45-2 mechanism, with it's shell on, IMHO may increase the noise slightly:

/>For comparison, here is that mechanism without the shell:

/>
=========

Conclusion:  effect on loco noise by adding the shell, varies on a case-by-case basis.

[Big Train]

That is not a fix - it only masks the problem and it will also attract dirt to the gears (which can eventually jam them).

I've heard this caution before and it has never been a problem for me in the 30+ years of running N Scale. And I only know of one incident in our group where FOD (Foreign Object Damage) caused a jammed truck. That had nothing to do with excess grease....in fact, the truck had no grease at all.

Loose parts will vibrate or oscillate creating noise.   In the perfect world the entire drive-train (motor and worms) would all be perfectly aligned and riding in permanently mounted bearings. But because our toys are not made to that type of tight tolerances, some play in the bearing block cradles has to be incorporated. It is not ideal solution, but a compromise.  The phosphor/bronze bearing block springs are one such compromise. Properly adjusted they allow the bearing to move but it they keep it from loosely flopping around in the bearing cradle.

Some backlash is needed and I think that allows some "self alignment" of the rotating driveshaft. But how much is too loose? Has anyone investigated this and found a solution? Like I said before, these aren't Omega watches, but this type of mechanism has been around for 15 years or so and the manufacturers have had a long time to fine tune things.

Which is OK with me....until I need to open them up to install decoders. If it weren't for that, I'd have no reason to mess with them and they would be silently running on the layout.

[atsf_arizona]

I made some photos of what I interpreted to be Ron Bearden's "spring in the universal joint" modification.

Note:  I experimented with a 2nd SD45-2 mechanism and applied these springs.  It did not get quieter,
as a matter of fact, it may be a bit noiser.  Thus, as far as I can tell at this early stage, this modification should be a
*judicious* application, don't do it to every loco just because you can.   Rather, the specific amount of tension, freeplay, interaction
of the truck and the worm gear, the alignment of the entire drivetrain, etc. I believe needs to be taken into consideration. 
Only when you determine that there is too much free play in the worm bearing from a backwords and forwards standpoint....
then you could apply this technique in attempt to address it.

Here's the photos:







Inside view of the flywheel, where the universal joint fits:



So, what I am getting so far is:

• Unfortunately, there doesn't appear to be any scientific tools that we've identified (yet), that can help us scientifically diagnose the noisy mechanisms
• Rather, it simply takes a good logical understanding of the principles of how these mechanisms are put together, what the possible sources of noise may be, and then.....
• Logically using the tips/techniques listed here to methodically identify what may be producing the noise, then methodically eliminate noise producing possibilities...  until you arrive at (hopefully) a better running split frame N scale diesel mechanism
• Each locomotive could have it's own "personality" in terms of the particular combination of fit, finish, tolerances, alignments, etc.  Remember that and treat each loco with a fresh eye if necessary
• Very experienced modelers here have reported that tuning N scale locomotive mechanisms is more art than science, and it's unfortunately not unusual to dis-assemble one of these locos and then find out it is more noisy after re-assembly.   It does seem to be a bit unpredictable whether a loco will be more noisy after re-assembly

Once we have the foundational information that this thread is helping us all document, then we each individually need to invest the time necessary to learn to apply the knowledge.  OK, I'm cool with that.   

I still hope that we will come across some sort of diagnostic tools or laser tools or something to help us instrument this all. 

Any ideas?  Something we could adapt from other disciplines, hobbies, or professions to apply here?

In meantime, I will go back to trying the techniques in this thread on these two locos and see what further happens......

Other comments/questions?  Or answers to previous questions?   

Thx all, keep it coming, I'll keep monitoring.

[peteski]

I've heard this caution before and it has never been a problem for me in the 30+ years of running N Scale. And I only know of one incident in our group where FOD (Foreign Object Damage) caused a jammed truck. That had nothing to do with excess grease....in fact, the truck had no grease at all.

Of course, piece of ballast can get into plastic gears without any lubricant (possibly attracted by a static charge generated by the teeth meshing), but loading the gearbox with grease IMO increases a probability of more foreign objects getting picked up and mixed into the grease and eventually ending up in between the teeth.

I have serviced dozens of N scale locomotives and I find plenty of debris stuck in the teeth of lubricated gears.  Much more often than in un-lubricaded gears.

In the end it is a question of following the best recommended practices.

[atsf_arizona]

I just watched your video of the B40 mechanism with Alco trucks.  You asked what that whine was.

Of course I can't be sure, but it sounds like the worm is riding too tight on the double drive gear in the truck tower.
You said you made some kind of mods in order to make this work on RSD shells.   Did you mod the truck to chassis interface.
If you did, that my best guess is that the chassis needs to be raise about a millimeter.  You could try this by adding a shim glued to the bottom of the chassis at the gear tower opening.

I also encountered this when working on my Kumata book.  I member of this board (Bob Gilmore?) sent me his brass Centipede.  One problem I noticed was the drive gear was pressing its neighbod gear to tight.  When I bent the gear frame to give a tiny gap, the noise dropped significantly.

So this issue may not apply to what you did, but it does add to the discussion in general.
We want the worm TEETH to move the gear TEETH. 
We don't want the teeth of either to hit the the bottom of the gap between the teeth.

That's a good thought.  I will have to try this as well.

To answer your question on whether I modified the truck to chassis interface, yes, I had to.   I have a photo essay
on what I needed to do to the Atlas B40-8 mechanism to get the Atlas C628 trucks to fit, see here:

 http://www.pbase.com/atsf_arizona/n_scale_rsd15&page=all

And in particular for the truck to chassis interface:







So, Ron, it may be that I need to make the slot vertically deeper yet in the 2nd photo, perhaps that will relieve some pressure on the
gear tower to worm gear.

I wish there was some way to "x-ray" the mechanism and see if this is the case that you suspect, or somehow to instrument it.... but again, I gather this is going to be trial and error.    I'll report back what happens once I try this.

I gather that is why this is a hobby and not rocket science. 

I still hope that we will come across some sort of diagnostic tools or laser tools or something to help us instrument all of this. 

Any ideas?  Some tools, probes, scopes we could adapt from other disciplines, hobbies, or professions to apply here?

[jagged ben]

I have to say it will take a fair amount of evidence to convince me that the less than smooth running of an undamaged IM SD45-2 would be due primarily to anything other than those horrible Atlas designed trucks that Daniel quite rightly excoriated back on page one.   Those of us who model second and early third generation diesels in N scale really deserve a replacement option for that truck.

[mecgp7]

That seems to imply that you think that a motor itself is a source of majority of the noise.  The way I see it, the motor itself is very seldom the source of noise/vibrations. I find that much more often that the vibrations are generated by mechanical components other than the motor.

I have a dozen or so Atlas motors all from GP7/9 chassis. Testing them after pulling them from the chassis and the difference in the noise produced between them is huge. Some are just loud, others are quiet until you put them back in the chassis and they get noisy which tells me the noise is due to the rest of the drive train as you say. Point is, if the motor is noisy on its own, it will never be quiet in a chassis. I consider it a "bad" motor even though it operates just fine.
I always swap out motors as part of the process as well. If the chassis is quiet with another motor then I consider the original the problem. I'm not saying it can't be fixed, but sometimes the solution is motor specific.

[narrowminded]

Would this be a suitable test meter? http://www.tenma.com/products/product/72-8170
Below that unit in cost seem to be mostly compact "pocket-size" meters, with limited choice of measuring ranges, while above that are the "professional" type meters, such as those by Simpson, at ten times the price.

I just used an analog panel meter wired with a DPDT switch on the output for direction change without reversing the meter.  I also have a 15v voltage meter attached.  These should be available on Ebay for under $5 shipped.

So I looked it up.  Here's exactly the one I'm using although mine came from a different seller.  You may even find a better price than $5.   (Reminder to self: Don't go into the panel meter business.)

http://www.ebay.com/itm/DC-0-100MA-Analog-Amp-Meter-Ammeter-Current-Panel-DC-100MA-/162186441111?var=&hash=item25c310b197

You may want a slightly broader range for larger locos and a search should get you what you need.  Get one that's pretty close for your actual needed range so that it's easy to see the dial and the movement.  For my purposes I'm looking for 0 through about 60mA  with an odd excursion nearer to 90mA or the ever present opportunity to oops which pegs the thing.

[u18b]

OK, John.
That's what I was worried about.

Getting/keeping locos quiet is hard enough without tinkering (as I myself do a lot).

When I look at this photo....


it looks to me like the teeth are different between to two trucks.

It could be the angle of the photo, but please go back and check the tooth depth.

If I am correct, then my hunch earlier was probably also correct.

If the teeth in the main double drive gear which interfaces the worm on the Alco truck is longer, then you are at risk of those teeth bottoming out in the grooves of the B40 worm.

Now, Atlas uses a lot of interchangeable parts.  So it could be the photo.

But check the tooth depth between those two trucks.
Here is a need for a new scientific measuring device!

 

[u18b]

By the way, on the spring in the flywheel trick....

I found that the length of the spring was absolutely critical.
Noise did not diminish much until it was just right.
But this is the very ESSENCE of trial and error to find that.

But second, long after I did that thread, I found that IM F units had the same problem of lateral slop.
The solution that worked perfectly was to slide the doughnut out ward until the slop was taken up.

Truthfully, I'm not a big fan of the doughnut u-joint that IM likes so much.
WAY to much can go wrong (including cracked parts).

I think the flywheel with the hex hole is so very much better and tried and true.

If you really wanted to spare no expense and get this thing running as quietly as possible, then replace the flywheels (or whole motor) and the u-joint system.

[randgust]

I have swapped an Intermountain SD45-2 chassis with a Kato SD40-2 chassis successfully.   I had to lightly 'buzz out' the inside of the shell with a Dremel and cut a bit of the walkway edge off.   Other than that, nothing much.

I did that primarily not for noise, but for getting a perfectly matched Kato mechanism to MU with the other eight Kato six-axle mechanisms it may be paired with.

I've put C30-7 mechs under Trix U30 shells and U30CG shells, and yes, the one Kato SD40 is also under a Lima FP45.   Two more Kato mechs are under shortened Lima shells to make F45's.   

The only obnoxious noisy six-axle split-frame I think I've ever owned was the Atlas SD50 from the clicking out of the front truck.

I've got some really rumbly Model Power RSD15 chassis but nothing in heaven or earth can fix that.   I've already got them as quiet as they'll ever get by swapping motors and tuning them.

[rrjim1]

I have a dozen or so Atlas motors all from GP7/9 chassis. Testing them after pulling them from the chassis and the difference in the noise produced between them is huge. Some are just loud, others are quiet until you put them back in the chassis and they get noisy which tells me the noise is due to the rest of the drive train as you say. Point is, if the motor is noisy on its own, it will never be quiet in a chassis. I consider it a "bad" motor even though it operates just fine.
I always swap out motors as part of the process as well. If the chassis is quiet with another motor then I consider the original the problem. I'm not saying it can't be fixed, but sometimes the solution is motor specific.

The motors that are noisy are not balanced.  If you balance these motors/flywheels you would be surprised how smooth and quiet the will run. I have done this to several Atlas motors and flywheels. Learned this back in the 1960s with Aurora slot cars, a balanced armature reduced the noise with the three brass gears.   

[atsf_arizona]

OK, John.
That's what I was worried about.

Getting/keeping locos quiet is hard enough without tinkering (as I myself do a lot).

When I look at this photo....


it looks to me like the teeth are different between to two trucks.

It could be the angle of the photo, but please go back and check the tooth depth.

If I am correct, then my hunch earlier was probably also correct.

If the teeth in the main double drive gear which interfaces the worm on the Alco truck is longer, then you are at risk of those teeth bottoming out in the grooves of the B40 worm.

Now, Atlas uses a lot of interchangeable parts.  So it could be the photo.

But check the tooth depth between those two trucks.
Here is a need for a new scientific measuring device!

Ron, you appear to have found a "gear teeth bottoming out on the worm" issue in both of my noisy mechanisms, just by your diagnosis of the sound in the Youtube:  http://www.tubechop.com/watch/8728439  (amazing on your part.)   

———————

1)  I don't think it is a difference in teeth count or the height of the teeth, the photo angle above just made it appear that way.  I went back and compared with calipers, the height on both the original 4-axle truck and the C628 truck,  from the mounting pad to the top of the teeth.  That measured exactly the same. 

( I will add photos later to make this clear).


2)  BUT what I did discover after I mounted power-pack alligator clips to the side of the B40-8 mechanism, and then checked in the
vertical direction by pushing the truck up and down vertically, listening for sound the teeth bottoming out in the worm and the associated binding.

===> I discovered that on this Atlas B40-8 / C628 truck mechanism on the *rear* truck only
===> Using *either* the original 2 axle truck or the C628 truck…
===> On the rear truck end only, *both* trucks were bottoming out on the worm. 

===> I went back and checked the much quieter 1st version of this mechanism with C628 trucks.   
===>  Same result:  the front truck doesn’t bottom out, the rear truck only if I push it in just a bit, does bottom out
(but not enough to cause a noise!) 


The front truck C628 truck on both Atlas B40-8 frames, for some reason, does not bottom out.  Perhaps frame tolerances?

So, clearly, the rear truck's gear teeth (on both the orig 2 axle and the C628 truck) were hitting the worm.


3) To test this theory on the noisy 2nd B40-8 mechanism, I put a .020" shim on the bottom of the rear truck chassis mounting pads, then repeated the above vertical-bottoming out tests as well as running this on the layout:   

YOUTUBE ( I really apologize for the shaky video, I did this fast while on break from family duties) :

 

4)  Result, noticeably quieter!   Appears the gear teeth no longer bottom out on the worm.

I will have to test later, whether a .010" shim is enough, or if the .020" is required.   

It appears you, Ron Bearden, by the sound alone in a YouTube video, identified that gear teeth were bottoming out on the worm. Amazing and thank you.

=======================

Conclusion:

This "vertical test - making sure gear teeth do not bottom out in the worm" appears to be a good test that we all should check for, in our noisy locos.   


=========================

Later I will show an update where the same "gear teeth bottoming out on worm gear"  issue was present on the noisy IM SD45-2 mechanism in this thread.   I'll show the quietness that resulted from solving that, in a YouTube  (later on when I can get to it as I have family visiting).

The cause turned out to be a different problem, within the IM SD60-style trucks. 

Stay tuned, you all are going to love seeing what I found there, because it and the solution may apply to many of our noisy IM mechanisms......  (teaser..... )